From didrocks at ubuntu.com Wed Jun 1 06:23:57 2016 From: didrocks at ubuntu.com (Didier Roche) Date: Wed, 1 Jun 2016 08:23:57 +0200 Subject: Docs for snapcraft In-Reply-To: <574E1D61.1050303@ubuntu.com> References: <574D6A46.2040105@ubuntu.com> <574E1D61.1050303@ubuntu.com> Message-ID: <574E7F7D.3040802@ubuntu.com> Le 01/06/2016 01:25, Sergio Schvezov a écrit : > El 31/05/16 a las 07:41, Mark Shuttleworth escribió: >> Hi folks > Hi there. > > I'm going to break the ice here or try to. Hey, > >> It's great to see the level of interest in making snaps! I've been >> tracking comments and a lot of people seem to be stuck on snapcraft, so >> I think we can put some effort into improving those docs. > We landed a change which will come in 2.10 which on `snapcraft init` > prepares something you can snap right away. I've been on the crossroads > on this one but it figures that someone doing `snapcraft init` is just > getting started with the tool. So now you can > > $ snapcraft init > $ snapcraft > > And get a snap right away. I want to make this more guided as we move > along (gamification). > > It is still a template to the keen eye, so it is still of use to someone > in the know how wanting to ramp up a new project. > >> But I wondered if we shouldn't instead focus up front on explaining the >> snap format itself. That way, people could make snaps manually >> initially, then switch to using the tools they like best for making the >> snap. > What to expect of a snap and how it integrates with the rest of the > system would be really ideal. I would paint it with user stories ("I > write desktop apps ...", "I work on embedded systems over i2c..", etc.) > so we don't go into length about technical details that are not relevant > to some people. Yes everyone uses `interfaces`, but introducing the > concept at the right time makes things easier to grasp because it > matters to the persons use case at that time. I agree with Sergio that we should concentrate on the tools people will spend the most time with at the main focus, and build our getting started story around it. Developers focus on tasks, like "I want to build an YYY app", and having this as clear titles in the documentation, learning notions steps by steps and complexifying the use case (still reusing a large part of familiar concepts, learnt in previous sections) is the key to not have people lost in the journey. The gamification and instant gratification is the compelling case for this: 1. do this 2. run and enjoy 3. explain the new notion we just introduced 4. goto 1 > >> Personally I think snapcraft is amazing, but it does create an extra >> layer of abstraction to push through, which may be confusing to someone >> just starting out. >> >> Thoughts? > My thoughts are biased towards trying to use snapcraft for everything > but we should not block on people wanting to do whatever they want > during their creative process. Of course, explaining the base concept (file system and such) is important, but that can happen once we have 3-4 success of the virtuous loop I explained above and having the base concepts nicely shaped in developer's head. Then, we can introduce a bug for instance as the next step, and see how to debug/inspect it. This is when the snapcraft lifecycle concept, and the snap/ directory can be introduced, exploring this way the snap (and not snapcraft) concepts like meta/snap.yaml, wrapper, and file system… Does this make sense? Didier From mark at ubuntu.com Wed Jun 1 06:30:35 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Wed, 1 Jun 2016 07:30:35 +0100 Subject: Docs for snapcraft In-Reply-To: <574E7F7D.3040802@ubuntu.com> References: <574D6A46.2040105@ubuntu.com> <574E1D61.1050303@ubuntu.com> <574E7F7D.3040802@ubuntu.com> Message-ID: <574E810B.2020505@ubuntu.com> On 01/06/16 07:23, Didier Roche wrote: > >>> Personally I think snapcraft is amazing, but it does create an extra >>> layer of abstraction to push through, which may be confusing to someone >>> just starting out. >>> >>> Thoughts? >> My thoughts are biased towards trying to use snapcraft for everything >> but we should not block on people wanting to do whatever they want >> during their creative process. > Of course, explaining the base concept (file system and such) is > important, but that can happen once we have 3-4 success of the virtuous > loop I explained above and having the base concepts nicely shaped in > developer's head. > Then, we can introduce a bug for instance as the next step, and see how > to debug/inspect it. This is when the snapcraft lifecycle concept, and > the snap/ directory can be introduced, exploring this way the snap (and > not snapcraft) concepts like meta/snap.yaml, wrapper, and file system… Right, we agree on the basics. Here's the challenge - the smartest people don't sit down to write a simple snap. They want to make a snap of the thing they care about, which is probably big and ugly inside because it's been around enough for someone to care about it. So, in that environment, learning snapcraft is a big layer of indirection, and worse, if you hit the limits of snapcraft and have to start writing a plugin, you are spending time and effort on something you don't care about in order to get to something you DO care about. That same person could probably MANUALLY construct a snap, as long as they know what the constraints are. They can manually build their code, they can build and copy, they can jiggle things to work. I love snapcraft and believe we will make it perfect. But right now, I see a lot of people hitting its limits and being baffled as to what it is doing and why. Smart people saying "I give up because I can't even get a bash script to work in a snap". That's a problem we must face head on, not deny. Mark From didrocks at ubuntu.com Wed Jun 1 06:34:33 2016 From: didrocks at ubuntu.com (Didier Roche) Date: Wed, 1 Jun 2016 08:34:33 +0200 Subject: Docs for snapcraft In-Reply-To: <574E810B.2020505@ubuntu.com> References: <574D6A46.2040105@ubuntu.com> <574E1D61.1050303@ubuntu.com> <574E7F7D.3040802@ubuntu.com> <574E810B.2020505@ubuntu.com> Message-ID: <574E81F9.6020104@ubuntu.com> Le 01/06/2016 08:30, Mark Shuttleworth a écrit : > On 01/06/16 07:23, Didier Roche wrote: >> >>>> Personally I think snapcraft is amazing, but it does create an extra >>>> layer of abstraction to push through, which may be confusing to someone >>>> just starting out. >>>> >>>> Thoughts? >>> My thoughts are biased towards trying to use snapcraft for everything >>> but we should not block on people wanting to do whatever they want >>> during their creative process. >> Of course, explaining the base concept (file system and such) is >> important, but that can happen once we have 3-4 success of the virtuous >> loop I explained above and having the base concepts nicely shaped in >> developer's head. >> Then, we can introduce a bug for instance as the next step, and see how >> to debug/inspect it. This is when the snapcraft lifecycle concept, and >> the snap/ directory can be introduced, exploring this way the snap (and >> not snapcraft) concepts like meta/snap.yaml, wrapper, and file system… > > Right, we agree on the basics. Here's the challenge - the smartest > people don't sit down to write a simple snap. They want to make a snap > of the thing they care about, which is probably big and ugly inside > because it's been around enough for someone to care about it. > > So, in that environment, learning snapcraft is a big layer of > indirection, and worse, if you hit the limits of snapcraft and have to > start writing a plugin, you are spending time and effort on something > you don't care about in order to get to something you DO care about. > > That same person could probably MANUALLY construct a snap, as long as > they know what the constraints are. They can manually build their code, > they can build and copy, they can jiggle things to work. > > I love snapcraft and believe we will make it perfect. But right now, I > see a lot of people hitting its limits and being baffled as to what it > is doing and why. Smart people saying "I give up because I can't even > get a bash script to work in a snap". That's a problem we must face head > on, not deny. Fair enough, and I understand your feeling there. I'm letting the floor opened for others to comment :) Cheers, Didier From pierre.equoy at canonical.com Wed Jun 1 07:52:44 2016 From: pierre.equoy at canonical.com (Pierre Equoy) Date: Wed, 1 Jun 2016 15:52:44 +0800 Subject: =?UTF-8?Q?=E2=80=9CDirectory_=27=2E=27_is_not_installable=2E_File_=27setup=2Epy=27?= =?UTF-8?Q?_not_found=2E=E2=80=9D_when_trying_to_build_a_snap?= Message-ID: Hi there, I've been discussing with the author of Rapid Photo Downloader [1] to get a snap of it. I believe it is a good candidate because it's a Python3 app that uses some libraries that are not packaged in Debian (python-gphoto2), thus making it pretty hard to package in the traditional way (PPA...). I currently uses the following snapcraft.yaml [2] but I'm running into this issue when trying to build it: python3 /tmp/rapid/parts/python-gphoto2/install/usr/bin/pip3 install --target /tmp/rapid/parts/python-gphoto2/install/usr/lib/python3.5/site-packages . Directory '.' is not installable. File 'setup.py' not found. (full log here [3]) I'm using snapcraft 2.9 on Xenial. Any idea where to look at? Thanks! [1] http://www.damonlynch.net/rapid/index.html [2] http://pastebin.ubuntu.com/16885221/ [3] http://pastebin.ubuntu.com/16885228/ -- Pierre Equoy QA & Certification Engineer | Canonical www.canonical.com | www.ubuntu.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From didrocks at ubuntu.com Wed Jun 1 08:48:53 2016 From: didrocks at ubuntu.com (Didier Roche) Date: Wed, 1 Jun 2016 10:48:53 +0200 Subject: =?UTF-8?Q?Re:_=e2=80=9cDirectory_'.'_is_not_installable._File_'setu?= =?UTF-8?Q?p.py'_not_found.=e2=80=9d_when_trying_to_build_a_snap?= In-Reply-To: References: Message-ID: <574EA175.2010601@ubuntu.com> Le 01/06/2016 09:52, Pierre Equoy a écrit : > Hi there, Hey Pierre, > > I've been discussing with the author of Rapid Photo Downloader [1] to > get a snap of it. I believe it is a good candidate because it's a > Python3 app that uses some libraries that are not packaged in Debian > (python-gphoto2), thus making it pretty hard to package in the > traditional way (PPA...). > > I currently uses the following snapcraft.yaml [2] but I'm running into > this issue when trying to build it: > > python3 /tmp/rapid/parts/python-gphoto2/install/usr/bin/pip3 install --target /tmp/rapid/parts/python-gphoto2/install/usr/lib/python3.5/site-packages . > Directory '.' is not installable. File 'setup.py' not found. > > (full log here [3]) > > I'm using snapcraft 2.9 on Xenial. > Any idea where to look at? It seems to me that you have two parts, and none of them have a source: (snapcraft help sources) to locate the code. The plugin: python3 expects that the code is reachable locally or via source: and will execute setup.py from it. Here, it seems that you are staging the debian packages from the ubuntu archive only, so what you should do it using the nil plugin instead (which will do nothing but staging the code from the packages and their dependencies). Then, you can add the addtional part for Rapid Photo Downloader code itself. Does this make sense? Cheers, Didier From vinzjobard at gmail.com Wed Jun 1 09:28:26 2016 From: vinzjobard at gmail.com (Vincent JOBARD) Date: Wed, 01 Jun 2016 09:28:26 +0000 Subject: Docs for snapcraft In-Reply-To: <574E81F9.6020104@ubuntu.com> References: <574D6A46.2040105@ubuntu.com> <574E1D61.1050303@ubuntu.com> <574E7F7D.3040802@ubuntu.com> <574E810B.2020505@ubuntu.com> <574E81F9.6020104@ubuntu.com> Message-ID: Hi there, I'm not a specialist but as evangelist this is a resume of my exchange with many devs, on IT forums, social networks and so on, since snaps were announced to personnal desktop. Devs don't want to manage with another package format where they have to manage all the dependances, seccomp filters and App Armor profile. Devs want to devs, they don't want to package, except maybe those who targets Snappy Ubuntu Core as a base plateform for innovation. But as I understand, in this particular case, lots of them belong to Canonical partners, and have particular relationship with Canonical devs, that could help. Other devs will only package their apps in snaps if it's very easy to do and if it doesn't take so much time, and maybe be integrated into their integration processus (I push my code on my git repo and the snap will be automatically generated, is generally what devs wanted to embrace snap technology) So I think, as evangelist, that snapcraft have to be a primary target to promote snaps. Else, non pro-Ubuntu devs will never use it Cheers Winael On Wed, Jun 1, 2016 at 8:35 AM Didier Roche wrote: > Le 01/06/2016 08:30, Mark Shuttleworth a écrit : > > On 01/06/16 07:23, Didier Roche wrote: > >> > >>>> Personally I think snapcraft is amazing, but it does create an extra > >>>> layer of abstraction to push through, which may be confusing to > someone > >>>> just starting out. > >>>> > >>>> Thoughts? > >>> My thoughts are biased towards trying to use snapcraft for everything > >>> but we should not block on people wanting to do whatever they want > >>> during their creative process. > >> Of course, explaining the base concept (file system and such) is > >> important, but that can happen once we have 3-4 success of the virtuous > >> loop I explained above and having the base concepts nicely shaped in > >> developer's head. > >> Then, we can introduce a bug for instance as the next step, and see how > >> to debug/inspect it. This is when the snapcraft lifecycle concept, and > >> the snap/ directory can be introduced, exploring this way the snap (and > >> not snapcraft) concepts like meta/snap.yaml, wrapper, and file system… > > > > Right, we agree on the basics. Here's the challenge - the smartest > > people don't sit down to write a simple snap. They want to make a snap > > of the thing they care about, which is probably big and ugly inside > > because it's been around enough for someone to care about it. > > > > So, in that environment, learning snapcraft is a big layer of > > indirection, and worse, if you hit the limits of snapcraft and have to > > start writing a plugin, you are spending time and effort on something > > you don't care about in order to get to something you DO care about. > > > > That same person could probably MANUALLY construct a snap, as long as > > they know what the constraints are. They can manually build their code, > > they can build and copy, they can jiggle things to work. > > > > I love snapcraft and believe we will make it perfect. But right now, I > > see a lot of people hitting its limits and being baffled as to what it > > is doing and why. Smart people saying "I give up because I can't even > > get a bash script to work in a snap". That's a problem we must face head > > on, not deny. > > Fair enough, and I understand your feeling there. > > I'm letting the floor opened for others to comment :) > Cheers, > Didier > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > -------------- next part -------------- An HTML attachment was scrubbed... URL: From reinhard.pointner at gmail.com Wed Jun 1 10:01:34 2016 From: reinhard.pointner at gmail.com (Reinhard Pointner) Date: Wed, 1 Jun 2016 18:01:34 +0800 Subject: Docs for snapcraft In-Reply-To: References: <574D6A46.2040105@ubuntu.com> <574E1D61.1050303@ubuntu.com> <574E7F7D.3040802@ubuntu.com> <574E810B.2020505@ubuntu.com> <574E81F9.6020104@ubuntu.com> Message-ID: Hi, If you want to target the Ubuntu Software Center / App Store then Snappy is a given. If you like it or not. (Personally, I like snappy, though it's still a bit rough on the edges). Snappy (thanks to snapcraft) is not difficult and quite easy to get a hang of even in the current state. Wix/MSI is a nightmare in comparison even if you copy and paste everything from stackoverflow. The main issue is the lack of "real" examples. Hello World examples don't cut it for real desktop apps (that crash with all kinds of weird UI framework issues when run in confined mode). More than documentation, more "real world" examples that show what is currently possible and what isn't, would be far more helpful. That being said, snappy for the desktop is under extremely active development now, and there should be weekly/bi-weekly updates for snapd now. There's a lot of major show stoppers for snappy desktop apps, that either make things hard or impossible, but now that people start using it, these things will get figured out over the next few months. Cheers, Reinhard On 1 June 2016 at 17:28, Vincent JOBARD wrote: > Hi there, > > I'm not a specialist but as evangelist this is a resume of my exchange > with many devs, on IT forums, social networks and so on, since snaps were > announced to personnal desktop. > > Devs don't want to manage with another package format where they have to > manage all the dependances, seccomp filters and App Armor profile. > > Devs want to devs, they don't want to package, except maybe those who > targets Snappy Ubuntu Core as a base plateform for innovation. But as I > understand, in this particular case, lots of them belong to Canonical > partners, and have particular relationship with Canonical devs, that could > help. > > Other devs will only package their apps in snaps if it's very easy to do > and if it doesn't take so much time, and maybe be integrated into their > integration processus (I push my code on my git repo and the snap will be > automatically generated, is generally what devs wanted to embrace snap > technology) > > So I think, as evangelist, that snapcraft have to be a primary target to > promote snaps. Else, non pro-Ubuntu devs will never use it > > Cheers > Winael > > On Wed, Jun 1, 2016 at 8:35 AM Didier Roche wrote: > >> Le 01/06/2016 08:30, Mark Shuttleworth a écrit : >> > On 01/06/16 07:23, Didier Roche wrote: >> >> >> >>>> Personally I think snapcraft is amazing, but it does create an extra >> >>>> layer of abstraction to push through, which may be confusing to >> someone >> >>>> just starting out. >> >>>> >> >>>> Thoughts? >> >>> My thoughts are biased towards trying to use snapcraft for everything >> >>> but we should not block on people wanting to do whatever they want >> >>> during their creative process. >> >> Of course, explaining the base concept (file system and such) is >> >> important, but that can happen once we have 3-4 success of the virtuous >> >> loop I explained above and having the base concepts nicely shaped in >> >> developer's head. >> >> Then, we can introduce a bug for instance as the next step, and see how >> >> to debug/inspect it. This is when the snapcraft lifecycle concept, and >> >> the snap/ directory can be introduced, exploring this way the snap (and >> >> not snapcraft) concepts like meta/snap.yaml, wrapper, and file system… >> > >> > Right, we agree on the basics. Here's the challenge - the smartest >> > people don't sit down to write a simple snap. They want to make a snap >> > of the thing they care about, which is probably big and ugly inside >> > because it's been around enough for someone to care about it. >> > >> > So, in that environment, learning snapcraft is a big layer of >> > indirection, and worse, if you hit the limits of snapcraft and have to >> > start writing a plugin, you are spending time and effort on something >> > you don't care about in order to get to something you DO care about. >> > >> > That same person could probably MANUALLY construct a snap, as long as >> > they know what the constraints are. They can manually build their code, >> > they can build and copy, they can jiggle things to work. >> > >> > I love snapcraft and believe we will make it perfect. But right now, I >> > see a lot of people hitting its limits and being baffled as to what it >> > is doing and why. Smart people saying "I give up because I can't even >> > get a bash script to work in a snap". That's a problem we must face head >> > on, not deny. >> >> Fair enough, and I understand your feeling there. >> >> I'm letting the floor opened for others to comment :) >> Cheers, >> Didier >> >> -- >> Snapcraft mailing list >> Snapcraft at lists.ubuntu.com >> Modify settings or unsubscribe at: >> https://lists.ubuntu.com/mailman/listinfo/snapcraft >> > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From alan.pope at canonical.com Wed Jun 1 10:14:33 2016 From: alan.pope at canonical.com (Alan Pope) Date: Wed, 1 Jun 2016 11:14:33 +0100 Subject: Docs for snapcraft In-Reply-To: References: <574D6A46.2040105@ubuntu.com> <574E1D61.1050303@ubuntu.com> <574E7F7D.3040802@ubuntu.com> <574E810B.2020505@ubuntu.com> <574E81F9.6020104@ubuntu.com> Message-ID: Hi, On 1 June 2016 at 11:01, Reinhard Pointner wrote: > The main issue is the lack of "real" examples. Hello World examples don't > cut it for real desktop apps (that crash with all kinds of weird UI > framework issues when run in confined mode). More than documentation, more > "real world" examples that show what is currently possible and what isn't, > would be far more helpful. > I agree! So we (Community Team) made "snappy playpen[2]" [1] as a side project to allow us to learn the ropes, push the boundaries of snapcraft/snapd and provide some 'real' examples. The goal isn't for *us* to upload all these snaps to the store, and we're not planning on snappifying/snappying/snapcrafting (whatever the verb is) everything, but it's more as a learning exercise. We can of course provide these as patches to upstream projects so *they* can upload to the store. There's a couple of simple ones like ffmpeg and moon-buggy, and some more complex ones like Atom but as time goes on, and as we fix them, we're adding more and more which should cover a bunch of use cases. Thanks to didrocks we have CI testing the creation of each snap per pull request. Working example contributions welcome. [1] https://github.com/ubuntu/snappy-playpen [2] https://en.wikipedia.org/wiki/Playpen Cheers, -- Alan Pope Community Manager Canonical - Ubuntu Engineering and Services +44 (0) 7973 620 164 alan.pope at canonical.com http://ubuntu.com/ From alan.pope at canonical.com Wed Jun 1 10:18:31 2016 From: alan.pope at canonical.com (Alan Pope) Date: Wed, 1 Jun 2016 11:18:31 +0100 Subject: Docs for snapcraft In-Reply-To: <574E81F9.6020104@ubuntu.com> References: <574D6A46.2040105@ubuntu.com> <574E1D61.1050303@ubuntu.com> <574E7F7D.3040802@ubuntu.com> <574E810B.2020505@ubuntu.com> <574E81F9.6020104@ubuntu.com> Message-ID: Hi, On 1 June 2016 at 07:34, Didier Roche wrote: > Fair enough, and I understand your feeling there. > > I'm letting the floor opened for others to comment :) My 2p. I've been 'playing' with snapcraft recently, packaging up other people's projects. Not for upload to the store, but to learn and exercise the boundaries of the tools. One task I find myself doing repeatedly is creating the initial snapcraft.yaml by copy and pasting data from elsewhere. While this is only a one time task (per snap), it could be streamlined. The data already exists for most projects inside a debian/control file or an RPM spec (hand-wavy because i don't know RPM) file. Rather than 'snapcraft init' only adding the basic boilerplate, I'd love to see it extended to yank data from existing places. So for example:- # I already have a deb (long standing project) and want to make a snap # This would pre-fill as much as it can for the first basic metadata stanza snapcraft init flightgear_3.4.0-3.dsc # I know my flightgear app has some dependencies # This would add one part per add, and maybe set the original flightgear part above to be after: [simgear,fgdata] snapcraft part add simgear_3.4.0-3.dsc snapcraft part add http://example.com/fgdata_3.4.0-3.tar.gz I'd picture the above to interrogate the debian/control file or whatever and pluck out all the build-packages/stage-packages to fill that out for me. While this would initially just help the transition from deb to snap, the developer could easily add in the necessary source-type, source and plugin to build from upstream git to crank out nightly/dev builds. That's only part of the story of course. The big blocker I find with most of my snap adventures is the opengl plugin and other desktop integration. Command line tools like ffmpeg, and qt apps like qtox were pretty straightforward (with convoluted/verbose launcher scripts). But anything graphically significant like mame, flightgear, kodi & love2d seem to be blocked on our opengl plugin not working quite right, or themes/fonts or other desktop integration things not being quite right. Some help in that regard would be welcome. That said, I'm really quite enjoying playing with snappy and snapcraft. It's tremendous fun. Cheers, -- Alan Pope Community Manager Canonical - Ubuntu Engineering and Services +44 (0) 7973 620 164 alan.pope at canonical.com http://ubuntu.com/ From mark at ubuntu.com Wed Jun 1 10:24:47 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Wed, 1 Jun 2016 12:24:47 +0200 Subject: Docs for snapcraft In-Reply-To: References: <574D6A46.2040105@ubuntu.com> <574E1D61.1050303@ubuntu.com> <574E7F7D.3040802@ubuntu.com> <574E810B.2020505@ubuntu.com> <574E81F9.6020104@ubuntu.com> Message-ID: <574EB7EF.2000503@ubuntu.com> OK, it sounds to me like most folks are in favor of snapcraft as the recommended up-front way to snap glory :) Thank you! We'll broaden out the examples (and make them easy to Google for) and straighten out the desktop integration issues (already in progress :)). Mark From dave.chiluk at canonical.com Wed Jun 1 20:44:24 2016 From: dave.chiluk at canonical.com (Dave Chiluk) Date: Wed, 1 Jun 2016 15:44:24 -0500 Subject: Docs for snapcraft In-Reply-To: <574D6A46.2040105@ubuntu.com> References: <574D6A46.2040105@ubuntu.com> Message-ID: <574F4928.8080905@canonical.com> On 05/31/2016 05:41 AM, Mark Shuttleworth wrote: > Personally I think snapcraft is amazing, but it does create an extra > layer of abstraction to push through, which may be confusing to someone > just starting out. Having out-of-date documentation/guides out there also creates an extra level of confusion. These two come to mind. https://developer.ubuntu.com/en/snappy/guides/system-updates/ https://developer.ubuntu.com/en/snappy/guides/porting/ As far as I've gathered from conversations with people, both of these no longer apply, but both of them provide the exact kind of information I'd like to know. Updating these guides for 16.04 would be awesome? Another guide request would be for how the system consumes snaps. - What daemons/scripts run to mount the snaps. - How are the accounted for (a/b for individual snaps)? I shouldn't have to roll a base image using ubuntu-device-flash just to tear it apart to understand how it's intended to work. Thanks, Dave. From ogra at ubuntu.com Wed Jun 1 21:57:10 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Wed, 01 Jun 2016 23:57:10 +0200 Subject: Docs for snapcraft In-Reply-To: <574F4928.8080905@canonical.com> References: <574D6A46.2040105@ubuntu.com> Message-ID: <32383d77-309b-485d-96f4-cac03d02cb44@ubuntu.com> hi, Am Mittwoch, 1. Juni 2016 22:44:24 CEST schrieb Dave Chiluk : > > https://developer.ubuntu.com/en/snappy/guides/porting/ > > As far as I've gathered from conversations with people, both of these no > longer apply, but both of them provide the exact kind of information I'd > like to know. Updating these guides for 16.04 would be awesome? > Updating the porting guide is definitely on the TODO list, once we produce actual images for the 16 release ciao oli -- Mit Dekko von meinem Ubuntu-Gerät gesendet From loic.minier at ubuntu.com Thu Jun 2 14:22:03 2016 From: loic.minier at ubuntu.com (=?UTF-8?B?TG/Dr2MgTWluaWVy?=) Date: Thu, 2 Jun 2016 16:22:03 +0200 Subject: Fwd: FreeCAD snap on ubuntu Message-ID: Hi, I wanted to share this testimony from Jean-Marie Verdun (vejmarie) on how he successfully created a snap for the FreeCAD desktop app :-) Cheers, - Loïc ---------- Forwarded message ---------- From: Jean-Marie Verdun Cc: ocp-telcos at lists.opencompute.org, " opencompute-server at lists.opencompute.org" < opencompute-server at lists.opencompute.org>, "ocp-europe at lists.opencompute.org" Hi Loic, Thanks for the test. I will have a look to the python issue that you face. I soon have identified from where it comes. It is probably due to the way I am building the package which is not the right one at the present time. Currently the process is the following: - I am starting a fresh new vagrant xenial VM - I am installing on it the build environment - The I start to build FreeCAD dependency with a build target within /opt/local/FreeCAD-0.17 which is quite unusual - I create deb packages based on incremental changes within this directory ( there is 3 majors libraries there, the libMED, VTK, OCCT ) - Then I compile FreeCAD and install it into /opt/local/FreeCAD-0.17 - I create the .deb - I generate the snapcraft.yaml based on the "binary" files created within the target and use the copy tool from snapcraft to stage the snap - Then I kick the snapcraft command to build the snap The usual process will be normally to add the dependencies within the snap and let snapcraft build them, but the options used and the complexity of them make the whole process quite complex. I discussed all of this with the developpers on IRC, and we are trying to improve the process. Regarding the issues I faced here they are: - locales - GDK and GTK errors - DRI error for 3D engines - Python error - Temporary files location error (the devmode stuff) Hopefully all of them have been addressed using the proper application launcher wrapper through environment variables. I have been helped for the locale (thanks to orga) Here is my launcher I had to add FREECAD_USER_DATA to FreeCAD source tree to relocate the .FreeCAD directory. This is the only modification which was needed upstream. So the way python is installed currently is insane as it is done by copying files by hand (while I should probably add the dependancy within the parts area into snapcraft.yaml) The other issue I am facing now is to insert all of this stuff within FreeCAD CI which is using travis. travis is still providing trusty as build environment, this is getting old stuff, and I had to switch to vagrant, but the build is currently done in single cpu due to a kernel bugs with Virtualbox, so it is taking a crazy amount of time. Everything will be improved in the near futur, but I am soon quite happy to see this stuff working (in the end we have a full 3D engine there) vejmarie #!/bin/bash export I18NPATH=$SNAP/usr/share/i18n export LOCPATH=$SNAP_USER_DATA LANG=en_US ENC=UTF-8 LOC="$LANG.$ENC" # generate a locale so we get properly working charsets and graphics if [ ! -e $SNAP_USER_DATA/$LOC ]; then $SNAP/usr/bin/localedef --prefix=$SNAP_USER_DATA -f $ENC -i $LANG $SNAP_USER_DATA/$LOC fi export LC_ALL=$LOC export LANG=$LOC export LANGUAGE=${LANG%_*} env >& $SNAP_USER_DATA/env export LIBGL_DRIVERS_PATH=$SNAP/usr/lib/x86_64-linux-gnu/dri export GTK_PATH=$SNAP/usr/lib/x86_64-linux-gnu/gtk-2.0/modules export LD_LIBRARY_PATH=$SNAP/usr/lib/x86_64-linux-gnu/gtk-2.0/modules:$LD_LIBRARY_PATH export GTK_DATA_PREFIX=$SNAP_USER_DATA export GTK_EXE_PREFIX=$SNAP/usr export GDK_PIXBUF_MODULE_FILE=$SNAP/usr/lib/x86_64-linux-gnu/gdk-pixbuf-2.0/2.10.0/loaders.cache export PYTHONHOME="$SNAP/usr" export PYTHONPATH="$SNAP/usr" export XDG_DATA_DIR="$SNAP/usr/share/glib-2.0/schemas" export GSETTINGS_SCHEMA_DIR="$SNAP/usr/share/glib-2.0/schemas" export LANG=en_US.UTF-8 export FREECAD_USER_DATA="$SNAP_USER_DATA" export XDG_CONFIG_HOME="$SNAP_USER_DATA" exec "$SNAP/opt/local/FreeCAD-0.17/bin/FreeCAD" -u $SNAP_USER_DATA/user.cfg -s $SNAP_USER_DATA/system.cfg "$@" Le 02/06/2016 à 00:12, Loïc Minier a écrit : Hi, On Tue, May 31, 2016 at 11:21 PM, Jean-Marie Verdun < jean-marie.verdun at splitted-desktop.com> wrote: > > https://myapps.developer.ubuntu.com/dev/click-apps/share/09951e1397a86ca9243bebcd715373197c702839fe5c5f9b8f1b5db81767306df0eb07698f8a6b3ac0c4/ > Worked like a charm and without devmode! I even removed ~/.FreeCAD and some other files to get to as clean a state as before my first install, and this time things went flawlessly. > You still need to configure the interfaces whatever happens FreeCAD needs > write access to a directory and read access to get input files. > Yup > The shared link is the same than before can you confirm me that it will > download Rev 2 ? > I did get rev 3 actually, which I guess you uploaded in the mean time. :-) > If it works I will propose some adaptation upstream to the FreeCAD team to > properly support snappy. > Awesome :-) If you have the time to write a bit about how you initially built FreeCAD and how you created the snap, I'd love to share these with the snappy community and developers (both to share your tips and to smoothen the parts which were hard to achieve). PS: I did get a couple of tracebacks when using the plot tool – I'm guessing missing bits for the full feature set to be available: =========== Running the Python command 'Plot_Axes' failed: Traceback (most recent call last): File "/snap/freecad/100001/opt/local/FreeCAD-0.17/Mod/Plot/PlotGui.py", line 58, in Activated import plotAxes [...] File "/snap/freecad/100001/usr/lib/python2.7/dist-packages/matplotlib/font_manager.py", line 1026, in __init__ paths = [os.path.join(rcParams['datapath'], 'fonts', 'ttf'), File "/snap/freecad/100001/usr/lib/python2.7/posixpath.py", line 70, in join elif path == '' or path.endswith('/'): 'NoneType' object has no attribute 'endswith'Running the Python command 'Plot_Axes' failed: [...] File "/snap/freecad/100001/usr/lib/python2.7/dist-packages/matplotlib/font_manager.py", line 1026, in __init__ paths = [os.path.join(rcParams['datapath'], 'fonts', 'ttf'), File "/snap/freecad/100001/usr/lib/python2.7/posixpath.py", line 70, in join elif path == '' or path.endswith('/'): 'NoneType' object has no attribute 'endswith' =========== Cheers, - Loïc -------------- next part -------------- An HTML attachment was scrubbed... URL: From david.calle at canonical.com Thu Jun 2 20:35:12 2016 From: david.calle at canonical.com (=?UTF-8?Q?David_Call=c3=a9?=) Date: Thu, 2 Jun 2016 22:35:12 +0200 Subject: Snap! Community weekly update: W22 Message-ID: <57509880.8020509@canonical.com> Here is an update of some of the things the Canonical Community team together with some other people were working on this past week. You will receive it every thursday, including some highlights of what happened in the snap community! ## Documentation - dev.ubuntu.com/desktop / - Hints at devmode usefulness in get-started and examples - Moved all examples to their github.com/ubuntu/snappy-playpen version - dev.ubuntu.com/snappy /* - Update of all /guides/* pages (snapd documentation) - Update of all /build-apps/* pages (snapcraft documentation) - Reorg of /guides landing page - Update of /guides/channels - Removal of /guides/frameworks and /guides/hashesyaml-file - Refresh of /support content - Publication of the Ubuntu Core 16 - Security Whitepaper - Successful testing of /mir-snaps against Snapcraft 2.9 ## Snaps examples - Krita (upstream snap) released to the Ubuntu Store - Good progress on VLC (both master and stable) in strict confinement - Addition of an Atom snap to snappy-playpen (thanks cwayne!) - CI fixes merged into snappy-playpen (thanks didrocks!) - Addition of DOSBox to snappy-playpen (thanks Alan!) - Extended snappy-playpen docs, getting ready for announce. ## Community highlights - Krita upstream have announced their first Krita snap along the 3.0 release , which is now available in Ubuntu Software or via “snap find” CLI - Default Ubuntu MATE applications snapping in progress upstream - A MonoGame community member is working on an app to generate MonoGame snaps - The team behind Nabto has snapped their IoT communication stack Cheers, David -------------- next part -------------- An HTML attachment was scrubbed... URL: From woodrow.shen at canonical.com Fri Jun 3 06:59:26 2016 From: woodrow.shen at canonical.com (Woodrow Shen) Date: Fri, 3 Jun 2016 14:59:26 +0800 Subject: No subject Message-ID: Hi all, I made roseapple-pi gadget snap[1] & kernel snap[2] by referencing some examples[3][4], then I tried to use u-d-f[5] to build my own snappy image. However, I got the weird errors from it as below when u-d-f was running: ... Installing roseapple-pi_1.0_armhf.snap Installing roseapple-pi-kernel_3.10.37_armhf.snap failed to install "roseapple-pi-kernel_3.10.37_armhf.snap" from "edge": roseapple-pi-kernel_3.10.37_armhf.snap failed to install: failed to install kernel can not extract kernel assets: cannot determine bootloader ... I also cant' figure out where is it from udf code branch [6]. Any thoughts about this error ? Thanks, [1] https://github.com/xapp-le/SnappyUbuntuCore/tree/master/builder [2] https://github.com/xapp-le/kernel/tree/Ubuntu-Snappy-Core [3] https://git.launchpad.net/~p-pisati/ubuntu/+source/linux/?h=x-raspi2 [4] https://git.launchpad.net/~canonical-hwe-team/+git/hikey_linux/tree/snapcraft.yaml [5] https://people.canonical.com/~mvo/all-snaps/ [6] https://code.launchpad.net/~snappy-dev/goget-ubuntu-touch/all-snaps -- Woodrow Shen Software Engineer, Canonical ltd. UES | CE | PC & Core, Taipei -------------- next part -------------- An HTML attachment was scrubbed... URL: From woodrow.shen at canonical.com Fri Jun 3 07:01:05 2016 From: woodrow.shen at canonical.com (Woodrow Shen) Date: Fri, 3 Jun 2016 15:01:05 +0800 Subject: ubuntu-device-flash errors for building armhf image Message-ID: Edit subject. On Fri, Jun 3, 2016 at 2:59 PM, Woodrow Shen wrote: > Hi all, > > I made roseapple-pi gadget snap[1] & kernel snap[2] by referencing some > examples[3][4], then I tried to use u-d-f[5] to build my own snappy image. > However, I got the weird errors from it as below when u-d-f was running: > > ... > Installing roseapple-pi_1.0_armhf.snap > Installing roseapple-pi-kernel_3.10.37_armhf.snap > failed to install "roseapple-pi-kernel_3.10.37_armhf.snap" from "edge": > roseapple-pi-kernel_3.10.37_armhf.snap failed to install: failed to install > kernel can not extract kernel assets: cannot determine bootloader > ... > > I also cant' figure out where is it from udf code branch [6]. > > Any thoughts about this error ? > Thanks, > > [1] https://github.com/xapp-le/SnappyUbuntuCore/tree/master/builder > [2] https://github.com/xapp-le/kernel/tree/Ubuntu-Snappy-Core > [3] https://git.launchpad.net/~p-pisati/ubuntu/+source/linux/?h=x-raspi2 > [4] > https://git.launchpad.net/~canonical-hwe-team/+git/hikey_linux/tree/snapcraft.yaml > [5] https://people.canonical.com/~mvo/all-snaps/ > [6] https://code.launchpad.net/~snappy-dev/goget-ubuntu-touch/all-snaps > -- > Woodrow Shen > Software Engineer, Canonical ltd. > UES | CE | PC & Core, Taipei > -- Woodrow Shen Software Engineer, Canonical ltd. UES | CE | PC & Core, Taipei -------------- next part -------------- An HTML attachment was scrubbed... URL: From didrocks at ubuntu.com Fri Jun 3 09:19:45 2016 From: didrocks at ubuntu.com (Didier Roche) Date: Fri, 3 Jun 2016 11:19:45 +0200 Subject: ubuntu-device-flash errors for building armhf image In-Reply-To: References: Message-ID: <57514BB1.1080308@ubuntu.com> Le 03/06/2016 09:01, Woodrow Shen a écrit : > Edit subject. Hey! Thanks for reaching out to us, let's try to fix this all together :) It seems that your kernel snap is uploaded to the edge channel from the store and you can reach it out. I think then the image try to determine the bootloader you included in your image and depending on this, copying (in uboot) or linking it (in grub). The error message just says that it can't determine this bootloader, how did you assemble your snaps and included your bootloaded? Maybe Oliver or Sergio (for the snapcraft part) may help there? Cheers, Didier > > On Fri, Jun 3, 2016 at 2:59 PM, Woodrow Shen > > wrote: > > Hi all, > > I made roseapple-pi gadget snap[1] & kernel snap[2] by referencing > some examples[3][4], then I tried to use u-d-f[5] to build my own > snappy image. However, I got the weird errors from it as below > when u-d-f was running: > > ... > Installing roseapple-pi_1.0_armhf.snap > Installing roseapple-pi-kernel_3.10.37_armhf.snap > failed to install "roseapple-pi-kernel_3.10.37_armhf.snap" from > "edge": roseapple-pi-kernel_3.10.37_armhf.snap failed to install: > failed to install kernel can not extract kernel assets: cannot > determine bootloader > ... > > I also cant' figure out where is it from udf code branch [6]. > > Any thoughts about this error ? > Thanks, > > [1] https://github.com/xapp-le/SnappyUbuntuCore/tree/master/builder > [2] https://github.com/xapp-le/kernel/tree/Ubuntu-Snappy-Core > [3] https://git.launchpad.net/~p-pisati/ubuntu/+source/linux/?h=x-raspi2 > > [4] https://git.launchpad.net/~canonical-hwe-team/+git/hikey_linux/tree/snapcraft.yaml > > [5] https://people.canonical.com/~mvo/all-snaps/ > > [6] https://code.launchpad.net/~snappy-dev/goget-ubuntu-touch/all-snaps > > -- > Woodrow Shen > Software Engineer, Canonical ltd. > UES | CE | PC & Core, Taipei > > > > > -- > Woodrow Shen > Software Engineer, Canonical ltd. > UES | CE | PC & Core, Taipei > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From daniel.holbach at canonical.com Fri Jun 3 09:20:57 2016 From: daniel.holbach at canonical.com (Daniel Holbach) Date: Fri, 3 Jun 2016 11:20:57 +0200 Subject: Announcing the Snappy Playpen Message-ID: <57514BF9.4070204@canonical.com> Hello everybody, Snappy Playpen is now on Github [1] and it's where we want to work together on snapping new software. This will provide excellent examples to new users of snapcraft, we will be able to document best practices, learn from each other and create an incubator for new snaps to be added to the store. Snappy Playpen won't be a collection of production-ready snaps, we are treating it a bit like a combination of research project and documented best practices. If you are curious, just check out our main github page [1] and read the docs there. It's easy and we're quite accessible. Find us on gitter [2], IRC [3] or this mailing list to find out how to get involved. You can get started at any time and contribute whatever you feel makes sense, but we want to host themed "sprint" weeks as well. If you have suggestions (e.g. a IoT-related week, a KDE-related week, server app, etc.), let us know. For those weeks we will make sure we have experts there to help us figure this out together. Next week will be our first Snappy Playpen sprint and it will be a "free for all" week. This will help us to figure out the details and learn about what you all exactly want to do. On Tuesday, 7th June 2016, we will make a big push and make sure our snapd and snapcraft engineers are there to answer questions and help figure out solutions together. Mark the day in your calendar and check out our docs to find out how to get started. WHAT: Snappy playpen sprint WHEN: Tuesday, 7th June 2016 all day WHERE: Join us on gitter [2] or IRC [3] We are looking forward to seeing you there and your PRs come in! Have a great day, Daniel [1] https://github.com/ubuntu/snappy-playpen [2] https://gitter.im/ubuntu/snappy-playpen [3] http://webchat.freenode.net/?channels=snappy From ogra at ubuntu.com Fri Jun 3 09:38:08 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Fri, 03 Jun 2016 11:38:08 +0200 Subject: ubuntu-device-flash errors for building armhf image In-Reply-To: References: Message-ID: <1464946688.30150.122.camel@anubis> hi, > > On Fri, Jun 3, 2016 at 2:59 PM, Woodrow Shen > wrote: > Hi all, > > > I made roseapple-pi gadget snap[1] there are multiple issues with your gadget snap ... in your snap.yaml you define two files that do not exist: ... - path: boot-assets/bootloader.bin ... - path: boot-assets/u-boot.bin we also do not support any text files for configuring the bootloader any more, you need to create a uboot.env (and adjust your uboot build time configuration to use and find it): #define CONFIG_ENV_SIZE SZ_128K #define CONFIG_ENV_IS_IN_FAT #define FAT_ENV_INTERFACE "mmc" #define FAT_ENV_DEVICE_AND_PART "0:1" #define CONFIG_SYS_REDUNDAND_ENVIRONMENT (in include/configs/$your-board.h, note that FAT_ENV_DEVICE_AND_PART needs to point to your vfat partition that holds the uboot.env file) the uboot.env needs to be created using the mkenvimage command either from your uboot tree (it gets built in the tools directory) or from the u-boot-tools package in the archive: mkenvimage -r -s 131072 -o uboot.env uEnv.txt (note the options are not optional ;) -r and -s 131072 are mandatory) and change "- path: boot-assets/uEnv.txt" to "- path: boot-assets/uboot.env" in your snap.yaml first see if it works better with the adjusted gadget snap and the very latest ubuntu-device-flash from http://people.canonical.com/~mvo/all-snaps/ if there are still issues then we can go through your kernel snap ... ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From woodrow.shen at canonical.com Fri Jun 3 10:12:15 2016 From: woodrow.shen at canonical.com (Woodrow Shen) Date: Fri, 3 Jun 2016 18:12:15 +0800 Subject: ubuntu-device-flash errors for building armhf image In-Reply-To: <1464946688.30150.122.camel@anubis> References: <1464946688.30150.122.camel@anubis> Message-ID: Hi Ogra, Thanks your help a lot !! The uboot.env is a file I miss, and now I can build an image successfully. ^0^/ I also wonder why don't we provide some gold examples / guides to instruct how to build the gadget snap for the specific arch, especially the gadget couldn't created by snapcraft. Cheers, Woodrow, On Fri, Jun 3, 2016 at 5:38 PM, Oliver Grawert wrote: > hi, > > > > > On Fri, Jun 3, 2016 at 2:59 PM, Woodrow Shen > > wrote: > > Hi all, > > > > > > I made roseapple-pi gadget snap[1] > > there are multiple issues with your gadget snap ... > in your snap.yaml you define two files that do not exist: > ... > - path: boot-assets/bootloader.bin > ... > - path: boot-assets/u-boot.bin > > we also do not support any text files for configuring the bootloader any > more, you need to create a uboot.env (and adjust your uboot build time > configuration to use and find it): > > #define CONFIG_ENV_SIZE SZ_128K > #define CONFIG_ENV_IS_IN_FAT > #define FAT_ENV_INTERFACE "mmc" > #define FAT_ENV_DEVICE_AND_PART "0:1" > #define CONFIG_SYS_REDUNDAND_ENVIRONMENT > > (in include/configs/$your-board.h, note that FAT_ENV_DEVICE_AND_PART > needs to point to your vfat partition that holds the uboot.env file) > > the uboot.env needs to be created using the mkenvimage command either > from your uboot tree (it gets built in the tools directory) or from the > u-boot-tools package in the archive: > > mkenvimage -r -s 131072 -o uboot.env uEnv.txt > > (note the options are not optional ;) -r and -s 131072 are mandatory) > and change "- path: boot-assets/uEnv.txt" to "- path: > boot-assets/uboot.env" in your snap.yaml > > first see if it works better with the adjusted gadget snap and the very > latest ubuntu-device-flash from > http://people.canonical.com/~mvo/all-snaps/ > > if there are still issues then we can go through your kernel snap ... > > ciao > oli > > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > > -- Woodrow Shen Software Engineer, Canonical ltd. UES | CE | PC & Core, Taipei -------------- next part -------------- An HTML attachment was scrubbed... URL: From ogra at ubuntu.com Fri Jun 3 10:21:03 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Fri, 03 Jun 2016 12:21:03 +0200 Subject: ubuntu-device-flash errors for building armhf image In-Reply-To: References: <1464946688.30150.122.camel@anubis> Message-ID: <1464949263.17816.2.camel@anubis> hi, Am Freitag, den 03.06.2016, 18:12 +0800 schrieb Woodrow Shen: > Hi Ogra, > > > Thanks your help a lot !! > The uboot.env is a file I miss, and now I can build an image > successfully. ^0^/ > > > I also wonder why don't we provide some gold examples / guides to > instruct how to build the gadget snap for the specific arch, i'm actually working on this, but until the new image creation tool is ready (we will drop ubuntu-device-flash) the gadget and kernel specifications can still change. once everything is finalized we will get a proper step by step porting guide ... ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From zygmunt.krynicki at canonical.com Fri Jun 3 10:30:07 2016 From: zygmunt.krynicki at canonical.com (Zygmunt Krynicki) Date: Fri, 03 Jun 2016 12:30:07 +0200 Subject: ubuntu-device-flash errors for building armhf image In-Reply-To: References: <1464946688.30150.122.camel@anubis> Message-ID: On Friday, 3 June 2016 12:12:15 CEST, Woodrow Shen wrote: > Hi Ogra, > > Thanks your help a lot !! > The uboot.env is a file I miss, and now I can build an image successfully. > ^0^/ > > I also wonder why don't we provide some gold examples / guides to instruct > how to build the gadget snap for the specific arch, especially the gadget > couldn't created by snapcraft. I am sure we will do exactly that it is just that right now gadget snap are not finalized and the tooling around it is not fully implemented. I plan to provide sources for building gadget snaps for all supported platforms. From there you can fork and tweak as little or as much as you want. Best regards ZK > > Cheers, > Woodrow, > > On Fri, Jun 3, 2016 at 5:38 PM, Oliver Grawert wrote: > >> hi, >> >> > >> > On Fri, Jun 3, 2016 at 2:59 PM, Woodrow Shen >> > wrote: >> > Hi all, >> > >> > >> > I made roseapple-pi gadget snap[1] >> >> there are multiple issues with your gadget snap ... >> in your snap.yaml you define two files that do not exist: >> ... >> - path: boot-assets/bootloader.bin >> ... >> - path: boot-assets/u-boot.bin >> >> we also do not support any text files for configuring the bootloader any >> more, you need to create a uboot.env (and adjust your uboot build time >> configuration to use and find it): >> >> #define CONFIG_ENV_SIZE SZ_128K >> #define CONFIG_ENV_IS_IN_FAT >> #define FAT_ENV_INTERFACE "mmc" >> #define FAT_ENV_DEVICE_AND_PART "0:1" >> #define CONFIG_SYS_REDUNDAND_ENVIRONMENT >> >> (in include/configs/$your-board.h, note that FAT_ENV_DEVICE_AND_PART >> needs to point to your vfat partition that holds the uboot.env file) >> >> the uboot.env needs to be created using the mkenvimage command either >> from your uboot tree (it gets built in the tools directory) or from the >> u-boot-tools package in the archive: >> >> mkenvimage -r -s 131072 -o uboot.env uEnv.txt >> >> (note the options are not optional ;) -r and -s 131072 are mandatory) >> and change "- path: boot-assets/uEnv.txt" to "- path: >> boot-assets/uboot.env" in your snap.yaml >> >> first see if it works better with the adjusted gadget snap and the very >> latest ubuntu-device-flash from >> http://people.canonical.com/~mvo/all-snaps/ >> >> if there are still issues then we can go through your kernel snap ... >> >> ciao >> oli >> >> >> -- >> Snapcraft mailing list >> Snapcraft at lists.ubuntu.com >> Modify settings or unsubscribe at: >> https://lists.ubuntu.com/mailman/listinfo/snapcraft >> >> > > -- Wysłano programem Dekko z urządzenia Ubuntu From mabnhdev at gmail.com Fri Jun 3 13:21:10 2016 From: mabnhdev at gmail.com (MikeB) Date: Fri, 3 Jun 2016 09:21:10 -0400 Subject: Relocating snapcraft output directories Message-ID: I work in a development environment where sources are kept on a filesystem that is automatically backed up fairly often and we place our build output/generated objects on fast build servers that are not backed up. Having build output on the backed-up filesystem is very much discouraged. With snapcraft, I find myself keeping my snap directory on the build server and manually backing up my work-in-progress whenever I remember to do it. Also, for a similar reason, I feel that locating custom plugins in parts/plugins is unfortunate. Other than parts/plugins, the parts directory only contains generated files. It would be nice if the plugins directory could be located in a source area rather than a build output area. Would it be possible to enhance snapcraft to allow all generated objects - parts, stage, snap, and the generated snap to be located on a different filesystem? Regards, Mike -------------- next part -------------- An HTML attachment was scrubbed... URL: From loic.minier at ubuntu.com Fri Jun 3 14:21:11 2016 From: loic.minier at ubuntu.com (=?UTF-8?B?TG/Dr2MgTWluaWVy?=) Date: Fri, 3 Jun 2016 16:21:11 +0200 Subject: Relocating snapcraft output directories In-Reply-To: References: Message-ID: On Fri, Jun 3, 2016 at 3:21 PM, MikeB wrote: > Also, for a similar reason, I feel that locating custom plugins in > parts/plugins is unfortunate. Other than parts/plugins, the parts > directory only contains generated files. It would be nice if the plugins > directory could be located in a source area rather than a build output area. > I have to second this; I often rm-rf-ed parts/ after forgetting I had a plugin there :-) > Would it be possible to enhance snapcraft to allow all generated objects - > parts, stage, snap, and the generated snap to be located on a different > filesystem? > Can't speak on whether we want this, but in any case there is a range of strategies you can put in place: - bind-mounts or symlinks (haven't tested symlinks, but pretty sure bind-mounts would work) - or rsync your bits back and forth before/after builds - this also works well if you keep your snapcraft bits in a git repo and just commit/pull/push regularly Cheers, - Loïc -------------- next part -------------- An HTML attachment was scrubbed... URL: From didrocks at ubuntu.com Fri Jun 3 14:56:19 2016 From: didrocks at ubuntu.com (Didier Roche) Date: Fri, 3 Jun 2016 16:56:19 +0200 Subject: Relocating snapcraft output directories In-Reply-To: References: Message-ID: <57519A93.9070109@ubuntu.com> Le 03/06/2016 16:21, Loïc Minier a écrit : > On Fri, Jun 3, 2016 at 3:21 PM, MikeB > wrote: > > Also, for a similar reason, I feel that locating custom plugins in > parts/plugins is unfortunate. Other than parts/plugins, the parts > directory only contains generated files. It would be nice if the > plugins directory could be located in a source area rather than a > build output area. > > > I have to second this; I often rm-rf-ed parts/ after forgetting I had > a plugin there :-) Can I triple this? (not related to any beer brand)! It often see some developers rm -rf parts/ and then thought "oh, I forgot that I had a plugin" (no, I won't drop names ;)). So yeah, having a separate directory (maybe not "plugin" because all of project may have a plugin/ directory right away would be nice! Didier -------------- next part -------------- An HTML attachment was scrubbed... URL: From kyle.fazzari at canonical.com Fri Jun 3 15:05:35 2016 From: kyle.fazzari at canonical.com (Kyle Fazzari) Date: Fri, 3 Jun 2016 11:05:35 -0400 Subject: Relocating snapcraft output directories In-Reply-To: References: Message-ID: <57519CBF.8080402@canonical.com> On 06/03/2016 09:21 AM, MikeB wrote: > I work in a development environment where sources are kept on a > filesystem that is automatically backed up fairly often and we place our > build output/generated objects on fast build servers that are not backed > up. Having build output on the backed-up filesystem is very much > discouraged. > > With snapcraft, I find myself keeping my snap directory on the build > server and manually backing up my work-in-progress whenever I remember > to do it. > > Also, for a similar reason, I feel that locating custom plugins in > parts/plugins is unfortunate. Other than parts/plugins, the parts > directory only contains generated files. It would be nice if the > plugins directory could be located in a source area rather than a build > output area. I agree with this, it's not the most user-friendly to place the plugins. It also makes the cleaning logic more fragile. I suspect it was put here originally because "parts" was already a directory used by snapcraft; we'll have to give some thought as to where such things should go. > Would it be possible to enhance snapcraft to allow all generated objects > - parts, stage, snap, and the generated snap to be located on a > different filesystem? This I'm not so sure about. There are ways you can accomplish this without specific snapcraft support (I see Loïc mentioned a few). -- Kyle Fazzari (kyrofa) Software Engineer Canonical Ltd. kyle at canonical.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 836 bytes Desc: OpenPGP digital signature URL: From jacob.ryan.humphrey at gmail.com Fri Jun 3 16:37:22 2016 From: jacob.ryan.humphrey at gmail.com (Jacob Humphrey) Date: Fri, 3 Jun 2016 11:37:22 -0500 Subject: Relocating snapcraft output directories In-Reply-To: <57519CBF.8080402@canonical.com> References: <57519CBF.8080402@canonical.com> Message-ID: > I suspect it was put here > originally because "parts" was already a directory used by snapcraft; > we'll have to give some thought as to where such things should go. How about we put it under a snapcraft_plugins directory? That would be very clear as to its purpose and unlikely to get accidentally deleted. It would also be unlikely that a project already has this directory -Jacob Humphrey -------------- next part -------------- An HTML attachment was scrubbed... URL: From uframer at gmail.com Sun Jun 5 16:12:39 2016 From: uframer at gmail.com (Jiao Ye) Date: Mon, 6 Jun 2016 00:12:39 +0800 Subject: Is Ubuntu Core gone? Message-ID: Dear all, I don't know whether this is the right mail list to ask this question. Please kindly correct me if I'm wrong. Today I noticed that the ubuntu core releases in " http://cdimage.ubuntu.com/ubuntu-core/releases/16.04/release/" are all gone. Is this a temporary action or Canonical just stopped releasing them? -- So many things to do, so little time. -------------- next part -------------- An HTML attachment was scrubbed... URL: From gustavo.niemeyer at canonical.com Sun Jun 5 18:07:22 2016 From: gustavo.niemeyer at canonical.com (Gustavo Niemeyer) Date: Sun, 5 Jun 2016 15:07:22 -0300 Subject: Is Ubuntu Core gone? In-Reply-To: References: Message-ID: Hi Jiao, These images were put there by mistake. We're still working on getting the 16.04 images ready for releasing. They're definitely coming, though. On Sun, Jun 5, 2016 at 1:12 PM, Jiao Ye wrote: > Dear all, > > I don't know whether this is the right mail list to ask this question. > Please kindly correct me if I'm wrong. > > Today I noticed that the ubuntu core releases in " > http://cdimage.ubuntu.com/ubuntu-core/releases/16.04/release/" are all > gone. Is this a temporary action or Canonical just stopped releasing them? > > -- > So many things to do, so little time. > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > > -- gustavo @ http://niemeyer.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From kristijan.zic at linux.com Sun Jun 5 18:10:06 2016 From: kristijan.zic at linux.com (=?UTF-8?Q?Kristijan_=C5=BDic?=) Date: Sun, 05 Jun 2016 18:10:06 +0000 Subject: Is Ubuntu Core gone? In-Reply-To: References: Message-ID: Will Ubuntu Core be supported on Raspberry Pi 3? On Sun, 5 Jun 2016 8:08 pm Gustavo Niemeyer, wrote: > Hi Jiao, > > These images were put there by mistake. We're still working on getting the > 16.04 images ready for releasing. > > They're definitely coming, though. > > On Sun, Jun 5, 2016 at 1:12 PM, Jiao Ye wrote: > >> Dear all, >> >> I don't know whether this is the right mail list to ask this question. >> Please kindly correct me if I'm wrong. >> >> Today I noticed that the ubuntu core releases in " >> http://cdimage.ubuntu.com/ubuntu-core/releases/16.04/release/" are all >> gone. Is this a temporary action or Canonical just stopped releasing them? >> >> -- >> So many things to do, so little time. >> >> -- >> Snapcraft mailing list >> Snapcraft at lists.ubuntu.com >> Modify settings or unsubscribe at: >> https://lists.ubuntu.com/mailman/listinfo/snapcraft >> >> > > > -- > gustavo @ http://niemeyer.net > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > -------------- next part -------------- An HTML attachment was scrubbed... URL: From steve.langasek at canonical.com Sun Jun 5 19:20:21 2016 From: steve.langasek at canonical.com (Steve Langasek) Date: Sun, 5 Jun 2016 12:20:21 -0700 Subject: Is Ubuntu Core gone? In-Reply-To: References: Message-ID: <20160605192021.GA10348@virgil.dodds.net> On Mon, Jun 06, 2016 at 12:12:39AM +0800, Jiao Ye wrote: > I don't know whether this is the right mail list to ask this question. > Please kindly correct me if I'm wrong. > Today I noticed that the ubuntu core releases in " > http://cdimage.ubuntu.com/ubuntu-core/releases/16.04/release/" are all > gone. Is this a temporary action or Canonical just stopped releasing them? There has not yet been a release of snappy Ubuntu Core based on Ubuntu 16.04 LTS. The images that you saw there were of a different type, previously referred to as "ubuntu core" but now renamed to "ubuntu base" to eliminate confusion. -- Steve Langasek Give me a lever long enough and a Free OS Debian Developer to set it on, and I can move the world. Ubuntu Developer http://www.debian.org/ slangasek at ubuntu.com vorlon at debian.org -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: not available URL: From gustavo.niemeyer at canonical.com Sun Jun 5 20:28:55 2016 From: gustavo.niemeyer at canonical.com (Gustavo Niemeyer) Date: Sun, 5 Jun 2016 17:28:55 -0300 Subject: Is Ubuntu Core gone? In-Reply-To: References: Message-ID: Yes, definitely! On Jun 5, 2016 3:10 PM, "Kristijan Žic" wrote: > Will Ubuntu Core be supported on Raspberry Pi 3? > > On Sun, 5 Jun 2016 8:08 pm Gustavo Niemeyer, < > gustavo.niemeyer at canonical.com> wrote: > >> Hi Jiao, >> >> These images were put there by mistake. We're still working on getting >> the 16.04 images ready for releasing. >> >> They're definitely coming, though. >> >> On Sun, Jun 5, 2016 at 1:12 PM, Jiao Ye wrote: >> >>> Dear all, >>> >>> I don't know whether this is the right mail list to ask this question. >>> Please kindly correct me if I'm wrong. >>> >>> Today I noticed that the ubuntu core releases in " >>> http://cdimage.ubuntu.com/ubuntu-core/releases/16.04/release/" are all >>> gone. Is this a temporary action or Canonical just stopped releasing them? >>> >>> -- >>> So many things to do, so little time. >>> >>> -- >>> Snapcraft mailing list >>> Snapcraft at lists.ubuntu.com >>> Modify settings or unsubscribe at: >>> https://lists.ubuntu.com/mailman/listinfo/snapcraft >>> >>> >> >> >> -- >> gustavo @ http://niemeyer.net >> -- >> Snapcraft mailing list >> Snapcraft at lists.ubuntu.com >> Modify settings or unsubscribe at: >> https://lists.ubuntu.com/mailman/listinfo/snapcraft >> > -------------- next part -------------- An HTML attachment was scrubbed... URL: From gustavo.niemeyer at canonical.com Sun Jun 5 20:37:12 2016 From: gustavo.niemeyer at canonical.com (Gustavo Niemeyer) Date: Sun, 5 Jun 2016 17:37:12 -0300 Subject: Is Ubuntu Core gone? In-Reply-To: <20160605192021.GA10348@virgil.dodds.net> References: <20160605192021.GA10348@virgil.dodds.net> Message-ID: At some point I believe we actually had some real Ubuntu Core images up there, but these were not supposed to be published as they are still very immature. On Jun 5, 2016 4:21 PM, "Steve Langasek" wrote: > On Mon, Jun 06, 2016 at 12:12:39AM +0800, Jiao Ye wrote: > > I don't know whether this is the right mail list to ask this question. > > Please kindly correct me if I'm wrong. > > > Today I noticed that the ubuntu core releases in " > > http://cdimage.ubuntu.com/ubuntu-core/releases/16.04/release/" are all > > gone. Is this a temporary action or Canonical just stopped releasing > them? > > There has not yet been a release of snappy Ubuntu Core based on Ubuntu > 16.04 LTS. The images that you saw there were of a different type, > previously referred to as "ubuntu core" but now renamed to "ubuntu base" to > eliminate confusion. > > -- > Steve Langasek Give me a lever long enough and a Free OS > Debian Developer to set it on, and I can move the world. > Ubuntu Developer http://www.debian.org/ > slangasek at ubuntu.com vorlon at debian.org > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From ogra at ubuntu.com Sun Jun 5 22:09:36 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Mon, 06 Jun 2016 00:09:36 +0200 Subject: Is Ubuntu Core gone? In-Reply-To: References: <20160605192021.GA10348@virgil.dodds.net> Message-ID: <1465164576.4270.3.camel@ubuntu.com> hi, Am Sonntag, den 05.06.2016, 17:37 -0300 schrieb Gustavo Niemeyer: > At some point I believe we actually had some real Ubuntu Core images > up there, but these were not supposed to be published as they are > still very immature. the images steve refers to were the ubuntu-core tarballs (a tarball to be used to create a build env or as base for images) that we used to produce under the ubuntu-core name until recently ... they now got renamed to ubuntu-base. the only Snappy Ubuntu Core .img files we ever released on cdimage are 15.04 ones. (for 16 we only produce the snaps publically atm (which will likely change soon with direct store uploads from cdimage). ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From scarlett.gately.clark at gmail.com Wed Jun 1 10:24:02 2016 From: scarlett.gately.clark at gmail.com (Scarlett Clark) Date: Wed, 1 Jun 2016 03:24:02 -0700 Subject: Docs for snapcraft In-Reply-To: References: <574D6A46.2040105@ubuntu.com> <574E1D61.1050303@ubuntu.com> <574E7F7D.3040802@ubuntu.com> <574E810B.2020505@ubuntu.com> <574E81F9.6020104@ubuntu.com> Message-ID: On Jun 1, 2016 03:02, "Reinhard Pointner" wrote: > > Hi, > > If you want to target the Ubuntu Software Center / App Store then Snappy is a given. If you like it or not. (Personally, I like snappy, though it's still a bit rough on the edges). Snappy (thanks to snapcraft) is not difficult and quite easy to get a hang of even in the current state. Wix/MSI is a nightmare in comparison even if you copy and paste everything from stackoverflow. > > The main issue is the lack of "real" examples. Hello World examples don't cut it for real desktop apps (that crash with all kinds of weird UI framework issues when run in confined mode). More than documentation, more "real world" examples that show what is currently possible and what isn't, would be far more helpful. > > That being said, snappy for the desktop is under extremely active development now, and there should be weekly/bi-weekly updates for snapd now. There's a lot of major show stoppers for snappy desktop apps, that either make things hard or impossible, but now that people start using it, these things will get figured out over the next few months. > > Cheers, > Reinhard > > > On 1 June 2016 at 17:28, Vincent JOBARD wrote: >> >> Hi there, >> >> I'm not a specialist but as evangelist this is a resume of my exchange with many devs, on IT forums, social networks and so on, since snaps were announced to personnal desktop. >> >> Devs don't want to manage with another package format where they have to manage all the dependances, seccomp filters and App Armor profile. >> >> Devs want to devs, they don't want to package, except maybe those who targets Snappy Ubuntu Core as a base plateform for innovation. But as I understand, in this particular case, lots of them belong to Canonical partners, and have particular relationship with Canonical devs, that could help. >> >> Other devs will only package their apps in snaps if it's very easy to do and if it doesn't take so much time, and maybe be integrated into their integration processus (I push my code on my git repo and the snap will be automatically generated, is generally what devs wanted to embrace snap technology) >> >> So I think, as evangelist, that snapcraft have to be a primary target to promote snaps. Else, non pro-Ubuntu devs will never use it >> >> Cheers >> Winael >> >> On Wed, Jun 1, 2016 at 8:35 AM Didier Roche wrote: >>> >>> Le 01/06/2016 08:30, Mark Shuttleworth a écrit : >>> > On 01/06/16 07:23, Didier Roche wrote: >>> >> >>> >>>> Personally I think snapcraft is amazing, but it does create an extra >>> >>>> layer of abstraction to push through, which may be confusing to someone >>> >>>> just starting out. >>> >>>> >>> >>>> Thoughts? >>> >>> My thoughts are biased towards trying to use snapcraft for everything >>> >>> but we should not block on people wanting to do whatever they want >>> >>> during their creative process. >>> >> Of course, explaining the base concept (file system and such) is >>> >> important, but that can happen once we have 3-4 success of the virtuous >>> >> loop I explained above and having the base concepts nicely shaped in >>> >> developer's head. >>> >> Then, we can introduce a bug for instance as the next step, and see how >>> >> to debug/inspect it. This is when the snapcraft lifecycle concept, and >>> >> the snap/ directory can be introduced, exploring this way the snap (and >>> >> not snapcraft) concepts like meta/snap.yaml, wrapper, and file system… >>> > >>> > Right, we agree on the basics. Here's the challenge - the smartest >>> > people don't sit down to write a simple snap. They want to make a snap >>> > of the thing they care about, which is probably big and ugly inside >>> > because it's been around enough for someone to care about it. >>> > >>> > So, in that environment, learning snapcraft is a big layer of >>> > indirection, and worse, if you hit the limits of snapcraft and have to >>> > start writing a plugin, you are spending time and effort on something >>> > you don't care about in order to get to something you DO care about. >>> > >>> > That same person could probably MANUALLY construct a snap, as long as >>> > they know what the constraints are. They can manually build their code, >>> > they can build and copy, they can jiggle things to work. >>> > >>> > I love snapcraft and believe we will make it perfect. But right now, I >>> > see a lot of people hitting its limits and being baffled as to what it >>> > is doing and why. Smart people saying "I give up because I can't even >>> > get a bash script to work in a snap". That's a problem we must face head >>> > on, not deny. >>> >>> Fair enough, and I understand your feeling there. >>> >>> I'm letting the floor opened for others to comment :) >>> Cheers, >>> Didier >>> >>> -- >>> Snapcraft mailing list >>> Snapcraft at lists.ubuntu.com >>> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft >> >> >> -- >> Snapcraft mailing list >> Snapcraft at lists.ubuntu.com >> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft >> > > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft > Docs would be great, though I was able to find my way well enough. My issues are as stated above where I am in need of creating my own plugin and it is very difficult to debug. Logging or a debug mode would be great. Also logs to debug odd UI issues I could not seem to locate. Otherwise snapcraft is coming along great. My packages are far from simple :) Cheers Scarlett -------------- next part -------------- An HTML attachment was scrubbed... URL: From chris.wayne at canonical.com Wed Jun 1 22:18:22 2016 From: chris.wayne at canonical.com (Chris Wayne) Date: Wed, 1 Jun 2016 18:18:22 -0400 Subject: Docs for snapcraft In-Reply-To: <32383d77-309b-485d-96f4-cac03d02cb44@ubuntu.com> References: <574D6A46.2040105@ubuntu.com> <574F4928.8080905@canonical.com> <32383d77-309b-485d-96f4-cac03d02cb44@ubuntu.com> Message-ID: Do we have an ETA on when we will produce actual images for series 16? On Wed, Jun 1, 2016 at 5:57 PM, Oliver Grawert wrote: > > hi, > Am Mittwoch, 1. Juni 2016 22:44:24 CEST schrieb Dave Chiluk < > dave.chiluk at canonical.com>: > >> >> https://developer.ubuntu.com/en/snappy/guides/porting/ >> >> As far as I've gathered from conversations with people, both of these no >> longer apply, but both of them provide the exact kind of information I'd >> like to know. Updating these guides for 16.04 would be awesome? >> >> Updating the porting guide is definitely on the TODO list, once we > produce actual images for the 16 release > > ciao > oli > > > > -- > Mit Dekko von meinem Ubuntu-Gerät gesendet > > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > -------------- next part -------------- An HTML attachment was scrubbed... URL: From peter.chen at lemaker.com Mon Jun 6 11:02:54 2016 From: peter.chen at lemaker.com (peter.chen at lemaker.com) Date: Mon, 6 Jun 2016 19:02:54 +0800 Subject: snappy: Command not found References: , Message-ID: <2016060619025454498725@lemaker.com> Hi, Woodrow I fork you project at https://github.com/xapp-le/SnappyUbuntuCore before, and i have made snappy 16 image based on the kernel snap and gadget snap refer to the Hikey, However, the system can not work. and stay in the uboot stage, as follows: The environment file of the uboot uEnv.txt is same as your project which located at https://github.com/xapp-le/SnappyUbuntuCore/tree/master/builder/gadget/boot-assets . I kown you had updated your project a few days ago, could the snappy 16 work on the roseapple-pi ? Why need to burn the boot_fix.bin to the snappy image by using the makefile object fix-bootflag? Looking forward your reply. Thanks Best Regards Peter Chen LeMaker Developer Team Making Innovation Easy. Email: support at lemaker.org (Technical Support) product at lemaker.org (Product Distribution) Website: http://www.lemaker.org/ From: Woodrow Shen Date: 2016-05-31 19:32 To: Anthony Wong CC: peter.chen at lemaker.com; Liming Wang; 张腾; snapcraft Subject: Re: snappy: Command not found Hi Peter, The work from https://github.com/xapp-le/SnappyUbuntuCore is created by me, and I'm sorry to make you confused because the current master branch isn't ported completely yet, and I'll continue to do the porting based on 3.11 kernel. Woodrow On Tue, May 31, 2016 at 6:27 PM, Anthony Wong wrote: Hi Peter, It will be easier if you follow how we enabled the hikey board for demo, because that's for Ubuntu Core 16. You can find the kernel and the associated snapcraft.yaml to build the kernel snap at https://code.launchpad.net/~canonical-hwe-team/+git/hikey_linux and the gadget snap at https://code.launchpad.net/~canonical-hwe-team/+git/hikey_gadget_snap. More details on bringing the board up can be found here. Happy snapping! Anthony -- Anthony Wong Engineering Manager, Hardware Enablement Canonical Ltd. www.canonical.com On 31 May 2016 at 17:03, peter.chen at lemaker.com wrote: Hi, ZK Thank you . Best Regards Peter Chen Making Innovation Easy LeMaker Team -- The Professional Makers for Hardware and Software Customization. Address: B1002, SIAT campus, Shenzhen University Town, Shenzhen, China Post Code: 518055 Tel: 0755-36330749 Email: support at lemaker.org (Technical Support) product at lemaker.org (Product Distribution) http://www.lemaker.org/ From: Zygmunt Krynicki Date: 2016-05-31 16:59 To: peter.chen at lemaker.com CC: snapcraft; 张腾 Subject: Re: Re: snappy: Command not found On Tue, May 31, 2016 at 10:58 AM, peter.chen at lemaker.com wrote: Hi, ZK The project use the command below in the SnappyUbuntuCore/builder/gadget.mk : snappy: snappy build gadget However, the newer command snap didn't support the same option: snap build gadget And i should how to use it ? Thanks. There are some subtle differences between the two. Let me check that repository out and get back to you. Best regards ZK -- Snapcraft mailing list Snapcraft at lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft -- Snapcraft mailing list Snapcraft at lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft -- Woodrow Shen Software Engineer, Canonical ltd. UES | CE | PC & Core, Taipei -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: Catch.jpg Type: image/jpeg Size: 135732 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: 6594_6594_6594_c(06-06-18-39-08).jpg Type: image/jpeg Size: 6594 bytes Desc: not available URL: From ogra at ubuntu.com Mon Jun 6 14:12:25 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Mon, 06 Jun 2016 16:12:25 +0200 Subject: snappy: Command not found In-Reply-To: <2016060619025454498725@lemaker.com> References: , <2016060619025454498725@lemaker.com> Message-ID: <1465222345.8064.5.camel@ubuntu.com> hi, Am Montag, den 06.06.2016, 19:02 +0800 schrieb peter.chen at lemaker.com: > Hi, Woodrow > I fork you project at https://github.com/xapp-le/SnappyUbuntuCore bef > ore, and i have made snappy 16  image based on the kernel snap and > gadget snap refer to the Hikey, However, the system can not work. and > stay in the uboot stage, as follows: > > > The environment file of the uboot uEnv.txt is same as your project > which located at https://github.com/xapp-le/SnappyUbuntuCore/tree/mas > ter/builder/gadget/boot-assets . > I kown you had updated your project a few days ago, could the snappy > 16 work on the roseapple-pi ?  > Why need to burn the boot_fix.bin to the snappy image by using the > makefile object fix-bootflag?  the uEnv.txt wont be respected or used by ubuntu-device-flash and snapd.  when the image is assembled ubuntu-device-flash injects the proper paths (for kernel, initrd.img and dtbs) into uboot.env, you need to make your uboot use an env file instead of a txt file ... please see the thread "ubuntu-device-flash errors for building armhf image" on this mailing list where i walk woodrow through the process to create a proper uboot.env ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From sergio.schvezov at canonical.com Mon Jun 6 14:28:22 2016 From: sergio.schvezov at canonical.com (Sergio Schvezov) Date: Mon, 06 Jun 2016 11:28:22 -0300 Subject: Docs for snapcraft In-Reply-To: References: <574D6A46.2040105@ubuntu.com> <574E1D61.1050303@ubuntu.com> <574E7F7D.3040802@ubuntu.com> <574E810B.2020505@ubuntu.com> <574E81F9.6020104@ubuntu.com> Message-ID: El miércoles, 1 de junio de 2016 07h'24:02 ART, Scarlett Clark escribió: > Docs would be great, though I was able to find my way well enough. My > issues are as stated above where I am in need of creating my own plugin and > it is very difficult to debug. Logging or a debug mode would be great. Also > logs to debug odd UI issues I could not seem to locate. Otherwise snapcraft > is coming along great. My packages are far from simple :) Hey, this is good to hear. Just a pointer on the custom plugin thing, try it with --debug and you will get more information and python tracebacks. 2.10 is in the oven and making it's may through xenial-proposed, should reach all users soon and will send a proper announcement, there is much more polishing on getting started and user friendly errors in it. Cheers Sergio -- Enviado con Dekko desde mi dispositivo Ubuntu From mhall119 at ubuntu.com Mon Jun 6 14:36:31 2016 From: mhall119 at ubuntu.com (Michael Hall) Date: Mon, 6 Jun 2016 10:36:31 -0400 Subject: Docs for snapcraft In-Reply-To: References: <574D6A46.2040105@ubuntu.com> <574E1D61.1050303@ubuntu.com> <574E7F7D.3040802@ubuntu.com> <574E810B.2020505@ubuntu.com> <574E81F9.6020104@ubuntu.com> Message-ID: <57558A6F.8070909@ubuntu.com> On 06/06/2016 10:28 AM, Sergio Schvezov wrote: > El miércoles, 1 de junio de 2016 07h'24:02 ART, Scarlett Clark > escribió: >> Docs would be great, though I was able to find my way well enough. My >> issues are as stated above where I am in need of creating my own >> plugin and >> it is very difficult to debug. Logging or a debug mode would be great. >> Also >> logs to debug odd UI issues I could not seem to locate. Otherwise >> snapcraft >> is coming along great. My packages are far from simple :) > > Hey, this is good to hear. Just a pointer on the custom plugin thing, > try it with --debug and you will get more information and python > tracebacks. > > 2.10 is in the oven and making it's may through xenial-proposed, should > reach all users soon and will send a proper announcement, there is much > more polishing on getting started and user friendly errors in it. > > Cheers > Sergio > > There is also snappy-debug, which is very useful for debugging runtime problems with your snap. If you haven't tried it yet, hit me up on IRC and I'll show you how. Michael Hall mhall119 at ubuntu.com From w.andrew.keech at gmail.com Mon Jun 6 23:05:48 2016 From: w.andrew.keech at gmail.com (Andrew Keech) Date: Mon, 6 Jun 2016 16:05:48 -0700 Subject: Snappy Playpen: very incomplete gimp-from-git snap Message-ID: <575601CC.7010200@gmail.com> hey everyone, I've got a few hours into a snap of development GIMP, which I have used builds of for years from ppa:otto-kesselgulash/gimp-edge, and successfully built myself in the past with the intention of contributing to GEGL. Turns out that I'm likely still years away from having C skills that would be up to the task. So, in attempting to contribute in other ways I really want to get the momentum going for snappy. Pretty sure the question had come up in one of the community Q&A's before, and mhall said that it was too weird a project to just start on for the team. Which is true, I fully expect this to soak up weeks worth of poking at unless Thorsten Stettin/ some-other-Debian-or-Ubuntu-user-who-is-a-GIMP-developer steps up. I've been using Ubuntu since 2014, watching the community Q&A's on youtube; really love the community team and all you guys do, I believe Ubuntu is entering a new era after years of unacknowledged hard work. Thank you Mark! Thanks everyone else! Thanks so much for this github repo! -Andy PS: not sure it's really complete enough for a merge request yet, so here's a link to the repo https://github.com/wandrewkeech/snappy-playpen From leo.arias at canonical.com Tue Jun 7 01:38:32 2016 From: leo.arias at canonical.com (Leo Arias) Date: Mon, 6 Jun 2016 19:38:32 -0600 Subject: Snappy Playpen: very incomplete gimp-from-git snap In-Reply-To: <575601CC.7010200@gmail.com> References: <575601CC.7010200@gmail.com> Message-ID: <57562598.3010000@canonical.com> Hello Andy, welcome to snapcraft! Your contribution is great. We'll look at it in tomorrow's sprint for sure; I hope you can join us. pura vida. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: OpenPGP digital signature URL: From peter.chen at lemaker.com Tue Jun 7 01:51:20 2016 From: peter.chen at lemaker.com (peter.chen at lemaker.com) Date: Tue, 7 Jun 2016 09:51:20 +0800 Subject: snappy: Command not found References: , , <2016060619025454498725@lemaker.com>, <1465222345.8064.5.camel@ubuntu.com> Message-ID: <2016060709511982709523@lemaker.com> Hi, I have read the thread which you mentioned last week, the uEnv.txt file has been converted to the uboot.env by the tool mkenvimage before creating the gadget snap. Finally, the snappy 16 image has been made successfully by the ubuntu-device-flash, but there is a problem which the uboot can't load the kernel initrd dtbs when the system started. Do you have any better suggestions ? Thanks. Best Regards Peter Chen LeMaker Developer Team Making Innovation Easy. Email: support at lemaker.org (Technical Support) product at lemaker.org (Product Distribution) Website: http://www.lemaker.org/ From: Oliver Grawert Date: 2016-06-06 22:12 To: snapcraft Subject: Re: Re: snappy: Command not found hi, Am Montag, den 06.06.2016, 19:02 +0800 schrieb peter.chen at lemaker.com: > Hi, Woodrow > I fork you project at https://github.com/xapp-le/SnappyUbuntuCore bef > ore, and i have made snappy 16 image based on the kernel snap and > gadget snap refer to the Hikey, However, the system can not work. and > stay in the uboot stage, as follows: > > > The environment file of the uboot uEnv.txt is same as your project > which located at https://github.com/xapp-le/SnappyUbuntuCore/tree/mas > ter/builder/gadget/boot-assets . > I kown you had updated your project a few days ago, could the snappy > 16 work on the roseapple-pi ? > Why need to burn the boot_fix.bin to the snappy image by using the > makefile object fix-bootflag? the uEnv.txt wont be respected or used by ubuntu-device-flash and snapd. when the image is assembled ubuntu-device-flash injects the proper paths (for kernel, initrd.img and dtbs) into uboot.env, you need to make your uboot use an env file instead of a txt file ... please see the thread "ubuntu-device-flash errors for building armhf image" on this mailing list where i walk woodrow through the process to create a proper uboot.env ciao oli -- Snapcraft mailing list Snapcraft at lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft -------------- next part -------------- An HTML attachment was scrubbed... URL: From woodrow.shen at canonical.com Tue Jun 7 03:17:50 2016 From: woodrow.shen at canonical.com (Woodrow Shen) Date: Tue, 7 Jun 2016 11:17:50 +0800 Subject: snappy: Command not found In-Reply-To: <2016060709511982709523@lemaker.com> References: <2016060619025454498725@lemaker.com> <1465222345.8064.5.camel@ubuntu.com> <2016060709511982709523@lemaker.com> Message-ID: Hi Peter, *boot_fix.bin *is workaround to make booting normally without active boot flag. I suspect above uboot error messages is missing the value of ${snappy_kernel} when uboot needs to load the files, like: loadkernel=fatload mmc ${mmcdev}:${mmcpart} ${kernel_addr} ${snappy_kernel}/${kernel_file} Wonder if u-d-f has the capability to set ${snappy_kernel} to uboot.env as what I check the git source from https://github.com/snapcore/snapd/blob/master/partition/uboot.go On Tue, Jun 7, 2016 at 9:51 AM, peter.chen at lemaker.com < peter.chen at lemaker.com> wrote: > Hi, > I have read the thread which you mentioned last week, the* uEnv.txt *file > has been converted to the *uboot.env* by the tool *mkenvimage *before creating > the gadget snap. > Finally, the snappy 16 image has been made successfully by the > *ubuntu-device-flash*, but there is a problem which the uboot can't load > the *kernel* *initrd dtbs *when the system started. > Do you have any better suggestions ? Thanks. > > Best Regards > ------------------------------ > > *Peter Chen* > > > *LeMaker **Developer Team* > > > > *Making Innovation Easy. * > > Email: > support at lemaker.org (Technical Support) > product at lemaker.org (Product Distribution) > > Website: http://www.lemaker.org/ > > > > > > > > > > *From:* Oliver Grawert > *Date:* 2016-06-06 22:12 > *To:* snapcraft > *Subject:* Re: Re: snappy: Command not found > hi, > Am Montag, den 06.06.2016, 19:02 +0800 schrieb peter.chen at lemaker.com: > > Hi, Woodrow > > I fork you project at https://github.com/xapp-le/SnappyUbuntuCore bef > > ore, and i have made snappy 16 image based on the kernel snap and > > gadget snap refer to the Hikey, However, the system can not work. and > > stay in the uboot stage, as follows: > > > > > > The environment file of the uboot uEnv.txt is same as your project > > which located at https://github.com/xapp-le/SnappyUbuntuCore/tree/mas > > ter/builder/gadget/boot-assets . > > I kown you had updated your project a few days ago, could the snappy > > 16 work on the roseapple-pi ? > > Why need to burn the boot_fix.bin to the snappy image by using the > > makefile object fix-bootflag? > > the uEnv.txt wont be respected or used by ubuntu-device-flash and > snapd. > when the image is assembled ubuntu-device-flash injects the proper > paths (for kernel, initrd.img and dtbs) into uboot.env, you need to > make your uboot use an env file instead of a txt file ... > > please see the thread "ubuntu-device-flash errors for building armhf > image" on this mailing list where i walk woodrow through the process to > create a proper uboot.env > > ciao > oli > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > > -- Woodrow Shen Software Engineer, Canonical ltd. UES | CE | PC & Core, Taipei -------------- next part -------------- An HTML attachment was scrubbed... URL: From mark at ubuntu.com Tue Jun 7 03:40:25 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Tue, 7 Jun 2016 04:40:25 +0100 Subject: Snappy Playpen: very incomplete gimp-from-git snap In-Reply-To: <575601CC.7010200@gmail.com> References: <575601CC.7010200@gmail.com> Message-ID: <57564229.7010405@ubuntu.com> On 07/06/16 00:05, Andrew Keech wrote: > I've got a few hours into a snap of development GIMP, which I have > used builds of for years from ppa:otto-kesselgulash/gimp-edge, and > successfully built myself in the past with the intention of > contributing to GEGL. Turns out that I'm likely still years away from > having C skills that would be up to the task. So, in attempting to > contribute in other ways I really want to get the momentum going for > snappy. Rocking! Thank you Andrew. The development GIMP is a great choice as I'm sure the snap experience will encourage more folks to track active development, giving the GIMP faster feedback. It takes many of us to change the world, you'll find lots of people here who have made it possible and who will help out if you get stuck. Welcome aboard! Mark From vinzjobard at ubuntu.com Tue Jun 7 08:07:25 2016 From: vinzjobard at ubuntu.com (Winael) Date: Tue, 07 Jun 2016 08:07:25 +0000 Subject: [Snappy-Playpen] Need some help to create Etherpad-snap Message-ID: Hi Every one, Since Playpen is beginnig, it will be more easely to work on this litthe project. I'm trying to build an etherpad-lite snap, but I'm stuck for many reasons : - I'm a totally noob that wants to learn - To be able to work, I must copy entire folder from `parts/etherpad-lite/src` to the `./snap` folder and I didn't find how to do it There will be lots of work to do (very useful to learn) because I planned to add/replace dirtydb by a mariadb database, and add a configuration gui for webdm. The goal is very simple. Learn a lot and sharing knowledge from basics to inspire people to make others snaps from the Framacloud project, a project lead by a French Organization that promote Free Software and an ungoglelize Internet program full of Free services (etherpad-lite, ethercal, tinytiny-rss...). Because French LoCo works a lot with this organization, animate Snapcraft workshop around their services could be nice. You can find a branch for this project : https://github.com/Winael/snappy-playpen-1/tree/etherpad-lite/etherpad-lite And a Telegram Group : https://telegram.me/EtherpadLitePlaypenSnappy Thx for your help guys Cheers Winael -------------- next part -------------- An HTML attachment was scrubbed... URL: From daniel.holbach at canonical.com Tue Jun 7 08:17:45 2016 From: daniel.holbach at canonical.com (Daniel Holbach) Date: Tue, 7 Jun 2016 10:17:45 +0200 Subject: Snappy Playpen Kickoff today Message-ID: <57568329.5010900@canonical.com> Hello everybody, if you haven't read it already, here's the announce of the Snappy Playpen: https://developer.ubuntu.com/en/blog/2016/06/03/announcing-snappy-playpen/ Basically we're hanging out together in - https://gitter.im/ubuntu/snappy-playpen or - #snappy on Freenode and work together on bringing new snaps to https://github.com/ubuntu/snappy-playpen - we'll learn from each other and document best practices together. We're looking forward to seeing you all today and in the coming days! Have a great day, Daniel From daniel.holbach at canonical.com Tue Jun 7 08:19:19 2016 From: daniel.holbach at canonical.com (Daniel Holbach) Date: Tue, 7 Jun 2016 10:19:19 +0200 Subject: [Snappy-Playpen] Need some help to create Etherpad-snap In-Reply-To: References: Message-ID: <57568387.5000901@canonical.com> Salut Vincent, thanks a lot for your help and interest. On 07.06.2016 10:07, Winael wrote: > - To be able to work, I must copy entire folder from > `parts/etherpad-lite/src` to the `./snap` folder and I didn't find how > to do it Alan pointed out this earlier, maybe it will help? http://bazaar.launchpad.net/~popey/+junk/flightgear/view/head:/snapcraft.yaml Good to see you on gitter already! :-) Have a great day, Daniel From ogra at ubuntu.com Tue Jun 7 09:17:41 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Tue, 07 Jun 2016 11:17:41 +0200 Subject: snappy: Command not found In-Reply-To: References: <2016060619025454498725@lemaker.com> <1465222345.8064.5.camel@ubuntu.com> <2016060709511982709523@lemaker.com> Message-ID: <1465291061.17816.15.camel@anubis> hi, Am Dienstag, den 07.06.2016, 11:17 +0800 schrieb Woodrow Shen: > boot_fix.bin is workaround to make booting normally without active > boot flag. > I suspect above uboot error messages is missing the value > of ${snappy_kernel} when uboot needs to load the files, like: > loadkernel=fatload mmc ${mmcdev}:${mmcpart} ${kernel_addr} > ${snappy_kernel}/${kernel_file} exactly, u-d-f creates an image and then installs the kernel snap inside. after it installed it knows the exact path to the binaries, which is /snap/$packagename/$revision and this path is set to snappy_kernel= inside the uboot.env file. every time you upgrade or roll-back the kernel this variable is adjusted alongside, if a boot fails the value automatically falls back to the last revision that worked and it reboots. ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From vasilisc777 at gmail.com Tue Jun 7 10:43:11 2016 From: vasilisc777 at gmail.com (Alekseenko Vasilii) Date: Tue, 7 Jun 2016 13:43:11 +0300 Subject: questions Message-ID: <5756A53F.70200@gmail.com> Sorry for my english. ---------------- 1) snap packages classified in Ubuntu Store like deb (Main, Restricted, Universe, Multiverse)? 2) May I compile app and send in Ubuntu Store without open source? 3) Application in snap package requested "Network Interfaces". Maybe app later fetch evil-library from evil-host? 4) Where guarantee what author packed application rather than evil-man packed with backdoor? 5) Evidently, Snap packages are all-sufficient. Average snap should be bigger than average deb. Why Krita in snap package ~100 мб, but Krita in deb package with their dependency ~500 mb? 6) What is analogous to command "apt-get autoclean" ? Who clears folder /var/lib/snapd/snaps/ ? 7) Sqiud snap package now available. When LAMP/LEMP turn is? 8) I don't need older version application. How I may remove older revision? ---------------- -- Best regards, Vasilisc From mark at ubuntu.com Tue Jun 7 11:14:38 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Tue, 7 Jun 2016 20:14:38 +0900 Subject: questions In-Reply-To: <5756A53F.70200@gmail.com> References: <5756A53F.70200@gmail.com> Message-ID: <5756AC9E.7010404@ubuntu.com> Hi Vasili On 07/06/16 19:43, Alekseenko Vasilii wrote: > 1) snap packages classified in Ubuntu Store like deb (Main, > Restricted, Universe, Multiverse)? No, they are all the same. We would like to tag some of them to indicate source availability etc, patches welcome. > 2) May I compile app and send in Ubuntu Store without open source? Yes. > 3) Application in snap package requested "Network Interfaces". Maybe > app later fetch evil-library from evil-host? > Possibly, but then it will run inside the same confinement as the app. It could to evil things with the data you feed the app, but should not see any other app's data. > 4) Where guarantee what author packed application rather than evil-man > packed with backdoor? We're using GPG signatures. > 5) Evidently, Snap packages are all-sufficient. Average snap should be > bigger than average deb. Why Krita in snap package ~100 мб, but Krita > in deb package with their dependency ~500 mb? Good question :) > 6) What is analogous to command "apt-get autoclean" ? Who clears > folder /var/lib/snapd/snaps/ ? The snaps are removed automatically when they are no longer needed for rollbacks. > 7) Sqiud snap package now available. When LAMP/LEMP turn is? When you submit it! Or when someone else does. > 8) I don't need older version application. How I may remove older > revision? The system gardens this automatically for you :) Mark From nhaines at ubuntu.com Tue Jun 7 11:19:16 2016 From: nhaines at ubuntu.com (Nathan Haines) Date: Tue, 7 Jun 2016 04:19:16 -0700 Subject: questions In-Reply-To: <5756AC9E.7010404@ubuntu.com> References: <5756A53F.70200@gmail.com> <5756AC9E.7010404@ubuntu.com> Message-ID: <5756ADB4.60409@ubuntu.com> On 06/07/2016 04:14 AM, Mark Shuttleworth wrote: >> 5) Evidently, Snap packages are all-sufficient. Average snap should be >> bigger than average deb. Why Krita in snap package ~100 мб, but Krita >> in deb package with their dependency ~500 mb? Oh, you know this one, Mark! It's because deb packages are uncompressed during installation and stored around the filesystem, whereas a snap is a heavily-compressed squashfs that's mounted in place without being uncompressed on disk. :) -- Nathan Haines Ubuntu - http://www.ubuntu.com/ From sergio.schvezov at canonical.com Tue Jun 7 11:23:53 2016 From: sergio.schvezov at canonical.com (Sergio Schvezov) Date: Tue, 07 Jun 2016 08:23:53 -0300 Subject: questions In-Reply-To: <5756A53F.70200@gmail.com> Message-ID: <2f697d62-810b-40ef-9e3d-08b7f8cf1f00@canonical.com> El martes, 7 de junio de 2016 07h'43:11 ART, Alekseenko Vasilii escribió: > Sorry for my english. > > ---------------- > > 1) snap packages classified in Ubuntu Store like deb (Main, Restricted, > Universe, Multiverse)? It is all one big bucket. > 2) May I compile app and send in Ubuntu Store without open source? Yes, the software licenses of the software you use may require you to share the source as determined in those licenses. > 3) Application in snap package requested "Network Interfaces". Maybe app > later fetch evil-library from evil-host? True, but you the evilness would be confined to that snap. To be fair, nothing prevents bundling evil-library in evil-snap from the start. > 4) Where guarantee what author packed application rather than evil-man > packed with backdoor? > 5) Evidently, Snap packages are all-sufficient. Average snap should be > bigger than average deb. Why Krita in snap package ~100 мб, but Krita in > deb package with their dependency ~500 mb? It could be related to fine tuning of the libraries used. The packages in the Ubuntu archive have almost everything turned on. > 6) What is analogous to command "apt-get autoclean" ? Who clears folder > /var/lib/snapd/snaps/ ? snapd, it garbage collects. > 7) Sqiud snap package now available. When LAMP/LEMP turn is? When someone gets to it 😌 > 8) I don't need older version application. How I may remove older revision? It would be garbage collected by the system. -- Enviado con Dekko desde mi dispositivo Ubuntu From mark at ubuntu.com Tue Jun 7 11:26:12 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Tue, 7 Jun 2016 20:26:12 +0900 Subject: questions In-Reply-To: <5756ADB4.60409@ubuntu.com> References: <5756A53F.70200@gmail.com> <5756AC9E.7010404@ubuntu.com> <5756ADB4.60409@ubuntu.com> Message-ID: <5756AF54.9080504@ubuntu.com> On 07/06/16 20:19, Nathan Haines wrote: > On 06/07/2016 04:14 AM, Mark Shuttleworth wrote: >>> 5) Evidently, Snap packages are all-sufficient. Average snap should be >>> bigger than average deb. Why Krita in snap package ~100 мб, but Krita >>> in deb package with their dependency ~500 mb? > > Oh, you know this one, Mark! > > It's because deb packages are uncompressed during installation and > stored around the filesystem, whereas a snap is a heavily-compressed > squashfs that's mounted in place without being uncompressed on disk. :) Yes indeed :) Also, Krita doesn't need all the files in all the deps. Especially as more of the krita deps are expressed as source parts, the package size will come down. Mark From ogra at ubuntu.com Tue Jun 7 11:38:43 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Tue, 07 Jun 2016 13:38:43 +0200 Subject: questions In-Reply-To: <5756AF54.9080504@ubuntu.com> References: <5756A53F.70200@gmail.com> <5756AC9E.7010404@ubuntu.com> <5756ADB4.60409@ubuntu.com> <5756AF54.9080504@ubuntu.com> Message-ID: <1465299523.30748.1.camel@ubuntu.com> hi, Am Dienstag, den 07.06.2016, 20:26 +0900 schrieb Mark Shuttleworth: > On 07/06/16 20:19, Nathan Haines wrote: > > > > On 06/07/2016 04:14 AM, Mark Shuttleworth wrote: > > > > > > > > > > > 5) Evidently, Snap packages are all-sufficient. Average snap > > > > should be > > > > bigger than average deb. Why Krita in snap package ~100 мб, but > > > > Krita > > > > in deb package with their dependency ~500 mb? > > Oh, you know this one, Mark! > > > > It's because deb packages are uncompressed during installation and > > stored around the filesystem, whereas a snap is a heavily- > > compressed > > squashfs that's mounted in place without being uncompressed on > > disk.  :) > Yes indeed :) > > Also, Krita doesn't need all the files in all the deps. Especially as > more of the krita deps are expressed as source parts, the package > size > will come down. i belive snapcraft does not install any recommends by default, for some packages that can make a huge difference (along with the admittedly stronger compression in a snap) ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From svij at ubuntu.com Tue Jun 7 11:41:03 2016 From: svij at ubuntu.com (Sujeevan (svij) Vijayakumaran) Date: Tue, 7 Jun 2016 13:41:03 +0200 Subject: Installing ubuntu-core snap inside a LXD container fails Message-ID: <76b68477-104c-3582-2f9b-3f69868fbb8b@ubuntu.com> Hello! I'm trying to install a snap inside a freshly created LXD container, which isn't possible: ----- ubuntu at snap-test:~$ sudo snap install taskwarrior_2.5.1-snap0_amd64.snap 64.75 MB / 64.75 MB [========================================================================] 100.00 % 5.38 MB/s error: cannot perform the following tasks: - Mount snap "ubuntu-core" ([start snap-ubuntu\x2dcore-122.mount] failed with exit status 1: Job for snap-ubuntu\x2dcore-122.mount failed. See "systemctl status "snap-ubuntu\\x2dcore-122.mount"" and "journalctl -xe" for details. ) ubuntu at snap-test:~$ systemctl status "snap-ubuntu\\x2dcore-122.mount" ● snap-ubuntu\x2dcore-122.mount - Squashfs mount unit for ubuntu-core Loaded: loaded (/etc/systemd/system/snap-ubuntu\x2dcore-122.mount; enabled; vendor preset: enabled) Active: failed (Result: exit-code) since Tue 2016-06-07 11:32:13 UTC; 16s ago Where: /snap/ubuntu-core/122 What: /var/lib/snapd/snaps/ubuntu-core_122.snap Process: 8584 ExecMount=/bin/mount /var/lib/snapd/snaps/ubuntu-core_122.snap /snap/ubuntu-core/122 (code=exited, status=32) Jun 07 11:32:13 snap-test systemd[1]: Mounting Squashfs mount unit for ubuntu-core... Jun 07 11:32:13 snap-test mount[8584]: mount: /snap/ubuntu-core/122: mount failed: Unknown error -1 Jun 07 11:32:13 snap-test systemd[1]: snap-ubuntu\x2dcore-122.mount: Mount process exited, code=exited status=32 Jun 07 11:32:13 snap-test systemd[1]: Failed to mount Squashfs mount unit for ubuntu-core. Jun 07 11:32:13 snap-test systemd[1]: snap-ubuntu\x2dcore-122.mount: Unit entered failed state. ------ It works fine if it's not inside a LXD container. Is this an issue with LXD or with snapd or am I missing something? Also: Why is there an "ugly" \x2d (which seems to be a hypen-minus) in the systemd-file rather than a "normal" minus? Cheers, Sujeevan From svij at ubuntu.com Tue Jun 7 12:37:30 2016 From: svij at ubuntu.com (Sujeevan (svij) Vijayakumaran) Date: Tue, 7 Jun 2016 14:37:30 +0200 Subject: Installing ubuntu-core snap inside a LXD container fails In-Reply-To: <5756BEDF.6010402@ubuntu.com> References: <76b68477-104c-3582-2f9b-3f69868fbb8b@ubuntu.com> <5756BEDF.6010402@ubuntu.com> Message-ID: Hi, Am 07.06.2016 um 14:32 schrieb Mark Shuttleworth: > On 07/06/16 20:41, Sujeevan (svij) Vijayakumaran wrote: >> It works fine if it's not inside a LXD container. Is this an issue with >> LXD or with snapd or am I missing something? > > That's because snapd tickles the kernel in a way that blows LXD 2.0's > mind :) but it's a top priority for the LXD crowd to fix that. Good catch! > > status: confirmed Good to know! Then I'll just have to wait. :-) Cheers Sujeevan PS: You forgot to answer to the Mailinglist :) PPS: Are you coming to UbuCon Europe? ;) From svij at ubuntu.com Tue Jun 7 12:49:14 2016 From: svij at ubuntu.com (Sujeevan (svij) Vijayakumaran) Date: Tue, 7 Jun 2016 14:49:14 +0200 Subject: Man-pages in snaps Message-ID: <062e78c4-151f-a53e-f807-38c7c8c18963@ubuntu.com> Hello, I started creating a snap for the CLI task manager taskwarrior (https://taskwarrior.org/). This project has really good man pages. Anyway there is currently no support to access man-pages from snaps: https://bugs.launchpad.net/snappy/+bug/1575593 Since it doesn't make much sense to publish an CLI app without the good man pages: Is there any other ways to make it possible that the user can access the man pages of a snap? The tools "task help" command doesnt help here since is doesn’t show everything. Cheers, Sujeevan From david.planella at ubuntu.com Tue Jun 7 13:15:33 2016 From: david.planella at ubuntu.com (David Planella) Date: Tue, 7 Jun 2016 15:15:33 +0200 Subject: Snappy Playpen: very incomplete gimp-from-git snap In-Reply-To: <575601CC.7010200@gmail.com> References: <575601CC.7010200@gmail.com> Message-ID: Hi Andy, Depending on your time zone and availability you might have not seen this yet, but we're at https://gitter.im/ubuntu/snappy-playpen hacking on snaps together across a couple of different time zones during today. So do come along and join us! :) What you've started there is awesome, and while really appreciate the kind words, it's people like you who make Ubuntu what it is. Thanks! Looking forward to meeting you at the playpen and to continuing the GIMP conversation on the list :) Cheers, David. On Tue, Jun 7, 2016 at 1:05 AM, Andrew Keech wrote: > hey everyone, > > I've got a few hours into a snap of development GIMP, which I have used > builds of for years from ppa:otto-kesselgulash/gimp-edge, and successfully > built myself in the past with the intention of contributing to GEGL. Turns > out that I'm likely still years away from having C skills that would be up > to the task. So, in attempting to contribute in other ways I really want to > get the momentum going for snappy. Pretty sure the question had come up in > one of the community Q&A's before, and mhall said that it was too weird a > project to just start on for the team. Which is true, I fully expect this > to soak up weeks worth of poking at unless Thorsten Stettin/ > some-other-Debian-or-Ubuntu-user-who-is-a-GIMP-developer steps up. > > I've been using Ubuntu since 2014, watching the community Q&A's on > youtube; really love the community team and all you guys do, I believe > Ubuntu is entering a new era after years of unacknowledged hard work. > Thank you Mark! Thanks everyone else! Thanks so much for this github repo! > > -Andy > > PS: not sure it's really complete enough for a merge request yet, so > here's a link to the repo https://github.com/wandrewkeech/snappy-playpen > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > -------------- next part -------------- An HTML attachment was scrubbed... URL: From peter.chen at lemaker.com Tue Jun 7 13:35:37 2016 From: peter.chen at lemaker.com (peter.chen at lemaker.com) Date: Tue, 7 Jun 2016 21:35:37 +0800 Subject: snappy: Command not found [solved] References: , , <2016060619025454498725@lemaker.com>, <1465222345.8064.5.camel@ubuntu.com>, <2016060709511982709523@lemaker.com>, , <1465291061.17816.15.camel@anubis> Message-ID: <2016060721353741830828@lemaker.com> Hi, All Thanks for your help, The problem which the uboot can't load the kernel has been fixed, and the snappy ubuntu 16 is ported to the Lemaker Guitar successfully now. The problem mainly related to uboot and i have committed the patch to our github at https://github.com/LeMaker/u-boot-actions/commit/1a2f52c79652bd33002a6aee4e318db2061e4879 , Hope this can help others. Best Regards Peter Chen LeMaker Developer Team Making Innovation Easy. Email: support at lemaker.org (Technical Support) product at lemaker.org (Product Distribution) Website: http://www.lemaker.org/ From: Oliver Grawert Date: 2016-06-07 17:17 To: Woodrow Shen CC: peter.chen at lemaker.com; snapcraft Subject: Re: Re: snappy: Command not found hi, Am Dienstag, den 07.06.2016, 11:17 +0800 schrieb Woodrow Shen: > boot_fix.bin is workaround to make booting normally without active > boot flag. > I suspect above uboot error messages is missing the value > of ${snappy_kernel} when uboot needs to load the files, like: > loadkernel=fatload mmc ${mmcdev}:${mmcpart} ${kernel_addr} > ${snappy_kernel}/${kernel_file} exactly, u-d-f creates an image and then installs the kernel snap inside. after it installed it knows the exact path to the binaries, which is /snap/$packagename/$revision and this path is set to snappy_kernel= inside the uboot.env file. every time you upgrade or roll-back the kernel this variable is adjusted alongside, if a boot fails the value automatically falls back to the last revision that worked and it reboots. ciao oli -------------- next part -------------- An HTML attachment was scrubbed... URL: From claudioandre.br at gmail.com Tue Jun 7 14:14:36 2016 From: claudioandre.br at gmail.com (=?UTF-8?Q?Claudio_Andr=C3=A9?=) Date: Tue, 7 Jun 2016 11:14:36 -0300 Subject: Installing ubuntu-core snap inside a LXD container fails In-Reply-To: <76b68477-104c-3582-2f9b-3f69868fbb8b@ubuntu.com> References: <76b68477-104c-3582-2f9b-3f69868fbb8b@ubuntu.com> Message-ID: 2016-06-07 9:37 GMT-03:00 Sujeevan (svij) Vijayakumaran : > > > That's because snapd tickles the kernel in a way that blows[...] > That is interesting. Where I can find more about the 'tickles'? My motivation: - I'm seeing closed source software behaves badly, e.g, AMD software prints. Internal Error: as failed Codegen phase failed compilation. If I know what the 'tickles' mean, I might be able to do something. Thanks. 2016-06-07 8:41 GMT-03:00 Sujeevan (svij) Vijayakumaran : > Hello! > > I'm trying to install a snap inside a freshly created LXD container, > which isn't possible: > > ----- > ubuntu at snap-test:~$ sudo snap install taskwarrior_2.5.1-snap0_amd64.snap > 64.75 MB / 64.75 MB > [========================================================================] > 100.00 % 5.38 MB/s > > error: cannot perform the following tasks: > - Mount snap "ubuntu-core" ([start snap-ubuntu\x2dcore-122.mount] failed > with exit status 1: Job for snap-ubuntu\x2dcore-122.mount failed. See > "systemctl status "snap-ubuntu\\x2dcore-122.mount"" and "journalctl -xe" > for details. > ) > ubuntu at snap-test:~$ systemctl status "snap-ubuntu\\x2dcore-122.mount" > ● snap-ubuntu\x2dcore-122.mount - Squashfs mount unit for ubuntu-core > Loaded: loaded (/etc/systemd/system/snap-ubuntu\x2dcore-122.mount; > enabled; vendor preset: enabled) > Active: failed (Result: exit-code) since Tue 2016-06-07 11:32:13 UTC; > 16s ago > Where: /snap/ubuntu-core/122 > What: /var/lib/snapd/snaps/ubuntu-core_122.snap > Process: 8584 ExecMount=/bin/mount > /var/lib/snapd/snaps/ubuntu-core_122.snap /snap/ubuntu-core/122 > (code=exited, status=32) > > Jun 07 11:32:13 snap-test systemd[1]: Mounting Squashfs mount unit for > ubuntu-core... > Jun 07 11:32:13 snap-test mount[8584]: mount: /snap/ubuntu-core/122: > mount failed: Unknown error -1 > Jun 07 11:32:13 snap-test systemd[1]: snap-ubuntu\x2dcore-122.mount: > Mount process exited, code=exited status=32 > Jun 07 11:32:13 snap-test systemd[1]: Failed to mount Squashfs mount > unit for ubuntu-core. > Jun 07 11:32:13 snap-test systemd[1]: snap-ubuntu\x2dcore-122.mount: > Unit entered failed state. > ------ > > It works fine if it's not inside a LXD container. Is this an issue with > LXD or with snapd or am I missing something? > > Also: Why is there an "ugly" \x2d (which seems to be a hypen-minus) in > the systemd-file rather than a "normal" minus? > > Cheers, > Sujeevan > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > -------------- next part -------------- An HTML attachment was scrubbed... URL: From reinhard.pointner at gmail.com Tue Jun 7 14:56:54 2016 From: reinhard.pointner at gmail.com (Reinhard Pointner) Date: Tue, 7 Jun 2016 22:56:54 +0800 Subject: Debugging `confinement: strict` issues Message-ID: Hi all, Does anyone have any hints on how to debug issues that only happen in non-devmode? Like logging when certain "interfaces" are used? I'm working on snapping a Java application, and I've got `plugs: [home, network]` in my snapcraft.yaml but the app can't write any files in the user home (home plug should give the app write access right? also, what about network shares mounted by the user?). Also, when making http requests that thread will just deadlock (no exceptions or and timeouts not respected, so i suspect lower level network api issues). Cheers, Reinhard -------------- next part -------------- An HTML attachment was scrubbed... URL: From zygmunt.krynicki at canonical.com Tue Jun 7 15:01:35 2016 From: zygmunt.krynicki at canonical.com (Zygmunt Krynicki) Date: Tue, 7 Jun 2016 17:01:35 +0200 Subject: Debugging `confinement: strict` issues In-Reply-To: References: Message-ID: You should see useful hints in the system log (use journalctl to see it). Most of the errors boil down to one of two cases: forbidden system call (the message will just contain the system call number along with some techno babble) and forbidden filesystem path (the message will indicate the path, the word DENIED and the operation that was requested but got denied). If you find those you know where to start. Please share your findings. Mounted network shares won't be visible AFAIR. Home will be auto-connected as of the next version but for now you need to connect it manually with "snap connect yoursnapname:home ubuntu-core:home". Best regards. ZK On Tue, Jun 7, 2016 at 4:56 PM, Reinhard Pointner wrote: > Hi all, > > Does anyone have any hints on how to debug issues that only happen in > non-devmode? Like logging when certain "interfaces" are used? > > I'm working on snapping a Java application, and I've got `plugs: [home, > network]` in my snapcraft.yaml but the app can't write any files in the user > home (home plug should give the app write access right? also, what about > network shares mounted by the user?). Also, when making http requests that > thread will just deadlock (no exceptions or and timeouts not respected, so i > suspect lower level network api issues). > > Cheers, > Reinhard > > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > From reinhard.pointner at gmail.com Tue Jun 7 15:27:58 2016 From: reinhard.pointner at gmail.com (Reinhard Pointner) Date: Tue, 7 Jun 2016 23:27:58 +0800 Subject: Debugging `confinement: strict` issues In-Reply-To: References: Message-ID: 1. Alright, home not working by default right now explains the first issue. Here's the syslog. Makes sense: audit[3155]: AVC apparmor="DENIED" operation="rename_src" profile="snap.filebot.filebot" name="/home/reinhard/Documents/test.mp3" pid=3155 comm="java" requested_mask="wrd" denied_mask="wrd" fsuid=1000 ouid=1000 2. The network issue seems more intricate. So here's the Java stack trace for a thread that deadlocks: java.net.Inet4AddressImpl.lookupAllHostAddr(Native Method) java.net.InetAddress$2.lookupAllHostAddr(InetAddress.java:928) java.net.InetAddress.getAddressesFromNameService(InetAddress.java:1323) java.net.InetAddress.getAllByName0(InetAddress.java:1276) java.net.InetAddress.getAllByName(InetAddress.java:1192) java.net.InetAddress.getAllByName(InetAddress.java:1126) java.net.InetAddress.getByName(InetAddress.java:1076) sun.net.www.protocol.http.HttpURLConnection$5.run(HttpURLConnection.java:959) These snippets from the syslog might be relevant: Fails to access some /proc stuff: audit[3207]: AVC apparmor="DENIED" operation="open" profile="snap.filebot.filebot" name="/proc/3206/net/if_inet6" pid=3207 comm="java" requested_mask="r" denied_mask="r" fsuid=1000 ouid=0 audit[3207]: AVC apparmor="DENIED" operation="open" profile="snap.filebot.filebot" name="/proc/3206/net/ipv6_route" pid=3207 comm="java" requested_mask="r" denied_mask="r" fsuid=1000 ouid=0 audit[3207]: AVC apparmor="DENIED" operation="open" profile="snap.filebot.filebot" name="/proc/3206/net/if_inet6" pid=3207 comm="java" requested_mask="r" denied_mask="r" fsuid=1000 ouid=0 No idea what this does: audit: type=1326 audit(1465311994.620:61): auid=4294967295 uid=1000 gid=1000 ses=4294967295 pid=3233 comm="java" exe="/snap/filebot/100001/oracle-java/jre/bin/java" sig=31 arch=c000003e syscall=49 compat=0 ip=0x7f3a02393837 code=0x0 The Java stack trace says something about IPv4 but the syslog says things about IPv6. This is a bit strange. Cheers, Reinhard On 7 June 2016 at 23:01, Zygmunt Krynicki wrote: > You should see useful hints in the system log (use journalctl to see > it). Most of the errors boil down to one of two cases: forbidden > system call (the message will just contain the system call number > along with some techno babble) and forbidden filesystem path (the > message will indicate the path, the word DENIED and the operation that > was requested but got denied). > > If you find those you know where to start. Please share your findings. > > Mounted network shares won't be visible AFAIR. Home will be > auto-connected as of the next version but for now you need to connect > it manually with "snap connect yoursnapname:home ubuntu-core:home". > > Best regards. > ZK > > On Tue, Jun 7, 2016 at 4:56 PM, Reinhard Pointner > wrote: > > Hi all, > > > > Does anyone have any hints on how to debug issues that only happen in > > non-devmode? Like logging when certain "interfaces" are used? > > > > I'm working on snapping a Java application, and I've got `plugs: [home, > > network]` in my snapcraft.yaml but the app can't write any files in the > user > > home (home plug should give the app write access right? also, what about > > network shares mounted by the user?). Also, when making http requests > that > > thread will just deadlock (no exceptions or and timeouts not respected, > so i > > suspect lower level network api issues). > > > > Cheers, > > Reinhard > > > > > > -- > > Snapcraft mailing list > > Snapcraft at lists.ubuntu.com > > Modify settings or unsubscribe at: > > https://lists.ubuntu.com/mailman/listinfo/snapcraft > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From zygmunt.krynicki at canonical.com Tue Jun 7 16:05:12 2016 From: zygmunt.krynicki at canonical.com (Zygmunt Krynicki) Date: Tue, 7 Jun 2016 18:05:12 +0200 Subject: Debugging `confinement: strict` issues In-Reply-To: References: Message-ID: On Tue, Jun 7, 2016 at 5:27 PM, Reinhard Pointner wrote: > 1. > Alright, home not working by default right now explains the first issue. > > Here's the syslog. Makes sense: > audit[3155]: AVC apparmor="DENIED" operation="rename_src" > profile="snap.filebot.filebot" name="/home/reinhard/Documents/test.mp3" > pid=3155 comm="java" requested_mask="wrd" denied_mask="wrd" fsuid=1000 > ouid=1000 > > > 2. > The network issue seems more intricate. > > So here's the Java stack trace for a thread that deadlocks: > java.net.Inet4AddressImpl.lookupAllHostAddr(Native Method) > java.net.InetAddress$2.lookupAllHostAddr(InetAddress.java:928) > java.net.InetAddress.getAddressesFromNameService(InetAddress.java:1323) > java.net.InetAddress.getAllByName0(InetAddress.java:1276) > java.net.InetAddress.getAllByName(InetAddress.java:1192) > java.net.InetAddress.getAllByName(InetAddress.java:1126) > java.net.InetAddress.getByName(InetAddress.java:1076) > sun.net.www.protocol.http.HttpURLConnection$5.run(HttpURLConnection.java:959) > > These snippets from the syslog might be relevant: > > Fails to access some /proc stuff: > audit[3207]: AVC apparmor="DENIED" operation="open" > profile="snap.filebot.filebot" name="/proc/3206/net/if_inet6" pid=3207 > comm="java" requested_mask="r" denied_mask="r" fsuid=1000 ouid=0 > audit[3207]: AVC apparmor="DENIED" operation="open" > profile="snap.filebot.filebot" name="/proc/3206/net/ipv6_route" pid=3207 > comm="java" requested_mask="r" denied_mask="r" fsuid=1000 ouid=0 > audit[3207]: AVC apparmor="DENIED" operation="open" > profile="snap.filebot.filebot" name="/proc/3206/net/if_inet6" pid=3207 > comm="java" requested_mask="r" denied_mask="r" fsuid=1000 ouid=0 Those will go away after you add the "network-bind" plug as instructed below. > No idea what this does: > audit: type=1326 audit(1465311994.620:61): auid=4294967295 uid=1000 gid=1000 > ses=4294967295 pid=3233 comm="java" > exe="/snap/filebot/100001/oracle-java/jre/bin/java" sig=31 arch=c000003e > syscall=49 compat=0 ip=0x7f3a02393837 code=0x0 This is syscall 49 -- aka bind, add the network-bind plug to your app and you should be good to go. After those two you should be good to go. Best regards ZK From mark at ubuntu.com Tue Jun 7 16:31:06 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Wed, 8 Jun 2016 01:31:06 +0900 Subject: Installing ubuntu-core snap inside a LXD container fails In-Reply-To: References: <76b68477-104c-3582-2f9b-3f69868fbb8b@ubuntu.com> Message-ID: <5756F6CA.3000702@ubuntu.com> On 07/06/16 23:14, Claudio André wrote: > 2016-06-07 9:37 GMT-03:00 Sujeevan (svij) > Vijayakumaran >: > > > That's because snapd tickles the kernel in a way that blows[...] > > > That is interesting. Where I can find more about the 'tickles'? My > motivation: > - I'm seeing closed source software behaves badly, e.g, AMD software > prints. > > Internal Error: as failed > Codegen phase failed compilation. > > If I know what the 'tickles' mean, I might be able to do something. > Thanks. I think it has to do with nesting kernel structures related to security (LXD sets those up, then snapd wants to tweak them again), but a much clearer answer would come from Stephane cc'd. Mark > > 2016-06-07 8:41 GMT-03:00 Sujeevan (svij) Vijayakumaran > >: > > Hello! > > I'm trying to install a snap inside a freshly created LXD container, > which isn't possible: > > ----- > ubuntu at snap-test:~$ sudo snap install > taskwarrior_2.5.1-snap0_amd64.snap > 64.75 MB / 64.75 MB > [========================================================================] > 100.00 % 5.38 MB/s > > error: cannot perform the following tasks: > - Mount snap "ubuntu-core" ([start snap-ubuntu\x2dcore-122.mount] > failed > with exit status 1: Job for snap-ubuntu\x2dcore-122.mount failed. See > "systemctl status "snap-ubuntu\\x2dcore-122.mount"" and > "journalctl -xe" > for details. > ) > ubuntu at snap-test:~$ systemctl status "snap-ubuntu\\x2dcore-122.mount" > ● snap-ubuntu\x2dcore-122.mount - Squashfs mount unit for ubuntu-core > Loaded: loaded (/etc/systemd/system/snap-ubuntu\x2dcore-122.mount; > enabled; vendor preset: enabled) > Active: failed (Result: exit-code) since Tue 2016-06-07 > 11:32:13 UTC; > 16s ago > Where: /snap/ubuntu-core/122 > What: /var/lib/snapd/snaps/ubuntu-core_122.snap > Process: 8584 ExecMount=/bin/mount > /var/lib/snapd/snaps/ubuntu-core_122.snap /snap/ubuntu-core/122 > (code=exited, status=32) > > Jun 07 11:32:13 snap-test systemd[1]: Mounting Squashfs mount unit for > ubuntu-core... > Jun 07 11:32:13 snap-test mount[8584]: mount: /snap/ubuntu-core/122: > mount failed: Unknown error -1 > Jun 07 11:32:13 snap-test systemd[1]: snap-ubuntu\x2dcore-122.mount: > Mount process exited, code=exited status=32 > Jun 07 11:32:13 snap-test systemd[1]: Failed to mount Squashfs mount > unit for ubuntu-core. > Jun 07 11:32:13 snap-test systemd[1]: snap-ubuntu\x2dcore-122.mount: > Unit entered failed state. > ------ > > It works fine if it's not inside a LXD container. Is this an issue > with > LXD or with snapd or am I missing something? > > Also: Why is there an "ugly" \x2d (which seems to be a hypen-minus) in > the systemd-file rather than a "normal" minus? > > Cheers, > Sujeevan > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From stephane.graber at canonical.com Tue Jun 7 16:36:31 2016 From: stephane.graber at canonical.com (=?iso-8859-1?Q?St=E9phane?= Graber) Date: Tue, 7 Jun 2016 12:36:31 -0400 Subject: Installing ubuntu-core snap inside a LXD container fails In-Reply-To: <5756F6CA.3000702@ubuntu.com> References: <76b68477-104c-3582-2f9b-3f69868fbb8b@ubuntu.com> <5756F6CA.3000702@ubuntu.com> Message-ID: <20160607163631.GW26506@castiana> On Wed, Jun 08, 2016 at 01:31:06AM +0900, Mark Shuttleworth wrote: > On 07/06/16 23:14, Claudio André wrote: > > 2016-06-07 9:37 GMT-03:00 Sujeevan (svij) > > Vijayakumaran >: > > > > > That's because snapd tickles the kernel in a way that blows[...] > > > > > > That is interesting. Where I can find more about the 'tickles'? My > > motivation: > > - I'm seeing closed source software behaves badly, e.g, AMD software > > prints. > > > > Internal Error: as failed > > Codegen phase failed compilation. > > > > If I know what the 'tickles' mean, I might be able to do something. > > Thanks. > > I think it has to do with nesting kernel structures related to security > (LXD sets those up, then snapd wants to tweak them again), but a much > clearer answer would come from Stephane cc'd. > > Mark Hi, Containers cannot mount squashfs file systems, or setup loop mounts or setup apparmor profiles. All of which are required to get a snap running. All of the above are blocked by the kernel as unsafe to use by an unprivileged user (which you are in a LXD container). My team is actively working on some kernel changes and some changes to snapd itself to allow this to work in the near future. Stéphane > > > > > > 2016-06-07 8:41 GMT-03:00 Sujeevan (svij) Vijayakumaran > > >: > > > > Hello! > > > > I'm trying to install a snap inside a freshly created LXD container, > > which isn't possible: > > > > ----- > > ubuntu at snap-test:~$ sudo snap install > > taskwarrior_2.5.1-snap0_amd64.snap > > 64.75 MB / 64.75 MB > > [========================================================================] > > 100.00 % 5.38 MB/s > > > > error: cannot perform the following tasks: > > - Mount snap "ubuntu-core" ([start snap-ubuntu\x2dcore-122.mount] > > failed > > with exit status 1: Job for snap-ubuntu\x2dcore-122.mount failed. See > > "systemctl status "snap-ubuntu\\x2dcore-122.mount"" and > > "journalctl -xe" > > for details. > > ) > > ubuntu at snap-test:~$ systemctl status "snap-ubuntu\\x2dcore-122.mount" > > ● snap-ubuntu\x2dcore-122.mount - Squashfs mount unit for ubuntu-core > > Loaded: loaded (/etc/systemd/system/snap-ubuntu\x2dcore-122.mount; > > enabled; vendor preset: enabled) > > Active: failed (Result: exit-code) since Tue 2016-06-07 > > 11:32:13 UTC; > > 16s ago > > Where: /snap/ubuntu-core/122 > > What: /var/lib/snapd/snaps/ubuntu-core_122.snap > > Process: 8584 ExecMount=/bin/mount > > /var/lib/snapd/snaps/ubuntu-core_122.snap /snap/ubuntu-core/122 > > (code=exited, status=32) > > > > Jun 07 11:32:13 snap-test systemd[1]: Mounting Squashfs mount unit for > > ubuntu-core... > > Jun 07 11:32:13 snap-test mount[8584]: mount: /snap/ubuntu-core/122: > > mount failed: Unknown error -1 > > Jun 07 11:32:13 snap-test systemd[1]: snap-ubuntu\x2dcore-122.mount: > > Mount process exited, code=exited status=32 > > Jun 07 11:32:13 snap-test systemd[1]: Failed to mount Squashfs mount > > unit for ubuntu-core. > > Jun 07 11:32:13 snap-test systemd[1]: snap-ubuntu\x2dcore-122.mount: > > Unit entered failed state. > > ------ > > > > It works fine if it's not inside a LXD container. Is this an issue > > with > > LXD or with snapd or am I missing something? > > > > Also: Why is there an "ugly" \x2d (which seems to be a hypen-minus) in > > the systemd-file rather than a "normal" minus? > > > > Cheers, > > Sujeevan > > > > -- > > Snapcraft mailing list > > Snapcraft at lists.ubuntu.com > > Modify settings or unsubscribe at: > > https://lists.ubuntu.com/mailman/listinfo/snapcraft > > > > > > > > > -- Stéphane Graber Ubuntu developer http://www.canonical.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: not available URL: From svij at ubuntu.com Tue Jun 7 16:42:13 2016 From: svij at ubuntu.com (Sujeevan Vijayakumaran) Date: Tue, 7 Jun 2016 18:42:13 +0200 Subject: Installing ubuntu-core snap inside a LXD container fails In-Reply-To: <20160607163631.GW26506@castiana> References: <76b68477-104c-3582-2f9b-3f69868fbb8b@ubuntu.com> <5756F6CA.3000702@ubuntu.com> <20160607163631.GW26506@castiana> Message-ID: <72b71791-b5ae-d903-d029-8eec6d253163@ubuntu.com> Great, thanks for the clarification! Am 07.06.2016 um 18:36 schrieb Stéphane Graber: > On Wed, Jun 08, 2016 at 01:31:06AM +0900, Mark Shuttleworth wrote: >> On 07/06/16 23:14, Claudio André wrote: >>> 2016-06-07 9:37 GMT-03:00 Sujeevan (svij) >>> Vijayakumaran >: >>> >>> > That's because snapd tickles the kernel in a way that blows[...] >>> >>> >>> That is interesting. Where I can find more about the 'tickles'? My >>> motivation: >>> - I'm seeing closed source software behaves badly, e.g, AMD software >>> prints. >>> >>> Internal Error: as failed >>> Codegen phase failed compilation. >>> >>> If I know what the 'tickles' mean, I might be able to do something. >>> Thanks. >> >> I think it has to do with nesting kernel structures related to security >> (LXD sets those up, then snapd wants to tweak them again), but a much >> clearer answer would come from Stephane cc'd. >> >> Mark > > Hi, > > Containers cannot mount squashfs file systems, or setup loop mounts or > setup apparmor profiles. All of which are required to get a snap > running. > > All of the above are blocked by the kernel as unsafe to use by an > unprivileged user (which you are in a LXD container). > > > My team is actively working on some kernel changes and some changes to > snapd itself to allow this to work in the near future. > > Stéphane > >> >> >>> >>> 2016-06-07 8:41 GMT-03:00 Sujeevan (svij) Vijayakumaran >>> >: >>> >>> Hello! >>> >>> I'm trying to install a snap inside a freshly created LXD container, >>> which isn't possible: >>> >>> ----- >>> ubuntu at snap-test:~$ sudo snap install >>> taskwarrior_2.5.1-snap0_amd64.snap >>> 64.75 MB / 64.75 MB >>> [========================================================================] >>> 100.00 % 5.38 MB/s >>> >>> error: cannot perform the following tasks: >>> - Mount snap "ubuntu-core" ([start snap-ubuntu\x2dcore-122.mount] >>> failed >>> with exit status 1: Job for snap-ubuntu\x2dcore-122.mount failed. See >>> "systemctl status "snap-ubuntu\\x2dcore-122.mount"" and >>> "journalctl -xe" >>> for details. >>> ) >>> ubuntu at snap-test:~$ systemctl status "snap-ubuntu\\x2dcore-122.mount" >>> ● snap-ubuntu\x2dcore-122.mount - Squashfs mount unit for ubuntu-core >>> Loaded: loaded (/etc/systemd/system/snap-ubuntu\x2dcore-122.mount; >>> enabled; vendor preset: enabled) >>> Active: failed (Result: exit-code) since Tue 2016-06-07 >>> 11:32:13 UTC; >>> 16s ago >>> Where: /snap/ubuntu-core/122 >>> What: /var/lib/snapd/snaps/ubuntu-core_122.snap >>> Process: 8584 ExecMount=/bin/mount >>> /var/lib/snapd/snaps/ubuntu-core_122.snap /snap/ubuntu-core/122 >>> (code=exited, status=32) >>> >>> Jun 07 11:32:13 snap-test systemd[1]: Mounting Squashfs mount unit for >>> ubuntu-core... >>> Jun 07 11:32:13 snap-test mount[8584]: mount: /snap/ubuntu-core/122: >>> mount failed: Unknown error -1 >>> Jun 07 11:32:13 snap-test systemd[1]: snap-ubuntu\x2dcore-122.mount: >>> Mount process exited, code=exited status=32 >>> Jun 07 11:32:13 snap-test systemd[1]: Failed to mount Squashfs mount >>> unit for ubuntu-core. >>> Jun 07 11:32:13 snap-test systemd[1]: snap-ubuntu\x2dcore-122.mount: >>> Unit entered failed state. >>> ------ >>> >>> It works fine if it's not inside a LXD container. Is this an issue >>> with >>> LXD or with snapd or am I missing something? >>> >>> Also: Why is there an "ugly" \x2d (which seems to be a hypen-minus) in >>> the systemd-file rather than a "normal" minus? >>> >>> Cheers, >>> Sujeevan >>> >>> -- >>> Snapcraft mailing list >>> Snapcraft at lists.ubuntu.com >>> Modify settings or unsubscribe at: >>> https://lists.ubuntu.com/mailman/listinfo/snapcraft >>> >>> >>> >>> >> > > > From jamie at canonical.com Tue Jun 7 17:22:27 2016 From: jamie at canonical.com (Jamie Strandboge) Date: Tue, 07 Jun 2016 12:22:27 -0500 Subject: Debugging `confinement: strict` issues In-Reply-To: References: Message-ID: <1465320147.3393.95.camel@canonical.com> On Tue, 2016-06-07 at 22:56 +0800, Reinhard Pointner wrote: > Hi all, > > Does anyone have any hints on how to debug issues that only happen in > non-devmode? Like logging when certain "interfaces" are used? > > I'm working on snapping a Java application, and I've got `plugs: [home, > network]` in my snapcraft.yaml but the app can't write any files in the > user home (home plug should give the app write access right? also, what > about network shares mounted by the user?). Also, when making http requests > that thread will just deadlock (no exceptions or and timeouts not > respected, so i suspect lower level network api issues). > In addition to what Zygmunt said elsewhere in the thread, there is a snap in the store called 'snappy-debug' that will get better over time but is already useful. $ sudo snap install snappy-debug --devmode # after 2.0.7 lands, drop --devmode $ sudo snap connect snappy-debug:log-observe ubuntu-core:log-observe $ sudo /snap/bin/snappy-debug.security scanlog This scanlog command will tail the syslog, resolving seccomp syscall numbers and make suggestions on what interfaces to use, changes to make to your snap to work within the sandbox, etc. Can filter by snap with: $ sudo /snap/bin/snappy-debug.security scanlog -- Jamie Strandboge | http://www.canonical.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: This is a digitally signed message part URL: From victor.palau at canonical.com Tue Jun 7 19:03:20 2016 From: victor.palau at canonical.com (Victor Palau) Date: Tue, 7 Jun 2016 20:03:20 +0100 Subject: Man-pages in snaps In-Reply-To: <062e78c4-151f-a53e-f807-38c7c8c18963@ubuntu.com> References: <062e78c4-151f-a53e-f807-38c7c8c18963@ubuntu.com> Message-ID: Hi, could you not include man in your snap too? Thanks, Victor On Tue, Jun 7, 2016 at 1:49 PM, Sujeevan (svij) Vijayakumaran < svij at ubuntu.com> wrote: > Hello, > > I started creating a snap for the CLI task manager taskwarrior > (https://taskwarrior.org/). This project has really good man pages. > Anyway there is currently no support to access man-pages from snaps: > https://bugs.launchpad.net/snappy/+bug/1575593 > > Since it doesn't make much sense to publish an CLI app without the good > man pages: Is there any other ways to make it possible that the user can > access the man pages of a snap? The tools "task help" command doesnt > help here since is doesn’t show everything. > > Cheers, > Sujeevan > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > -------------- next part -------------- An HTML attachment was scrubbed... URL: From aps337 at gmail.com Tue Jun 7 19:30:22 2016 From: aps337 at gmail.com (Adam) Date: Tue, 7 Jun 2016 15:30:22 -0400 Subject: Snapcraft as post-build operation? Message-ID: Hi All, Pretty excited about the concept of snaps. I have an application I want to port over, but I'm a little confused about where to start. Seems like part of the concept is to integrate the build machinery for a project and make it a "slave" to Snapcraft. Unfortunately the application I want to port over, a database backed web server based application has some pretty complex build machinery that's evolved over many years. I can't help but think that it makes more sense, for my application at least, to think of Snapcraft like an application installer build that happens post-build. None of the examples seem to fit that model though. Am I trying to put a square peg in a round hole? If what I am doing seems feasible, where do I start? Can I setup the yaml file to just cp files over from my output directory and then build up the contents of my snap that way? Thanks for any insight and help you can provide. From mark at ubuntu.com Tue Jun 7 19:35:54 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Wed, 8 Jun 2016 04:35:54 +0900 Subject: Snapcraft as post-build operation? In-Reply-To: References: Message-ID: <5757221A.1090800@ubuntu.com> On 08/06/16 04:30, Adam wrote: > Am I trying to put a square peg in a round hole? If what I am doing > seems feasible, where do I start? Can I setup the yaml file to just cp > files over from my output directory and then build up the contents of > my snap that way? Yes, you can definitely do that. The simplest way to think of a snap is as a directory with ALL your app files and dependencies underneath it. There's also a set of directories you can write to. The rest of the system is read-only to your app, so the most tricky piece will be teaching your app only to write where it is allowed to write, and to look for dependencies underneath its own directory. Snapcraft does its best to juggle things for apps following common conventions so that all of this magically happens without changing the source code, but it does have its limits particularly for ornery in-house codebases. We have always managed to *make* a snap under your circumstances, it just sometimes requires a little twisting of the bits. Is that enough guidance for you to get started? Mark From janssen at parc.com Wed Jun 8 01:46:08 2016 From: janssen at parc.com (Bill Janssen) Date: Tue, 07 Jun 2016 18:46:08 -0700 Subject: errant escape codes sent to stdout Message-ID: <25062.1465350368@parc.com> When I run "snapcraft init" in an Emacs shell, I get this: ~$ snapcraft init Created snapcraft.yaml. ~$ TERM is "dumb". Seems sub-optimal and hard to read. Some way to turn this off? Interestingly, "snapcraft --help" doesn't do this. Bill From leo.arias at canonical.com Wed Jun 8 02:03:57 2016 From: leo.arias at canonical.com (Leo Arias) Date: Tue, 7 Jun 2016 20:03:57 -0600 Subject: errant escape codes sent to stdout In-Reply-To: <25062.1465350368@parc.com> References: <25062.1465350368@parc.com> Message-ID: <57577D0D.6030003@canonical.com> Hello! On 2016-06-07 19:46, Bill Janssen wrote: > When I run "snapcraft init" in an Emacs shell, I get this: > > ~$ snapcraft init > Created snapcraft.yaml. > ~$ > > TERM is "dumb". > > Seems sub-optimal and hard to read. Some way to turn this off? > > Interestingly, "snapcraft --help" doesn't do this. Those are the colors we display on step messages. We should remove the fancy stuff if the terminal doesn't support it. Can you please report a bug in https://bugs.launchpad.net/snapcraft ? Thank you! -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: OpenPGP digital signature URL: From fcole90 at gmail.com Wed Jun 8 04:07:54 2016 From: fcole90 at gmail.com (Fabio Colella) Date: Wed, 08 Jun 2016 04:07:54 +0000 Subject: Snapcraft as post-build operation? In-Reply-To: <5757221A.1090800@ubuntu.com> References: <5757221A.1090800@ubuntu.com> Message-ID: An in addition, if you are really not able to use snapcraft for building your project, you could prepare your binaries and then use the copy plugin to put them inside the snap together with some script to handle environment variables and/or architecture Cheers, Fabio On Tue, 7 Jun 2016, 21:37 Mark Shuttleworth, wrote: > On 08/06/16 04:30, Adam wrote: > > Am I trying to put a square peg in a round hole? If what I am doing > > seems feasible, where do I start? Can I setup the yaml file to just cp > > files over from my output directory and then build up the contents of > > my snap that way? > > Yes, you can definitely do that. > > The simplest way to think of a snap is as a directory with ALL your app > files and dependencies underneath it. > > There's also a set of directories you can write to. > > The rest of the system is read-only to your app, so the most tricky > piece will be teaching your app only to write where it is allowed to > write, and to look for dependencies underneath its own directory. > Snapcraft does its best to juggle things for apps following common > conventions so that all of this magically happens without changing the > source code, but it does have its limits particularly for ornery > in-house codebases. > > We have always managed to *make* a snap under your circumstances, it > just sometimes requires a little twisting of the bits. > > Is that enough guidance for you to get started? > > Mark > > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > -------------- next part -------------- An HTML attachment was scrubbed... URL: From didrocks at ubuntu.com Wed Jun 8 07:12:08 2016 From: didrocks at ubuntu.com (Didier Roche) Date: Wed, 8 Jun 2016 09:12:08 +0200 Subject: Man-pages in snaps In-Reply-To: <062e78c4-151f-a53e-f807-38c7c8c18963@ubuntu.com> References: <062e78c4-151f-a53e-f807-38c7c8c18963@ubuntu.com> Message-ID: <5757C548.3050204@ubuntu.com> Le 07/06/2016 14:49, Sujeevan (svij) Vijayakumaran a écrit : > Hello, > > I started creating a snap for the CLI task manager taskwarrior > (https://taskwarrior.org/). This project has really good man pages. > Anyway there is currently no support to access man-pages from snaps: > https://bugs.launchpad.net/snappy/+bug/1575593 > > Since it doesn't make much sense to publish an CLI app without the good > man pages: Is there any other ways to make it possible that the user can > access the man pages of a snap? The tools "task help" command doesnt > help here since is doesn’t show everything. Hey Sujeevan, The desktop team indeed opened this bug, and I'm sure the snappy team will respond to it quickly. For now, you can't really workaround that (well, telling people to man /snaps/snap_name/current/… isn't an option ;)) However, thinking about this, you may be able to add a "man" command in between (so that autocompletion show it) in your snapcraft.yaml, which is executing "man /snaps/snap_name/current/…" as a workaround. That way you will have (if your snap_name matches your command name): $ taskwarrior $ taskwarrior.man Does this fit the bill until we get proper manpage support in snapd? Cheers, Didier From simon.fels at canonical.com Wed Jun 8 08:46:04 2016 From: simon.fels at canonical.com (Simon Fels) Date: Wed, 8 Jun 2016 10:46:04 +0200 Subject: Auto-connection of slots/plugs for preinstalled snaps Message-ID: <5757DB4C.9020203@canonical.com> Hey everyone, with putting several snaps into place (network-manager, modem-manager, ...) we're now moving towards getting them preinstalled on customer devices. The missing bit for this is that something has to connect all the plugs/slots either at pre installation or first-boot time. What is the current plan to solve this? regards, Simon From ogra at ubuntu.com Wed Jun 8 08:59:28 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Wed, 08 Jun 2016 10:59:28 +0200 Subject: Man-pages in snaps In-Reply-To: <5757C548.3050204@ubuntu.com> References: <062e78c4-151f-a53e-f807-38c7c8c18963@ubuntu.com> <5757C548.3050204@ubuntu.com> Message-ID: <1465376368.17816.20.camel@anubis> hi, Am Mittwoch, den 08.06.2016, 09:12 +0200 schrieb Didier Roche: > Le 07/06/2016 14:49, Sujeevan (svij) Vijayakumaran a écrit : > > Hello, > > > > I started creating a snap for the CLI task manager taskwarrior > > (https://taskwarrior.org/). This project has really good man pages. > > Anyway there is currently no support to access man-pages from snaps: > > https://bugs.launchpad.net/snappy/+bug/1575593 > > > > Since it doesn't make much sense to publish an CLI app without the good > > man pages: Is there any other ways to make it possible that the user can > > access the man pages of a snap? The tools "task help" command doesnt > > help here since is doesn’t show everything. > > Hey Sujeevan, > > The desktop team indeed opened this bug, and I'm sure the snappy team > will respond to it quickly. > For now, you can't really workaround that (well, telling people to man > /snaps/snap_name/current/… isn't an option ;)) > > However, thinking about this, you may be able to add a "man" command in > between (so that autocompletion show it) in your snapcraft.yaml, which > is executing "man /snaps/snap_name/current/…" as a workaround. > That way you will have (if your snap_name matches your command name): > > $ taskwarrior > $ taskwarrior.man > > Does this fit the bill until we get proper manpage support in snapd? there is no man command installed in in the core snap. you'd rather ship the man binary (and it's deps) inside your snap as well and make your taskwarrior.man command use it. ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From gustavo.niemeyer at canonical.com Wed Jun 8 09:21:50 2016 From: gustavo.niemeyer at canonical.com (Gustavo Niemeyer) Date: Wed, 8 Jun 2016 06:21:50 -0300 Subject: Man-pages in snaps In-Reply-To: <1465376368.17816.20.camel@anubis> References: <062e78c4-151f-a53e-f807-38c7c8c18963@ubuntu.com> <5757C548.3050204@ubuntu.com> <1465376368.17816.20.camel@anubis> Message-ID: Yeah, that's indeed not great. We need to teach snapd to prepare a directory for mans and play with /etc/manpath.config. On Wed, Jun 8, 2016 at 5:59 AM, Oliver Grawert wrote: > hi, > Am Mittwoch, den 08.06.2016, 09:12 +0200 schrieb Didier Roche: > > Le 07/06/2016 14:49, Sujeevan (svij) Vijayakumaran a écrit : > > > Hello, > > > > > > I started creating a snap for the CLI task manager taskwarrior > > > (https://taskwarrior.org/). This project has really good man pages. > > > Anyway there is currently no support to access man-pages from snaps: > > > https://bugs.launchpad.net/snappy/+bug/1575593 > > > > > > Since it doesn't make much sense to publish an CLI app without the good > > > man pages: Is there any other ways to make it possible that the user > can > > > access the man pages of a snap? The tools "task help" command doesnt > > > help here since is doesn’t show everything. > > > > Hey Sujeevan, > > > > The desktop team indeed opened this bug, and I'm sure the snappy team > > will respond to it quickly. > > For now, you can't really workaround that (well, telling people to man > > /snaps/snap_name/current/… isn't an option ;)) > > > > However, thinking about this, you may be able to add a "man" command in > > between (so that autocompletion show it) in your snapcraft.yaml, which > > is executing "man /snaps/snap_name/current/…" as a workaround. > > That way you will have (if your snap_name matches your command name): > > > > $ taskwarrior > > $ taskwarrior.man > > > > Does this fit the bill until we get proper manpage support in snapd? > > there is no man command installed in in the core snap. > you'd rather ship the man binary (and it's deps) inside your snap as > well and make your taskwarrior.man command use it. > > ciao > oli > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > > -- gustavo @ http://niemeyer.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From mark at ubuntu.com Wed Jun 8 09:55:59 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Wed, 8 Jun 2016 18:55:59 +0900 Subject: Great release of snapcraft 2.10 Message-ID: <5757EBAF.2020604@ubuntu.com> Just to say it was a pleasure watching through the bug system as fixes were made and verified in the 2.10 release of snapcraft. Appreciate the pace and the attention to detail! Mark -------------- next part -------------- An HTML attachment was scrubbed... URL: From john.lenton at canonical.com Wed Jun 8 10:18:37 2016 From: john.lenton at canonical.com (John Lenton) Date: Wed, 8 Jun 2016 11:18:37 +0100 Subject: Installing ubuntu-core snap inside a LXD container fails In-Reply-To: <76b68477-104c-3582-2f9b-3f69868fbb8b@ubuntu.com> References: <76b68477-104c-3582-2f9b-3f69868fbb8b@ubuntu.com> Message-ID: On 7 June 2016 at 12:41, Sujeevan (svij) Vijayakumaran wrote: > Also: Why is there an "ugly" \x2d (which seems to be a hypen-minus) in > the systemd-file rather than a "normal" minus? that's what systemd-escape outputs as the filename for a mountpoint that contains U+002D HYPHEN-MINUS, which is your regular - as present in for example "ubuntu-core". From john.lenton at canonical.com Wed Jun 8 10:58:45 2016 From: john.lenton at canonical.com (John Lenton) Date: Wed, 8 Jun 2016 11:58:45 +0100 Subject: errant escape codes sent to stdout In-Reply-To: <57577D0D.6030003@canonical.com> References: <25062.1465350368@parc.com> <57577D0D.6030003@canonical.com> Message-ID: FWIW, "tput colors" might be what you want to do to check (unless you want to talk to terminfo from python itself). In gnome-terminal: $ tput colors 256 linux console: $ tput colors 8 emacs shell: $ tput colors -1 emacs term: $ tput colors 8 Bill, as a workaround instead of M-x shell, do M-x term. On 8 June 2016 at 03:03, Leo Arias wrote: > Hello! > > On 2016-06-07 19:46, Bill Janssen wrote: >> When I run "snapcraft init" in an Emacs shell, I get this: >> >> ~$ snapcraft init >> [0;32mCreated snapcraft.yaml. [0m >> ~$ >> >> TERM is "dumb". >> >> Seems sub-optimal and hard to read. Some way to turn this off? >> >> Interestingly, "snapcraft --help" doesn't do this. > > Those are the colors we display on step messages. We should remove the > fancy stuff if the terminal doesn't support it. > Can you please report a bug in https://bugs.launchpad.net/snapcraft ? > > Thank you! > > > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft > From mark at ubuntu.com Wed Jun 8 11:11:07 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Wed, 8 Jun 2016 20:11:07 +0900 Subject: Auto-connection of slots/plugs for preinstalled snaps In-Reply-To: <5757DB4C.9020203@canonical.com> References: <5757DB4C.9020203@canonical.com> Message-ID: <5757FD4B.6040509@ubuntu.com> Hi Simon We need two "at-boot" stories. One is a first-boot story, things like filesystem expansion and the device instantiation aspects of cloud-init (user creation for example). The other is a console experience, so before there are users one can do things like network config. Steve, Loic and Ryan cc'd are the folks to touch base with on these fronts. Mark On 08/06/16 17:46, Simon Fels wrote: > Hey everyone, > > with putting several snaps into place (network-manager, modem-manager, > ...) we're now moving towards getting them preinstalled on customer > devices. The missing bit for this is that something has to connect all > the plugs/slots either at pre installation or first-boot time. > > What is the current plan to solve this? > > regards, > Simon > From gustavo.niemeyer at canonical.com Wed Jun 8 11:30:59 2016 From: gustavo.niemeyer at canonical.com (Gustavo Niemeyer) Date: Wed, 8 Jun 2016 08:30:59 -0300 Subject: Auto-connection of slots/plugs for preinstalled snaps In-Reply-To: <5757FD4B.6040509@ubuntu.com> References: <5757DB4C.9020203@canonical.com> <5757FD4B.6040509@ubuntu.com> Message-ID: Simon, we'll enable you to specify in a declarative way which interfaces you want connected and what attributes they should have the first time the device boots. This will override auto-connection completely, since it's the device itself telling what it wants. On Wed, Jun 8, 2016 at 8:11 AM, Mark Shuttleworth wrote: > Hi Simon > > We need two "at-boot" stories. > > One is a first-boot story, things like filesystem expansion and the > device instantiation aspects of cloud-init (user creation for example). > > The other is a console experience, so before there are users one can do > things like network config. > > Steve, Loic and Ryan cc'd are the folks to touch base with on these fronts. > > Mark > > On 08/06/16 17:46, Simon Fels wrote: > > Hey everyone, > > > > with putting several snaps into place (network-manager, modem-manager, > > ...) we're now moving towards getting them preinstalled on customer > > devices. The missing bit for this is that something has to connect all > > the plugs/slots either at pre installation or first-boot time. > > > > What is the current plan to solve this? > > > > regards, > > Simon > > > > > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > -- gustavo @ http://niemeyer.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From simon.fels at canonical.com Wed Jun 8 11:40:44 2016 From: simon.fels at canonical.com (Simon Fels) Date: Wed, 8 Jun 2016 13:40:44 +0200 Subject: Auto-connection of slots/plugs for preinstalled snaps In-Reply-To: References: <5757DB4C.9020203@canonical.com> <5757FD4B.6040509@ubuntu.com> Message-ID: <5758043C.2040807@canonical.com> On 08.06.2016 13:30, Gustavo Niemeyer wrote: > > Simon, we'll enable you to specify in a declarative way which interfaces you > want connected and what attributes they should have the first time the device boots. > > This will override auto-connection completely, since it's the device itself > telling what it wants. > > On Wed, Jun 8, 2016 at 8:11 AM, Mark Shuttleworth > wrote: > > Hi Simon > > We need two "at-boot" stories. > > One is a first-boot story, things like filesystem expansion and the > device instantiation aspects of cloud-init (user creation for example). > > The other is a console experience, so before there are users one can do > things like network config. > > Steve, Loic and Ryan cc'd are the folks to touch base with on these fronts. Thanks guys. Any time line for when this will be ready? regards, Simon > Mark > > On 08/06/16 17:46, Simon Fels wrote: > > Hey everyone, > > > > with putting several snaps into place (network-manager, modem-manager, > > ...) we're now moving towards getting them preinstalled on customer > > devices. The missing bit for this is that something has to connect all > > the plugs/slots either at pre installation or first-boot time. > > > > What is the current plan to solve this? > > > > regards, > > Simon > > > > > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > > > > > -- > gustavo @ http://niemeyer.net > From gustavo.niemeyer at canonical.com Wed Jun 8 11:52:24 2016 From: gustavo.niemeyer at canonical.com (Gustavo Niemeyer) Date: Wed, 8 Jun 2016 08:52:24 -0300 Subject: Auto-connection of slots/plugs for preinstalled snaps In-Reply-To: <5758043C.2040807@canonical.com> References: <5757DB4C.9020203@canonical.com> <5757FD4B.6040509@ubuntu.com> <5758043C.2040807@canonical.com> Message-ID: We're not too far, and the feature itself is not a lot of work, but probably takes a few more weeks as we churn through the pre-requirements. This is related to the configuration story (depends on hooks whch is in progress), new gadget specification, and new first-boot declarative conventions. On Wed, Jun 8, 2016 at 8:40 AM, Simon Fels wrote: > On 08.06.2016 13:30, Gustavo Niemeyer wrote: > > > > Simon, we'll enable you to specify in a declarative way which interfaces > you > > want connected and what attributes they should have the first time the > device boots. > > > > This will override auto-connection completely, since it's the device > itself > > telling what it wants. > > > > On Wed, Jun 8, 2016 at 8:11 AM, Mark Shuttleworth > > wrote: > > > > Hi Simon > > > > We need two "at-boot" stories. > > > > One is a first-boot story, things like filesystem expansion and the > > device instantiation aspects of cloud-init (user creation for > example). > > > > The other is a console experience, so before there are users one can > do > > things like network config. > > > > Steve, Loic and Ryan cc'd are the folks to touch base with on these > fronts. > > Thanks guys. Any time line for when this will be ready? > > regards, > Simon > > > Mark > > > > On 08/06/16 17:46, Simon Fels wrote: > > > Hey everyone, > > > > > > with putting several snaps into place (network-manager, > modem-manager, > > > ...) we're now moving towards getting them preinstalled on > customer > > > devices. The missing bit for this is that something has to > connect all > > > the plugs/slots either at pre installation or first-boot time. > > > > > > What is the current plan to solve this? > > > > > > regards, > > > Simon > > > > > > > > > > > -- > > Snapcraft mailing list > > Snapcraft at lists.ubuntu.com > > Modify settings or unsubscribe at: > > https://lists.ubuntu.com/mailman/listinfo/snapcraft > > > > > > > > > > -- > > gustavo @ http://niemeyer.net > > > > -- gustavo @ http://niemeyer.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From woodrow.shen at canonical.com Wed Jun 8 14:31:30 2016 From: woodrow.shen at canonical.com (Woodrow Shen) Date: Wed, 8 Jun 2016 22:31:30 +0800 Subject: snappy: Command not found [solved] In-Reply-To: <2016060721353741830828@lemaker.com> References: <2016060619025454498725@lemaker.com> <1465222345.8064.5.camel@ubuntu.com> <2016060709511982709523@lemaker.com> <1465291061.17816.15.camel@anubis> <2016060721353741830828@lemaker.com> Message-ID: Hi Peter, Thanks your source code from github, and today I also verified the porting ( https://github.com/xapp-le/SnappyUbuntuCore) of ubuntu-core 16 for roseapple-pi. Now it can work on ubuntu-core 16 successfully. :) Woodrow On Tue, Jun 7, 2016 at 9:35 PM, peter.chen at lemaker.com < peter.chen at lemaker.com> wrote: > Hi, All > Thanks for your help, The problem which the uboot can't load the kernel > has been fixed, and the snappy ubuntu 16 is ported to the Lemaker Guitar > successfully now. > The problem mainly related to uboot and i have committed the patch to our > github at > https://github.com/LeMaker/u-boot-actions/commit/1a2f52c79652bd33002a6aee4e318db2061e4879 > , > Hope this can help others. > > Best Regards > ------------------------------ > > *Peter Chen* > > > *LeMaker **Developer Team* > > > > *Making Innovation Easy. * > > Email: > support at lemaker.org (Technical Support) > product at lemaker.org (Product Distribution) > > Website: http://www.lemaker.org/ > > > > > > > > > > *From:* Oliver Grawert > *Date:* 2016-06-07 17:17 > *To:* Woodrow Shen > *CC:* peter.chen at lemaker.com; snapcraft > *Subject:* Re: Re: snappy: Command not found > hi, > Am Dienstag, den 07.06.2016, 11:17 +0800 schrieb Woodrow Shen: > > > > boot_fix.bin is workaround to make booting normally without active > > boot flag. > > I suspect above uboot error messages is missing the value > > of ${snappy_kernel} when uboot needs to load the files, like: > > > loadkernel=fatload mmc ${mmcdev}:${mmcpart} ${kernel_addr} > > ${snappy_kernel}/${kernel_file} > > exactly, u-d-f creates an image and then installs the kernel snap > inside. after it installed it knows the exact path to the binaries, > which is /snap/$packagename/$revision and this path is set to > snappy_kernel= inside the uboot.env file. every time you upgrade or > roll-back the kernel this variable is adjusted alongside, if a boot > fails the value automatically falls back to the last revision that > worked and it reboots. > > ciao > oli > > -- Woodrow Shen Software Engineer, Canonical ltd. UES | CE | PC & Core, Taipei -------------- next part -------------- An HTML attachment was scrubbed... URL: From daniel.holbach at canonical.com Wed Jun 8 15:15:38 2016 From: daniel.holbach at canonical.com (Daniel Holbach) Date: Wed, 8 Jun 2016 17:15:38 +0200 Subject: Blog post: Snappy Playpen Kickoff Highlights Message-ID: <5758369A.20801@canonical.com> Hello everybody, David Callé wrote up a nice report of what happened at the Snappy Playpen Kickoff yesterday. We are all still busy working on snaps and collaborating together. Check out the full post here: https://developer.ubuntu.com/en/blog/2016/06/08/snappy-playpen-kickoff-highlights/ Have a great day, Daniel From janssen at parc.com Wed Jun 8 16:19:09 2016 From: janssen at parc.com (Bill Janssen) Date: Wed, 08 Jun 2016 09:19:09 -0700 Subject: errant escape codes sent to stdout In-Reply-To: References: <25062.1465350368@parc.com> <57577D0D.6030003@canonical.com> Message-ID: <12825.1465402749@parc.com> John Lenton wrote: > FWIW, "tput colors" might be what you want to do to check (unless you > want to talk to terminfo from python itself). > In gnome-terminal: > $ tput colors > 256 > linux console: > $ tput colors > 8 > emacs shell: > $ tput colors > -1 > emacs term: > $ tput colors > 8 > > Bill, as a workaround instead of M-x shell, do M-x term. Thanks for the info, and the suggestion, John, but I do many other things with the Emacs shell, besides snapcraft. I can't switch to Emacs term just to fix this bit of clumsiness. I'll see if I can file a bug report, though. Bill > > On 8 June 2016 at 03:03, Leo Arias wrote: > > Hello! > > > > On 2016-06-07 19:46, Bill Janssen wrote: > >> When I run "snapcraft init" in an Emacs shell, I get this: > >> > >> ~$ snapcraft init > >> [0;32mCreated snapcraft.yaml. [0m > >> ~$ > >> > >> TERM is "dumb". > >> > >> Seems sub-optimal and hard to read. Some way to turn this off? > >> > >> Interestingly, "snapcraft --help" doesn't do this. > > > > Those are the colors we display on step messages. We should remove the > > fancy stuff if the terminal doesn't support it. > > Can you please report a bug in https://bugs.launchpad.net/snapcraft ? > > > > Thank you! > > > > > > > > -- > > Snapcraft mailing list > > Snapcraft at lists.ubuntu.com > > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft > > > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft From mark at ubuntu.com Wed Jun 8 19:41:17 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Thu, 9 Jun 2016 04:41:17 +0900 Subject: snappy: Command not found [solved] In-Reply-To: References: <2016060619025454498725@lemaker.com> <1465222345.8064.5.camel@ubuntu.com> <2016060709511982709523@lemaker.com> <1465291061.17816.15.camel@anubis> <2016060721353741830828@lemaker.com> Message-ID: <575874DD.20609@ubuntu.com> On 08/06/16 23:31, Woodrow Shen wrote: > Thanks your source code from github, and today I also verified the > porting (https://github.com/xapp-le/SnappyUbuntuCore) of ubuntu-core > 16 for roseapple-pi. Now it can work on ubuntu-core 16 successfully. :) Let us know how it works, Peter! Mark From jamie at canonical.com Wed Jun 8 19:51:51 2016 From: jamie at canonical.com (Jamie Strandboge) Date: Wed, 08 Jun 2016 14:51:51 -0500 Subject: Man-pages in snaps In-Reply-To: References: <062e78c4-151f-a53e-f807-38c7c8c18963@ubuntu.com> <5757C548.3050204@ubuntu.com> <1465376368.17816.20.camel@anubis> Message-ID: <1465415511.3074.13.camel@canonical.com> On Wed, 2016-06-08 at 06:21 -0300, Gustavo Niemeyer wrote: > Yeah, that's indeed not great. > > We need to teach snapd to prepare a directory for mans and play with > /etc/manpath.config. > Once man is in the os snap, this would be comfortable for users indeed. I do think it worth noting that this would allow untrusted input (the man page) to be fed into an unconfined process ('/usr/bin/man' from the user's shell), but I think this design idea is fine if the review tools verify the man page (just like we do with desktop files). If there is a bug for this, please add a review tools task and I'll add a check. > On Wed, Jun 8, 2016 at 5:59 AM, Oliver Grawert wrote: > > > > > hi, > > Am Mittwoch, den 08.06.2016, 09:12 +0200 schrieb Didier Roche: > > > > > > Le 07/06/2016 14:49, Sujeevan (svij) Vijayakumaran a écrit : > > > > > > > > Hello, > > > > > > > > I started creating a snap for the CLI task manager taskwarrior > > > > (https://taskwarrior.org/). This project has really good man pages. > > > > Anyway there is currently no support to access man-pages from snaps: > > > > https://bugs.launchpad.net/snappy/+bug/1575593 > > > > > > > > Since it doesn't make much sense to publish an CLI app without the good > > > > man pages: Is there any other ways to make it possible that the user > > can > > > > > > > > > > > access the man pages of a snap? The tools "task help" command doesnt > > > > help here since is doesn’t show everything. > > > Hey Sujeevan, > > > > > > The desktop team indeed opened this bug, and I'm sure the snappy team > > > will respond to it quickly. > > > For now, you can't really workaround that (well, telling people to man > > > /snaps/snap_name/current/… isn't an option ;)) > > > > > > However, thinking about this, you may be able to add a "man" command in > > > between (so that autocompletion show it) in your snapcraft.yaml, which > > > is executing "man /snaps/snap_name/current/…" as a workaround. > > > That way you will have (if your snap_name matches your command name): > > > > > > $ taskwarrior > > > $ taskwarrior.man > > > > > > Does this fit the bill until we get proper manpage support in snapd? > > there is no man command installed in in the core snap. > > you'd rather ship the man binary (and it's deps) inside your snap as > > well and make your taskwarrior.man command use it. > > > > ciao > >         oli > > > > -- > > Snapcraft mailing list > > Snapcraft at lists.ubuntu.com > > Modify settings or unsubscribe at: > > https://lists.ubuntu.com/mailman/listinfo/snapcraft > > > > > > --  > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/s > napcraft -- Jamie Strandboge | http://www.canonical.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: This is a digitally signed message part URL: From mark at ubuntu.com Wed Jun 8 20:29:16 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Thu, 9 Jun 2016 05:29:16 +0900 Subject: Man-pages in snaps In-Reply-To: <1465415511.3074.13.camel@canonical.com> References: <062e78c4-151f-a53e-f807-38c7c8c18963@ubuntu.com> <5757C548.3050204@ubuntu.com> <1465376368.17816.20.camel@anubis> <1465415511.3074.13.camel@canonical.com> Message-ID: <5758801C.5020903@ubuntu.com> On 09/06/16 04:51, Jamie Strandboge wrote: > Once man is in the os snap, this would be comfortable for users indeed. > > I do think it worth noting that this would allow untrusted input (the man page) > to be fed into an unconfined process ('/usr/bin/man' from the user's shell) Why not require "snap install man" ? Mark -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: OpenPGP digital signature URL: From ogra at ubuntu.com Wed Jun 8 20:32:24 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Wed, 08 Jun 2016 22:32:24 +0200 Subject: Man-pages in snaps In-Reply-To: <5758801C.5020903@ubuntu.com> References: <062e78c4-151f-a53e-f807-38c7c8c18963@ubuntu.com> <5757C548.3050204@ubuntu.com> <1465376368.17816.20.camel@anubis> <1465415511.3074.13.camel@canonical.com> <5758801C.5020903@ubuntu.com> Message-ID: <1465417944.7154.6.camel@ubuntu.com> hi, Am Donnerstag, den 09.06.2016, 05:29 +0900 schrieb Mark Shuttleworth: > On 09/06/16 04:51, Jamie Strandboge wrote: > > > > Once man is in the os snap, this would be comfortable for users > > indeed. > > > > I do think it worth noting that this would allow untrusted input > > (the man page) > > to be fed into an unconfined process ('/usr/bin/man' from the > > user's shell) > Why not require "snap install man" ? that would be much better than wasting space for all the deps that man pulls into the core snap. (we currently explicitly remove all documentation and its tools from the snap, only leaving the copyright files in /usr/share/doc) ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From jamie at canonical.com Wed Jun 8 20:39:26 2016 From: jamie at canonical.com (Jamie Strandboge) Date: Wed, 08 Jun 2016 15:39:26 -0500 Subject: Man-pages in snaps In-Reply-To: <5758801C.5020903@ubuntu.com> References: <062e78c4-151f-a53e-f807-38c7c8c18963@ubuntu.com> <5757C548.3050204@ubuntu.com> <1465376368.17816.20.camel@anubis> <1465415511.3074.13.camel@canonical.com> <5758801C.5020903@ubuntu.com> Message-ID: <1465418366.3074.30.camel@canonical.com> On Thu, 2016-06-09 at 05:29 +0900, Mark Shuttleworth wrote: > On 09/06/16 04:51, Jamie Strandboge wrote: > > > > Once man is in the os snap, this would be comfortable for users indeed. > > > > I do think it worth noting that this would allow untrusted input (the man > > page) > > to be fed into an unconfined process ('/usr/bin/man' from the user's shell) > Why not require "snap install man" ? > This is indeed possible and better since man then runs under confinement. We take Gustavo's idea of creating a man page directory that snapd puts the man pages in, then we add a 'man-pages' interface to allow reading that directory. The 'man' snap 'plugs: [ man-pages ]'. What is also nice about this approach is that we can have man-page-only snaps such as those that were stripped out of the os snap, manpages, manpages-dev, translated man pages, man pages for snappy itself, etc. -- Jamie Strandboge | http://www.canonical.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: This is a digitally signed message part URL: From sergio.schvezov at canonical.com Thu Jun 9 02:04:22 2016 From: sergio.schvezov at canonical.com (Sergio Schvezov) Date: Wed, 08 Jun 2016 23:04:22 -0300 Subject: ANN: snapcraft 2.10 is now available In-Reply-To: <5758B759.8030403@ubuntu.com> Message-ID: <08f50689-5787-4e50-a3b6-5a796f528b8a@ubuntu.com> Hello snapcrafters! We are pleased to announce the release of version 2.10 of snapcraft: https://launchpad.net/snapcraft/+milestone/2.10 We are introducing many new features in this release: - Migrating to macaroons support for authentication. This will basically require people to relogin to upload, this move allows us to build more features to talk to the store APIs for a better developer experience. - `snapcraft init` now produces a valid template that you can just run against. - Support for zip files as a source type which most plugins can consume. - We renamed `strip` to `prime`. The former wasn't the best name for a lifecycle step. Use of `strip` will print deprecation warnings. - Initial backend support to work on the parts ecosystem (more on this later, this is just a teaser). - An assumes field which will be used by snapd the assert certain features are supported by the system for that snap to work properly. Many more ongoing tasks and bugs have been fixed, mostly polish related to outputs and error messages; by the way, bugs are always welcome when an error message does not make sense. There have been tiny but nice improvements to the go and nodejs plugins as well. Refer to the milestone page for the complete list. To consume the latest snapcraft release on Xenial Xerus (16.04), we suggest you install the snapcraft package from the Ubuntu Archives: sudo apt update sudo apt install snapcraft To get the source for this release check it out at https://github.com/ubuntu-core/snapcraft/releases/tag/2.10 A great place to collaborate and discuss features, bugs and ideas on snapcraft is snapcraft at lists.ubuntu.com mailing list or directly in the #snappy channel on irc.freenode.net. To file bugs, please go to https://bugs.launchpad.net/snapcraft/+filebug. Last but not least the team thanks all the external contributions, keep them coming! Happy snapcrafting, - Sergio and the team -- Enviado con Dekko desde mi dispositivo Ubuntu From david.calle at canonical.com Thu Jun 9 16:06:15 2016 From: david.calle at canonical.com (=?UTF-8?Q?David_Call=c3=a9?=) Date: Thu, 9 Jun 2016 18:06:15 +0200 Subject: Snap! Community weekly update: W23 Message-ID: <575993F7.9070408@canonical.com> Here is an update on the highlights in the snap community this week: ## Snap examples - The first snappy Playpen event was a blast! With more than 40 people participating and a dozen of new branches, we felt it was a great learning process for all involved and we gathered many new examples: - Leafpad - SMPlayer - Galculator - Plank - Tinyproxy - Heroku CLI - And a creativity winner: a snap for the Ubuntu Kylin icon theme, which bundles commands to enable/disable the theme There are more branches still being worked on (for example a git build of GIMP !). Check out the full details in our full recap of the Snappy Playpen kickoff . ## Documentation - General updates along the new snapd and snapcraft releases - guides/interfaces - guides/autoupdate - Ongoing content review with the Web Team ## Community highlights - Vincent Jobard has posted a video tutorial in French about Snapcraft - Jean-Marie Verdun wrote a blog post for the French-speaking community about snapping FreeCAD - The Mycroft team is working on snapping their core AI engine and their Text-to-Speech Mimic engine Cheers, David -------------- next part -------------- An HTML attachment was scrubbed... URL: From leo.arias at canonical.com Thu Jun 9 16:11:01 2016 From: leo.arias at canonical.com (Leo Arias) Date: Thu, 9 Jun 2016 10:11:01 -0600 Subject: Snap! Community weekly update: W23 In-Reply-To: <575993F7.9070408@canonical.com> References: <575993F7.9070408@canonical.com> Message-ID: <57599515.3060708@canonical.com> Thank you community team! -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: OpenPGP digital signature URL: From fcole90 at gmail.com Thu Jun 9 19:28:28 2016 From: fcole90 at gmail.com (Fabio Colella) Date: Thu, 9 Jun 2016 21:28:28 +0200 Subject: Snap! Community weekly update: W23 In-Reply-To: <575993F7.9070408@canonical.com> References: <575993F7.9070408@canonical.com> Message-ID: Very glad the Ubuntu Kylin icon theme snap was appreciated 😊 On 9 June 2016 at 18:06, David Callé wrote: > Here is an update on the highlights in the snap community this week: > > ## Snap examples > > - The first snappy Playpen event was a blast! With more than 40 people > participating and a dozen of new branches, we felt it was a great learning > process for all involved and we gathered many new examples: > - Leafpad > - SMPlayer > - Galculator > - Plank > - Tinyproxy > - Heroku CLI > - And a creativity winner: a snap for the Ubuntu Kylin icon theme, > which bundles commands to enable/disable the theme > > There are more branches still being worked on (for example a git build of > GIMP !). > Check out the full details in our full recap of the Snappy Playpen kickoff > > . > > ## Documentation > > - General updates along the new snapd and snapcraft releases > - guides/interfaces > > - guides/autoupdate > > - Ongoing content review with the Web Team > > ## Community highlights > > - Vincent Jobard has posted a video tutorial > in French about Snapcraft > - Jean-Marie Verdun wrote a blog post > > for the French-speaking community about snapping FreeCAD > - The Mycroft team is working on snapping their core AI engine > and their Text-to-Speech > Mimic engine > > Cheers, > David > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From manik at canonical.com Thu Jun 9 20:41:20 2016 From: manik at canonical.com (Manik Taneja) Date: Thu, 9 Jun 2016 13:41:20 -0700 Subject: Snap! Community weekly update: W23 In-Reply-To: <575993F7.9070408@canonical.com> References: <575993F7.9070408@canonical.com> Message-ID: Great work guys. This is super useful in bootstrapping the ecosystem. Cheers, Manik On Thu, Jun 9, 2016 at 9:06 AM, David Callé wrote: > Here is an update on the highlights in the snap community this week: > > ## Snap examples > > - The first snappy Playpen event was a blast! With more than 40 people > participating and a dozen of new branches, we felt it was a great learning > process for all involved and we gathered many new examples: > - Leafpad > - SMPlayer > - Galculator > - Plank > - Tinyproxy > - Heroku CLI > - And a creativity winner: a snap for the Ubuntu Kylin icon theme, > which bundles commands to enable/disable the theme > > There are more branches still being worked on (for example a git build of > GIMP !). > Check out the full details in our full recap of the Snappy Playpen kickoff > > . > > ## Documentation > > - General updates along the new snapd and snapcraft releases > - guides/interfaces > > - guides/autoupdate > > - Ongoing content review with the Web Team > > ## Community highlights > > - Vincent Jobard has posted a video tutorial > in French about Snapcraft > - Jean-Marie Verdun wrote a blog post > > for the French-speaking community about snapping FreeCAD > - The Mycroft team is working on snapping their core AI engine > and their Text-to-Speech > Mimic engine > > Cheers, > David > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From daniel.holbach at canonical.com Fri Jun 10 10:06:04 2016 From: daniel.holbach at canonical.com (Daniel Holbach) Date: Fri, 10 Jun 2016 12:06:04 +0200 Subject: Second all-day Snappy Playpen event Message-ID: <575A910C.9060005@canonical.com> Hello hello, In our first Snappy Playpen event last Tuesday we simply wanted to bring people together, invite them to get to know the team, get started together and see how things go. It went great, check out the report [1]! Next week, on Tuesday, 14th June, we want to meet up and snap software together again. Obviously you can join the playpen gitter channel [2] or #snappy on Freenode [3] (or contribute to Snappy Playpen [4]) at any time, but on Tuesday we want to get everyone together and make another push to get good stuff landed together. This time we want to especially extend the invitation to all upstreams who are interested in getting their software snapped. If you are interested and need help, join us and we will figure out things together. If you still need to be convinced, here are a few reasons why this might make sense for your project: - Just run snapcraft upload to upload a snap to the store. (Maybe even hook it up with your CI?) - No lengthy review process. Publication within seconds. - Use different channels (stable, beta, edge) to ship different versions of your software to different audiences. - Build instructions in snapcraft.yaml are very simple, all nice and declarative. - Millions of Ubuntu 16.04 users can easily install your software through the software center. We would also like to invite all Ubuntu flavours to participate. If you want to play around with snaps, we will help you get started. WHAT: Snappy playpen [4] sprint WHEN: Tuesday, 14th June 2016 all day WHERE: Join us on gitter [2] or IRC [3] [1] https://developer.ubuntu.com/en/blog/2016/06/08/snappy-playpen-kickoff-highlights/ [2] https://gitter.im/ubuntu/snappy-playpen [3] http://webchat.freenode.net/?channels=snappy [4] https://github.com/ubuntu/snappy-playpen Have a great day, Daniel From daniel.holbach at canonical.com Fri Jun 10 13:35:31 2016 From: daniel.holbach at canonical.com (Daniel Holbach) Date: Fri, 10 Jun 2016 15:35:31 +0200 Subject: Snap docs blitz Message-ID: <575AC223.4020401@canonical.com> Hello everybody, I quickly went over the following projects in Launchpad: - snappy (for snapd) - snapcraft - developer-ubuntu-com and tagged all bugs related to snap docs with 'snap-docs'. If you find any other bugs which are relevant, please use the tag too. This gives us one URL to look at for organising a docs blitz and it is: https://bugs.launchpad.net/bugs/+bugs?field.tag=snap-docs Let's all make an effort to get these bugs fixed. Thanks everyone! Have a great day, Daniel From gustavo at niemeyer.net Fri Jun 10 17:45:55 2016 From: gustavo at niemeyer.net (Gustavo Niemeyer) Date: Fri, 10 Jun 2016 14:45:55 -0300 Subject: Snap docs blitz In-Reply-To: <575AC223.4020401@canonical.com> References: <575AC223.4020401@canonical.com> Message-ID: Sweet, thanks Daniel! On Fri, Jun 10, 2016 at 10:35 AM, Daniel Holbach < daniel.holbach at canonical.com> wrote: > Hello everybody, > > I quickly went over the following projects in Launchpad: > > - snappy (for snapd) > - snapcraft > - developer-ubuntu-com > > and tagged all bugs related to snap docs with 'snap-docs'. If you find > any other bugs which are relevant, please use the tag too. > > This gives us one URL to look at for organising a docs blitz and it is: > > https://bugs.launchpad.net/bugs/+bugs?field.tag=snap-docs > > Let's all make an effort to get these bugs fixed. > > Thanks everyone! > > Have a great day, > Daniel > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > -- gustavo @ http://niemeyer.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From claudioandre.br at gmail.com Sun Jun 12 21:14:36 2016 From: claudioandre.br at gmail.com (=?UTF-8?Q?Claudio_Andr=C3=A9?=) Date: Sun, 12 Jun 2016 18:14:36 -0300 Subject: Packaging for Ubuntu on Windows [off-topic] Message-ID: Dear, it is a kind of RFC. Sorry for the noise. Stealing ideas from what snapcraft does, I just packaged (via zip file) one application to run on Ubuntu on Windows. And it is working [1]. If possible, **in the future**, I would like to use snapcraft to help me in this task, since I did, basically, what snapcraft does. Well, it might worth it. Please, take it seriously at your own risk, I'm just thinking loud. BTW: I'm just thinking about a zip file package. Thank you. [1] https://cloud.githubusercontent.com/assets/1702923/15993386/d3dd434c-30ba-11e6-98c6-0ceb2a77d3f9.png -------------- next part -------------- An HTML attachment was scrubbed... URL: From mark at ubuntu.com Sun Jun 12 21:23:03 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Sun, 12 Jun 2016 22:23:03 +0100 Subject: Packaging for Ubuntu on Windows [off-topic] In-Reply-To: References: Message-ID: <575DD2B7.9010603@ubuntu.com> Yes, this is plausible, we'll need to see how the Microsoft LX team feels about the relevant security and container-related syscalls. I imagine they will want Docker to work, which will mean quite a significant portion will get done. Mark On 12/06/16 22:14, Claudio André wrote: > Dear, it is a kind of RFC. Sorry for the noise. > > Stealing ideas from what snapcraft does, I just packaged (via zip > file) one application to run on Ubuntu on Windows. And it is working [1]. > > If possible, **in the future**, I would like to use snapcraft to help > me in this task, since I did, basically, what snapcraft does. Well, it > might worth it. > > Please, take it seriously at your own risk, I'm just thinking loud. > BTW: I'm just thinking about a zip file package. > Thank you. > > [1] > https://cloud.githubusercontent.com/assets/1702923/15993386/d3dd434c-30ba-11e6-98c6-0ceb2a77d3f9.png > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From woodrow.shen at canonical.com Mon Jun 13 11:20:40 2016 From: woodrow.shen at canonical.com (Woodrow Shen) Date: Mon, 13 Jun 2016 19:20:40 +0800 Subject: confinement failure when uploading to store Message-ID: Hi all, I have one simple problem about uploading kernel snap to store as I used snapcraft to create my kernel snap. I got the following error after the upload was done: 'confinement' should only be specified with 'type: app' lint-snap-v2_confinement_valid But 'confinement' flag will be added by snapcraft automatically, if I don't do that manually. So is it expected for this behavior ? -- Woodrow Shen Software Engineer, Canonical ltd. UES | CE | PC & Core, Taipei -------------- next part -------------- An HTML attachment was scrubbed... URL: From jamie at canonical.com Mon Jun 13 12:57:45 2016 From: jamie at canonical.com (Jamie Strandboge) Date: Mon, 13 Jun 2016 15:57:45 +0300 Subject: confinement failure when uploading to store In-Reply-To: References: Message-ID: <1465822665.4257.1.camel@canonical.com> On Mon, 2016-06-13 at 19:20 +0800, Woodrow Shen wrote: > Hi all, > > I have one simple problem about uploading kernel snap to store as I used > snapcraft to create my kernel snap. I got the following error after the > upload was done: > > 'confinement' should only be specified with 'type: app' > lint-snap-v2_confinement_valid > > But 'confinement' flag will be added by snapcraft automatically, if I don't > do that manually. So is it expected for this behavior ? > This sounds like a bug in snapcraft (I've CC'd Sergio to be sure). The confinement flag only makes sense in the context of app snaps AFAIK, so snapcraft shouldn't be adding it to kernel snaps. -- Jamie Strandboge | http://www.canonical.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: This is a digitally signed message part URL: From gustavo.niemeyer at canonical.com Mon Jun 13 13:41:18 2016 From: gustavo.niemeyer at canonical.com (Gustavo Niemeyer) Date: Mon, 13 Jun 2016 10:41:18 -0300 Subject: confinement failure when uploading to store In-Reply-To: <1465822665.4257.1.camel@canonical.com> References: <1465822665.4257.1.camel@canonical.com> Message-ID: I think it makes sense in general, actually. The kernel is special because the image itself is of course not confined, but if/when we introduce hooks for example, we'll want these to be properly confined, declaring any interfaces they require to run explicitly as usual for everything else. Same idea for the gadget snap. That being the case, having development mode for them also makes sense. On Mon, Jun 13, 2016 at 9:57 AM, Jamie Strandboge wrote: > On Mon, 2016-06-13 at 19:20 +0800, Woodrow Shen wrote: > > Hi all, > > > > I have one simple problem about uploading kernel snap to store as I used > > snapcraft to create my kernel snap. I got the following error after the > > upload was done: > > > > 'confinement' should only be specified with 'type: app' > > lint-snap-v2_confinement_valid > > > > But 'confinement' flag will be added by snapcraft automatically, if I > don't > > do that manually. So is it expected for this behavior ? > > > > This sounds like a bug in snapcraft (I've CC'd Sergio to be sure). The > confinement flag only makes sense in the context of app snaps AFAIK, so > snapcraft shouldn't be adding it to kernel snaps. > > -- > Jamie Strandboge | http://www.canonical.com > > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > > -- gustavo @ http://niemeyer.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From jamie at canonical.com Mon Jun 13 15:16:44 2016 From: jamie at canonical.com (Jamie Strandboge) Date: Mon, 13 Jun 2016 18:16:44 +0300 Subject: confinement failure when uploading to store In-Reply-To: References: <1465822665.4257.1.camel@canonical.com> Message-ID: <1465831004.4257.17.camel@canonical.com> On Mon, 2016-06-13 at 10:41 -0300, Gustavo Niemeyer wrote: > I think it makes sense in general, actually.  The kernel is special because > the image itself is of course not confined, but if/when we introduce hooks > for example, we'll want these to be properly confined, declaring any > interfaces they require to run explicitly as usual for everything else. > Same idea for the gadget snap. > > That being the case, having development mode for them also makes sense. > Ok, I'll update the tools accordingly. -- Jamie Strandboge | http://www.canonical.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: This is a digitally signed message part URL: From daniel.holbach at canonical.com Mon Jun 13 15:46:21 2016 From: daniel.holbach at canonical.com (Daniel Holbach) Date: Mon, 13 Jun 2016 17:46:21 +0200 Subject: Second all-day Snappy Playpen event In-Reply-To: <575A910C.9060005@canonical.com> References: <575A910C.9060005@canonical.com> Message-ID: <575ED54D.5040001@canonical.com> Hey hey, just a quick reminder for everyone, this is tomorrow! :-) See you there, Daniel On 10.06.2016 12:06, Daniel Holbach wrote: > Hello hello, > > In our first Snappy Playpen event last Tuesday we simply wanted to bring > people together, invite them to get to know the team, get started > together and see how things go. It went great, check out the report [1]! > > Next week, on Tuesday, 14th June, we want to meet up and snap software > together again. Obviously you can join the playpen gitter channel [2] or > #snappy on Freenode [3] (or contribute to Snappy Playpen [4]) at any > time, but on Tuesday we want to get everyone together and make another > push to get good stuff landed together. > > This time we want to especially extend the invitation to all upstreams > who are interested in getting their software snapped. If you are > interested and need help, join us and we will figure out things > together. If you still need to be convinced, here are a few reasons why > this might make sense for your project: > > - Just run snapcraft upload to upload a snap to the store. (Maybe > even hook it up with your CI?) > - No lengthy review process. Publication within seconds. > - Use different channels (stable, beta, edge) to ship different > versions of your software to different audiences. > - Build instructions in snapcraft.yaml are very simple, all nice > and declarative. > - Millions of Ubuntu 16.04 users can easily install your > software through the software center. > > We would also like to invite all Ubuntu flavours to participate. If you > want to play around with snaps, we will help you get started. > > WHAT: Snappy playpen [4] sprint > WHEN: Tuesday, 14th June 2016 all day > WHERE: Join us on gitter [2] or IRC [3] > > > [1] > https://developer.ubuntu.com/en/blog/2016/06/08/snappy-playpen-kickoff-highlights/ > [2] https://gitter.im/ubuntu/snappy-playpen > [3] http://webchat.freenode.net/?channels=snappy > [4] https://github.com/ubuntu/snappy-playpen > > Have a great day, > Daniel > From michael.vogt at canonical.com Mon Jun 13 20:04:41 2016 From: michael.vogt at canonical.com (Michael Vogt) Date: Mon, 13 Jun 2016 22:04:41 +0200 Subject: ANN: snapd 2.0.8 available in 16.04 Message-ID: <20160613200441.GA6538@bod> Snapd 2.0.8 available in 16.04 ============================== The snappy team is very happy to announce that the 2.0.8 release is now available in 16.04 via "xenial-updates". The 2.0.8 release contains a number of improvements and fixes over the previous 2.0.5 release that was availalbe before. The highlights: - automatically refresh all installed snaps daily - new `snap refresh` and `snap refresh --list` commands - new `snap try` command - support for fcitx and mozc input methods in the unity7 interface - add global gsettings interface - autoconnect the "home" interface - add "socketcall" to allowed syscalls in the "network-bind" inteface - new "cups" interface - make snapd read /etc/environment (allows easier global http_proxy settings) - allow AppMenu and launcher API for unity7 - add new location-observer interface - allow updates to snapd via updates to the OS snap - fix bug about "app names" being too lenient A more verbose changelog is available https://github.com/snapcore/snapd/blob/master/debian/changelog and the full details can be found here: https://github.com/snapcore/snapd/tree/2.0.8 We hope you like it as much as we do. If you find any issues, please let us know via: http://bugs.launchpad.net/snappy Thanks! Michael (on behalf of the snappy team) From tsimonq2 at ubuntu.com Mon Jun 13 20:31:24 2016 From: tsimonq2 at ubuntu.com (Simon Quigley) Date: Mon, 13 Jun 2016 15:31:24 -0500 Subject: ANN: snapd 2.0.8 available in 16.04 In-Reply-To: <20160613200441.GA6538@bod> References: <20160613200441.GA6538@bod> Message-ID: <575F181C.9030303@ubuntu.com> Greetings Michael, Sorry to hijack this thread, but there's a bit of a versioning issue here. snapd is version 2.0.8 in 16.04, but it is still at version 2.0.*2* in Yakkety: < tsimonq2> !info snapd yakkety < ubottu> snapd (source: snapd): Tool to interact with Ubuntu Core Snappy.. In component main, is optional. Version 2.0.2 (yakkety), package size 2745 kB, installed size 14700 kB < tsimonq2> !info snapd xenial < ubottu> snapd (source: snapd): Tool to interact with Ubuntu Core Snappy.. In component main, is optional. Version 2.0.8 (xenial), package size 3581 kB, installed size 17156 kB When will 2.0.8 be in Yakkety, and wouldn't this sort of thing be in Yakkety first? I use Yakkety, and I'm stuck compiling from source or using the package from Xenial if I want to have the improvements that came with 2.0.8. Have a nice day! -- Simon Quigley tsimonq2 at ubuntu.com tsimonq2 on Freenode From chris.wayne at canonical.com Mon Jun 13 23:17:23 2016 From: chris.wayne at canonical.com (Chris Wayne) Date: Mon, 13 Jun 2016 19:17:23 -0400 Subject: ANN: snapd 2.0.8 available in 16.04 In-Reply-To: <20160613200441.GA6538@bod> References: <20160613200441.GA6538@bod> Message-ID: Nice, great stuff! Do we have any indication on when these changes may land in an os snap? Thanks Chris On Mon, Jun 13, 2016 at 4:04 PM, Michael Vogt wrote: > Snapd 2.0.8 available in 16.04 > ============================== > > The snappy team is very happy to announce that the 2.0.8 release is > now available in 16.04 via "xenial-updates". > > The 2.0.8 release contains a number of improvements and fixes over the > previous 2.0.5 release that was availalbe before. The highlights: > > - automatically refresh all installed snaps daily > - new `snap refresh` and `snap refresh --list` commands > - new `snap try` command > - support for fcitx and mozc input methods in the unity7 interface > - add global gsettings interface > - autoconnect the "home" interface > - add "socketcall" to allowed syscalls in the "network-bind" inteface > - new "cups" interface > - make snapd read /etc/environment (allows easier global http_proxy > settings) > - allow AppMenu and launcher API for unity7 > - add new location-observer interface > - allow updates to snapd via updates to the OS snap > - fix bug about "app names" being too lenient > > A more verbose changelog is available > https://github.com/snapcore/snapd/blob/master/debian/changelog and the > full details can be found here: > https://github.com/snapcore/snapd/tree/2.0.8 > > We hope you like it as much as we do. If you find any issues, please > let us know via: > http://bugs.launchpad.net/snappy > > Thanks! > Michael (on behalf of the snappy team) > > > > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > -------------- next part -------------- An HTML attachment was scrubbed... URL: From mark at ubuntu.com Tue Jun 14 04:25:31 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Tue, 14 Jun 2016 05:25:31 +0100 Subject: ANN: snapd 2.0.8 available in 16.04 In-Reply-To: References: <20160613200441.GA6538@bod> Message-ID: <575F873B.2010807@ubuntu.com> On 14/06/16 00:17, Chris Wayne wrote: > Nice, great stuff! Do we have any indication on when these changes > may land in an os snap? I think they are automatically landed in OS snaps, because those are built from the Ubuntu archives (with a few additional archives for example to get the latest snapd). One of the nice changes in 2.0.8 is that snapd will use the snapd from inside the OS snap, which means snapd-on-classic is updated whenever the os-snap is updated. Mark From simon.fels at canonical.com Tue Jun 14 07:07:22 2016 From: simon.fels at canonical.com (Simon Fels) Date: Tue, 14 Jun 2016 09:07:22 +0200 Subject: ANN: snapd 2.0.8 available in 16.04 In-Reply-To: <575F873B.2010807@ubuntu.com> References: <20160613200441.GA6538@bod> <575F873B.2010807@ubuntu.com> Message-ID: <575FAD2A.1060406@canonical.com> On 14.06.2016 06:25, Mark Shuttleworth wrote: > On 14/06/16 00:17, Chris Wayne wrote: >> Nice, great stuff! Do we have any indication on when these changes >> may land in an os snap? > > I think they are automatically landed in OS snaps, because those are > built from the Ubuntu archives (with a few additional archives for > example to get the latest snapd). > > One of the nice changes in 2.0.8 is that snapd will use the snapd from > inside the OS snap, which means snapd-on-classic is updated whenever the > os-snap is updated. @Chris: New OS snaps are produced daily on http://cdimage.ubuntu.com/ubuntu-core/xenial/daily-preinstalled/current/ which will include whatever is available in xenial. AFAIK they are just not yet pushed into the store automatically after passing a set of tests. regards, Simon From ogra at ubuntu.com Tue Jun 14 07:23:33 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Tue, 14 Jun 2016 09:23:33 +0200 Subject: ANN: snapd 2.0.8 available in 16.04 In-Reply-To: <575F873B.2010807@ubuntu.com> References: <20160613200441.GA6538@bod> <575F873B.2010807@ubuntu.com> Message-ID: <1465889013.22110.23.camel@ubuntu.com> hi, Am Dienstag, den 14.06.2016, 05:25 +0100 schrieb Mark Shuttleworth: > On 14/06/16 00:17, Chris Wayne wrote: > > > > Nice, great stuff!  Do we have any indication on when these changes > > may land in an os snap? > I think they are automatically landed in OS snaps, because those are > built from the Ubuntu archives (with a few additional archives for > example to get the latest snapd). > > One of the nice changes in 2.0.8 is that snapd will use the snapd > from > inside the OS snap, which means snapd-on-classic is updated whenever > the > os-snap is updated. note that the submit process to the store is still a manual thing, daily builds are actually on the cdimage url that simon pointed to but are not yet magically appearing in the edge channel in the store (that is in progress) ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From jian.luo.cn at gmail.com Tue Jun 14 09:49:54 2016 From: jian.luo.cn at gmail.com (Jian LUO) Date: Tue, 14 Jun 2016 11:49:54 +0200 Subject: Is the new ubuntu-image tool ready for test? In-Reply-To: References: Message-ID: Hi list, Can I find the source on GitHub or Launchpad yet? I'd love to make some tests and give feedbacks, especially on seeding U-Boot on x86 arch. Jian -------------- next part -------------- An HTML attachment was scrubbed... URL: From zygmunt.krynicki at canonical.com Tue Jun 14 09:52:19 2016 From: zygmunt.krynicki at canonical.com (Zygmunt Krynicki) Date: Tue, 14 Jun 2016 11:52:19 +0200 Subject: Is the new ubuntu-image tool ready for test? In-Reply-To: References: Message-ID: Hi The new ubuntu-image script is not ready for testing yet. If you would like to help with development feel free to contact me either here or in private (or ping me on #snappy on freenode) Best regards ZK (zyga on IRC) On Tue, Jun 14, 2016 at 11:49 AM, Jian LUO wrote: > Hi list, > > Can I find the source on GitHub or Launchpad yet? I'd love to make some > tests and give feedbacks, especially on seeding U-Boot on x86 arch. > > Jian > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From jenny.murphy at episensor.com Tue Jun 14 11:15:48 2016 From: jenny.murphy at episensor.com (Jenny Murphy) Date: Tue, 14 Jun 2016 12:15:48 +0100 Subject: help with ant and copy plugins for Java application Message-ID: Hi, I am trying to build up some competency with snapcraft for java applications. Currently I have a hello world project which is based on https://github.com/ubuntu-core/snapcraft/blob/1.x/examples/java-hello-world/snapcraft.yaml . I have extended in that HelloWorld.java references a method from another package in an external jar file (HellowWorld-helper.jar) I edited the snapcraft.yaml to use the copy plugin to copy the jar file to the jar directory. It can be seen there after the snap stage. However in bin.wrapper.wrapper , the external jar file is not added to the classpath. So the application fails on running. How do I fix this? -- *Jenny Murphy* *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* jenny.murphy at episensor.com t | +353 (0) 61 512 511 w | http://www.episensor.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From sergio.schvezov at canonical.com Tue Jun 14 12:33:38 2016 From: sergio.schvezov at canonical.com (Sergio Schvezov) Date: Tue, 14 Jun 2016 09:33:38 -0300 Subject: help with ant and copy plugins for Java application In-Reply-To: Message-ID: El martes, 14 de junio de 2016 08h'15:48 ART, Jenny Murphy escribió: > Hi, > I am trying to build up some competency with snapcraft for java > applications. > > Currently I have a hello world project which is based on > https://github.com/ubuntu-core/snapcraft/blob/1.x/examples/java-hello-world/snapcraft.yaml > . > > I have extended in that HelloWorld.java references a method from another > package in an external jar file (HellowWorld-helper.jar) > > I edited the snapcraft.yaml to use the copy plugin to copy the jar file to > the jar directory. > It can be seen there after the snap stage. > > However in bin.wrapper.wrapper , the external jar file is not added to the > classpath. So the application fails on running. > > How do I fix this? First of all, thank you for giving this a try. I'll start with a couple of details and follow on with a solution. First of all, `bin.wrapper.bin` is an implementation detail, likely to go away when we implement allowing `apps` (and `plugins`) to define environment variables (https://pad.lv/1583259) which would do the right thing on a snappy system. How would this look like in the not so distant future: ... apps: bin: command: .... environment: CLASSPATH: $CLASSPATH:$SNAP/path_to_jar/jar.jar parts: ... Now on to your problem. The java related plugins have logic to export environment variables, but each part (driven by a plugin), is a private world of its own. Since you used the copy plugin, the logic in the copy plugin is just to copy. How do we solve in the immediate future? Well you would need to create something similar to that wrapper and export the required CLASSPATH. I'll cut off here to not go into to much details, but I am happy to follow up on this. Cheers Sergio -- Enviado con Dekko desde mi dispositivo Ubuntu From michael.vogt at canonical.com Tue Jun 14 15:38:34 2016 From: michael.vogt at canonical.com (Michael Vogt) Date: Tue, 14 Jun 2016 17:38:34 +0200 Subject: ANN: snapd 2.0.8 available in 16.04 In-Reply-To: <575F181C.9030303@ubuntu.com> References: <20160613200441.GA6538@bod> <575F181C.9030303@ubuntu.com> Message-ID: <20160614153834.GF27715@bod> On Mon, Jun 13, 2016 at 03:31:24PM -0500, Simon Quigley wrote: > Greetings Michael, Hello Simon, thanks for your mail. > Sorry to hijack this thread, but there's a bit of a versioning issue > here. snapd is version 2.0.8 in 16.04, but it is still at version > 2.0.*2* in Yakkety: > > < tsimonq2> !info snapd yakkety > < ubottu> snapd (source: snapd): Tool to interact with Ubuntu Core > Snappy.. In component main, is optional. Version 2.0.2 (yakkety), > package size 2745 kB, installed size 14700 kB > > < tsimonq2> !info snapd xenial > < ubottu> snapd (source: snapd): Tool to interact with Ubuntu Core > Snappy.. In component main, is optional. Version 2.0.8 (xenial), package > size 3581 kB, installed size 17156 kB > > When will 2.0.8 be in Yakkety, and wouldn't this sort of thing be in > Yakkety first? I use Yakkety, and I'm stuck compiling from source or > using the package from Xenial if I want to have the improvements that > came with 2.0.8. Version 2.0.8+16.10 is in yakkety-proposed. However there are autopkgtest regression that prevent the 2.0.8 version from entering yakkety itself. The testsuite is exactly the same as the one that runs successfully on 16.04 and the errors in yakkety look not at all related to snapd. I strongly suspect its something in yakkety but so far I was not able to isolate it (did some runs in a yakkety VM, I can reproduce the failure but made no progress on finding the rootcasue so far). Cheers, Michael From mark at ubuntu.com Tue Jun 14 17:14:03 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Tue, 14 Jun 2016 18:14:03 +0100 Subject: Welcome to... Snapcraft.io! Message-ID: <57603B5B.4000108@ubuntu.com> Hi everybody Just a brief note to say (with apologies :) that we have aliased this list to snapcraft at lists.snapcraft.io, please do update your address books! http://snapcraft.io is the new home for all snappy crafters. We thought it would be appropriate to find a neutral space on the interwebs to celebrate the fact at snaps should now work, unmodified, on Arch, CentOS, Debian, Elementary, Fedora, Gentoo, RHEL, SUSE... and of course the whole family of *buntu's. snapd packages have landed in those distros or been submitted for inclusion, instructions are on the site, and I'm delighted to say that all the ISVs that I've talked to who are moving to snaps for IoT or the desktop or the cloud, are very excited to have those snaps "Just Work" on all these other Linux distros too. We all know that snaps are fantastic, it's wonderful to be confident that a great snap can be used and appreciated by users regardless of their preference for Linux distro. This is a once in a lifetime opportunity to defragment Linux apps, we're proud to be part of it and open to sharing the benefits with everyone else who makes Linux possible. Mark -------------- next part -------------- An HTML attachment was scrubbed... URL: From fcole90 at gmail.com Tue Jun 14 18:07:25 2016 From: fcole90 at gmail.com (Fabio Colella) Date: Tue, 14 Jun 2016 20:07:25 +0200 Subject: Welcome to... Snapcraft.io! In-Reply-To: <57603B5B.4000108@ubuntu.com> References: <57603B5B.4000108@ubuntu.com> Message-ID: That's awesome, I really hope it helps to improve the software distribution on all Linux distros. Now it would be great to have snapcraft as a snap package too, so that's easy also to package software in other distros as well. One note too, the tutorial should not assume the user to be on Ubuntu 16.04, but *just* to have the latest snapd package installed in his/her favourite distibution ;) On 14 June 2016 at 19:14, Mark Shuttleworth wrote: > Hi everybody > > Just a brief note to say (with apologies :) that we have aliased this list > to snapcraft at lists.snapcraft.io, please do update your address books! > > http://snapcraft.io is the new home for all snappy crafters. We thought > it would be appropriate to find a neutral space on the interwebs to > celebrate the fact at snaps should now work, unmodified, on Arch, CentOS, > Debian, Elementary, Fedora, Gentoo, RHEL, SUSE... and of course the whole > family of *buntu's. > > snapd packages have landed in those distros or been submitted for > inclusion, instructions are on the site, and I'm delighted to say that all > the ISVs that I've talked to who are moving to snaps for IoT or the desktop > or the cloud, are very excited to have those snaps "Just Work" on all these > other Linux distros too. We all know that snaps are fantastic, it's > wonderful to be confident that a great snap can be used and appreciated by > users regardless of their preference for Linux distro. This is a once in a > lifetime opportunity to defragment Linux apps, we're proud to be part of it > and open to sharing the benefits with everyone else who makes Linux > possible. > > Mark > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From peter.polz at gmail.com Tue Jun 14 19:10:32 2016 From: peter.polz at gmail.com (Peter Polz) Date: Tue, 14 Jun 2016 21:10:32 +0200 Subject: No subject Message-ID: -------------- next part -------------- An HTML attachment was scrubbed... URL: From sergio.schvezov at canonical.com Tue Jun 14 19:32:04 2016 From: sergio.schvezov at canonical.com (Sergio Schvezov) Date: Tue, 14 Jun 2016 16:32:04 -0300 Subject: ANN: snapcraft 2.11 is now available Message-ID: <57605BB4.6090900@ubuntu.com> Hello snapcrafters! We are pleased to announce the release of version 2.11 of snapcraft: https://launchpad.net/snapcraft/+milestone/2.11 This release mainly enables the `snapcraft tour` command which is mentioned in the recently announced new landing page for this snappy world http://snapcraft.io/create/ As usual with every release we try to lower the amount of bugs, this one includes many testing related fixes and improvements. To consume the latest snapcraft release on Xenial Xerus (16.04), we suggest you install the snapcraft package from the Ubuntu Archives: sudo apt update sudo apt install snapcraft To get the source for this release check it out at https://github.com/ubuntu-core/snapcraft/releases/tag/2.11 A great place to collaborate and discuss features, bugs and ideas on snapcraft is snapcraft at lists.ubuntu.com mailing list or directly in the #snappy channel on irc.freenode.net. To file bugs, please go to https://bugs.launchpad.net/snapcraft/+filebug. Happy snapcrafting, - Sergio and the team -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: OpenPGP digital signature URL: From jian.luo.cn at gmail.com Tue Jun 14 21:34:44 2016 From: jian.luo.cn at gmail.com (Jian LUO) Date: Tue, 14 Jun 2016 23:34:44 +0200 Subject: apparmor="DENIED" with ecrypted Home directory Message-ID: Hi, I got these error in a new xenial installation on my laptop. daniel at virgo:~$ pciutils.lspci ; journalctl | tail -n3 failed to create user data directory. errmsg: Permission denied Jun 14 23:02:23 virgo audit[7661]: AVC apparmor="DENIED" operation="open" profile="/usr/bin/ubuntu-core-launcher" name="/home/.ecryptfs/daniel/.Private/" pid=7661 comm="ubuntu-core-lau" requested_mask="r" denied_mask="r" fsuid=1000 ouid=1000 Jun 14 23:02:23 virgo kernel: audit: type=1400 audit(1465938143.001:272): apparmor="DENIED" operation="open" profile="/usr/bin/ubuntu-core-launcher" name="/home/.ecryptfs/daniel/.Private/" pid=7661 comm="ubuntu-core-lau" requested_mask="r" denied_mask="r" fsuid=1000 ouid=1000 Jun 14 23:02:23 virgo kernel: ecryptfs_dir_open: Error attempting to initialize the lower file for the dentry with name [/]; rc = [-13] I've tested with atom, hello-world and pciutils from the store w/ or w/o --devmode. Google search leads me to these two Bugs #359338 [1] and #1574526 [2]. But doubt that this one is also related. Should I fire a new one? Jian --- daniel at virgo:~$ uname -a ; dpkg -l snapd|tail -n1 Linux virgo 4.4.0-24-generic #43-Ubuntu SMP Wed Jun 8 19:27:37 UTC 2016 x86_64 x86_64 x86_64 GNU/Linux ii snapd 2.0.8 amd64 Tool to interact with Ubuntu Core Snappy. daniel at virgo:~$ snap list Name Version Rev Developer Notes atom-cwayne 1.9.0 1 cwayne18 devmode hello-world 6.1 26 canonical - pciutils 3.3.1-1 1 woodrow devmode ubuntu-core 16.04+20160531.11-56 122 canonical - [1] https://bugs.launchpad.net/ubuntu/+source/ecryptfs-utils/+bug/ [2] https://bugs.launchpad.net/snappy/+bug/1574526 -------------- next part -------------- An HTML attachment was scrubbed... URL: From jamie at canonical.com Wed Jun 15 02:44:52 2016 From: jamie at canonical.com (Jamie Strandboge) Date: Wed, 15 Jun 2016 05:44:52 +0300 Subject: apparmor="DENIED" with ecrypted Home directory In-Reply-To: References: Message-ID: <1465958692.15807.23.camel@canonical.com> On Tue, 2016-06-14 at 23:34 +0200, Jian LUO wrote: > Hi, > > I got these error in a new xenial installation on my laptop. > > daniel at virgo:~$ pciutils.lspci ; journalctl | tail -n3 > failed to create user data directory. errmsg: Permission denied > Jun 14 23:02:23 virgo audit[7661]: AVC apparmor="DENIED" operation="open" > profile="/usr/bin/ubuntu-core-launcher" > name="/home/.ecryptfs/daniel/.Private/" pid=7661 comm="ubuntu-core-lau" > requested_mask="r" denied_mask="r" fsuid=1000 ouid=1000 > Jun 14 23:02:23 virgo kernel: audit: type=1400 audit(1465938143.001:272): > apparmor="DENIED" operation="open" profile="/usr/bin/ubuntu-core-launcher" > name="/home/.ecryptfs/daniel/.Private/" pid=7661 comm="ubuntu-core-lau" > requested_mask="r" denied_mask="r" fsuid=1000 ouid=1000 > Jun 14 23:02:23 virgo kernel: ecryptfs_dir_open: Error attempting to > initialize the lower file for the dentry with name [/]; rc = [-13] > > I've tested with atom, hello-world and pciutils from the store w/ or w/o > --devmode. > > Google search leads me to these two Bugs #359338 [1] and #1574526 [2]. But > doubt that this one is also related. Should I fire a new one? > This is the open bug: https://bugs.launchpad.net/snappy/+bug/1574556 It should be fixed soon-- it is fixed in 16.10 and a 16.04 upload (with quite a few other changes) is coming. > Jian > --- > daniel at virgo:~$ uname -a ; dpkg -l snapd|tail -n1 > Linux virgo 4.4.0-24-generic #43-Ubuntu SMP Wed Jun 8 19:27:37 UTC 2016 > x86_64 x86_64 x86_64 GNU/Linux > ii  snapd          2.0.8        amd64        Tool to interact with Ubuntu > Core Snappy. > > daniel at virgo:~$ snap list > Name         Version               Rev  Developer  Notes > atom-cwayne  1.9.0                 1    cwayne18   devmode > hello-world  6.1                   26   canonical  - > pciutils     3.3.1-1               1    woodrow    devmode > ubuntu-core  16.04+20160531.11-56  122  canonical  - > > [1] https://bugs.launchpad.net/ubuntu/+source/ecryptfs-utils/+bug/ > [2] https://bugs.launchpad.net/snappy/+bug/1574526 > --  > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/s > napcraft -- Jamie Strandboge | http://www.canonical.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: This is a digitally signed message part URL: From mark at ubuntu.com Wed Jun 15 03:56:22 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Wed, 15 Jun 2016 04:56:22 +0100 Subject: Welcome to... Snapcraft.io! In-Reply-To: References: <57603B5B.4000108@ubuntu.com> Message-ID: <5760D1E6.3090802@ubuntu.com> On 14/06/16 19:07, Fabio Colella wrote: > One note too, the tutorial should not assume the user to be on Ubuntu > 16.04, but *just* to have the latest snapd package installed in > his/her favourite distibution ;) Yes, we can update that now. Also, it would be great to see patches to snapcraft that let it build from anyh Linux, using LXD containers for the actual build environment, but letting people use whatever Linux they like for the workstation. Mark From vasilisc777 at gmail.com Wed Jun 15 07:35:46 2016 From: vasilisc777 at gmail.com (Alekseenko Vasilii) Date: Wed, 15 Jun 2016 10:35:46 +0300 Subject: jar in snap Message-ID: <57610552.6060802@gmail.com> Sorry for my english. Please help me. I have a working Java application (*.jar). How do I pack it in the snap? I have LT-3.4-SNAPSHOT/ snapcraft.yaml In snapcraft.yaml ---- name: lt version: 3.4 summary: LT description: LT confinement: strict architectures: [amd64, i386] apps: lt: command: java -jar lt.jar plugs: [home] parts: lt: plugin: jdk source: ./LT-3.4-SNAPSHOT ---- But command unsquashfs -l *.snap > report.txt show me what snap contain only java without programm LT. -- Best regards, Vasilisc From jenny.murphy at episensor.com Wed Jun 15 10:45:13 2016 From: jenny.murphy at episensor.com (Jenny Murphy) Date: Wed, 15 Jun 2016 11:45:13 +0100 Subject: help with ant and copy plugins for Java application In-Reply-To: References: Message-ID: Hi, To follow up on this issue I solved it by extending the wrapper script. I add the required external jars to the existing class-path in wrapper. So wrapper looks like : #!/bin/sh export CLASSPATH=$SNAP_APP_PATH/jar/HelloWorld-helper.jar:$CLASSPATH java oata.HelloWorld This works fine. I have two examples now derived from the original java-hello-world example. Example 1 builds two jars and lets snapcraft do the work by adding the two jars that it builds to the class-path. Example 2 builds one jar and then takes care to add the required external jar to the class-path in the wrapper script. I can provide both if you think they might be of interest to other developers. Regards, Jenny On 14 June 2016 at 13:33, Sergio Schvezov wrote: > El martes, 14 de junio de 2016 08h'15:48 ART, Jenny Murphy < > jenny.murphy at episensor.com> escribió: > >> Hi, >> I am trying to build up some competency with snapcraft for java >> applications. >> >> Currently I have a hello world project which is based on >> >> https://github.com/ubuntu-core/snapcraft/blob/1.x/examples/java-hello-world/snapcraft.yaml >> . >> >> I have extended in that HelloWorld.java references a method from another >> package in an external jar file (HellowWorld-helper.jar) >> >> I edited the snapcraft.yaml to use the copy plugin to copy the jar file to >> the jar directory. >> It can be seen there after the snap stage. >> >> However in bin.wrapper.wrapper , the external jar file is not added to the >> classpath. So the application fails on running. >> >> How do I fix this? >> > > First of all, thank you for giving this a try. I'll start with a couple of > details and follow on with a solution. > > First of all, `bin.wrapper.bin` is an implementation detail, likely to go > away when we implement allowing `apps` (and `plugins`) to define > environment variables (https://pad.lv/1583259) which would do the right > thing on a snappy system. How would this look like in the not so distant > future: > > ... > apps: > bin: > command: .... > environment: > CLASSPATH: $CLASSPATH:$SNAP/path_to_jar/jar.jar > > parts: > ... > > Now on to your problem. The java related plugins have logic to export > environment variables, but each part (driven by a plugin), is a private > world of its own. Since you used the copy plugin, the logic in the copy > plugin is just to copy. How do we solve in the immediate future? Well you > would need to create something similar to that wrapper and export the > required CLASSPATH. > > I'll cut off here to not go into to much details, but I am happy to follow > up on this. > > Cheers > Sergio > > > -- > Enviado con Dekko desde mi dispositivo Ubuntu > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > -- *Jenny Murphy* *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* jenny.murphy at episensor.com t | +353 (0) 61 512 511 w | http://www.episensor.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From didrocks at ubuntu.com Wed Jun 15 14:29:02 2016 From: didrocks at ubuntu.com (Didier Roche) Date: Wed, 15 Jun 2016 16:29:02 +0200 Subject: jar in snap In-Reply-To: <57610552.6060802@gmail.com> References: <57610552.6060802@gmail.com> Message-ID: <5761662E.9050209@ubuntu.com> Le 15/06/2016 09:35, Alekseenko Vasilii a écrit : > Sorry for my english. Please help me. I have a working Java > application (*.jar). How do I pack it in the snap? > I have > LT-3.4-SNAPSHOT/ > snapcraft.yaml Hi, Thanks for looking into snapcraft! I'm not an expert on the jdk plugin myself, (I CCed Sergio who knows more than I about it). I'm a little bit puzzled about the command you are trying. Do you have the content (a github repo or a list of files if closed source) of LT-3.4-SNAPSHOT/? That will help us looking at it. Also, please attach your report.txt so that we can see the list of files around. Thanks! Didier > > In snapcraft.yaml > ---- > name: lt > version: 3.4 > summary: LT > description: LT > confinement: strict > architectures: [amd64, i386] > > apps: > lt: > command: java -jar lt.jar > plugs: [home] > > parts: > lt: > plugin: jdk > source: ./LT-3.4-SNAPSHOT > ---- > But command > unsquashfs -l *.snap > report.txt > show me what snap contain only java without programm LT. > From sergio.schvezov at canonical.com Wed Jun 15 14:31:10 2016 From: sergio.schvezov at canonical.com (Sergio Schvezov) Date: Wed, 15 Jun 2016 11:31:10 -0300 Subject: jar in snap In-Reply-To: <5761662E.9050209@ubuntu.com> References: <57610552.6060802@gmail.com> <5761662E.9050209@ubuntu.com> Message-ID: <576166AE.4060009@canonical.com> El 15/06/16 a las 11:29, Didier Roche escribió: > Le 15/06/2016 09:35, Alekseenko Vasilii a écrit : >> Sorry for my english. Please help me. I have a working Java >> application (*.jar). How do I pack it in the snap? >> I have >> LT-3.4-SNAPSHOT/ >> snapcraft.yaml > > Hi, > > Thanks for looking into snapcraft! I'm not an expert on the jdk plugin > myself, (I CCed Sergio who knows more than I about it). > I'm a little bit puzzled about the command you are trying. > > Do you have the content (a github repo or a list of files if closed > source) of LT-3.4-SNAPSHOT/? > That will help us looking at it. Also, please attach your report.txt so > that we can see the list of files around. So without knowing what is in LT-3.4-SNAPSHOT I can say that the jdk plugin just puts the jdk into the snap, if your intention is to build something that is in that directory then you probably want the ant or maven plugin to drive this. From manik at canonical.com Wed Jun 15 16:20:48 2016 From: manik at canonical.com (Manik Taneja) Date: Wed, 15 Jun 2016 09:20:48 -0700 Subject: test Message-ID: -------------- next part -------------- An HTML attachment was scrubbed... URL: From john.agosta at canonical.com Wed Jun 15 18:09:55 2016 From: john.agosta at canonical.com (John Agosta) Date: Wed, 15 Jun 2016 12:09:55 -0600 Subject: snapping a set of python2 scripts from launchpad Message-ID: <576199F3.90009@canonical.com> Hi: I suspect this should be strait forward, but have been fumbling with this ... I have a set of python2 scripts sitting in a launchpad.net bzr branch that I am trying to snap up using snapcraft. These are essentially just a set of files (scripts) that have never been placed into any formal packaging structure. I can correctly build the python part using the structure: parts: lptools: plugin: python2 source: lp:~jagosta/my-program/lp_tools Where I am stuck is with defining the apps:, thus creating the commands that I would like to be placed into the bin/ directory of the snap. thanks, -- John Agosta From janssen at parc.com Wed Jun 15 19:49:17 2016 From: janssen at parc.com (Bill Janssen) Date: Wed, 15 Jun 2016 12:49:17 -0700 Subject: snapcraft conda plug-in? Message-ID: <29338.1466020157@parc.com> I've got a big package, multiple programs, that I assemble in one directory under /opt/goodstuff, using conda as the packaging system -- the goodstuff directory tree is a conda 'env'. It seems that this should be easy to re-package as a snap. However, conda does a few things that have to be changed, or undone, for snap: 1) It mungs shebangs; that is, a header like "#!/usr/bin/env python" gets changed to "#!/opt/goodstuff/bin/python". I think that would have to be undone; what snapcraft stage should that occur in? 2) It symlinks in some of the binaries; for example, /opt/goodstuff/bin/conda is a symlink to /opt/conda/bin/conda. Those would have to be reified (copied), perhaps in the "pull" stage? 3) It rewrites RPATH on some platforms (OS X), but perhaps not on Linux, I'll have to check. That should be redone to accord with snap. 4) It's not a completely clean package. There are some dependencies, like "python-dbus", that are hard to build, and therefore installed via apt-get. Seems to me that the way to this is with a conda plug-in for snapcraft that would take a conda env and package it; "copy" with fixes for conda. Any suggestions about this? What would be a good plug-in to use as a template? Bill From svij at ubuntu.com Wed Jun 15 20:14:04 2016 From: svij at ubuntu.com (Sujeevan Vijayakumaran) Date: Wed, 15 Jun 2016 22:14:04 +0200 Subject: Man-pages in snaps In-Reply-To: <5757C548.3050204@ubuntu.com> References: <062e78c4-151f-a53e-f807-38c7c8c18963@ubuntu.com> <5757C548.3050204@ubuntu.com> Message-ID: <4fefd795-db7b-41f0-71e1-34a1d4f89c97@ubuntu.com> Hi Didier, Am 08.06.2016 um 09:12 schrieb Didier Roche: > Le 07/06/2016 14:49, Sujeevan (svij) Vijayakumaran a écrit : > The desktop team indeed opened this bug, and I'm sure the snappy team > will respond to it quickly. > For now, you can't really workaround that (well, telling people to man > /snaps/snap_name/current/… isn't an option ;)) > > However, thinking about this, you may be able to add a "man" command in > between (so that autocompletion show it) in your snapcraft.yaml, which > is executing "man /snaps/snap_name/current/…" as a workaround. > That way you will have (if your snap_name matches your command name): While this sounds like a good workaround, this doesn't work. Atleast, as I understood, "Man /snap/taskwarrior/…" should work, but it doesn't find any man pages. If I'm not completely wrong, than snapcraft doesn't include the man pages which where provided by the taskwarrior package. Are those man pages explicitly excluded from snapcraft? This is my snapcraft.yaml: https://github.com/svijee/taskwarrior-snap/blob/master/snapcraft.yaml Cheers, Sujeevan From ogra at ubuntu.com Thu Jun 16 06:17:00 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Thu, 16 Jun 2016 08:17:00 +0200 Subject: Welcome to... Snapcraft.io! In-Reply-To: <57603B5B.4000108@ubuntu.com> References: <57603B5B.4000108@ubuntu.com> Message-ID: <1466057820.5531.3.camel@ubuntu.com> hi, Am Dienstag, den 14.06.2016, 18:14 +0100 schrieb Mark Shuttleworth: >  > http://snapcraft.io is the new home for all snappy crafters. could we have a redirect for "www.snapcraft.io" to "snapcraft.io" in the DNS ? currently that does not resolve to anything and some (specifically mobile i think) browsers auto-add www to the url entry. ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From nhaines at ubuntu.com Thu Jun 16 06:22:02 2016 From: nhaines at ubuntu.com (Nathan Haines) Date: Wed, 15 Jun 2016 23:22:02 -0700 Subject: Welcome to... Snapcraft.io! In-Reply-To: <1466057820.5531.3.camel@ubuntu.com> References: <57603B5B.4000108@ubuntu.com> <1466057820.5531.3.camel@ubuntu.com> Message-ID: <5762458A.9010106@ubuntu.com> On 06/15/2016 11:17 PM, Oliver Grawert wrote: > could we have a redirect for "www.snapcraft.io" to "snapcraft.io" in > the DNS ? currently that does not resolve to anything and some > (specifically mobile i think) browsers auto-add www to the url entry. A little off-topic, but I first read the subject line (this time around) as "snapcraft.lol" and now I'm sad that isn't a thing. It's good to have a redirect for www.snapcraft.io (I do the opposite with my own web site). -- Nathan Haines Ubuntu - http://www.ubuntu.com/ From mark at ubuntu.com Thu Jun 16 06:28:37 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Thu, 16 Jun 2016 07:28:37 +0100 Subject: Welcome to... Snapcraft.io! In-Reply-To: <5762458A.9010106@ubuntu.com> References: <57603B5B.4000108@ubuntu.com> <1466057820.5531.3.camel@ubuntu.com> <5762458A.9010106@ubuntu.com> Message-ID: <57624715.3010009@ubuntu.com> Is this fixed or do we have a split horizon DNS - I see www.snapcraft.io and snapcraft.io both resolve to the same addresses. Mark From ogra at ubuntu.com Thu Jun 16 06:38:58 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Thu, 16 Jun 2016 08:38:58 +0200 Subject: Welcome to... Snapcraft.io! In-Reply-To: <57624715.3010009@ubuntu.com> References: <57603B5B.4000108@ubuntu.com> <1466057820.5531.3.camel@ubuntu.com> <5762458A.9010106@ubuntu.com> <57624715.3010009@ubuntu.com> Message-ID: <1466059138.5531.6.camel@ubuntu.com> hi, Am Donnerstag, den 16.06.2016, 07:28 +0100 schrieb Mark Shuttleworth: > Is this fixed or do we have a split horizon DNS - I see www.snapcraft > .io > and snapcraft.io both resolve to the same addresses. > hmm, we finally got a canonical WLAN at the sprint today, i switched to it and it now resolves (it did not from the formerly used hotel network), might be caused by some transparent proxy stuff the hotel uses. so, all fine i guess :) ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From zygmunt.krynicki at canonical.com Thu Jun 16 08:09:43 2016 From: zygmunt.krynicki at canonical.com (Zygmunt Krynicki) Date: Thu, 16 Jun 2016 10:09:43 +0200 Subject: snapcraft conda plug-in? In-Reply-To: <29338.1466020157@parc.com> References: <29338.1466020157@parc.com> Message-ID: On Wed, Jun 15, 2016 at 9:49 PM, Bill Janssen wrote: > I've got a big package, multiple programs, that I assemble in one > directory under /opt/goodstuff, using conda as the packaging system -- > the goodstuff directory tree is a conda 'env'. It seems that this > should be easy to re-package as a snap. However, conda does a few > things that have to be changed, or undone, for snap: > > 1) It mungs shebangs; that is, a header like "#!/usr/bin/env python" > gets changed to "#!/opt/goodstuff/bin/python". I think that would > have to be undone; what snapcraft stage should that occur in? > Hey Remember that your snap will most likely ship python by itself. The shebang line doesn't matter because the root filesystem at the time your snap will execute is not going to show much of the host distribution file system. > > 2) It symlinks in some of the binaries; for example, > /opt/goodstuff/bin/conda is a symlink to /opt/conda/bin/conda. Those > would have to be reified (copied), perhaps in the "pull" stage? > In general, you should install your snap to "anything you want" and that will show up under /snap/conda/current/. In this case you would have /snap/conda/current/opt/conda/bin/conda. You should try to simplify the structure though it matters little in practice. > > 3) It rewrites RPATH on some platforms (OS X), but perhaps not on > Linux, I'll have to check. That should be redone to accord with snap. > This should be harmless but note that currently snapcraft has a bug when executables use rpath. It is better to strip rpath and rely on LD_LIBRARY_PATH that snapcraft and snapd automatically arrange for you. > 4) It's not a completely clean package. There are some dependencies, > like "python-dbus", that are hard to build, and therefore installed > via apt-get. > That's okay. > > Seems to me that the way to this is with a conda plug-in for snapcraft > that would take a conda env and package it; "copy" with fixes for conda. > > Any suggestions about this? What would be a good plug-in to use as a > template? You can put a custom plugin alongside your snapcraft.yaml file. You can experiment with the make plugin since makefiles are familiar to many people and you can use that as a way to just put files in various places and see what happens. Oh and while you experiment, do use "snap try", it will be much faster than rebuilding the snap all the time. Best regards ZK > > Bill > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > -------------- next part -------------- An HTML attachment was scrubbed... URL: From didrocks at ubuntu.com Thu Jun 16 08:15:12 2016 From: didrocks at ubuntu.com (Didier Roche) Date: Thu, 16 Jun 2016 10:15:12 +0200 Subject: snapcraft conda plug-in? In-Reply-To: <29338.1466020157@parc.com> References: <29338.1466020157@parc.com> Message-ID: <57626010.2040005@ubuntu.com> Hey Bill! Thanks for joining the snapcraft unstoppable bandwagon :) Le 15/06/2016 21:49, Bill Janssen a écrit : > I've got a big package, multiple programs, that I assemble in one > directory under /opt/goodstuff, using conda as the packaging system -- > the goodstuff directory tree is a conda 'env'. It seems that this > should be easy to re-package as a snap. However, conda does a few > things that have to be changed, or undone, for snap: > > 1) It mungs shebangs; that is, a header like "#!/usr/bin/env python" > gets changed to "#!/opt/goodstuff/bin/python". I think that would > have to be undone; what snapcraft stage should that occur in? > > 2) It symlinks in some of the binaries; for example, > /opt/goodstuff/bin/conda is a symlink to /opt/conda/bin/conda. Those > would have to be reified (copied), perhaps in the "pull" stage? > > 3) It rewrites RPATH on some platforms (OS X), but perhaps not on > Linux, I'll have to check. That should be redone to accord with snap. I think you are on the right path with this one: I would (apart if Sergio and others have a better way), build my own "conda" plugin. This one can specify "after: [other_part_list]", so that it can do those kinds of changes in the other parts (parts//install/) directory and impacting them. For renaming files and directories, you can have a look at the filesets and reorg concepts from snapcraft (https://developer.ubuntu.com/en/snappy/build-apps/snapcraft-syntax/) Maybe Sergio would have a better idea on how to solve those? > > 4) It's not a completely clean package. There are some dependencies, > like "python-dbus", that are hard to build, and therefore installed > via apt-get. You can either rebuild them directly as another part of your snapcraft.yaml. As you say, some may be hard to build, you can use the "stage-packages" stenza to declare a list of package to pull from the ubuntu archives while building your snap (including its dependencies). That should hopefully help you getting started! > Seems to me that the way to this is with a conda plug-in for snapcraft > that would take a conda env and package it; "copy" with fixes for conda. > > Any suggestions about this? What would be a good plug-in to use as a > template? I would suggest to look at a simple plugin like the ant one : https://github.com/ubuntu-core/snapcraft/blob/master/snapcraft/plugins/ant.py (or copy plugin) I hope this gives you a good head's up. Do not hesitate if you have any further question or need help to ask around! Didier From didrocks at ubuntu.com Thu Jun 16 08:19:41 2016 From: didrocks at ubuntu.com (Didier Roche) Date: Thu, 16 Jun 2016 10:19:41 +0200 Subject: snapping a set of python2 scripts from launchpad In-Reply-To: <576199F3.90009@canonical.com> References: <576199F3.90009@canonical.com> Message-ID: <5762611D.9040806@ubuntu.com> Le 15/06/2016 20:09, John Agosta a écrit : > Hi: Hey John! Happy to help you there ;) > > I suspect this should be strait forward, but have been fumbling with > this ... I have a set of python2 scripts sitting in a launchpad.net > bzr branch that I am trying to snap up using snapcraft. These are > essentially just a set of files (scripts) that have never been placed > into any formal packaging structure. > > I can correctly build the python part using the structure: > > parts: > lptools: > plugin: python2 > source: lp:~jagosta/my-program/lp_tools > > Where I am stuck is with defining the apps:, thus creating the > commands that I would like to be placed into the bin/ directory of the > snap. Do you really need to have those scripts placed into the bin/ directory? apps: can define a command: directory, and you expose every entry point you want your user to get access to (see https://developer.ubuntu.com/en/snappy/build-apps/snapcraft-syntax/). Some examples of applications exposing multiple entry point is one of the application from the snapcraft.io tour (I encourage you to have a look here!): http://snapcraft.io/create/, in particular this section should give you your needed answers: "02-parts: Snapcraft makes snaps out of parts". |apps: hello: command: hello bash: command: bash | See that the 2 commands aren't in any particular directory, command just refers to a path to your executable, relative to the root of your snap. Hope this helps! Didier -------------- next part -------------- An HTML attachment was scrubbed... URL: From gustavo.niemeyer at canonical.com Thu Jun 16 14:32:57 2016 From: gustavo.niemeyer at canonical.com (Gustavo Niemeyer) Date: Thu, 16 Jun 2016 11:32:57 -0300 Subject: Leveling up snapd integration tests Message-ID: Hello all, Just out of the press, some details about how we are fixing and improving our integration test systems: http://blog.labix.org/2016/06/16/leveling-up-snapd-integration-tests Direct result of that is more stability for you wherever you are - any supported Linux distribution or native device - and more features too as we spend less time debugging. The post and documentation on GitHub has more details. Let us know if you have any questions. gustavo @ http://niemeyer.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From janssen at parc.com Thu Jun 16 15:05:40 2016 From: janssen at parc.com (Bill Janssen) Date: Thu, 16 Jun 2016 08:05:40 -0700 Subject: snapcraft conda plug-in? In-Reply-To: References: <29338.1466020157@parc.com> Message-ID: <27314.1466089540@parc.com> Zygmunt Krynicki wrote: > On Wed, Jun 15, 2016 at 9:49 PM, Bill Janssen wrote: > > > I've got a big package, multiple programs, that I assemble in one > > directory under /opt/goodstuff, using conda as the packaging system -- > > the goodstuff directory tree is a conda 'env'. It seems that this > > should be easy to re-package as a snap. However, conda does a few > > things that have to be changed, or undone, for snap: > > > > 1) It mungs shebangs; that is, a header like "#!/usr/bin/env python" > > gets changed to "#!/opt/goodstuff/bin/python". I think that would > > have to be undone; what snapcraft stage should that occur in? > > > > Hey > > Remember that your snap will most likely ship python by itself. Right. It's already present under /opt/goodstuff/bin/, so it will get copied into the snap. > The shebang line doesn't matter because the root filesystem at the > time your snap will execute is not going to show much of the host > distribution file system. So how is a shebang in a snap script resolved? Or are shebangs not resolved? That means I'd have to gen up a whole new line of launchers for my executables. > > 2) It symlinks in some of the binaries; for example, > > /opt/goodstuff/bin/conda is a symlink to /opt/conda/bin/conda. Those > > would have to be reified (copied), perhaps in the "pull" stage? > > > > In general, you should install your snap to "anything you want" and that I'm confused; I'm talking about building the snap squashfs, not installing it. I originally thought I could just use the "copy" plug-in, but that doesn't seem to work. When I figure up my Python script (now a command under /snap/bin), I get this error: $ /snap/bin/goodstuff ImportError: No module named site $ The shebang line (in /snap/goodstuff/x1/bin/goodstuff) is not modified; it still says, "#!/opt/goodstuff/bin/python". So I think this has to be changed. > will show up under /snap/conda/current/. In this case you would have > /snap/conda/current/opt/conda/bin/conda. You should try to simplify the > structure though it matters little in practice. Do you mean that I should just copy the entire conda install? Yes, that would work, I think. However, it would probably also copy lots of other stuff that I don't want. The thing is, conda already provides a way to assemble lots of parts into an integrated environment. I'd like to re-use that capability; I've already written the conda yaml for goodstuff. > > 3) It rewrites RPATH on some platforms (OS X), but perhaps not on > > Linux, I'll have to check. That should be redone to accord with snap. > > > > This should be harmless but note that currently snapcraft has a bug when > executables use rpath. What bug is that? > It is better to strip rpath and rely on LD_LIBRARY_PATH that snapcraft > and snapd automatically arrange for you. Yes, but... I tend to use systems like JModelica which has its own ideas about who should set LD_LIBRARY_PATH and what it should be set to. > > 4) It's not a completely clean package. There are some dependencies, > > like "python-dbus", that are hard to build, and therefore installed > > via apt-get. > > > > That's okay. Right. > > > > > > > Seems to me that the way to this is with a conda plug-in for snapcraft > > that would take a conda env and package it; "copy" with fixes for conda. > > > > Any suggestions about this? What would be a good plug-in to use as a > > template? > > > You can put a custom plugin alongside your snapcraft.yaml file. You can > experiment with the make plugin since makefiles are familiar to many people > and you can use that as a way to just put files in various places and see > what happens. > > Oh and while you experiment, do use "snap try", it will be much faster than > rebuilding the snap all the time. Thanks for the tip! Bill > > Best regards > ZK > > > > > > > Bill > > > > -- > > Snapcraft mailing list > > Snapcraft at lists.ubuntu.com > > Modify settings or unsubscribe at: > > https://lists.ubuntu.com/mailman/listinfo/snapcraft > > From wesley.mason at canonical.com Thu Jun 16 16:04:45 2016 From: wesley.mason at canonical.com (Wes Mason) Date: Thu, 16 Jun 2016 16:04:45 +0000 Subject: Welcome to... Snapcraft.io! In-Reply-To: <5762458A.9010106@ubuntu.com> References: <57603B5B.4000108@ubuntu.com> <1466057820.5531.3.camel@ubuntu.com> <5762458A.9010106@ubuntu.com> Message-ID: On Thu, 16 Jun 2016 at 07:22 Nathan Haines wrote: > On 06/15/2016 11:17 PM, Oliver Grawert wrote: > > could we have a redirect for "www.snapcraft.io" to "snapcraft.io" in > > the DNS ? currently that does not resolve to anything and some > > (specifically mobile i think) browsers auto-add www to the url entry. > > A little off-topic, but I first read the subject line (this time around) > as "snapcraft.lol" and now I'm sad that isn't a thing. > > Got you covered: http://snapcraft.lol/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From bret.barker at canonical.com Thu Jun 16 16:40:40 2016 From: bret.barker at canonical.com (Bret A. Barker) Date: Thu, 16 Jun 2016 12:40:40 -0400 Subject: Welcome to... Snapcraft.io! In-Reply-To: <1466059138.5531.6.camel@ubuntu.com> References: <57603B5B.4000108@ubuntu.com> <1466057820.5531.3.camel@ubuntu.com> <5762458A.9010106@ubuntu.com> <57624715.3010009@ubuntu.com> <1466059138.5531.6.camel@ubuntu.com> Message-ID: <20160616164040.GR71995@abitrandom.net> We got the www. DNS in place late last night. On Thu, Jun 16, 2016 at 08:38:58AM +0200, Oliver Grawert wrote: > hi, > Am Donnerstag, den 16.06.2016, 07:28 +0100 schrieb Mark Shuttleworth: > > Is this fixed or do we have a split horizon DNS - I see www.snapcraft > > .io > > and snapcraft.io both resolve to the same addresses. > > > hmm, we finally got a canonical WLAN at the sprint today, i switched to > it and it now resolves (it did not from the formerly used hotel > network), might be caused by some transparent proxy stuff the hotel > uses. > > so, all fine i guess :) > > ciao > oli From mark at ubuntu.com Thu Jun 16 18:27:05 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Thu, 16 Jun 2016 19:27:05 +0100 Subject: Welcome to... Snapcraft.io! In-Reply-To: <20160616164040.GR71995@abitrandom.net> References: <57603B5B.4000108@ubuntu.com> <1466057820.5531.3.camel@ubuntu.com> <5762458A.9010106@ubuntu.com> <57624715.3010009@ubuntu.com> <1466059138.5531.6.camel@ubuntu.com> <20160616164040.GR71995@abitrandom.net> Message-ID: <5762EF79.4040200@ubuntu.com> Thank you both :) From janssen at parc.com Thu Jun 16 18:32:46 2016 From: janssen at parc.com (Bill Janssen) Date: Thu, 16 Jun 2016 11:32:46 -0700 Subject: difference between stage and prime? Message-ID: <355.1466101966@parc.com> I'm trying to figure out what happens in the transition from stage to prime. My "stage" directory looks like a nice clean copy of my conda env. Good. But the "prime" directory has some cruft in it. For instance, it has a "home/wjanssen" directory tree, which is clearly *not* something I want in the snap. There's also an "opt/goodstuff/" tree, which contains only "opt/goodstuff/lib/libpython2.7.so.1.0". That's in addition to "lib/libpython2.7.so.1.0". Bill From john.agosta at canonical.com Thu Jun 16 19:37:09 2016 From: john.agosta at canonical.com (John Agosta) Date: Thu, 16 Jun 2016 13:37:09 -0600 Subject: snapping a set of python2 scripts from launchpad In-Reply-To: <5762611D.9040806@ubuntu.com> References: <576199F3.90009@canonical.com> <5762611D.9040806@ubuntu.com> Message-ID: <5762FFE5.1060906@canonical.com> An HTML attachment was scrubbed... URL: From joe.talbott at canonical.com Thu Jun 16 19:41:23 2016 From: joe.talbott at canonical.com (Joe Talbott) Date: Thu, 16 Jun 2016 15:41:23 -0400 Subject: snapping a set of python2 scripts from launchpad In-Reply-To: <5762FFE5.1060906@canonical.com> References: <576199F3.90009@canonical.com> <5762611D.9040806@ubuntu.com> <5762FFE5.1060906@canonical.com> Message-ID: Do you, by chance, have tabs in your YAML? Tabs aren't allowed in YAML. Joe On Thu, Jun 16, 2016 at 3:37 PM, John Agosta wrote: > On 06/16/2016 02:19 AM, Didier Roche wrote: > > Le 15/06/2016 20:09, John Agosta a écrit : > > Hi: > > > Hey John! > > Happy to help you there ;) > > > I suspect this should be strait forward, but have been fumbling with this > ... I have a set of python2 scripts sitting in a launchpad.net bzr branch > that I am trying to snap up using snapcraft. These are essentially just a > set of files (scripts) that have never been placed into any formal packaging > structure. > > I can correctly build the python part using the structure: > > parts: > lptools: > plugin: python2 > source: lp:~jagosta/my-program/lp_tools > > Where I am stuck is with defining the apps:, thus creating the commands that > I would like to be placed into the bin/ directory of the snap. > > > Do you really need to have those scripts placed into the bin/ directory? > apps: can define a command: directory, and you expose every entry point you > want your user to get access to (see > https://developer.ubuntu.com/en/snappy/build-apps/snapcraft-syntax/). > Some examples of applications exposing multiple entry point is one of the > application from the snapcraft.io tour (I encourage you to have a look > here!): http://snapcraft.io/create/, in particular this section should give > you your needed answers: "02-parts: Snapcraft makes snaps out of parts". > > apps: > hello: > command: hello > bash: > command: bash > > > See that the 2 commands aren't in any particular directory, command just > refers to a path to your executable, relative to the root of your snap. > > Hope this helps! > > Hi Didier, > My blocker is a schema error. I now see I get a similar error using the > example at http://snapcraft.io/create/, > > nam > e > : hello-debug > version: "2.10" > summary: GNU Hello with Bash for debugging > description: GNU hello prints a friendly greeting. > This is part of the snapcraft tour at https://snapcraft.io/create/ > > apps: > hello: > command: hello > bash: > command: bash > > parts: > gnu-hello: > plugin: autotools > source: http://ftp.gnu.org/gnu/hello/hello-2.10.tar.gz > gnu-bash: > plugin: autotools > configflags: ["--infodir=/var/bash/info"] > source: http://ftp.gnu.org/gnu/bash/bash-4.3.tar.gz > > I am running v2.11. The error is: > Issues while validating snapcraft.yaml: The 'hello' property does not match > the required schema: Additional properties are not allowed ('bash' was > unexpected) > > thanks. > > -- > John Agosta > > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > From john.agosta at canonical.com Thu Jun 16 19:54:47 2016 From: john.agosta at canonical.com (John Agosta) Date: Thu, 16 Jun 2016 13:54:47 -0600 Subject: snapping a set of python2 scripts from launchpad In-Reply-To: References: <576199F3.90009@canonical.com> <5762611D.9040806@ubuntu.com> <5762FFE5.1060906@canonical.com> Message-ID: <57630407.4070306@canonical.com> On 06/16/2016 01:41 PM, Joe Talbott wrote: > Do you, by chance, have tabs in your YAML? Tabs aren't allowed in YAML. > Hi Joe: I triple checked, no. In fact, I added a tab and found the parser does check and provides a nice error when it locates a tab: >> found character '\t' that cannot start any token on line 10 of snapcraft.yaml > > On Thu, Jun 16, 2016 at 3:37 PM, John Agosta wrote: >> On 06/16/2016 02:19 AM, Didier Roche wrote: >> >> Le 15/06/2016 20:09, John Agosta a écrit : >> >> Hi: >> >> >> Hey John! >> >> Happy to help you there ;) >> >> >> I suspect this should be strait forward, but have been fumbling with this >> ... I have a set of python2 scripts sitting in a launchpad.net bzr branch >> that I am trying to snap up using snapcraft. These are essentially just a >> set of files (scripts) that have never been placed into any formal packaging >> structure. >> >> I can correctly build the python part using the structure: >> >> parts: >> lptools: >> plugin: python2 >> source: lp:~jagosta/my-program/lp_tools >> >> Where I am stuck is with defining the apps:, thus creating the commands that >> I would like to be placed into the bin/ directory of the snap. >> >> >> Do you really need to have those scripts placed into the bin/ directory? >> apps: can define a command: directory, and you expose every entry point you >> want your user to get access to (see >> https://developer.ubuntu.com/en/snappy/build-apps/snapcraft-syntax/). >> Some examples of applications exposing multiple entry point is one of the >> application from the snapcraft.io tour (I encourage you to have a look >> here!): http://snapcraft.io/create/, in particular this section should give >> you your needed answers: "02-parts: Snapcraft makes snaps out of parts". >> >> apps: >> hello: >> command: hello >> bash: >> command: bash >> >> >> See that the 2 commands aren't in any particular directory, command just >> refers to a path to your executable, relative to the root of your snap. >> >> Hope this helps! >> >> Hi Didier, >> My blocker is a schema error. I now see I get a similar error using the >> example at http://snapcraft.io/create/, >> >> nam >> e >> : hello-debug >> version: "2.10" >> summary: GNU Hello with Bash for debugging >> description: GNU hello prints a friendly greeting. >> This is part of the snapcraft tour at https://snapcraft.io/create/ >> >> apps: >> hello: >> command: hello >> bash: >> command: bash >> >> parts: >> gnu-hello: >> plugin: autotools >> source: http://ftp.gnu.org/gnu/hello/hello-2.10.tar.gz >> gnu-bash: >> plugin: autotools >> configflags: ["--infodir=/var/bash/info"] >> source: http://ftp.gnu.org/gnu/bash/bash-4.3.tar.gz >> >> I am running v2.11. The error is: >> Issues while validating snapcraft.yaml: The 'hello' property does not match >> the required schema: Additional properties are not allowed ('bash' was >> unexpected) >> >> thanks. >> >> -- >> John Agosta >> >> >> -- >> Snapcraft mailing list >> Snapcraft at lists.ubuntu.com >> Modify settings or unsubscribe at: >> https://lists.ubuntu.com/mailman/listinfo/snapcraft >> -- John Agosta Program Manager, snappy Ubuntu Core and Internet of Things UES Commercial Device Operations Canonical USA, LTD John.Agosta at Canonical.com +1 (970) 217-5115 Ubuntu: creating the world’s best open source software platform From daniel.manrique at canonical.com Thu Jun 16 20:03:19 2016 From: daniel.manrique at canonical.com (Daniel Manrique) Date: Thu, 16 Jun 2016 16:03:19 -0400 Subject: snapping a set of python2 scripts from launchpad In-Reply-To: <57630407.4070306@canonical.com> References: <576199F3.90009@canonical.com> <5762611D.9040806@ubuntu.com> <5762FFE5.1060906@canonical.com> <57630407.4070306@canonical.com> Message-ID: I copied your snapcraft.yaml and ran it through snapcraft 2.11 and it worked \o/ Literally the only change was fixing the weird "nam e" thing at the beginning. After building and installing it, both hello-debug.bash and hello-debug.hello are available! So "works for me" :) On Thu, Jun 16, 2016 at 3:54 PM, John Agosta wrote: > On 06/16/2016 01:41 PM, Joe Talbott wrote: >> >> Do you, by chance, have tabs in your YAML? Tabs aren't allowed in YAML. >> > Hi Joe: > I triple checked, no. In fact, I added a tab and found the parser does > check and provides a nice error when it locates a tab: > >> found character '\t' that cannot start any token on line 10 of > snapcraft.yaml > >> >> On Thu, Jun 16, 2016 at 3:37 PM, John Agosta >> wrote: >>> >>> On 06/16/2016 02:19 AM, Didier Roche wrote: >>> >>> Le 15/06/2016 20:09, John Agosta a écrit : >>> >>> Hi: >>> >>> >>> Hey John! >>> >>> Happy to help you there ;) >>> >>> >>> I suspect this should be strait forward, but have been fumbling with this >>> ... I have a set of python2 scripts sitting in a launchpad.net bzr branch >>> that I am trying to snap up using snapcraft. These are essentially just a >>> set of files (scripts) that have never been placed into any formal >>> packaging >>> structure. >>> >>> I can correctly build the python part using the structure: >>> >>> parts: >>> lptools: >>> plugin: python2 >>> source: lp:~jagosta/my-program/lp_tools >>> >>> Where I am stuck is with defining the apps:, thus creating the commands >>> that >>> I would like to be placed into the bin/ directory of the snap. >>> >>> >>> Do you really need to have those scripts placed into the bin/ directory? >>> apps: can define a command: directory, and you expose every entry point >>> you >>> want your user to get access to (see >>> https://developer.ubuntu.com/en/snappy/build-apps/snapcraft-syntax/). >>> Some examples of applications exposing multiple entry point is one of the >>> application from the snapcraft.io tour (I encourage you to have a look >>> here!): http://snapcraft.io/create/, in particular this section should >>> give >>> you your needed answers: "02-parts: Snapcraft makes snaps out of parts". >>> >>> apps: >>> hello: >>> command: hello >>> bash: >>> command: bash >>> >>> >>> See that the 2 commands aren't in any particular directory, command just >>> refers to a path to your executable, relative to the root of your snap. >>> >>> Hope this helps! >>> >>> Hi Didier, >>> My blocker is a schema error. I now see I get a similar error using the >>> example at http://snapcraft.io/create/, >>> >>> nam >>> e >>> : hello-debug >>> version: "2.10" >>> summary: GNU Hello with Bash for debugging >>> description: GNU hello prints a friendly greeting. >>> This is part of the snapcraft tour at https://snapcraft.io/create/ >>> >>> apps: >>> hello: >>> command: hello >>> bash: >>> command: bash >>> >>> parts: >>> gnu-hello: >>> plugin: autotools >>> source: http://ftp.gnu.org/gnu/hello/hello-2.10.tar.gz >>> gnu-bash: >>> plugin: autotools >>> configflags: ["--infodir=/var/bash/info"] >>> source: http://ftp.gnu.org/gnu/bash/bash-4.3.tar.gz >>> >>> I am running v2.11. The error is: >>> Issues while validating snapcraft.yaml: The 'hello' property does not >>> match >>> the required schema: Additional properties are not allowed ('bash' was >>> unexpected) >>> >>> thanks. >>> >>> -- >>> John Agosta >>> >>> >>> -- >>> Snapcraft mailing list >>> Snapcraft at lists.ubuntu.com >>> Modify settings or unsubscribe at: >>> https://lists.ubuntu.com/mailman/listinfo/snapcraft >>> > > > -- > John Agosta > > Program Manager, snappy Ubuntu Core and Internet of Things > UES Commercial Device Operations > Canonical USA, LTD > John.Agosta at Canonical.com > +1 (970) 217-5115 > > Ubuntu: creating the world’s best open source software platform > > > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft From joe.talbott at canonical.com Thu Jun 16 20:06:22 2016 From: joe.talbott at canonical.com (Joe Talbott) Date: Thu, 16 Jun 2016 16:06:22 -0400 Subject: snapping a set of python2 scripts from launchpad In-Reply-To: References: <576199F3.90009@canonical.com> <5762611D.9040806@ubuntu.com> <5762FFE5.1060906@canonical.com> <57630407.4070306@canonical.com> Message-ID: Your example (as well as the one from snapcraft.io/create) work for me as well. Joe On Thu, Jun 16, 2016 at 4:03 PM, Daniel Manrique wrote: > I copied your snapcraft.yaml and ran it through snapcraft 2.11 and it worked \o/ > > Literally the only change was fixing the weird "nam e" thing at the beginning. > > After building and installing it, both hello-debug.bash and > hello-debug.hello are available! > > So "works for me" :) > > On Thu, Jun 16, 2016 at 3:54 PM, John Agosta wrote: >> On 06/16/2016 01:41 PM, Joe Talbott wrote: >>> >>> Do you, by chance, have tabs in your YAML? Tabs aren't allowed in YAML. >>> >> Hi Joe: >> I triple checked, no. In fact, I added a tab and found the parser does >> check and provides a nice error when it locates a tab: >> >> found character '\t' that cannot start any token on line 10 of >> snapcraft.yaml >> >>> >>> On Thu, Jun 16, 2016 at 3:37 PM, John Agosta >>> wrote: >>>> >>>> On 06/16/2016 02:19 AM, Didier Roche wrote: >>>> >>>> Le 15/06/2016 20:09, John Agosta a écrit : >>>> >>>> Hi: >>>> >>>> >>>> Hey John! >>>> >>>> Happy to help you there ;) >>>> >>>> >>>> I suspect this should be strait forward, but have been fumbling with this >>>> ... I have a set of python2 scripts sitting in a launchpad.net bzr branch >>>> that I am trying to snap up using snapcraft. These are essentially just a >>>> set of files (scripts) that have never been placed into any formal >>>> packaging >>>> structure. >>>> >>>> I can correctly build the python part using the structure: >>>> >>>> parts: >>>> lptools: >>>> plugin: python2 >>>> source: lp:~jagosta/my-program/lp_tools >>>> >>>> Where I am stuck is with defining the apps:, thus creating the commands >>>> that >>>> I would like to be placed into the bin/ directory of the snap. >>>> >>>> >>>> Do you really need to have those scripts placed into the bin/ directory? >>>> apps: can define a command: directory, and you expose every entry point >>>> you >>>> want your user to get access to (see >>>> https://developer.ubuntu.com/en/snappy/build-apps/snapcraft-syntax/). >>>> Some examples of applications exposing multiple entry point is one of the >>>> application from the snapcraft.io tour (I encourage you to have a look >>>> here!): http://snapcraft.io/create/, in particular this section should >>>> give >>>> you your needed answers: "02-parts: Snapcraft makes snaps out of parts". >>>> >>>> apps: >>>> hello: >>>> command: hello >>>> bash: >>>> command: bash >>>> >>>> >>>> See that the 2 commands aren't in any particular directory, command just >>>> refers to a path to your executable, relative to the root of your snap. >>>> >>>> Hope this helps! >>>> >>>> Hi Didier, >>>> My blocker is a schema error. I now see I get a similar error using the >>>> example at http://snapcraft.io/create/, >>>> >>>> nam >>>> e >>>> : hello-debug >>>> version: "2.10" >>>> summary: GNU Hello with Bash for debugging >>>> description: GNU hello prints a friendly greeting. >>>> This is part of the snapcraft tour at https://snapcraft.io/create/ >>>> >>>> apps: >>>> hello: >>>> command: hello >>>> bash: >>>> command: bash >>>> >>>> parts: >>>> gnu-hello: >>>> plugin: autotools >>>> source: http://ftp.gnu.org/gnu/hello/hello-2.10.tar.gz >>>> gnu-bash: >>>> plugin: autotools >>>> configflags: ["--infodir=/var/bash/info"] >>>> source: http://ftp.gnu.org/gnu/bash/bash-4.3.tar.gz >>>> >>>> I am running v2.11. The error is: >>>> Issues while validating snapcraft.yaml: The 'hello' property does not >>>> match >>>> the required schema: Additional properties are not allowed ('bash' was >>>> unexpected) >>>> >>>> thanks. >>>> >>>> -- >>>> John Agosta >>>> >>>> >>>> -- >>>> Snapcraft mailing list >>>> Snapcraft at lists.ubuntu.com >>>> Modify settings or unsubscribe at: >>>> https://lists.ubuntu.com/mailman/listinfo/snapcraft >>>> >> >> >> -- >> John Agosta >> >> Program Manager, snappy Ubuntu Core and Internet of Things >> UES Commercial Device Operations >> Canonical USA, LTD >> John.Agosta at Canonical.com >> +1 (970) 217-5115 >> >> Ubuntu: creating the world’s best open source software platform >> >> >> >> -- >> Snapcraft mailing list >> Snapcraft at lists.ubuntu.com >> Modify settings or unsubscribe at: >> https://lists.ubuntu.com/mailman/listinfo/snapcraft From john.agosta at canonical.com Thu Jun 16 20:12:41 2016 From: john.agosta at canonical.com (John Agosta) Date: Thu, 16 Jun 2016 14:12:41 -0600 Subject: snapping a set of python2 scripts from launchpad In-Reply-To: References: <576199F3.90009@canonical.com> <5762611D.9040806@ubuntu.com> <5762FFE5.1060906@canonical.com> <57630407.4070306@canonical.com> Message-ID: <57630839.5020907@canonical.com> Ok, we can go back to our regularly scheduled program...It now works using a different editor...I originally used "vim" to enter the yaml, and it must be placing in a special character that even ":set list" does not expose. Thanks! On 06/16/2016 02:06 PM, Joe Talbott wrote: > Your example (as well as the one from snapcraft.io/create) work for me as well. > > Joe > > On Thu, Jun 16, 2016 at 4:03 PM, Daniel Manrique > wrote: >> I copied your snapcraft.yaml and ran it through snapcraft 2.11 and it worked \o/ >> >> Literally the only change was fixing the weird "nam e" thing at the beginning. >> >> After building and installing it, both hello-debug.bash and >> hello-debug.hello are available! >> >> So "works for me" :) >> >> On Thu, Jun 16, 2016 at 3:54 PM, John Agosta wrote: >>> On 06/16/2016 01:41 PM, Joe Talbott wrote: >>>> Do you, by chance, have tabs in your YAML? Tabs aren't allowed in YAML. >>>> >>> Hi Joe: >>> I triple checked, no. In fact, I added a tab and found the parser does >>> check and provides a nice error when it locates a tab: >>> >> found character '\t' that cannot start any token on line 10 of >>> snapcraft.yaml >>> >>>> On Thu, Jun 16, 2016 at 3:37 PM, John Agosta >>>> wrote: >>>>> On 06/16/2016 02:19 AM, Didier Roche wrote: >>>>> >>>>> Le 15/06/2016 20:09, John Agosta a écrit : >>>>> >>>>> Hi: >>>>> >>>>> >>>>> Hey John! >>>>> >>>>> Happy to help you there ;) >>>>> >>>>> >>>>> I suspect this should be strait forward, but have been fumbling with this >>>>> ... I have a set of python2 scripts sitting in a launchpad.net bzr branch >>>>> that I am trying to snap up using snapcraft. These are essentially just a >>>>> set of files (scripts) that have never been placed into any formal >>>>> packaging >>>>> structure. >>>>> >>>>> I can correctly build the python part using the structure: >>>>> >>>>> parts: >>>>> lptools: >>>>> plugin: python2 >>>>> source: lp:~jagosta/my-program/lp_tools >>>>> >>>>> Where I am stuck is with defining the apps:, thus creating the commands >>>>> that >>>>> I would like to be placed into the bin/ directory of the snap. >>>>> >>>>> >>>>> Do you really need to have those scripts placed into the bin/ directory? >>>>> apps: can define a command: directory, and you expose every entry point >>>>> you >>>>> want your user to get access to (see >>>>> https://developer.ubuntu.com/en/snappy/build-apps/snapcraft-syntax/). >>>>> Some examples of applications exposing multiple entry point is one of the >>>>> application from the snapcraft.io tour (I encourage you to have a look >>>>> here!): http://snapcraft.io/create/, in particular this section should >>>>> give >>>>> you your needed answers: "02-parts: Snapcraft makes snaps out of parts". >>>>> >>>>> apps: >>>>> hello: >>>>> command: hello >>>>> bash: >>>>> command: bash >>>>> >>>>> >>>>> See that the 2 commands aren't in any particular directory, command just >>>>> refers to a path to your executable, relative to the root of your snap. >>>>> >>>>> Hope this helps! >>>>> >>>>> Hi Didier, >>>>> My blocker is a schema error. I now see I get a similar error using the >>>>> example at http://snapcraft.io/create/, >>>>> >>>>> nam >>>>> e >>>>> : hello-debug >>>>> version: "2.10" >>>>> summary: GNU Hello with Bash for debugging >>>>> description: GNU hello prints a friendly greeting. >>>>> This is part of the snapcraft tour at https://snapcraft.io/create/ >>>>> >>>>> apps: >>>>> hello: >>>>> command: hello >>>>> bash: >>>>> command: bash >>>>> >>>>> parts: >>>>> gnu-hello: >>>>> plugin: autotools >>>>> source: http://ftp.gnu.org/gnu/hello/hello-2.10.tar.gz >>>>> gnu-bash: >>>>> plugin: autotools >>>>> configflags: ["--infodir=/var/bash/info"] >>>>> source: http://ftp.gnu.org/gnu/bash/bash-4.3.tar.gz >>>>> >>>>> I am running v2.11. The error is: >>>>> Issues while validating snapcraft.yaml: The 'hello' property does not >>>>> match >>>>> the required schema: Additional properties are not allowed ('bash' was >>>>> unexpected) >>>>> >>>>> thanks. >>>>> >>>>> -- >>>>> John Agosta >>>>> >>>>> >>>>> -- >>>>> Snapcraft mailing list >>>>> Snapcraft at lists.ubuntu.com >>>>> Modify settings or unsubscribe at: >>>>> https://lists.ubuntu.com/mailman/listinfo/snapcraft >>>>> >>> >>> -- >>> John Agosta >>> >>> Program Manager, snappy Ubuntu Core and Internet of Things >>> UES Commercial Device Operations >>> Canonical USA, LTD >>> John.Agosta at Canonical.com >>> +1 (970) 217-5115 >>> >>> Ubuntu: creating the world’s best open source software platform >>> >>> >>> >>> -- >>> Snapcraft mailing list >>> Snapcraft at lists.ubuntu.com >>> Modify settings or unsubscribe at: >>> https://lists.ubuntu.com/mailman/listinfo/snapcraft -- John Agosta Program Manager, snappy Ubuntu Core and Internet of Things UES Commercial Device Operations Canonical USA, LTD John.Agosta at Canonical.com +1 (970) 217-5115 Ubuntu: creating the world’s best open source software platform From reinhard.pointner at gmail.com Fri Jun 17 05:18:31 2016 From: reinhard.pointner at gmail.com (Reinhard Pointner) Date: Fri, 17 Jun 2016 13:18:31 +0800 Subject: Uploading snaps to myapps.developer.ubuntu.com Message-ID: Hi snappers! How long does it take for a snap to become visible via the `snap` command after it has been uploaded and published via developer.ubuntu.com? I've uploaded a snap 24h ago (to beta and edge channels) and it's published according to the website. But snap says it can't find the package (i.e. `sudo snap install filebot --channel=edge`). Channel beta/edge don't appear in `snap find` afaik, but if I install by name, shouldn't it work even if it's not in the public listing? It's set as a paid package if that make a difference. Just to confirm, "Ubuntu Core" refers to normal Ubuntu (Desktop or headless server) computers? (I'm asking because https://myapps.developer.ubuntu.com/dev/click-apps/ seems to list my desktop snap, but the url says click apps which is for phones afaik) Cheers, Reinhard -------------- next part -------------- An HTML attachment was scrubbed... URL: From vasilisc777 at gmail.com Fri Jun 17 06:00:38 2016 From: vasilisc777 at gmail.com (Alekseenko Vasilii) Date: Fri, 17 Jun 2016 09:00:38 +0300 Subject: wrapper Message-ID: <57639206.3060109@gmail.com> What is the correct way to modify the file prime/command-program.wrapper ? I need to add their own variables and modify existing ones. -- Best regards, vasilisc From didrocks at ubuntu.com Fri Jun 17 06:11:29 2016 From: didrocks at ubuntu.com (Didier Roche) Date: Fri, 17 Jun 2016 08:11:29 +0200 Subject: difference between stage and prime? In-Reply-To: <355.1466101966@parc.com> References: <355.1466101966@parc.com> Message-ID: <57639491.3080500@ubuntu.com> Le 16/06/2016 20:32, Bill Janssen a écrit : > I'm trying to figure out what happens in the transition from stage to > prime. My "stage" directory looks like a nice clean copy of my conda > env. Good. But the "prime" directory has some cruft in it. For > instance, it has a "home/wjanssen" directory tree, which is clearly > *not* something I want in the snap. There's also an "opt/goodstuff/" > tree, which contains only "opt/goodstuff/lib/libpython2.7.so.1.0". > That's in addition to "lib/libpython2.7.so.1.0". Hey Bill, Kyle had a nice explanation for that on Wednesday. Basically (from what I got), not everything is copied from stage to prime, the copy is from parts//install to prime/ stage/ is used primarly when one of parts needs another one to build, like some headers: parts: foo: […] bar: plugins: autotools configflags: --use-foo-headers=$SNAPCRAFT_STAGE/usr/lib/header after: [foo] That way, bar will be able to use foo. There are some tricks with the organize (https://developer.ubuntu.com/en/snappy/build-apps/snapcraft-syntax/) keywords which organize stage/, but also needs to impact the install copy to prime/, which even Kyle wasn't fully sure about. That's why I'm CCing Sergio on that so that he can shed some lights (and we hope to be able to improve the documentation there as well). To filter what you ship in the finale snap, you can use the "snap" keyword (see same url above) and use something like: parts: foo: snap: - -home/wjanssen Cheers, Didier From didrocks at ubuntu.com Fri Jun 17 06:17:14 2016 From: didrocks at ubuntu.com (Didier Roche) Date: Fri, 17 Jun 2016 08:17:14 +0200 Subject: Uploading snaps to myapps.developer.ubuntu.com In-Reply-To: References: Message-ID: <576395EA.2000800@ubuntu.com> Le 17/06/2016 07:18, Reinhard Pointner a écrit : > Hi snappers! Hey Reinhard! > > How long does it take for a snap to become visible via the `snap` > command after it has been uploaded and published > via developer.ubuntu.com ? > > I've uploaded a snap 24h ago (to beta and edge channels) and it's > published according to the website. But snap says it can't find the > package (i.e. `sudo snap install filebot --channel=edge`). > > Channel beta/edge don't appear in `snap find` afaik, but if I install > by name, shouldn't it work even if it's not in the public listing? > It's set as a paid package if that make a difference. > It's normally instant (as soon as it says it's published, meaning that checks pass, which are automatic if you don't use restricted interfaces). I can't find your snap there either: https://uappexplorer.com/apps?q=filebot&sort=relevance&type=snappy or trying to install from the edge channel) It's either still in manual review mode (but the web interface is normally telling you this) or something happened. I'm ccing Natalia who will be able to check for this. > > > Just to confirm, "Ubuntu Core" refers to normal Ubuntu (Desktop or > headless server) computers? > > (I'm asking > because https://myapps.developer.ubuntu.com/dev/click-apps/ seems to > list my desktop snap, but the url says click apps which is for phones > afaik) "Ubuntu Core" refers to the snap packages, indeed. The url will be fixed at some point (the same store than click is used for snaps). Cheers, Didier -------------- next part -------------- An HTML attachment was scrubbed... URL: From didrocks at ubuntu.com Fri Jun 17 06:20:10 2016 From: didrocks at ubuntu.com (Didier Roche) Date: Fri, 17 Jun 2016 08:20:10 +0200 Subject: wrapper In-Reply-To: <57639206.3060109@gmail.com> References: <57639206.3060109@gmail.com> Message-ID: <5763969A.8040209@ubuntu.com> Le 17/06/2016 08:00, Alekseenko Vasilii a écrit : > What is the correct way to modify the file > prime/command-program.wrapper ? > I need to add their own variables and modify existing ones. > Hey Alekseenko, For now, so that you can snapcraft clean, build and so on again, you need to create another wrapper into your directory that you point from the "command:" stenza. This one can use the variables that prime/command-program.wrapper exports (it will be actually call by this one), so only file the delta of missing variables for you. In the near future, it's planned that snapcraft enable you to declare your own environment variables you want to add directly in snapcraft.yaml, which is going to remove this wrapper need. Cheers, Didier -------------- next part -------------- An HTML attachment was scrubbed... URL: From nhaines at ubuntu.com Fri Jun 17 07:57:44 2016 From: nhaines at ubuntu.com (Nathan Haines) Date: Fri, 17 Jun 2016 00:57:44 -0700 Subject: Trouble using sox in a snap. Message-ID: <5763AD78.6080908@ubuntu.com> Hello all, I'm having trouble running sox in a snap. I thought this one would be straightfoward now that we have a pulseaudio interface, but it turns out I'm pretty wrong. :) I was asked to file a bug, then encouraged to bring it up on the mailing list. The details are all in the bug, but basically I'm trying to run sox and generate whitenoise. The script works perfectly on a 16.04 LTS desktop install after running 'sudo apt install sox', but fails in a snap generated with the snapcraft.yaml file whether I use stage-packages: sox or various combinations of the additional packages in the snapcraft.yaml file. It doesn't work with --devmode, either. I attached a snap to the bug report, but one of the very happy features of snapcraft is that the snapcraft.yaml file and the "spacedrive" script should be enough to reproduce it. I suppose the only other thing of possible note is that I'm running snapcraft inside a xenial LXC container. Any help would be appreciated! I hope to upload this into the store eventually. It's *great* with headphones for getting work done! https://bugs.launchpad.net/bugs/1593558 -- Nathan Haines Ubuntu - http://www.ubuntu.com/ From seth.arnold at canonical.com Fri Jun 17 09:25:50 2016 From: seth.arnold at canonical.com (Seth Arnold) Date: Fri, 17 Jun 2016 12:25:50 +0300 Subject: difference between stage and prime? In-Reply-To: <355.1466101966@parc.com> References: <355.1466101966@parc.com> Message-ID: <20160617092550.GA12870@hunt> On Thu, Jun 16, 2016 at 11:32:46AM -0700, Bill Janssen wrote: > instance, it has a "home/wjanssen" directory tree, which is clearly > *not* something I want in the snap. There's also an "opt/goodstuff/" I believe the home/wjanssen part is: https://bugs.launchpad.net/ubuntu/+source/snapcraft/+bug/1592873 Thanks -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 473 bytes Desc: not available URL: From jamie at canonical.com Fri Jun 17 13:07:41 2016 From: jamie at canonical.com (Jamie Strandboge) Date: Fri, 17 Jun 2016 16:07:41 +0300 Subject: Uploading snaps to myapps.developer.ubuntu.com In-Reply-To: References: Message-ID: <1466168861.26796.84.camel@canonical.com> On Fri, 2016-06-17 at 13:18 +0800, Reinhard Pointner wrote: ... > I've uploaded a snap 24h ago (to beta and edge channels) and it's published > according to the website. But snap says it can't find the package (i.e. > `sudo snap install filebot --channel=edge`). > There is nothing up for manual review at this time. -- Jamie Strandboge | http://www.canonical.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: This is a digitally signed message part URL: From natalia.bidart at canonical.com Fri Jun 17 14:14:31 2016 From: natalia.bidart at canonical.com (Natalia) Date: Fri, 17 Jun 2016 11:14:31 -0300 Subject: Uploading snaps to myapps.developer.ubuntu.com In-Reply-To: <576395B0.4020409@canonical.com> References: <576395B0.4020409@canonical.com> Message-ID: On Fri, Jun 17, 2016 at 3:16 AM, Didier Roche wrote: > Le 17/06/2016 07:18, Reinhard Pointner a écrit : > > Hi snappers! > > > Hey Reinhard! > > > How long does it take for a snap to become visible via the `snap` command > after it has been uploaded and published via developer.ubuntu.com? > > I've uploaded a snap 24h ago (to beta and edge channels) and it's > published according to the website. But snap says it can't find the package > (i.e. `sudo snap install filebot --channel=edge`). > > Channel beta/edge don't appear in `snap find` afaik, but if I install by > name, shouldn't it work even if it's not in the public listing? It's set as > a paid package if that make a difference. > > It's normally instant (as soon as it says it's published, meaning that > checks pass, which are automatic if you don't use restricted interfaces). > I can't find your snap there either: > https://uappexplorer.com/apps?q=filebot&sort=relevance&type=snappy or > trying to install from the edge channel) > > It's either still in manual review mode (but the web interface is normally > telling you this) or something happened. I'm ccing Natalia who will be able > to check for this. > I checked in the store data and the snap is properly approved and published. But it uses "confinement: devmode", and there is still some work being done in the snap client to support installing devmode snaps directly from the store. So until that is completed, you will need to either sideload your snap, or submit a new revision with "confinement: strict" so "snap find" finds it. The devmode support in the snap client is almost ready, we expect it to be available for users by the end of next week. I hope this answers your question. Natalia. -------------- next part -------------- An HTML attachment was scrubbed... URL: From janssen at parc.com Fri Jun 17 15:36:49 2016 From: janssen at parc.com (Bill Janssen) Date: Fri, 17 Jun 2016 08:36:49 -0700 Subject: difference between stage and prime? In-Reply-To: <57639491.3080500@ubuntu.com> References: <355.1466101966@parc.com> <57639491.3080500@ubuntu.com> Message-ID: <30364.1466177809@parc.com> Didier Roche wrote: > To filter what you ship in the finale snap, you can use the "snap" > keyword (see same url above) and use something like: > parts: > foo: > snap: > - -home/wjanssen Ah, that's useful. Thanks. I'd seen that in the docs, but forgotten it. Of course, that means I need to predict what's going to be in the prime/ that I'll want to leave behind :-). Any idea why home/wjanssen is put in there? Bill From daniel.holbach at canonical.com Fri Jun 17 15:39:15 2016 From: daniel.holbach at canonical.com (Daniel Holbach) Date: Fri, 17 Jun 2016 17:39:15 +0200 Subject: Week 3 of the Snappy Playpen Message-ID: <576419A3.7090706@canonical.com> Hello everybody, The Snappy Playpen is hosted in github and we meet in both #snappy on Freenode and our gitter channel. We are hanging out there most of the time, but next week on Tuesday, 21st June we will get all experts in one room and together we will make a push to get both Electron based apps and server apps snapped. Obviously you can bring whatever own app you are interested in. Particularly if you are an upstream of a project, we're keen to help you get started. Snaps are a beautiful and simple way to get your app out to users, so let's make this happen together. If you are curious and want to take a first look, go to https://snapcraft.io and we'll take care of the questions together. WHAT: Snappy playpen sprint WHEN: Tuesday, 21st June 2016 all day WHERE: Join us on gitter or IRC Links: - https://gitter.im/ubuntu/snappy-playpen - https://github.com/ubuntu/snappy-playpen Have a great day, Daniel From mark at ubuntu.com Fri Jun 17 16:11:42 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Fri, 17 Jun 2016 17:11:42 +0100 Subject: Trouble using sox in a snap. In-Reply-To: <5763AD78.6080908@ubuntu.com> References: <5763AD78.6080908@ubuntu.com> Message-ID: <5764213E.8050403@ubuntu.com> On 17/06/16 08:57, Nathan Haines wrote: > I suppose the only other thing of possible note is that I'm running > snapcraft inside a xenial LXC container. I can say for sure that a snap won't install in a container - it's a limitation we'll address in this cycle - but have you tried outside the container? Mark From janssen at parc.com Fri Jun 17 16:49:22 2016 From: janssen at parc.com (Bill Janssen) Date: Fri, 17 Jun 2016 09:49:22 -0700 Subject: version of YAML being used in snapcraft.yaml? Message-ID: <32719.1466182162@parc.com> Is it 1.0, 1.1, or 1.2? Bill From janssen at parc.com Fri Jun 17 17:42:13 2016 From: janssen at parc.com (Bill Janssen) Date: Fri, 17 Jun 2016 10:42:13 -0700 Subject: accessing D-Bus Secret Service API from a snap? Message-ID: <1618.1466185333@parc.com> My application stores access credentials in the Secret Service. How do I enable access to this? Presumably this would be a plug or an interface (still confused on the distinction between the two of them). Or should I just install dbus-python into my Anaconda Python in my snap, and use it? Bill From mark at ubuntu.com Fri Jun 17 17:55:44 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Fri, 17 Jun 2016 18:55:44 +0100 Subject: accessing D-Bus Secret Service API from a snap? In-Reply-To: <1618.1466185333@parc.com> References: <1618.1466185333@parc.com> Message-ID: <576439A0.7080407@ubuntu.com> On 17/06/16 18:42, Bill Janssen wrote: > My application stores access credentials in the Secret Service. How do > I enable access to this? Presumably this would be a plug or an > interface (still confused on the distinction between the two of them). > Or should I just install dbus-python into my Anaconda Python in my snap, > and use it? > An "interface" is the name for a "type of connection". Think of it as a class or a type. The plug/slot are the consumer/provider, and they have names on both sides. Here's a (semi-hypothetical) example. An interface "disk" might be defined that is a way for a snap to claim a disk. Now the core snap could say it offers a 'disk interface' called 'directly-attached'. So that's a slot on the core snap called 'directly-attached' of type 'disk'. But you could also have a snap which is a client for an EMC SAN. That would have a slot called, say 'san' of type 'disk'. When you install the EMC snap on the system, there are thus two *slots* of interface 'disk', one that serves up directly attached disks on the machine, the other that serves up remote SAN disks. Now, imagine you have a snap which uses a disk. It would have a *plug* of the 'disk' interface (type), let's call it 'volume'. When you install the snap, you can 'connect the volume plug to the core:directly-attached or emc:san slot' Does this make sense? This is one of the many capabilities that really distinguish snappy from the things that have gone before. AppImage for example is an excellent desktop application container, but it wouldn't handle this sort of use case. snapd has heuistics for which of these plugs and slots can be auto-connected. For example, when you install a desktop app you are probably automatically connecting a "home" plug/slot/interface which gives that app access to most of your home directory but NOT access to things like ~/.ssh/ Was that useful? Mark From gustavo.niemeyer at canonical.com Fri Jun 17 18:11:32 2016 From: gustavo.niemeyer at canonical.com (Gustavo Niemeyer) Date: Fri, 17 Jun 2016 15:11:32 -0300 Subject: version of YAML being used in snapcraft.yaml? In-Reply-To: <32719.1466182162@parc.com> References: <32719.1466182162@parc.com> Message-ID: Hi Bill, We use multiple languages and multiple parsers, and I believe they use largely 1.1 and perhaps some details of 1.2. Is there anything in particular you are worried about? On Fri, Jun 17, 2016 at 1:49 PM, Bill Janssen wrote: > Is it 1.0, 1.1, or 1.2? > > Bill > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > -- gustavo @ http://niemeyer.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From janssen at parc.com Fri Jun 17 18:30:46 2016 From: janssen at parc.com (Bill Janssen) Date: Fri, 17 Jun 2016 11:30:46 -0700 Subject: accessing D-Bus Secret Service API from a snap? In-Reply-To: <576439A0.7080407@ubuntu.com> References: <1618.1466185333@parc.com> <576439A0.7080407@ubuntu.com> Message-ID: <2812.1466188246@parc.com> Mark Shuttleworth wrote: > Was that useful? Yes, thanks, Mark. Let me read that back: A "plug" goes in a "slot" to access the "interface" exported by the slot's snap. So, does this mean I have to write a snap to access the Secret Service API and provide it as a slot? And presumably also design an "interface" which that slot exports? Seems like a core kind of thing. Bill From evan.dandrea at canonical.com Fri Jun 17 18:51:04 2016 From: evan.dandrea at canonical.com (Evan Dandrea) Date: Fri, 17 Jun 2016 18:51:04 +0000 Subject: wrapper In-Reply-To: <5763969A.8040209@ubuntu.com> References: <57639206.3060109@gmail.com> <5763969A.8040209@ubuntu.com> Message-ID: Didier's suggestion of a wrapper script is a good one and the most straightforward. Another option, if you're using a custom plugin in parts/plugins/x-yourplugin.py, is to use the env() function to add environment variables. For example: https://github.com/ubuntu-core/snapcraft/blob/master/snapcraft/plugins/jdk.py#L26 -------------- next part -------------- An HTML attachment was scrubbed... URL: From kyle.fazzari at canonical.com Fri Jun 17 18:56:21 2016 From: kyle.fazzari at canonical.com (Kyle Fazzari) Date: Fri, 17 Jun 2016 14:56:21 -0400 Subject: difference between stage and prime? In-Reply-To: <20160617092550.GA12870@hunt> References: <355.1466101966@parc.com> <20160617092550.GA12870@hunt> Message-ID: <576447D5.1030903@canonical.com> On 06/17/2016 05:25 AM, Seth Arnold wrote: > On Thu, Jun 16, 2016 at 11:32:46AM -0700, Bill Janssen wrote: >> instance, it has a "home/wjanssen" directory tree, which is clearly >> *not* something I want in the snap. There's also an "opt/goodstuff/" > > I believe the home/wjanssen part is: > https://bugs.launchpad.net/ubuntu/+source/snapcraft/+bug/1592873 Which I believe is probably a dupe of: https://bugs.launchpad.net/snapcraft/+bug/1570895 I'm working on a fix for it as we speak. -- Kyle Fazzari (kyrofa) Software Engineer Canonical Ltd. kyle at canonical.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 836 bytes Desc: OpenPGP digital signature URL: From nhaines at ubuntu.com Fri Jun 17 20:24:56 2016 From: nhaines at ubuntu.com (Nathan Haines) Date: Fri, 17 Jun 2016 13:24:56 -0700 Subject: Trouble using sox in a snap. In-Reply-To: <5764213E.8050403@ubuntu.com> References: <5763AD78.6080908@ubuntu.com> <5764213E.8050403@ubuntu.com> Message-ID: <57645C98.7010000@ubuntu.com> On 06/17/2016 09:11 AM, Mark Shuttleworth wrote: > On 17/06/16 08:57, Nathan Haines wrote: >> I suppose the only other thing of possible note is that I'm running >> snapcraft inside a xenial LXC container. > > I can say for sure that a snap won't install in a container - it's a > limitation we'll address in this cycle - but have you tried outside the > container? Oh, I'm installing the snap on bare metal--just doing the snapcraft step inside the container. (I don't think that should make a difference, but I thought it worth mentioning because I know snaps don't run in containers.) -- Nathan Haines Ubuntu - http://www.ubuntu.com/ From david.calle at canonical.com Sat Jun 18 13:05:32 2016 From: david.calle at canonical.com (=?UTF-8?Q?David_Call=c3=a9?=) Date: Sat, 18 Jun 2016 15:05:32 +0200 Subject: Snap! Community weekly update: W24 Message-ID: <5765471C.2090708@canonical.com> Here is an update on the highlights in the snap community this week: We’ll start with a big shout out to everyone who has contributed to enable snapd on other distros: your support is amazing and whatever your distro of choice, company and areas of interest are, we’ll try to provide the resources for you to make the most of this effort. Thanks as well to everyone involved in building snapcraft.io . ## Snaps This week we ran another Snappy Playpen which was as exciting and busy as the first! We landed several new snaps examples: - Tyrant Unleashed Optimizer - mpv - Imagemagick6-stable - Keepassx - Consul - Dcos-cli - Deis workflow And a lot of them are in progress: - Fritzing - Gimp - Imagemagick7 - Sylpheed - Pidgin - TeXworks - MATE Desktop Thanks to all participants for this amazing feeling of working together, we are all learning a lot! If you are the upstream of one of the playpen snaps, get in touch on this list or on IRC #snappy, we’d love to hear your feedback and get you in the loop. You can find all the details of the event in the Playpen summary blog post by Daniel Holbach. ## Community highlights For this week’s highlights, we’d like to focus on a big piece of ongoing work in KDE. This link is probably the best way to show off what they are accomplishing, with no less than 18 KDE/Neon apps snapped or in progress, such as: - Digikam - Kate - Kdevelop - Kgpg Their READMEs are hinting at their status, so, do give them a try (clone repo, cd, snapcraft, snap install *.snap) and provide feedback. Furthermore David Edmundson has patched parts of KDE frameworks so that that KIO will work from inside of a snap. ## Documentation - Preparation of a ”docs blitz ” with a list of tagged bugs - Launched a snap FAQ based on top Ask Ubuntu questions - We have also started a roll of blog posts on snapd and snapcraft to keep you updated on the focus points of each release, keep an eye on dev.ubuntu.com/blog for more details! Cheers, David for the Community team -------------- next part -------------- An HTML attachment was scrubbed... URL: From janssen at parc.com Sat Jun 18 16:32:57 2016 From: janssen at parc.com (Bill Janssen) Date: Sat, 18 Jun 2016 09:32:57 -0700 Subject: version of YAML being used in snapcraft.yaml? In-Reply-To: References: <32719.1466182162@parc.com> Message-ID: <854.1466267577@parc.com> No, just trying to pin down details. I realize things are still in quite a bit of flux, but is there a written specification of this whole system somewhere? A design? Other than the code? Bill Gustavo Niemeyer wrote: > Hi Bill, > > We use multiple languages and multiple parsers, and I believe they use > largely 1.1 and perhaps some details of 1.2. > > Is there anything in particular you are worried about? > > On Fri, Jun 17, 2016 at 1:49 PM, Bill Janssen wrote: > > > Is it 1.0, 1.1, or 1.2? > > > > Bill > > > > -- > > Snapcraft mailing list > > Snapcraft at lists.ubuntu.com > > Modify settings or unsubscribe at: > > https://lists.ubuntu.com/mailman/listinfo/snapcraft > > > > > > -- > gustavo @ http://niemeyer.net From mark at ubuntu.com Sat Jun 18 17:53:48 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Sat, 18 Jun 2016 18:53:48 +0100 Subject: version of YAML being used in snapcraft.yaml? In-Reply-To: <854.1466267577@parc.com> References: <32719.1466182162@parc.com> <854.1466267577@parc.com> Message-ID: <57658AAC.4080502@ubuntu.com> On 18/06/16 17:32, Bill Janssen wrote: > No, just trying to pin down details. I realize things are still in > quite a bit of flux, but is there a written specification of this whole > system somewhere? A design? Other than the code? Most of what's written describes the whole behavior of an all-snap system. That comes from the origin of snaps inside Ubuntu Core. Now snaps as a standalone idea are spreading, I think the most useful thing we could create for developers is a detailed description of a single snap and the behavior of it. Snapcraft teaches you that indirectly, but I think a direct documentation of the snap is what's called for. Mark From ericoporto2008 at gmail.com Sun Jun 19 18:36:42 2016 From: ericoporto2008 at gmail.com (=?UTF-8?B?w4lyaWNvIFA=?=) Date: Sun, 19 Jun 2016 15:36:42 -0300 Subject: Is this mailing list working? Message-ID: Just a test. Thank you. -------------- next part -------------- An HTML attachment was scrubbed... URL: From kyle.fazzari at canonical.com Mon Jun 20 00:00:42 2016 From: kyle.fazzari at canonical.com (Kyle Fazzari) Date: Sun, 19 Jun 2016 20:00:42 -0400 Subject: Is this mailing list working? In-Reply-To: References: Message-ID: <5767322A.9070803@canonical.com> On 06/19/2016 02:36 PM, Érico P wrote: > Just a test. Thank you. Yes, we can hear you now! -- Kyle Fazzari (kyrofa) Software Engineer Canonical Ltd. kyle at canonical.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 836 bytes Desc: OpenPGP digital signature URL: From tsimonq2 at ubuntu.com Mon Jun 20 07:23:42 2016 From: tsimonq2 at ubuntu.com (Simon Quigley) Date: Mon, 20 Jun 2016 02:23:42 -0500 Subject: Snappy server source code? In-Reply-To: <1765962988.6315227.1466391838571.JavaMail.yahoo@mail.yahoo.com> References: <1765962988.6315227.1466391838571.JavaMail.yahoo.ref@mail.yahoo.com> <1765962988.6315227.1466391838571.JavaMail.yahoo@mail.yahoo.com> Message-ID: <576799FE.8080302@ubuntu.com> This is the best list to ask this: On 06/19/16 22:03, Raphael Costa wrote: > How would you go around creating your custom snap repository? I can't find the source code. > > Reasons are various, all other package managers in the wild have an open server. > -- Simon Quigley tsimonq2 at ubuntu.com tsimonq2 on Freenode From vasilisc777 at gmail.com Mon Jun 20 08:31:14 2016 From: vasilisc777 at gmail.com (Alekseenko Vasilii) Date: Mon, 20 Jun 2016 11:31:14 +0300 Subject: gtk app Message-ID: <5767A9D2.7080004@gmail.com> What is the right way in the packaging of the GTK program? Examples https://github.com/ubuntu/snappy-playpen/blob/hello-unity/hello-unity/snapcraft.yaml https://github.com/ubuntu/snappy-playpen/blob/master/galculator/snapcraft.yaml produce very big snap package. -- From mark at ubuntu.com Sun Jun 19 09:59:02 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Sun, 19 Jun 2016 10:59:02 +0100 Subject: accessing D-Bus Secret Service API from a snap? In-Reply-To: <2812.1466188246@parc.com> References: <1618.1466185333@parc.com> <576439A0.7080407@ubuntu.com> <2812.1466188246@parc.com> Message-ID: <57666CE6.8070703@ubuntu.com> On 17/06/16 19:30, Bill Janssen wrote: > Mark Shuttleworth wrote: > >> Was that useful? > Yes, thanks, Mark. Let me read that back: > > A "plug" goes in a "slot" to access the "interface" exported by the slot's snap. > > So, does this mean I have to write a snap to access the Secret Service > API and provide it as a slot? And presumably also design an "interface" > which that slot exports? Seems like a core kind of thing. Almost right. A "plug" and a "slot" must have the same "interface" to be connected at all. You can't connect a plug and a slot that have different interfaces. Perhaps we should call interfaces "shapes" - you would instinctively know you can only connect a plug to a slot of the same shape :) There is in progress an interface that allows select access to specific dbus namespaces. So for example, an interface that says "this snap wants to bind to the session dbus for org.gnome.foo.bar". If that would do for your purposes, then you have everything you need on the way. You'll just need to put the right plug in your snap and it will get connected automatically to the right slot, et voila. Chat with Zyga on IRC to get the details, I can't remember who was offering up that patch but Zyga would know. Mark From didrocks at ubuntu.com Mon Jun 20 09:23:18 2016 From: didrocks at ubuntu.com (Didier Roche) Date: Mon, 20 Jun 2016 11:23:18 +0200 Subject: gtk app In-Reply-To: <5767A9D2.7080004@gmail.com> References: <5767A9D2.7080004@gmail.com> Message-ID: <5767B606.7010101@ubuntu.com> Le 20/06/2016 10:31, Alekseenko Vasilii a écrit : > What is the right way in the packaging of the GTK program? Hey Alekseenko, > > Examples > https://github.com/ubuntu/snappy-playpen/blob/hello-unity/hello-unity/snapcraft.yaml > > https://github.com/ubuntu/snappy-playpen/blob/master/galculator/snapcraft.yaml > You are correct to have started there, especially the second link, where we are all working together and figuring out best practices on how to snap packages! > produce very big snap package. > Yeah, this is one of the tradeoff of snap packages. However, the net benefit is that you keep being in control of your own dependencies, that means that no of those will be able to break yourself in unattended fashion until you decide to upgrade them to a newer version. Cheers, Didier From alex_spataru at outlook.com Tue Jun 21 23:15:13 2016 From: alex_spataru at outlook.com (Alex Spataru) Date: Tue, 21 Jun 2016 18:15:13 -0500 Subject: Snapcraft and SDL2 issues Message-ID: Hello, I am creating a snap of my application, however, I am still having some issues: - My application uses SDL2 to get joystick input and generate sounds. When packaged as a snap, my application cannot access joystick information nor generate sound tones.- I launch several processes to obtain CPU usage and Battery status (which are later reported to the user). However, it seems that the "snapped" application cannot open external processes. For the moment, I use the following interfaces in my snap: - x11- network- pulseaudio- network-bind- system-observe- network-observe Here are some links that may be useful for anyone willing to help: - My snapcraft.yaml file: https://github.com/alex-spataru/snappy-playpen/blob/master/qdriverstation/snapcraft.yaml- A log of the files that my snapped application cannot read: http://paste.ubuntu.com/17591328/- My pull request to the playpen: https://github.com/ubuntu/snappy-playpen/pull/79 Is there something that I can change in my snapcraft.yaml file to fix my issue? Thanks in advance,--Alex Spataru -------------- next part -------------- An HTML attachment was scrubbed... URL: From alex_spataru at outlook.com Tue Jun 21 23:20:42 2016 From: alex_spataru at outlook.com (Alex Spataru) Date: Tue, 21 Jun 2016 18:20:42 -0500 Subject: Snapcraft and SDL2 issues In-Reply-To: References: Message-ID: Sorry for the previous message, it looks that my e-mail client messed up the text format... ### Original message I am creating a snap of my application, however, I am still having some issues: - My application uses SDL2 to get joystick input and generate sounds. When packaged as a snap, my application cannot access joystick information nor generate sound tones. - I launch several processes to obtain CPU usage and Battery status (which are later reported to the user). However, it seems that the "snapped" application cannot open external processes. For the moment, I use the following interfaces in my snap: - x11 - network - pulseaudio - network-bind - system-observe - network-observe Here are some links that may be useful for anyone willing to help: - My snapcraft.yaml file: https://github.com/alex-spataru/snappy-playpen/blob/master/qdriverstation/snapcraft.yaml - A log of the files that my snapped application cannot read: http://paste.ubuntu.com/17591328/ - My pull request to the playpen: https://github.com/ubuntu/snappy-playpen/pull/79 Is there something that I can change in my snapcraft.yaml file to fix my issue? Thanks in advance, --Alex Spataru From chris.willing at iinet.net.au Wed Jun 22 03:19:21 2016 From: chris.willing at iinet.net.au (Christoph Willing) Date: Wed, 22 Jun 2016 13:19:21 +1000 Subject: snapd without systemd? Message-ID: I'd like to run snapd on a machine without systemd. I've built all the code from https://github.com/snapcore/snapd after which the run-checks script seemed to run without error. However whenever I run snapd, the following error is returned: chris at d6:~/gowork/src$ sudo snapd Password: error: daemon does not handle 0 listeners right now, just one From the snapd.service and snapd.socket files, I gather that a listening socket needs to be set up at /run/snapd.socket. Can anyone advise if this is possible without systemd? Is it possible, for instance, to imitate the required listener socket functionality with the old inetd and, if so, any hints for an entry in inetd.conf & /etc/services? Thanks for any help with this, chris From zygmunt.krynicki at canonical.com Wed Jun 22 05:56:12 2016 From: zygmunt.krynicki at canonical.com (Zygmunt Krynicki) Date: Wed, 22 Jun 2016 07:56:12 +0200 Subject: snapd without systemd? In-Reply-To: References: Message-ID: <66C04E5D-9700-4363-8865-913D1AD4C4FF@canonical.com> > Wiadomość napisana przez Christoph Willing w dniu 22.06.2016, o godz. 05:19: > > I'd like to run snapd on a machine without systemd. I've built all the code from https://github.com/snapcore/snapd after which the run-checks script seemed to run without error. However whenever I run snapd, the following error is returned: Hi There are a number of things you’d have to change. Startup is just one of them. Systemd is used to handle services (background applications) as well as to mount squashfs filesystems on startup. You could certainly run snapd without systemd but those things would have to be tackled first with appropriate abstractions in the code. If you are interested in this then please let’s work together and make systemd an optional dependency. Best regards ZK From didrocks at ubuntu.com Wed Jun 22 06:07:37 2016 From: didrocks at ubuntu.com (Didier Roche) Date: Wed, 22 Jun 2016 08:07:37 +0200 Subject: Snapcraft and SDL2 issues In-Reply-To: References: Message-ID: <576A2B29.5020906@ubuntu.com> Le 22/06/2016 01:15, Alex Spataru a écrit : > Hello, Hey Alex! > > I am creating a snap of my application, however, I am still having > some issues: > > - My application uses SDL2 to get joystick input and generate sounds. > When packaged as a snap, my application cannot access joystick > information nor generate sound tones. Even in devmode? (in case you didn't start with this, you need to always try to install and develop your applicaiton in devmode, then you remove confinement once it works). Then, as you provided log for, "snappy-debug.security scanlog" is a good way to trap potential issues. * On sounds: from your logs, only /dev/shm/pulse-shm-3539825856 seems to be related to sound, Zygmunt, is there any known issue on this with the pulseaudio interface? * On joystick input: I guess that's some of the udev errors, do you have such interfaces planned? > - I launch several processes to obtain CPU usage and Battery status > (which are later reported to the user). However, it seems that the > "snapped" application cannot open external processes. Right, if those are accessing some part of the system outside of your confinement zone, it will. I think knowning what processes you are executed in which circumstances would help Jamie & Zygmunt (CCed here) to understand what kind of interfaces would be needed. Cheers, Didier > > For the moment, I use the following interfaces in my snap: > > - x11 > - network > - pulseaudio > - network-bind > - system-observe > - network-observe > > Here are some links that may be useful for anyone willing to help: > > - My /snapcraft.yaml > /file: https://github.com/alex-spataru/snappy-playpen/blob/master/qdriverstation/snapcraft.yaml > - A log of the files that my snapped application cannot > read: http://paste.ubuntu.com/17591328/ > - My pull request to the > playpen: https://github.com/ubuntu/snappy-playpen/pull/79 > > Is there something that I can change in my /snapcraft.yaml/ file to > fix my issue? > > Thanks in advance, > --Alex Spataru > > n -------------- next part -------------- An HTML attachment was scrubbed... URL: From vasilisc777 at gmail.com Wed Jun 22 06:18:45 2016 From: vasilisc777 at gmail.com (Alekseenko Vasilii) Date: Wed, 22 Jun 2016 09:18:45 +0300 Subject: snap home dir Message-ID: <576A2DC5.9020102@gmail.com> why program can't get listing /home/user/ directory? cat /var/log/syslog | grep audit Jun 22 08:56:42 vasilisc kernel: [ 248.550499] audit: type=1400 audit(1466575002.647:38): apparmor="DENIED" operation="open" profile="snap.languagetool.languagetool" name="/home/vasilisc/" pid=28374 comm="java" requested_mask="r" denied_mask="r" fsuid=1000 ouid=1000 Jun 22 08:56:52 vasilisc kernel: [ 258.066781] audit: type=1400 audit(1466575012.163:39): apparmor="DENIED" operation="open" profile="snap.languagetool.languagetool" name="/home/" pid=28675 comm="java" requested_mask="r" denied_mask="r" fsuid=1000 ouid=1000 =============== cat /snap/languagetool/current/meta/snap.yaml apps: languagetool: command: command-languagetool.wrapper plugs: - network - network-bind - x11 - home - unity7 architectures: - amd64 confinement: strict ---- From zygmunt.krynicki at canonical.com Wed Jun 22 06:23:17 2016 From: zygmunt.krynicki at canonical.com (Zygmunt Krynicki) Date: Wed, 22 Jun 2016 08:23:17 +0200 Subject: snap home dir In-Reply-To: <576A2DC5.9020102@gmail.com> References: <576A2DC5.9020102@gmail.com> Message-ID: <67AE4CA5-D909-4641-8552-E0B5906FE315@canonical.com> > Wiadomość napisana przez Alekseenko Vasilii w dniu 22.06.2016, o godz. 08:18: > > why program can't get listing /home/user/ directory? Hi Is the home interface connected? Best regards ZK From zygmunt.krynicki at canonical.com Wed Jun 22 06:26:28 2016 From: zygmunt.krynicki at canonical.com (Zygmunt Krynicki) Date: Wed, 22 Jun 2016 08:26:28 +0200 Subject: Snapcraft and SDL2 issues In-Reply-To: <576A2B29.5020906@ubuntu.com> References: <576A2B29.5020906@ubuntu.com> Message-ID: <8B7677B5-8214-41D4-ADA5-8F8D7AD3082B@canonical.com> > Wiadomość napisana przez Didier Roche w dniu 22.06.2016, o godz. 08:07: > > Le 22/06/2016 01:15, Alex Spataru a écrit : >> Hello, > > Hey Alex! >> >> I am creating a snap of my application, however, I am still having some issues: >> >> - My application uses SDL2 to get joystick input and generate sounds. When packaged as a snap, my application cannot access joystick information nor generate sound tones. > > Even in devmode? (in case you didn't start with this, you need to always try to install and develop your applicaiton in devmode, then you remove confinement once it works). > Then, as you provided log for, "snappy-debug.security scanlog" is a good way to trap potential issues. > > * On sounds: > from your logs, only /dev/shm/pulse-shm-3539825856 seems to be related to sound, Zygmunt, is there any known issue on this with the pulseaudio interface? I fixed two bugs related to sound yesterday. One affected all the systems other than ubuntu and debian. The other affected all systems. I suspect you may have to wait till …. today (today is the SRU day) to see this fixed. > * On joystick input: > I guess that's some of the udev errors, do you have such interfaces planned? Can you please file a bug and tag it with „snaps-interfaces” on http://bugs.launchpad.net/snappy > >> - I launch several processes to obtain CPU usage and Battery status (which are later reported to the user). However, it seems that the "snapped" application cannot open external processes. They can as long as they are inside the snap. You cannot rely on the host OS to have those applications. > > Right, if those are accessing some part of the system outside of your confinement zone, it will. I think knowning what processes you are executed in which circumstances would help Jamie & Zygmunt (CCed here) to understand what kind of interfaces would be needed. > > Cheers, > Didier > >> >> For the moment, I use the following interfaces in my snap: >> >> - x11 >> - network >> - pulseaudio >> - network-bind >> - system-observe >> - network-observe >> >> Here are some links that may be useful for anyone willing to help: >> >> - My snapcraft.yaml file: https://github.com/alex-spataru/snappy-playpen/blob/master/qdriverstation/snapcraft.yaml >> - A log of the files that my snapped application cannot read: http://paste.ubuntu.com/17591328/ >> - My pull request to the playpen: https://github.com/ubuntu/snappy-playpen/pull/79 >> >> Is there something that I can change in my snapcraft.yaml file to fix my issue? >> >> Thanks in advance, >> --Alex Spataru >> >> > n -------------- next part -------------- An HTML attachment was scrubbed... URL: From vasilisc777 at gmail.com Wed Jun 22 06:37:27 2016 From: vasilisc777 at gmail.com (Alekseenko Vasilii) Date: Wed, 22 Jun 2016 09:37:27 +0300 Subject: snap home Message-ID: <576A3227.1020605@gmail.com> Zygmunt, you have helped me a couple of months ago. My /home was link /home/vasilisc -> /mnt/datadisk/vasilisc I changed it for your advice mount | grep home /dev/sdc1 on /home type ext4 (rw,relatime,lazytime,nobarrier,commit=100,data=ordered) -- From vasilisc777 at gmail.com Wed Jun 22 06:43:55 2016 From: vasilisc777 at gmail.com (Alekseenko Vasilii) Date: Wed, 22 Jun 2016 09:43:55 +0300 Subject: slot type not equal plugs type Message-ID: <576A33AB.8000102@gmail.com> For example # snap install nmap 6.25 MB / 6.27 MB [======================================================================================================================================================================>_] 99.68 % 1.02 MB/s # snap interfaces Slot Plug :firewall-control - :home - :locale-control - :log-observe - :mount-observe - :network nmap :network-bind nmap :network-control - :network-observe - :opengl - :snapd-control - :system-observe - :timeserver-control - :timezone-control - :unity7 - :x11 - - nmap:network-control From zygmunt.krynicki at canonical.com Wed Jun 22 06:47:11 2016 From: zygmunt.krynicki at canonical.com (Zygmunt Krynicki) Date: Wed, 22 Jun 2016 08:47:11 +0200 Subject: slot type not equal plugs type In-Reply-To: <576A33AB.8000102@gmail.com> References: <576A33AB.8000102@gmail.com> Message-ID: <95BD3565-98DA-4E36-B3A8-73FF67E0CE54@canonical.com> > Wiadomość napisana przez Alekseenko Vasilii w dniu 22.06.2016, o godz. 08:43: > > For example This is expected. Both slots and plugs have names. In addition, they also have types. The type and the name can be the same (e.g. „network” but doesn’t have to be. The way the output looks like depends on the name. It is hidden if it is the same as interface name. In that case you will just see the snap name, as you can see below. > > # snap install nmap > 6.25 MB / 6.27 MB [======================================================================================================================================================================>_] 99.68 % 1.02 MB/s > > # snap interfaces > Slot Plug > :firewall-control - > :home - > :locale-control - > :log-observe - > :mount-observe - > :network nmap > :network-bind nmap > :network-control - > :network-observe - > :opengl - > :snapd-control - > :system-observe - > :timeserver-control - > :timezone-control - > :unity7 - > :x11 - > - nmap:network-control > > > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft From vasilisc777 at gmail.com Wed Jun 22 06:55:05 2016 From: vasilisc777 at gmail.com (Alekseenko Vasilii) Date: Wed, 22 Jun 2016 09:55:05 +0300 Subject: slot type not equal plugs type In-Reply-To: <95BD3565-98DA-4E36-B3A8-73FF67E0CE54@canonical.com> References: <576A33AB.8000102@gmail.com> <95BD3565-98DA-4E36-B3A8-73FF67E0CE54@canonical.com> Message-ID: <576A3649.8010309@gmail.com> On 22.06.2016 09:47, Zygmunt Krynicki wrote: > snap interfaces What do I need to make the right conclusion from all this? nmap do not have ubuntu-core:network-observe my snap do not have ubuntu-core:home -- С уважением, Алексеенко Василий http://vasilisc.com/ ---- From zygmunt.krynicki at canonical.com Wed Jun 22 06:56:52 2016 From: zygmunt.krynicki at canonical.com (Zygmunt Krynicki) Date: Wed, 22 Jun 2016 08:56:52 +0200 Subject: slot type not equal plugs type In-Reply-To: <576A3649.8010309@gmail.com> References: <576A33AB.8000102@gmail.com> <95BD3565-98DA-4E36-B3A8-73FF67E0CE54@canonical.com> <576A3649.8010309@gmail.com> Message-ID: <1E3061CC-5527-4BB3-AF93-95E0534EA19F@canonical.com> > Wiadomość napisana przez Alekseenko Vasilii w dniu 22.06.2016, o godz. 08:55: > > On 22.06.2016 09:47, Zygmunt Krynicki wrote: >> snap interfaces > > What do I need to make the right conclusion from all this? > nmap do not have ubuntu-core:network-observe No it doesn't > > my snap do not have ubuntu-core:home > No it doesn’t The output of „snap interfaces” consists of a list of connected interfaces followed by the list of disconnected interfaces. I don’t see your snap mentioned in home. > -- > С уважением, > Алексеенко Василий > http://vasilisc.com/ > ---- > From vasilisc777 at gmail.com Wed Jun 22 07:00:44 2016 From: vasilisc777 at gmail.com (Alekseenko Vasilii) Date: Wed, 22 Jun 2016 10:00:44 +0300 Subject: slot type not equal plugs type In-Reply-To: <1E3061CC-5527-4BB3-AF93-95E0534EA19F@canonical.com> References: <576A33AB.8000102@gmail.com> <95BD3565-98DA-4E36-B3A8-73FF67E0CE54@canonical.com> <576A3649.8010309@gmail.com> <1E3061CC-5527-4BB3-AF93-95E0534EA19F@canonical.com> Message-ID: <576A379C.2010300@gmail.com> On 22.06.2016 09:56, Zygmunt Krynicki wrote: > >> Wiadomość napisana przez Alekseenko Vasilii w dniu 22.06.2016, o godz. 08:55: >> >> On 22.06.2016 09:47, Zygmunt Krynicki wrote: >>> snap interfaces >> >> What do I need to make the right conclusion from all this? >> nmap do not have ubuntu-core:network-observe > > No it doesn't >> >> my snap do not have ubuntu-core:home >> > > No it doesn’t > > The output of „snap interfaces” consists of a list of connected interfaces followed by the list of disconnected interfaces. I don’t see your snap mentioned in home. > >> -- >> С уважением, >> Алексеенко Василий >> http://vasilisc.com/ >> ---- >> > Very strange # snap interfaces Slot Plug :network nmap :network-bind nmap :network-control - - nmap:network-control My additional command # sudo snap connect nmap:network-control ubuntu-core:network-control [/] Connect nmap:network-control to ubuntu-core:network-control # snap interfaces Slot Plug :network nmap :network-bind nmap :network-control nmap From vasilisc777 at gmail.com Wed Jun 22 07:05:45 2016 From: vasilisc777 at gmail.com (Alekseenko Vasilii) Date: Wed, 22 Jun 2016 10:05:45 +0300 Subject: slot type not equal plugs type In-Reply-To: <576A379C.2010300@gmail.com> References: <576A33AB.8000102@gmail.com> <95BD3565-98DA-4E36-B3A8-73FF67E0CE54@canonical.com> <576A3649.8010309@gmail.com> <1E3061CC-5527-4BB3-AF93-95E0534EA19F@canonical.com> <576A379C.2010300@gmail.com> Message-ID: <576A38C9.7030508@gmail.com> On 22.06.2016 10:00, Alekseenko Vasilii wrote: > On 22.06.2016 09:56, Zygmunt Krynicki wrote: >> >>> Wiadomość napisana przez Alekseenko Vasilii w >>> dniu 22.06.2016, o godz. 08:55: >>> >>> On 22.06.2016 09:47, Zygmunt Krynicki wrote: >>>> snap interfaces >>> >>> What do I need to make the right conclusion from all this? >>> nmap do not have ubuntu-core:network-observe >> >> No it doesn't >>> >>> my snap do not have ubuntu-core:home >>> >> >> No it doesn’t >> >> The output of „snap interfaces” consists of a list of connected >> interfaces followed by the list of disconnected interfaces. I don’t >> see your snap mentioned in home. >> >>> -- >>> С уважением, >>> Алексеенко Василий >>> http://vasilisc.com/ >>> ---- >>> >> > > Very strange > > # snap interfaces > Slot Plug > :network nmap > :network-bind nmap > :network-control - > - nmap:network-control > > My additional command > # sudo snap connect nmap:network-control ubuntu-core:network-control > > [/] Connect nmap:network-control to ubuntu-core:network-control > > # snap interfaces > Slot Plug > :network nmap > :network-bind nmap > :network-control nmap > Poor users =( # snap install languagetool 165.25 MB / 165.27 MB [==================================================================================================================================================================>_] 99.99 % 1.61 MB/s # snap interfaces Slot Plug :home - :network languagetool,nmap :network-bind languagetool,nmap :network-control nmap :unity7 languagetool :x11 languagetool - languagetool:home # sudo snap connect languagetool:home ubuntu-core:home [-] Connect languagetool:home to ubuntu-core:home # snap interfaces Slot Plug :home languagetool :network languagetool,nmap :network-bind languagetool,nmap :network-control nmap :unity7 languagetool :x11 languagetool I fixed problem, but situation very bad. -- From zygmunt.krynicki at canonical.com Wed Jun 22 07:07:10 2016 From: zygmunt.krynicki at canonical.com (Zygmunt Krynicki) Date: Wed, 22 Jun 2016 09:07:10 +0200 Subject: slot type not equal plugs type In-Reply-To: <576A38C9.7030508@gmail.com> References: <576A33AB.8000102@gmail.com> <95BD3565-98DA-4E36-B3A8-73FF67E0CE54@canonical.com> <576A3649.8010309@gmail.com> <1E3061CC-5527-4BB3-AF93-95E0534EA19F@canonical.com> <576A379C.2010300@gmail.com> <576A38C9.7030508@gmail.com> Message-ID: <643FEF34-6B3E-48C1-99DF-67BF5E13E332@canonical.com> > Wiadomość napisana przez Alekseenko Vasilii w dniu 22.06.2016, o godz. 09:05: > > On 22.06.2016 10:00, Alekseenko Vasilii wrote: >> On 22.06.2016 09:56, Zygmunt Krynicki wrote: >>> >>>> Wiadomość napisana przez Alekseenko Vasilii w >>>> dniu 22.06.2016, o godz. 08:55: >>>> >>>> On 22.06.2016 09:47, Zygmunt Krynicki wrote: >>>>> snap interfaces >>>> >>>> What do I need to make the right conclusion from all this? >>>> nmap do not have ubuntu-core:network-observe >>> >>> No it doesn't >>>> >>>> my snap do not have ubuntu-core:home >>>> >>> >>> No it doesn’t >>> >>> The output of „snap interfaces” consists of a list of connected >>> interfaces followed by the list of disconnected interfaces. I don’t >>> see your snap mentioned in home. >>> >>>> -- >>>> С уважением, >>>> Алексеенко Василий >>>> http://vasilisc.com/ >>>> ---- >>>> >>> >> >> Very strange >> >> # snap interfaces >> Slot Plug >> :network nmap >> :network-bind nmap >> :network-control - >> - nmap:network-control >> >> My additional command >> # sudo snap connect nmap:network-control ubuntu-core:network-control >> >> [/] Connect nmap:network-control to ubuntu-core:network-control >> >> # snap interfaces >> Slot Plug >> :network nmap >> :network-bind nmap >> :network-control nmap >> > Poor users =( > Which version of snapd are you running? The home interface is now auto-connecting. > # snap install languagetool > 165.25 MB / 165.27 MB [==================================================================================================================================================================>_] 99.99 % 1.61 MB/s > > # snap interfaces > Slot Plug > :home - > :network languagetool,nmap > :network-bind languagetool,nmap > :network-control nmap > :unity7 languagetool > :x11 languagetool > - languagetool:home > > > # sudo snap connect languagetool:home ubuntu-core:home > [-] Connect languagetool:home to ubuntu-core:home > > # snap interfaces > Slot Plug > :home languagetool > :network languagetool,nmap > :network-bind languagetool,nmap > :network-control nmap > :unity7 languagetool > :x11 languagetool > > I fixed problem, but situation very bad. > -- > From vasilisc777 at gmail.com Wed Jun 22 07:08:55 2016 From: vasilisc777 at gmail.com (Alekseenko Vasilii) Date: Wed, 22 Jun 2016 10:08:55 +0300 Subject: slot type not equal plugs type In-Reply-To: <643FEF34-6B3E-48C1-99DF-67BF5E13E332@canonical.com> References: <576A33AB.8000102@gmail.com> <95BD3565-98DA-4E36-B3A8-73FF67E0CE54@canonical.com> <576A3649.8010309@gmail.com> <1E3061CC-5527-4BB3-AF93-95E0534EA19F@canonical.com> <576A379C.2010300@gmail.com> <576A38C9.7030508@gmail.com> <643FEF34-6B3E-48C1-99DF-67BF5E13E332@canonical.com> Message-ID: <576A3987.6090909@gmail.com> On 22.06.2016 10:07, Zygmunt Krynicki wrote: > >> Wiadomość napisana przez Alekseenko Vasilii w dniu 22.06.2016, o godz. 09:05: >> >> On 22.06.2016 10:00, Alekseenko Vasilii wrote: >>> On 22.06.2016 09:56, Zygmunt Krynicki wrote: >>>> >>>>> Wiadomość napisana przez Alekseenko Vasilii w >>>>> dniu 22.06.2016, o godz. 08:55: >>>>> >>>>> On 22.06.2016 09:47, Zygmunt Krynicki wrote: >>>>>> snap interfaces >>>>> >>>>> What do I need to make the right conclusion from all this? >>>>> nmap do not have ubuntu-core:network-observe >>>> >>>> No it doesn't >>>>> >>>>> my snap do not have ubuntu-core:home >>>>> >>>> >>>> No it doesn’t >>>> >>>> The output of „snap interfaces” consists of a list of connected >>>> interfaces followed by the list of disconnected interfaces. I don’t >>>> see your snap mentioned in home. >>>> >>>>> -- >>>>> С уважением, >>>>> Алексеенко Василий >>>>> http://vasilisc.com/ >>>>> ---- >>>>> >>>> >>> >>> Very strange >>> >>> # snap interfaces >>> Slot Plug >>> :network nmap >>> :network-bind nmap >>> :network-control - >>> - nmap:network-control >>> >>> My additional command >>> # sudo snap connect nmap:network-control ubuntu-core:network-control >>> >>> [/] Connect nmap:network-control to ubuntu-core:network-control >>> >>> # snap interfaces >>> Slot Plug >>> :network nmap >>> :network-bind nmap >>> :network-control nmap >>> >> Poor users =( >> > > Which version of snapd are you running? The home interface is now auto-connecting. > > >> # snap install languagetool >> 165.25 MB / 165.27 MB [==================================================================================================================================================================>_] 99.99 % 1.61 MB/s >> >> # snap interfaces >> Slot Plug >> :home - >> :network languagetool,nmap >> :network-bind languagetool,nmap >> :network-control nmap >> :unity7 languagetool >> :x11 languagetool >> - languagetool:home >> >> >> # sudo snap connect languagetool:home ubuntu-core:home >> [-] Connect languagetool:home to ubuntu-core:home >> >> # snap interfaces >> Slot Plug >> :home languagetool >> :network languagetool,nmap >> :network-bind languagetool,nmap >> :network-control nmap >> :unity7 languagetool >> :x11 languagetool >> >> I fixed problem, but situation very bad. >> -- >> > # apt-cache policy snapd snapd: Installed: 2.0.2 Candidate: 2.0.2 Version table: *** 2.0.2 500 500 http://fi.archive.ubuntu.com/ubuntu devel/main amd64 Packages 100 /var/lib/dpkg/status # lsb_release -a No LSB modules are available. Distributor ID: Ubuntu Description: Ubuntu Yakkety Yak (development branch) Release: 16.10 Codename: yakkety -- From zygmunt.krynicki at canonical.com Wed Jun 22 07:09:55 2016 From: zygmunt.krynicki at canonical.com (Zygmunt Krynicki) Date: Wed, 22 Jun 2016 09:09:55 +0200 Subject: slot type not equal plugs type In-Reply-To: <576A3987.6090909@gmail.com> References: <576A33AB.8000102@gmail.com> <95BD3565-98DA-4E36-B3A8-73FF67E0CE54@canonical.com> <576A3649.8010309@gmail.com> <1E3061CC-5527-4BB3-AF93-95E0534EA19F@canonical.com> <576A379C.2010300@gmail.com> <576A38C9.7030508@gmail.com> <643FEF34-6B3E-48C1-99DF-67BF5E13E332@canonical.com> <576A3987.6090909@gmail.com> Message-ID: <54AC38EF-879B-44D2-9559-C1F46A433DF1@canonical.com> > Wiadomość napisana przez Alekseenko Vasilii w dniu 22.06.2016, o godz. 09:08: > > On 22.06.2016 10:07, Zygmunt Krynicki wrote: >> >>> Wiadomość napisana przez Alekseenko Vasilii w dniu 22.06.2016, o godz. 09:05: >>> >>> On 22.06.2016 10:00, Alekseenko Vasilii wrote: >>>> On 22.06.2016 09:56, Zygmunt Krynicki wrote: >>>>> >>>>>> Wiadomość napisana przez Alekseenko Vasilii w >>>>>> dniu 22.06.2016, o godz. 08:55: >>>>>> >>>>>> On 22.06.2016 09:47, Zygmunt Krynicki wrote: >>>>>>> snap interfaces >>>>>> >>>>>> What do I need to make the right conclusion from all this? >>>>>> nmap do not have ubuntu-core:network-observe >>>>> >>>>> No it doesn't >>>>>> >>>>>> my snap do not have ubuntu-core:home >>>>>> >>>>> >>>>> No it doesn’t >>>>> >>>>> The output of „snap interfaces” consists of a list of connected >>>>> interfaces followed by the list of disconnected interfaces. I don’t >>>>> see your snap mentioned in home. >>>>> >>>>>> -- >>>>>> С уважением, >>>>>> Алексеенко Василий >>>>>> http://vasilisc.com/ >>>>>> ---- >>>>>> >>>>> >>>> >>>> Very strange >>>> >>>> # snap interfaces >>>> Slot Plug >>>> :network nmap >>>> :network-bind nmap >>>> :network-control - >>>> - nmap:network-control >>>> >>>> My additional command >>>> # sudo snap connect nmap:network-control ubuntu-core:network-control >>>> >>>> [/] Connect nmap:network-control to ubuntu-core:network-control >>>> >>>> # snap interfaces >>>> Slot Plug >>>> :network nmap >>>> :network-bind nmap >>>> :network-control nmap >>>> >>> Poor users =( >>> >> >> Which version of snapd are you running? The home interface is now auto-connecting. >> >> >>> # snap install languagetool >>> 165.25 MB / 165.27 MB [==================================================================================================================================================================>_] 99.99 % 1.61 MB/s >>> >>> # snap interfaces >>> Slot Plug >>> :home - >>> :network languagetool,nmap >>> :network-bind languagetool,nmap >>> :network-control nmap >>> :unity7 languagetool >>> :x11 languagetool >>> - languagetool:home >>> >>> >>> # sudo snap connect languagetool:home ubuntu-core:home >>> [-] Connect languagetool:home to ubuntu-core:home >>> >>> # snap interfaces >>> Slot Plug >>> :home languagetool >>> :network languagetool,nmap >>> :network-bind languagetool,nmap >>> :network-control nmap >>> :unity7 languagetool >>> :x11 languagetool >>> >>> I fixed problem, but situation very bad. >>> -- >>> >> > > # apt-cache policy snapd > snapd: > Installed: 2.0.2 > Candidate: 2.0.2 Ah, yakkety is currently lagging behind because of some propagation issues. You can try to install 2.0.8 from proposed, it fixes a lot of issues like this. > Version table: > *** 2.0.2 500 > 500 http://fi.archive.ubuntu.com/ubuntu devel/main amd64 Packages > 100 /var/lib/dpkg/status > > # lsb_release -a > No LSB modules are available. > Distributor ID: Ubuntu > Description: Ubuntu Yakkety Yak (development branch) > Release: 16.10 > Codename: yakkety > > > -- > From didrocks at ubuntu.com Wed Jun 22 07:10:27 2016 From: didrocks at ubuntu.com (Didier Roche) Date: Wed, 22 Jun 2016 09:10:27 +0200 Subject: slot type not equal plugs type In-Reply-To: <576A38C9.7030508@gmail.com> References: <576A33AB.8000102@gmail.com> <95BD3565-98DA-4E36-B3A8-73FF67E0CE54@canonical.com> <576A3649.8010309@gmail.com> <1E3061CC-5527-4BB3-AF93-95E0534EA19F@canonical.com> <576A379C.2010300@gmail.com> <576A38C9.7030508@gmail.com> Message-ID: <576A39E3.6010108@ubuntu.com> Le 22/06/2016 09:05, Alekseenko Vasilii a écrit : > On 22.06.2016 10:00, Alekseenko Vasilii wrote: >> On 22.06.2016 09:56, Zygmunt Krynicki wrote: >>> >>>> Wiadomość napisana przez Alekseenko Vasilii w >>>> dniu 22.06.2016, o godz. 08:55: >>>> >>>> On 22.06.2016 09:47, Zygmunt Krynicki wrote: >>>>> snap interfaces >>>> >>>> What do I need to make the right conclusion from all this? >>>> nmap do not have ubuntu-core:network-observe >>> >>> No it doesn't >>>> >>>> my snap do not have ubuntu-core:home >>>> >>> >>> No it doesn’t >>> >>> The output of „snap interfaces” consists of a list of connected >>> interfaces followed by the list of disconnected interfaces. I don’t >>> see your snap mentioned in home. >>> >>>> -- >>>> С уважением, >>>> Алексеенко Василий >>>> http://vasilisc.com/ >>>> ---- >>>> >>> >> >> Very strange >> >> # snap interfaces >> Slot Plug >> :network nmap >> :network-bind nmap >> :network-control - >> - nmap:network-control >> >> My additional command >> # sudo snap connect nmap:network-control ubuntu-core:network-control >> >> [/] Connect nmap:network-control to ubuntu-core:network-control >> >> # snap interfaces >> Slot Plug >> :network nmap >> :network-bind nmap >> :network-control nmap >> > Poor users =( > > # snap install languagetool > 165.25 MB / 165.27 MB > [==================================================================================================================================================================>_] > 99.99 % 1.61 MB/s > > # snap interfaces > Slot Plug > :home - > :network languagetool,nmap > :network-bind languagetool,nmap > :network-control nmap > :unity7 languagetool > :x11 languagetool > - languagetool:home > > > # sudo snap connect languagetool:home ubuntu-core:home > [-] Connect languagetool:home to ubuntu-core:home > > # snap interfaces > Slot Plug > :home languagetool > :network languagetool,nmap > :network-bind languagetool,nmap > :network-control nmap > :unity7 languagetool > :x11 languagetool > > I fixed problem, but situation very bad. Hey Alekseenko, Some interfaces are autoconnecting (as you can see here, network, network-bind, unity7, x11. Some aren't (like home). We are revisiting the home interface to get it autoconnected for now until we have a great visual interfaces to prompt users about unconnected plugs/slots and ask their permissions to do this connection. This is a security feature, and why we have finer grain security in snaps than traditional packaging system, where the package has access to everything right away. Cheers, Didier From zygmunt.krynicki at canonical.com Wed Jun 22 09:39:26 2016 From: zygmunt.krynicki at canonical.com (Zygmunt Krynicki) Date: Wed, 22 Jun 2016 11:39:26 +0200 Subject: FYI: launchpad project for ubuntu-core-launcher was renamed to snap-confine Message-ID: Hey This is just a quick update to ensure nobody goes reporting 404 errors. The snap-confine project now lives on https://launchpad.net/snap-confine (for bugs) and https://github.com/snapcore/snap-confine (for code) Best regards ZK From vasilisc777 at gmail.com Wed Jun 22 09:40:53 2016 From: vasilisc777 at gmail.com (Alekseenko Vasilii) Date: Wed, 22 Jun 2016 12:40:53 +0300 Subject: confinement devmode Message-ID: <576A5D25.1030106@gmail.com> May I upload snap with "confinement: devmode" into Ubuntu Store? -- From zygmunt.krynicki at canonical.com Wed Jun 22 09:59:01 2016 From: zygmunt.krynicki at canonical.com (Zygmunt Krynicki) Date: Wed, 22 Jun 2016 11:59:01 +0200 Subject: confinement devmode In-Reply-To: <576A5D25.1030106@gmail.com> References: <576A5D25.1030106@gmail.com> Message-ID: <3372A097-62AC-4861-B737-7F578E2CEAC4@canonical.com> > Wiadomość napisana przez Alekseenko Vasilii w dniu 22.06.2016, o godz. 11:40: > > May I upload snap with "confinement: devmode" into Ubuntu Store? Hi Yes, but you cannot publish it in the stable channel. Just beta/edge ZK From didrocks at ubuntu.com Wed Jun 22 10:11:03 2016 From: didrocks at ubuntu.com (Didier Roche) Date: Wed, 22 Jun 2016 12:11:03 +0200 Subject: confinement devmode In-Reply-To: <576A5D25.1030106@gmail.com> References: <576A5D25.1030106@gmail.com> Message-ID: <576A6437.6050907@ubuntu.com> Le 22/06/2016 11:40, Alekseenko Vasilii a écrit : > May I upload snap with "confinement: devmode" into Ubuntu Store? Hey! Yeah, it's possible to upload it to the Ubuntu Store. However, it won't appear in the stable channel. Also, be aware of the current limitations that's going to be lifted pretty soon. I'm going to quote Natalia on this (as per https://lists.ubuntu.com/archives/snapcraft/2016-June/000235.html): "[…] it uses "confinement: devmode", and there is still some work being done in the snap client to support installing devmode snaps directly from the store. So until that is completed, you will need to either sideload your snap, or submit a new revision with "confinement: strict" so "snap find" finds it. The devmode support in the snap client is almost ready, we expect it to be available for users by the end of next week." Cheers, Didier From john.lenton at canonical.com Wed Jun 22 10:36:20 2016 From: john.lenton at canonical.com (John Lenton) Date: Wed, 22 Jun 2016 11:36:20 +0100 Subject: confinement devmode In-Reply-To: <576A6437.6050907@ubuntu.com> References: <576A5D25.1030106@gmail.com> <576A6437.6050907@ubuntu.com> Message-ID: for the record, the initial work to get --devmode to the store is on master already so should be going out in today's release. There's still a lot of work to do, but that bit is done. On 22 June 2016 at 11:11, Didier Roche wrote: > Le 22/06/2016 11:40, Alekseenko Vasilii a écrit : >> May I upload snap with "confinement: devmode" into Ubuntu Store? > > Hey! > > Yeah, it's possible to upload it to the Ubuntu Store. However, it won't > appear in the stable channel. > Also, be aware of the current limitations that's going to be lifted > pretty soon. I'm going to quote Natalia on this (as per > https://lists.ubuntu.com/archives/snapcraft/2016-June/000235.html): > > "[…] it uses "confinement: devmode", and there is still some work being > done in the snap client to support installing devmode snaps directly > from the store. So until that is completed, you will need to either > sideload your snap, or submit a new revision with "confinement: strict" > so "snap find" finds it. The devmode support in the snap client is > almost ready, we expect it to be available for users by the end of next > week." > > Cheers, > Didier > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft From gustavo.niemeyer at canonical.com Wed Jun 22 12:57:14 2016 From: gustavo.niemeyer at canonical.com (Gustavo Niemeyer) Date: Wed, 22 Jun 2016 09:57:14 -0300 Subject: Snappy server source code? In-Reply-To: <576799FE.8080302@ubuntu.com> References: <1765962988.6315227.1466391838571.JavaMail.yahoo.ref@mail.yahoo.com> <1765962988.6315227.1466391838571.JavaMail.yahoo@mail.yahoo.com> <576799FE.8080302@ubuntu.com> Message-ID: Hey Simon, Raphael, To move a bit faster we've based the current backend on infrastructure we already had live to support the Ubuntu phone. It's all the same web of services, effectively, so a bit hard to just take out and re-deploy. That said, we really have just a few entry points on it right now, which means it shouldn't be too hard to implement an alternative. In fact, we have been discussing developing a new one from the ground up to support tests more easily. If you are interested on this, we can discuss further and see how we might get started. On Mon, Jun 20, 2016 at 4:23 AM, Simon Quigley wrote: > This is the best list to ask this: > > On 06/19/16 22:03, Raphael Costa wrote: > > How would you go around creating your custom snap repository? I can't > find the source code. > > > > Reasons are various, all other package managers in the wild have an open > server. > > > > -- > Simon Quigley > tsimonq2 at ubuntu.com > tsimonq2 on Freenode > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > -- gustavo @ http://niemeyer.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From vasilisc777 at gmail.com Wed Jun 22 13:17:45 2016 From: vasilisc777 at gmail.com (Alekseenko Vasilii) Date: Wed, 22 Jun 2016 16:17:45 +0300 Subject: snapd in lubuntu Message-ID: <576A8FF9.3040907@gmail.com> Sorry for my English. My first snap package is LanguageTool 3.4 (java app). I tested the snap only on my computer (Ubuntu 16.10 + Unity7). I upload snap into Ubuntu Store without a thorough test. I create virtual machines (Xubuntu, Kubuntu, Ubuntu GNOME, Lubuntu, MATE) and only Lubuntu 16.04 displays an error. URL screenshot http://itmages.ru/image/view/4456402/27529211 In Lubuntu 16.04 languagetool wants open file for write /home/vasilisc/.config/LanguageTool/gui.state BUT this impossible. standard wrapper command-languagetool.wrapper -------------------------- #!/bin/sh export PATH="$SNAP/bin:$SNAP/usr/bin:$PATH" export LD_LIBRARY_PATH="$LD_LIBRARY_PATH:$SNAP/lib:$SNAP/usr/lib:$SNAP/lib/x86_64-linux-gnu:$SNAP/usr/lib/x86_64-linux-gnu" export LD_LIBRARY_PATH="$SNAP/usr/lib/x86_64-linux-gnu/mesa:$LD_LIBRARY_PATH" export JAVA_HOME=$SNAP/usr/lib/jvm/default-java export PATH=$SNAP/usr/lib/jvm/default-java/bin:$SNAP/usr/lib/jvm/default-java/jre/bin:$PATH export LD_LIBRARY_PATH="$SNAP$SNAP/usr/lib/jvm/java-8-openjdk-amd64/bin/../lib/amd64/jli:$SNAP$SNAP/usr/lib/jvm/java-8-openjdk-amd64/jre/bin/../lib/amd64/jli:$SNAP$SNAP/usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64:$SNAP$SNAP/usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/jli:$SNAP/usr/lib/x86_64-linux-gnu:$LD_LIBRARY_PATH" LD_LIBRARY_PATH=$SNAP_LIBRARY_PATH:$LD_LIBRARY_PATH exec "$SNAP/usr/bin/run.sh" "$@" -------------------------- Call my run.sh #!/bin/sh # Not good, needed for fontconfig export XDG_DATA_HOME=$SNAP/usr/share # Font Config export FONTCONFIG_PATH=$SNAP/etc/fonts/config.d export FONTCONFIG_FILE=$SNAP/etc/fonts/fonts.conf export HOME=$SNAP_USER_DATA java -jar -Duser.home=$SNAP_USER_DATA $SNAP/usr/bin/languagetool.jar -------------------------- My run.sh define new HOME and user.home = $SNAP_USER_DATA How can we explain this behavior Lubuntu (version snapd 2.0.8)? From alex_spataru at outlook.com Wed Jun 22 16:20:37 2016 From: alex_spataru at outlook.com (Alex Spataru) Date: Wed, 22 Jun 2016 11:20:37 -0500 Subject: Snapcraft and SDL2 issues In-Reply-To: References: Message-ID: Hello Didier, Thanks for the information! I will try your suggestions and report back. I am using the following processes to obtain information about the system: - grep - awk - upower - bash (to launch the other processes) If you need to see how I use them, please check this code:         https://github.com/WinT-3794/QDriverStation/blob/master/src/utilities.cpp#L44 As a side-note, I don't know if its worth the hassle of creating an interface that allows applications to access these processes. I am sure that I can find other (and better) ways to get information regarding the CPU and Battery states, without the need of using an external program to do that. Thanks, —Alex Spataru From alex_spataru at outlook.com Wed Jun 22 21:37:50 2016 From: alex_spataru at outlook.com (Alex Spataru) Date: Wed, 22 Jun 2016 16:37:50 -0500 Subject: Updates regarding SDL2 issues In-Reply-To: References: , Message-ID: Hello, Setting the confinement of the package to "devmode" fixed all issues with SDL (Joystick & audio work now). However, the application is still not able to launch the processes used to get CPU & Battery information. As far as I know, I can obtain CPU information by reading /proc/stat, however, I am not sure if the security model would allow my application to do that. I guess that reading battery state will be similar to obtaining CPU information, in the sense that I should read some file in the system (in the case that snap allows me to). Thanks for your time, Alex Spataru -------------- next part -------------- An HTML attachment was scrubbed... URL: From didrocks at ubuntu.com Thu Jun 23 05:43:59 2016 From: didrocks at ubuntu.com (Didier Roche) Date: Thu, 23 Jun 2016 07:43:59 +0200 Subject: Updates regarding SDL2 issues In-Reply-To: References: Message-ID: <576B771F.5040204@ubuntu.com> Le 22/06/2016 23:37, Alex Spataru a écrit : > Hello, Setting the confinement of the package to "devmode" fixed all > issues with SDL (Joystick & audio work now). However, the application > is still not able to launch the processes used to get CPU & Battery > information. As far as I know, I can obtain CPU information by reading > /proc/stat, however, I am not sure if the security model would allow > my application to do that. I guess that reading battery state will be > similar to obtaining CPU information, in the sense that I should read > some file in the system (in the case that snap allows me to). Thanks > for your time, Alex Spataru > > As Zygmunt told you on the other thread, please file a bug and tag it with „snaps-interfaces” on http://bugs.launchpad.net/snappy listing the files you need to access in /proc. Cheers, Didier -------------- next part -------------- An HTML attachment was scrubbed... URL: From jamie at canonical.com Thu Jun 23 13:33:35 2016 From: jamie at canonical.com (Jamie Strandboge) Date: Thu, 23 Jun 2016 08:33:35 -0500 Subject: Updates regarding SDL2 issues In-Reply-To: <576B771F.5040204@ubuntu.com> References: <576B771F.5040204@ubuntu.com> Message-ID: <1466688815.5164.5.camel@canonical.com> On Thu, 2016-06-23 at 07:43 +0200, Didier Roche wrote: > Le 22/06/2016 23:37, Alex Spataru a écrit : > > > > Hello, Setting the confinement of the package to "devmode" fixed all > > issues with SDL (Joystick & audio work now). However, the application > > is still not able to launch the processes used to get CPU & Battery > > information. As far as I know, I can obtain CPU information by reading > > /proc/stat, however, I am not sure if the security model would allow > > my application to do that. I guess that reading battery state will be > > similar to obtaining CPU information, in the sense that I should read > > some file in the system (in the case that snap allows me to). Thanks > > for your time, Alex Spataru > > > > > As Zygmunt told you on the other thread, please file a bug and tag it > with „snaps-interfaces” on http://bugs.launchpad.net/snappy listing the > files you need to access in /proc. > Correction: please use the 'snapd-interface' tag -- Jamie Strandboge | http://www.canonical.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: This is a digitally signed message part URL: From mark at ubuntu.com Thu Jun 23 13:53:15 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Thu, 23 Jun 2016 14:53:15 +0100 Subject: Snappy server source code? In-Reply-To: <576799FE.8080302@ubuntu.com> References: <1765962988.6315227.1466391838571.JavaMail.yahoo.ref@mail.yahoo.com> <1765962988.6315227.1466391838571.JavaMail.yahoo@mail.yahoo.com> <576799FE.8080302@ubuntu.com> Message-ID: <576BE9CB.8010702@ubuntu.com> Hi folks Since this question has come up a few times and the common answer keeps missing the point, let me have a go. The snap *format* is not intrinsically tied to a store. You can stand up a snap on a system regardless of how it arrived at that system. So the current store implementation is not particularly relevant, and would not be a good starting point. The simplest approach would be to focus on delivering a snap to a system over HTTPS. Since there are no complex dependency maps, you don't need the same sort of sophisticated infrastructure that APT or Debs or RPM do, you just need a webserver and wget. I think we should do this in the snap code itself so it is more obvious to people, because folks are hung up about the delivery of snaps from Ubuntu. I would not expect other distros to want to fetch snaps from Ubuntu unless there were useful snaps for them there, snaps could easily be served from Debian.org. In a sense, snaps are being punished for being ahead - of course there is a sophisticated store in Ubuntu, we've been doing mobile and IoT and commercial stores for several years. But that's no reason to denigrate snaps themselves, quite the opposite. Mark From scarlett.gately.clark at gmail.com Thu Jun 9 19:36:27 2016 From: scarlett.gately.clark at gmail.com (Scarlett Clark) Date: Thu, 9 Jun 2016 12:36:27 -0700 Subject: Snap! Community weekly update: W23 In-Reply-To: References: <575993F7.9070408@canonical.com> Message-ID: On Jun 9, 2016 12:28, "Fabio Colella" wrote: > > Very glad the Ubuntu Kylin icon theme snap was appreciated 😊 > > On 9 June 2016 at 18:06, David Callé wrote: >> >> Here is an update on the highlights in the snap community this week: >> >> ## Snap examples >> >> - The first snappy Playpen event was a blast! With more than 40 people participating and a dozen of new branches, we felt it was a great learning process for all involved and we gathered many new examples: >> - Leafpad >> - SMPlayer >> - Galculator >> - Plank >> - Tinyproxy >> - Heroku CLI >> - And a creativity winner: a snap for the Ubuntu Kylin icon theme, which bundles commands to enable/disable the theme >> >> There are more branches still being worked on (for example a git build of GIMP!). Check out the full details in our full recap of the Snappy Playpen kickoff. >> >> ## Documentation >> >> - General updates along the new snapd and snapcraft releases >> - guides/interfaces >> - guides/autoupdate >> - Ongoing content review with the Web Team >> >> ## Community highlights >> >> - Vincent Jobard has posted a video tutorial in French about Snapcraft >> - Jean-Marie Verdun wrote a blog post for the French-speaking community about snapping FreeCAD >> - The Mycroft team is working on snapping their core AI engine and their Text-to-Speech Mimic engine Another community highlight worth mentioning is KDE snapification is in full swing. Cheers Scarlett >> >> Cheers, >> David >> >> -- >> Snapcraft mailing list >> Snapcraft at lists.ubuntu.com >> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft >> > > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft > -------------- next part -------------- An HTML attachment was scrubbed... URL: From jenny.murphy at episensor.com Tue Jun 14 11:11:42 2016 From: jenny.murphy at episensor.com (Jenny Murphy) Date: Tue, 14 Jun 2016 12:11:42 +0100 Subject: Help with ant plugin and copy plugin Message-ID: Hi, I am trying to build up some competency with snapcraft for java applications. Currently I have a hello world project which is based on https://github.com/ubuntu-core/snapcraft/blob/1.x/examples/java-hello-world/snapcraft.yaml . I have extended in that HelloWorld.java references a method from another package in an external jar file (HellowWorld-helper.jar) I edited the snapcraft.yaml to use the copy plugin to copy the jar file to the jar directory. It can be seen there after the snap stage. However in bin.wrapper.wrapper , the external jar file is not added to the classpath. So the application fails on running. How do I fix this? Jenny -- *Jenny Murphy* *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* jenny.murphy at episensor.com t | +353 (0) 61 512 511 w | http://www.episensor.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From max at kristenonline.de Thu Jun 23 18:34:06 2016 From: max at kristenonline.de (Max) Date: Thu, 23 Jun 2016 20:34:06 +0200 Subject: Snappy server source code? In-Reply-To: <576BE9CB.8010702@ubuntu.com> References: <576BE9CB.8010702@ubuntu.com> Message-ID: <576C2B9E.5000505@kristenonline.de> Hello, So, if an independent Distro wanted to supply their users with their own appstore they could just set up a webserver an fork the Ubuntu-appstore? Would this even be legal, or could you only put the FreeSoftware Apps in your store? And how would you make sure they are the correct Binaries, build from the correct source? In LP you can build "reproducable" Snaps, but if you would build an independent appstore, you would have to find all the different locations where the developers put their source and the corresponding .yaml files to be sure that your Snaps correctly and veritably build. Wouldn't it make sense for snaps under an OS Licenses to be obliged by snapcraft to upload their .yaml files, so someone could verify they pull from the right repo, e.g github? Or maybe upload the whole source code like in Debian, to avoid fork-ability issues. The Snap store is a really great distribution tool for proprietary apps ATM , but OS apps have slightly other requirements, and not being forced to upload your source code, could lead potentially to bad practice or even malware being included in snaps. Max PS: sry if you are seeing this twice -------------- next part -------------- An HTML attachment was scrubbed... URL: From max at kristenonline.de Thu Jun 23 18:23:05 2016 From: max at kristenonline.de (Max) Date: Thu, 23 Jun 2016 20:23:05 +0200 Subject: Snappy server source code? Message-ID: <576C2909.4090504@kristenonline.de> Hello, So, if an independent Distro, wanted to supply their users with their own appstore they could just set up a webserver an fork the Ubuntu-appstore? Would this even be legal, or could you only put the FreeSoftware Apps in your store? And how would you make sure they are the correct Binaries, build from the correct source? In LP you can build "reproducable" Snaps, but if you would build an independent appstore, you would have to find all the different locations where the developers put their source and the corresponding .yaml files to be sure that your Snaps correctly and veritably build. Wouldn't it make sense for snaps under an OS Licenses to be obliged by snapcraft to upload their .yaml files, so someone could verify they pull from the right repo, e.g github? Or maybe upload the whole source code like in Debian, to avoid fork-ability issues. The Snap store is a really great distribution tool for proprietary apps ATM , but OS apps have slightly other requirements, and not being forced to upload your source code, could lead potentially to bad practice or even malware being included in snaps. Max From mark at ubuntu.com Fri Jun 24 12:27:41 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Fri, 24 Jun 2016 13:27:41 +0100 Subject: Snappy server source code? In-Reply-To: <576C2909.4090504@kristenonline.de> References: <576C2909.4090504@kristenonline.de> Message-ID: <576D273D.8070707@ubuntu.com> On 23/06/16 19:23, Max wrote: > So, if an independent Distro, wanted to supply their users with their > own appstore > they could just set up a webserver an fork the Ubuntu-appstore? > Would this even be legal, or could you only put the FreeSoftware Apps > in your store? Yes it would be legal from the point of view of snap code. You might need the snap creator's permission, depending on the license of the source but there's every reason to expect the people who make snaps to want them in as many stores as possible. It would be your decision as to whether you want non-free snaps in your store; fine for most distros, but folks like FSF and Debian would likely filter snaps based on license. > > And how would you make sure they are the correct Binaries, build from the > correct source? In LP you can build "reproducable" Snaps, but if you > would build > an independent appstore, you would have to find all the different > locations where > the developers put their source and the corresponding .yaml files to > be sure that > your Snaps correctly and veritably build. This is in the nature of all bundled app formats. For open source snaps, you would likely want to build them yourself. > > Wouldn't it make sense for snaps under an OS Licenses to be obliged by > snapcraft > to upload their .yaml files, so someone could verify they pull from > the right repo, e.g > github? Or maybe upload the whole source code like in Debian, to avoid > fork-ability issues. Yes indeed, and that's what we see happening. Drop a snapcraft.yaml in your source tree and anybody / everybody can then build the snap for themselves on hardware they trust. > The Snap store is a really great distribution tool for proprietary > apps ATM , but OS apps have slightly > other requirements, and not being forced to upload your source code, > could lead potentially > to bad practice or even malware being included in snaps. Yes, agreed. Mark From daniel.holbach at canonical.com Fri Jun 24 12:28:19 2016 From: daniel.holbach at canonical.com (Daniel Holbach) Date: Fri, 24 Jun 2016 14:28:19 +0200 Subject: Pulling knowledge out of Snappy Playpen Message-ID: <576D2763.8060906@canonical.com> Hello everybody, the Snappy Playpen [1] has been up and running for about three weeks now and we've had many pieces of software added to it (and lots more in progress). I'm quite happy that after a first rush of new snaps, we are still seeing new snaps being added and since a few days we are seeing a steady stream of improvements and some consolidation happening in terms of common launchers, etc. and more generally a common understanding of problems. So I thought at this stage it would make sense to take a closer look at what we could already learn from the snaps in the repository. Here is a quick document I put together: https://github.com/ubuntu/snappy-playpen/wiki/Examples The feedback we received so far is that it's quite useful. Bearing in mind that some of the code examples will change over time, what do you think where we could host documentation like this or how could we make it more official? Have a great day, Daniel From mark at ubuntu.com Fri Jun 24 12:39:03 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Fri, 24 Jun 2016 13:39:03 +0100 Subject: Pulling knowledge out of Snappy Playpen In-Reply-To: <576D2763.8060906@canonical.com> References: <576D2763.8060906@canonical.com> Message-ID: <576D29E7.50705@ubuntu.com> On 24/06/16 13:28, Daniel Holbach wrote: > Bearing in mind that some of the code examples will change over time, > what do you think where we could host documentation like this or how > could we make it more official? This is a great bit of insight Daniel, we should definitely ensure a flow of this knowledge into the standard snapcraft documentation, faq's and website. Mark From jamie.bennett at canonical.com Fri Jun 24 12:42:50 2016 From: jamie.bennett at canonical.com (Jamie Bennett) Date: Fri, 24 Jun 2016 08:42:50 -0400 Subject: Pulling knowledge out of Snappy Playpen In-Reply-To: <576D2763.8060906@canonical.com> References: <576D2763.8060906@canonical.com> Message-ID: <576D2ACA.4030700@canonical.com> On 24/06/16 08:28, Daniel Holbach wrote: > Hello everybody, > > the Snappy Playpen [1] has been up and running for about three weeks now > and we've had many pieces of software added to it (and lots more in > progress). I'm quite happy that after a first rush of new snaps, we are > still seeing new snaps being added and since a few days we are seeing a > steady stream of improvements and some consolidation happening in terms > of common launchers, etc. and more generally a common understanding of > problems. > > So I thought at this stage it would make sense to take a closer look at > what we could already learn from the snaps in the repository. Here is a > quick document I put together: > > https://github.com/ubuntu/snappy-playpen/wiki/Examples > > > The feedback we received so far is that it's quite useful. This is super useful, thanks for putting it together Daniel. > Bearing in mind that some of the code examples will change over time, > what do you think where we could host documentation like this or how > could we make it more official? > > Have a great day, > Daniel > From ogra at ubuntu.com Fri Jun 24 13:02:13 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Fri, 24 Jun 2016 15:02:13 +0200 Subject: Snap! Community weekly update: W23 In-Reply-To: References: <575993F7.9070408@canonical.com> Message-ID: <1466773333.4177.1.camel@ubuntu.com> hi, Am Donnerstag, den 09.06.2016, 12:36 -0700 schrieb Scarlett Clark: >  > Another community highlight worth mentioning is KDE snapification is > in full swing.  > Cheers Scarlett  \o/ ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From mark at ubuntu.com Fri Jun 24 13:18:52 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Fri, 24 Jun 2016 14:18:52 +0100 Subject: Snap! Community weekly update: W23 In-Reply-To: References: <575993F7.9070408@canonical.com> Message-ID: <576D333C.5060603@ubuntu.com> On 09/06/16 20:36, Scarlett Clark wrote: > Another community highlight worth mentioning is KDE snapification is > in full swing. That's great news! Do you have a top 5 list of feature requests that would help the community in that regard? I know the content-sharing interface, which would let you publish a standard set of KDE framework libraries to be shared, is well in progress. What else would be a help? Mark From david.calle at canonical.com Fri Jun 24 14:41:27 2016 From: david.calle at canonical.com (=?UTF-8?Q?David_Call=c3=a9?=) Date: Fri, 24 Jun 2016 16:41:27 +0200 Subject: Snap! Community weekly update: W25 Message-ID: <576D4697.1030004@canonical.com> Here is an update on the highlights in the snap community this week. Make sure you follow the new Snapcraft social channels (twitter ,google+ ,facebook ) for daily updates. ## Snaps After the third Snappy Playpen event on tuesday, it’s time to take a step back and see what we have accomplished so far as a community. We have reached 30 examples in the playpen: command-line and GUI apps, using GTK, Qt, written in Python, Go, Perl, built with cmake, autotools, etc. The events, that were originally planned as weekly hack days, now stay active all the time, with 60 people on the gitter chat and 29 members who have contributed to the github repo . We have decided to open the wiki and started on: * Documenting best practices * Fleshing out a snapcraft template for new contributions * Having a dedicated page to keep track of the general progress We are also working on documenting best practices for syntax and snapcraft.yaml organization, findings of the community are applied progressively to all examples. Therefore, if you are looking for guidance to snap an app, there is probably already an example covering your custom build / launcher / env variables use case and someone knowledgeable about it. The easiest way to get in touch is to come to the gitter chat . ## Community highlights * Let’s have a big shout-out to Andy Keech for keeping a watchful eye on most of the snaps in the Playpen and applying guidelines and best practices to a lot of them, making him the top committer on the project! * The renowned Arch Wiki now has a Snapd page . Thanks to the russian and japanese translators who have started localizing it. * And last but not least, LimeSDR, the snappy based software-defined radio platform has reached its crowdfunding goal . ## Reading list * Installing Nextcloud can be a snap by kyrofa, a series of blog posts detailing the creation of the Nextcloud snap * Snapd 2.0.9: full snap confinement on Elementary 0.4 on the devportal blog * Leveling up snapd integration tests by niemeyer * Making your first contribution to snapd by zyga Cheers, David -------------- next part -------------- An HTML attachment was scrubbed... URL: From mhall119 at ubuntu.com Fri Jun 24 16:44:18 2016 From: mhall119 at ubuntu.com (Michael Hall) Date: Fri, 24 Jun 2016 12:44:18 -0400 Subject: Snap! Community weekly update: W23 In-Reply-To: <576D333C.5060603@ubuntu.com> References: <575993F7.9070408@canonical.com> <576D333C.5060603@ubuntu.com> Message-ID: <576D6362.6020509@ubuntu.com> Scarlett can probably fill in more, but these are the big ones (besides from the content interface) that I am aware of: 1) Builting Qt from upstream as a part. From what I understand they use a non-standard set of build scripts that Snapcraft's plugins can't "just work" with. However, many newer KDE apps (and the kde libs that will ultimately use the content framework) often want the latest and greatest version of Qt. 2) There is currently an issue with Phonon which is preventing any audio playback from KDE apps. It's not an apparmor issue, since it happens even when installing with --devmode. David Edmundson has tracked it down to a path issue within Phonon, and it's not clear yet if this can be fixed in the snap itself or if it will require patches to Phonon. Michael Hall mhall119 at ubuntu.com On 06/24/2016 09:18 AM, Mark Shuttleworth wrote: > On 09/06/16 20:36, Scarlett Clark wrote: >> Another community highlight worth mentioning is KDE snapification is >> in full swing. > > That's great news! Do you have a top 5 list of feature requests that > would help the community in that regard? > > I know the content-sharing interface, which would let you publish a > standard set of KDE framework libraries to be shared, is well in > progress. What else would be a help? > > Mark > > From scarlett.gately.clark at gmail.com Fri Jun 24 18:01:00 2016 From: scarlett.gately.clark at gmail.com (Scarlett Clark) Date: Fri, 24 Jun 2016 11:01:00 -0700 Subject: Snap! Community weekly update: W23 In-Reply-To: <576D6362.6020509@ubuntu.com> References: <575993F7.9070408@canonical.com> <576D333C.5060603@ubuntu.com> <576D6362.6020509@ubuntu.com> Message-ID: On Fri, Jun 24, 2016 at 9:44 AM, Michael Hall wrote: > Scarlett can probably fill in more, but these are the big ones (besides > from the content interface) that I am aware of: > > 1) Builting Qt from upstream as a part. From what I understand they use > a non-standard set of build scripts that Snapcraft's plugins can't "just > work" with. However, many newer KDE apps (and the kde libs that will > ultimately use the content framework) often want the latest and greatest > version of Qt. > > Yes big one ^^ > 2) There is currently an issue with Phonon which is preventing any audio > playback from KDE apps. It's not an apparmor issue, since it happens > even when installing with --devmode. David Edmundson has tracked it down > to a path issue within Phonon, and it's not clear yet if this can be > fixed in the snap itself or if it will require patches to Phonon. > I am working on a workaround for this. ^^ We are having issues with DBus in confinement, or lack thereof, rendering some apps non functional out of devmode. So we need a dbus interface to allow it. And apparmor profile diffs that I will need to get the interfaces into snappy when I have them ready. Eventually we will want a plasma plug, but this is future work and we are no where close to starting it. Cheers, Scarlett > > Michael Hall > mhall119 at ubuntu.com > > On 06/24/2016 09:18 AM, Mark Shuttleworth wrote: > > On 09/06/16 20:36, Scarlett Clark wrote: > >> Another community highlight worth mentioning is KDE snapification is > >> in full swing. > > > > That's great news! Do you have a top 5 list of feature requests that > > would help the community in that regard? > > > > I know the content-sharing interface, which would let you publish a > > standard set of KDE framework libraries to be shared, is well in > > progress. What else would be a help? > > > > Mark > > > > > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > -------------- next part -------------- An HTML attachment was scrubbed... URL: From daniel.holbach at canonical.com Mon Jun 27 07:51:20 2016 From: daniel.holbach at canonical.com (Daniel Holbach) Date: Mon, 27 Jun 2016 09:51:20 +0200 Subject: Swing/AWT app crashes Message-ID: <5770DAF8.4000605@canonical.com> Hello everybody, the OpenJDK demo in the Snappy Playpen has been known to crash: https://github.com/ubuntu/snappy-playpen/tree/master/openjdk-demo Now somebody else brought up a similar issue here: http://askubuntu.com/questions/791853/snap-package-for-java-swing-and-awt-application-crashes-when-it-is-executed Do we know what needs to be done to fix the issue? Or if it can be worked around? Have a great day, Daniel From vasilisc777 at gmail.com Mon Jun 27 08:00:49 2016 From: vasilisc777 at gmail.com (Alekseenko Vasilii) Date: Mon, 27 Jun 2016 11:00:49 +0300 Subject: Swing/AWT app crashes In-Reply-To: <5770DAF8.4000605@canonical.com> References: <5770DAF8.4000605@canonical.com> Message-ID: <5770DD31.3010204@gmail.com> 27.06.2016 10:51, Daniel Holbach пишет: > Hello everybody, > > the OpenJDK demo in the Snappy Playpen has been known to crash: > > https://github.com/ubuntu/snappy-playpen/tree/master/openjdk-demo > > Now somebody else brought up a similar issue here: > > > http://askubuntu.com/questions/791853/snap-package-for-java-swing-and-awt-application-crashes-when-it-is-executed > > Do we know what needs to be done to fix the issue? Or if it can be > worked around? > > Have a great day, > Daniel > My first java app in snap - LanguageTool standart wrapper call my run.sh run.sh #!/bin/sh # Not good, needed for fontconfig export XDG_DATA_HOME=$SNAP/usr/share # Font Config export FONTCONFIG_PATH=$SNAP/etc/fonts/config.d export FONTCONFIG_FILE=$SNAP/etc/fonts/fonts.conf export HOME=$SNAP_USER_DATA java -jar -Duser.home=$SNAP_USER_DATA $SNAP/usr/bin/languagetool.jar And JavaApp run successfully --- From jenny.murphy at episensor.com Mon Jun 27 08:24:55 2016 From: jenny.murphy at episensor.com (Jenny Murphy) Date: Mon, 27 Jun 2016 09:24:55 +0100 Subject: ANT plugin - location of jar files. Message-ID: Hi, I am building a java application under snapcraft using the ant plugin. My build.xml file puts the created jar files in a directory called *lib*. The ant plugin complains about this because it was expecting to find jar files in a directory called *jar*. Is it possible to tell the ant plugin to look elsewhere? I don't think the copy plugin is the solution because this errors at the end of the ant part build step. Thanks in advance for any suggestions. -- *Jenny Murphy* *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* jenny.murphy at episensor.com t | +353 (0) 61 512 511 w | http://www.episensor.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From didrocks at ubuntu.com Mon Jun 27 08:50:26 2016 From: didrocks at ubuntu.com (Didier Roche) Date: Mon, 27 Jun 2016 10:50:26 +0200 Subject: ANT plugin - location of jar files. In-Reply-To: References: Message-ID: <5770E8D2.5080703@ubuntu.com> Le 27/06/2016 10:24, Jenny Murphy a écrit : > Hi, > I am building a java application under snapcraft using the ant plugin. > My build.xml file puts the created jar files in a directory called > *lib*. The ant plugin complains about this because it was expecting to > find jar files in a directory called *jar*. > > Is it possible to tell the ant plugin to look elsewhere? > > I don't think the copy plugin is the solution because this errors at > the end of the ant part build step. > > Thanks in advance for any suggestions. > > Hey Jenny, I'm by no mean a java expert and it doesn't seem that "snapcraft help ant" shows up any options for achieving this. The line of the plugin itself seems guilty: "files = glob.glob(os.path.join(self.builddir, 'target', '*.jar'))" Before advising you to inheirt/create your own plugin, I'm CCing Sergio who might have another idea or may want to grow a plugin options for this. Cheers, Didier -------------- next part -------------- An HTML attachment was scrubbed... URL: From jenny.murphy at episensor.com Mon Jun 27 08:51:30 2016 From: jenny.murphy at episensor.com (Jenny Murphy) Date: Mon, 27 Jun 2016 09:51:30 +0100 Subject: ANT plugin - location of jar files. In-Reply-To: References: Message-ID: Apologies, correction to my previous email : I am building a java application under snapcraft using the ant plugin. My build.xml file puts the created jar files in a directory called *lib*. The ant plugin complains about this because it was expecting to find jar files in a directory called *target*. On 27 June 2016 at 09:24, Jenny Murphy wrote: > Hi, > I am building a java application under snapcraft using the ant plugin. > My build.xml file puts the created jar files in a directory called *lib*. > The ant plugin complains about this because it was expecting to find jar > files in a directory called *jar*. > > Is it possible to tell the ant plugin to look elsewhere? > > I don't think the copy plugin is the solution because this errors at the > end of the ant part build step. > > Thanks in advance for any suggestions. > > > > > -- > *Jenny Murphy* > *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* > jenny.murphy at episensor.com t | +353 (0) 61 > 512 511 w | http://www.episensor.com > -- *Jenny Murphy* *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* jenny.murphy at episensor.com t | +353 (0) 61 512 511 w | http://www.episensor.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From daniel.holbach at canonical.com Mon Jun 27 09:43:14 2016 From: daniel.holbach at canonical.com (Daniel Holbach) Date: Mon, 27 Jun 2016 11:43:14 +0200 Subject: Swing/AWT app crashes In-Reply-To: <5770DD31.3010204@gmail.com> References: <5770DAF8.4000605@canonical.com> <5770DD31.3010204@gmail.com> Message-ID: <5770F532.3000203@canonical.com> Hello Alekseenko, On 27.06.2016 10:00, Alekseenko Vasilii wrote: > 27.06.2016 10:51, Daniel Holbach пишет: >> Hello everybody, >> >> the OpenJDK demo in the Snappy Playpen has been known to crash: >> >> https://github.com/ubuntu/snappy-playpen/tree/master/openjdk-demo >> >> Now somebody else brought up a similar issue here: >> >> >> http://askubuntu.com/questions/791853/snap-package-for-java-swing-and-awt-application-crashes-when-it-is-executed >> >> Do we know what needs to be done to fix the issue? Or if it can be >> worked around? >> >> Have a great day, >> Daniel >> > My first java app in snap - LanguageTool > > standart wrapper call my run.sh > > run.sh > #!/bin/sh > # Not good, needed for fontconfig > export XDG_DATA_HOME=$SNAP/usr/share > # Font Config > export FONTCONFIG_PATH=$SNAP/etc/fonts/config.d > export FONTCONFIG_FILE=$SNAP/etc/fonts/fonts.conf > export HOME=$SNAP_USER_DATA > java -jar -Duser.home=$SNAP_USER_DATA $SNAP/usr/bin/languagetool.jar > > And JavaApp run successfully thanks a lot for sharing your wrapper, it works great. I just pushed a PR which makes use of it here: https://github.com/ubuntu/snappy-playpen/pull/122 Have a great day, Daniel From matt.aguirre at gmail.com Mon Jun 27 12:49:59 2016 From: matt.aguirre at gmail.com (Matthew Aguirre) Date: Mon, 27 Jun 2016 08:49:59 -0400 Subject: Swing/AWT app crashes In-Reply-To: <5770DD31.3010204@gmail.com> References: <5770DAF8.4000605@canonical.com> <5770DD31.3010204@gmail.com> Message-ID: <1467031799.27597.0@smtp.gmail.com> Wow! Thanks, I've been trying to figure that one out. -- Matt On Mon, Jun 27, 2016 at 4:00 AM, Alekseenko Vasilii wrote: > 27.06.2016 10:51, Daniel Holbach пишет: > > Hello everybody, > > > > the OpenJDK demo in the Snappy Playpen has been known to crash: > > > > https://github.com/ubuntu/snappy-playpen/tree/master/openjdk-demo > > > > Now somebody else brought up a similar issue here: > > > > > > > http://askubuntu.com/questions/791853/snap-package-for-java-swing-and-awt-application-crashes-when-it-is-executed > > > > Do we know what needs to be done to fix the issue? Or if it can be > > worked around? > > > > Have a great day, > > Daniel > > > My first java app in snap - LanguageTool > > standart wrapper call my run.sh > > run.sh > #!/bin/sh > # Not good, needed for fontconfig > export XDG_DATA_HOME=$SNAP/usr/share > # Font Config > export FONTCONFIG_PATH=$SNAP/etc/fonts/config.d > export FONTCONFIG_FILE=$SNAP/etc/fonts/fonts.conf > export HOME=$SNAP_USER_DATA > java -jar -Duser.home=$SNAP_USER_DATA $SNAP/usr/bin/languagetool.jar > > And JavaApp run successfully > > --- > > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft -------------- next part -------------- An HTML attachment was scrubbed... URL: From sergio.schvezov at canonical.com Mon Jun 27 15:22:36 2016 From: sergio.schvezov at canonical.com (Sergio Schvezov) Date: Mon, 27 Jun 2016 12:22:36 -0300 Subject: ANT plugin - location of jar files. In-Reply-To: <5770E8D2.5080703@ubuntu.com> References: <5770E8D2.5080703@ubuntu.com> Message-ID: <577144BC.70308@canonical.com> El 27/06/16 a las 05:50, Didier Roche escribió: > Le 27/06/2016 10:24, Jenny Murphy a écrit : >> Hi, >> I am building a java application under snapcraft using the ant plugin. >> My build.xml file puts the created jar files in a directory called >> *lib*. The ant plugin complains about this because it was expecting to >> find jar files in a directory called *jar*. >> >> Is it possible to tell the ant plugin to look elsewhere? >> >> I don't think the copy plugin is the solution because this errors at >> the end of the ant part build step. >> >> Thanks in advance for any suggestions. >> >> > > Hey Jenny, > I'm by no mean a java expert and it doesn't seem that "snapcraft help > ant" shows up any options for achieving this. > The line of the plugin itself seems guilty: "files = > glob.glob(os.path.join(self.builddir, 'target', '*.jar'))" > > Before advising you to inheirt/create your own plugin, I'm CCing Sergio > who might have another idea or may want to grow a plugin options for this. Hi Jenny, I would welcome a bug report on https://bugs.launchpad.net/snapcraft/+filebug about this, an industry standard to follow by default would be good if `target` isn't the right one. We can take it from there but if you know python, a PR on github would be welcome as well :-) Cheers Sergio From daniel.holbach at canonical.com Mon Jun 27 16:06:40 2016 From: daniel.holbach at canonical.com (Daniel Holbach) Date: Mon, 27 Jun 2016 18:06:40 +0200 Subject: Snappy Playpen: let's get your snap landed Message-ID: <57714F10.3020908@canonical.com> Hello everybody, tomorrow we are going to have another Snappy Playpen event in which we are going to be around in - #snappy on freenode and - https://gitter.im/ubuntu/snappy-playpen to help you snap your favourite piece of software and get it landed in the Snappy Playpen. If you need help with finishing touches or have some more basic questions, we are happy to help you out. Most of the current work in the Snappy Playpen is documented at https://github.com/ubuntu/snappy-playpen/wiki/Examples Some of us want to spend some time tomorrow as well to do a bit of cleanup in the repository and documenting currently known issues, so if you know of any, let us know and we'll make sure we are going to track them as bug reports. WHAT: Snappy Playpen Event WHERE: #snappy on Freenode or https://gitter.im/ubuntu/snappy-playpen WHEN: Tomorrow, 28th June 2016 Thanks a lot everyone, we are looking forward to tomorrow! Have a great day, Daniel From john.agosta at canonical.com Mon Jun 27 17:29:35 2016 From: john.agosta at canonical.com (John Agosta) Date: Mon, 27 Jun 2016 11:29:35 -0600 Subject: Python2: problem with importing lazr.restfulclient Message-ID: <5771627F.8070501@canonical.com> An HTML attachment was scrubbed... URL: From sergio.schvezov at canonical.com Mon Jun 27 17:55:01 2016 From: sergio.schvezov at canonical.com (Sergio Schvezov) Date: Mon, 27 Jun 2016 14:55:01 -0300 Subject: Python2: problem with importing lazr.restfulclient In-Reply-To: <5771627F.8070501@canonical.com> References: <5771627F.8070501@canonical.com> Message-ID: <57716875.1060806@ubuntu.com> El 27/06/16 a las 14:29, John Agosta escribió: > Hi: Hi there! > I am attempting to create a launchpad API part and have run into an > issue with the snap correctly resolving lazr.restfulclient. > The error I get is: > > Traceback (most recent call last): > File "/snap/lptools-john/x2/usr/bin/testreport.py", line 9, in > > from launchpadlib.launchpad import Launchpad > File > "/snap/lptools-john/x2/usr/lib/python2.7/dist-packages/launchpadlib/launchpad.py", > line 33, in > from lazr.restfulclient.resource import ( > ImportError: No module named lazr.restfulclient.resource I am not entirely sure what is up with this module coming from the debian package in the archives but Kyle (CCed) has had success with this by grabbing the package from pypi directly using the `python-packages` keyword. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: OpenPGP digital signature URL: From kyle.fazzari at canonical.com Mon Jun 27 17:59:37 2016 From: kyle.fazzari at canonical.com (Kyle Fazzari) Date: Mon, 27 Jun 2016 13:59:37 -0400 Subject: Python2: problem with importing lazr.restfulclient In-Reply-To: <57716875.1060806@ubuntu.com> References: <5771627F.8070501@canonical.com> <57716875.1060806@ubuntu.com> Message-ID: <57716989.4020303@canonical.com> On 06/27/2016 01:55 PM, Sergio Schvezov wrote: > > > El 27/06/16 a las 14:29, John Agosta escribió: >> Hi: > > Hi there! > >> I am attempting to create a launchpad API part and have run into an >> issue with the snap correctly resolving lazr.restfulclient. > >> The error I get is: >> >> Traceback (most recent call last): >> File "/snap/lptools-john/x2/usr/bin/testreport.py", line 9, in >> >> from launchpadlib.launchpad import Launchpad >> File >> "/snap/lptools-john/x2/usr/lib/python2.7/dist-packages/launchpadlib/launchpad.py", >> line 33, in >> from lazr.restfulclient.resource import ( >> ImportError: No module named lazr.restfulclient.resource > > I am not entirely sure what is up with this module coming from the > debian package in the archives but Kyle (CCed) has had success with this > by grabbing the package from pypi directly using the `python-packages` > keyword. Indeed. I've not investigated this enough to know for sure, but it seems that the lazr deb doesn't contain the correct __init__.py files. I suspect they're setup after the fact, e.g. in a postinst. Anyway, as Sergio said, I was able to work around it by using `python-packages` instead. My use-case was launchpadlib as well, so you might find this useful: https://github.com/kyrofa/cla-check/blob/master/snapcraft.yaml -- Kyle Fazzari (kyrofa) Software Engineer Canonical Ltd. kyle at canonical.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 836 bytes Desc: OpenPGP digital signature URL: From john.agosta at canonical.com Mon Jun 27 20:59:09 2016 From: john.agosta at canonical.com (John Agosta) Date: Mon, 27 Jun 2016 14:59:09 -0600 Subject: Python2: problem with importing lazr.restfulclient In-Reply-To: <57716989.4020303@canonical.com> References: <5771627F.8070501@canonical.com> <57716875.1060806@ubuntu.com> <57716989.4020303@canonical.com> Message-ID: <5771939D.4010908@canonical.com> An HTML attachment was scrubbed... URL: From vasilisc777 at gmail.com Tue Jun 28 06:39:05 2016 From: vasilisc777 at gmail.com (Alekseenko Vasilii) Date: Tue, 28 Jun 2016 09:39:05 +0300 Subject: audio player Message-ID: <57721B89.5090908@gmail.com> I packed audio-player into snap package and AppArmor denied access name="/dev/shm/pulse-shm-1137513322" pid=2926 comm="audio-player" requested_mask="r" denied_mask="r" snapcraft.xml have stage-packages: - libxcb-shm0 This bug not closed https://bugs.launchpad.net/snappy/+bug/1578217 How to use PulseAudio? -- From vasilisc777 at gmail.com Tue Jun 28 06:51:37 2016 From: vasilisc777 at gmail.com (Alekseenko Vasilii) Date: Tue, 28 Jun 2016 09:51:37 +0300 Subject: sorry Message-ID: <57721E79.6010108@gmail.com> My Ubuntu 16.10 (snapd 2.0.2) don't have slot pulseuadio and I made a mistake. Sorry $ LANG=C sudo apt-cache policy snapd snapd: Installed: 2.0.2 Candidate: 2.0.2 Version table: *** 2.0.2 500 500 http://fi.archive.ubuntu.com/ubuntu devel/main amd64 Packages 100 /var/lib/dpkg/status -- From jamie.bennett at canonical.com Tue Jun 28 07:56:05 2016 From: jamie.bennett at canonical.com (Jamie Bennett) Date: Tue, 28 Jun 2016 08:56:05 +0100 Subject: sorry In-Reply-To: <57721E79.6010108@gmail.com> References: <57721E79.6010108@gmail.com> Message-ID: <80C32020-B188-44A4-9F28-D2971032794A@canonical.com> Hey Alekseenko. A update to snapd 2.0.6 or higher on xenial will give you the pulseaudio interface. The package in Yakkety is not being updated at the moment due to an issue we are in the process of fixing. Regards, Jamie. > On 28 Jun 2016, at 07:51, Alekseenko Vasilii wrote: > > My Ubuntu 16.10 (snapd 2.0.2) don't have slot pulseuadio and I made a mistake. Sorry > > $ LANG=C sudo apt-cache policy snapd > snapd: > Installed: 2.0.2 > Candidate: 2.0.2 > Version table: > *** 2.0.2 500 > 500 http://fi.archive.ubuntu.com/ubuntu devel/main amd64 Packages > 100 /var/lib/dpkg/status > > > -- > > > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft From pawel.stolowski at canonical.com Tue Jun 28 08:44:11 2016 From: pawel.stolowski at canonical.com (Pawel Stolowski) Date: Tue, 28 Jun 2016 10:44:11 +0200 Subject: App denied access to /lib and crashes Message-ID: <577238DB.4030905@canonical.com> Hi, I've been trying to create a snap package for Scid-vs-PC (and old-style TCL/TK based app) but have only been able to get it working in devmode. In the strict mode it crashes in libtk8.6.so and the segfault appears right after a denied access to read "/lib", in dmesg which makes me think that tcl/tk doesn't handle such (unexpected) scenario very well. When running in the devmode I get: [ 4039.752903] audit: type=1400 audit(1467102032.459:56): apparmor="ALLOWED" operation="open" profile="snap.scid-vs-pc.scidvspc" name="/lib/" pid=18523 comm="tkscid" requested_mask="r" denied_mask="r" fsuid=1000 ouid=0 (and the app runs fine). I suspect that just making "/lib" readable to my snap would make that app happy, so a couple of questions: - can I somehow expose "/lib" in read-only mode to my snap under "strict" confinement? - or can I somehow simulate the presence of "/lib" (and let it be empty)? Cheers, Pawel From ogra at ubuntu.com Tue Jun 28 08:49:58 2016 From: ogra at ubuntu.com (Oliver Grawert) Date: Tue, 28 Jun 2016 10:49:58 +0200 Subject: sorry In-Reply-To: <57721E79.6010108@gmail.com> References: <57721E79.6010108@gmail.com> Message-ID: <1467103798.8355.18.camel@anubis> hi, Am Dienstag, den 28.06.2016, 09:51 +0300 schrieb Alekseenko Vasilii: > My Ubuntu 16.10 (snapd 2.0.2) don't have slot pulseuadio and I made a > mistake. Sorry > > $ LANG=C sudo apt-cache policy snapd > snapd: > Installed: 2.0.2 > Candidate: 2.0.2 > Version table: > *** 2.0.2 500 > 500 http://fi.archive.ubuntu.com/ubuntu devel/main amd64 Packages > 100 /var/lib/dpkg/status > i think there were no dependency changes in the last few version, so if it is urgent you can just grab the deb from https://launchpad.net/ubuntu/+source/snapd and install it by using "sudo dpkg -i ..." ciao oli -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: This is a digitally signed message part URL: From vasilisc777 at gmail.com Tue Jun 28 09:23:13 2016 From: vasilisc777 at gmail.com (Alekseenko Vasilii) Date: Tue, 28 Jun 2016 12:23:13 +0300 Subject: syscall 142 Message-ID: <57724201.6030805@gmail.com> Jun 28 12:15:22 vb kernel: [ 167.292913] audit: type=1326 audit(1467105322.080:24): auid=4294967295 uid=1000 gid=1000 ses=4294967295 pid=1805 comm="audioplayer-main" exe="/snap/audioplayer/x1/usr/bin/audioplayer" sig=31 arch=c000003e syscall=142 compat=0 ip=0x7f5aef3efca7 code=0x0 scmp_sys_resolver 142 sched_setparam how to allow the audioplayer - sched_setparam? From zygmunt.krynicki at canonical.com Tue Jun 28 09:50:37 2016 From: zygmunt.krynicki at canonical.com (Zygmunt Krynicki) Date: Tue, 28 Jun 2016 11:50:37 +0200 Subject: syscall 142 In-Reply-To: <57724201.6030805@gmail.com> References: <57724201.6030805@gmail.com> Message-ID: <2D847C08-E14E-4E70-81C6-33F880E17665@canonical.com> > Wiadomość napisana przez Alekseenko Vasilii w dniu 28.06.2016, o godz. 11:23: > > > Jun 28 12:15:22 vb kernel: [ 167.292913] audit: type=1326 audit(1467105322.080:24): auid=4294967295 uid=1000 gid=1000 ses=4294967295 pid=1805 comm="audioplayer-main" exe="/snap/audioplayer/x1/usr/bin/audioplayer" sig=31 arch=c000003e syscall=142 compat=0 ip=0x7f5aef3efca7 code=0x0 > > scmp_sys_resolver 142 > sched_setparam > > how to allow the audioplayer - sched_setparam? Currently this system call is not allowed. I’m not 100% sure but it looks like it is there to prevent processes from stalling the rest of the system. I think it would be really better if we didn’t kill the process but returned EPERM instead. DO you know if your audio player can handle that case? Best regards ZK From vasilisc777 at gmail.com Tue Jun 28 10:05:11 2016 From: vasilisc777 at gmail.com (Alekseenko Vasilii) Date: Tue, 28 Jun 2016 13:05:11 +0300 Subject: syscall 142 In-Reply-To: <2D847C08-E14E-4E70-81C6-33F880E17665@canonical.com> References: <57724201.6030805@gmail.com> <2D847C08-E14E-4E70-81C6-33F880E17665@canonical.com> Message-ID: <57724BD7.20200@gmail.com> 28.06.2016 12:50, Zygmunt Krynicki пишет: > Currently this system call is not allowed. I’m not 100% sure but it looks like it is there to prevent processes from stalling the rest of the system. > > I think it would be really better if we didn’t kill the process but returned EPERM instead. DO you know if your audio player can handle that case? I don't know =( It is the project 0xDEADBEEF https://github.com/Alexey-Yakovenko/deadbeef -- From vasilisc777 at gmail.com Tue Jun 28 10:13:52 2016 From: vasilisc777 at gmail.com (Vasilisc) Date: Tue, 28 Jun 2016 13:13:52 +0300 Subject: syscall 142 In-Reply-To: <2D847C08-E14E-4E70-81C6-33F880E17665@canonical.com> References: <57724201.6030805@gmail.com> <2D847C08-E14E-4E70-81C6-33F880E17665@canonical.com> Message-ID: <57724DE0.5000504@gmail.com> 28.06.2016 12:50, Zygmunt Krynicki пишет: > >> Wiadomość napisana przez Alekseenko Vasilii w dniu 28.06.2016, o godz. 11:23: >> >> >> Jun 28 12:15:22 vb kernel: [ 167.292913] audit: type=1326 audit(1467105322.080:24): auid=4294967295 uid=1000 gid=1000 ses=4294967295 pid=1805 comm="audioplayer-main" exe="/snap/audioplayer/x1/usr/bin/audioplayer" sig=31 arch=c000003e syscall=142 compat=0 ip=0x7f5aef3efca7 code=0x0 >> >> scmp_sys_resolver 142 >> sched_setparam >> >> how to allow the audioplayer - sched_setparam? > > Currently this system call is not allowed. I’m not 100% sure but it looks like it is there to prevent processes from stalling the rest of the system. > > I think it would be really better if we didn’t kill the process but returned EPERM instead. DO you know if your audio player can handle that case? > > Best regards > ZK > sched_setparam can be connected to threads? http://linux.die.net/man/2/sched_setparam Switch-off of threads can help? -- Best regards, vasilisc From vasilisc777 at gmail.com Tue Jun 28 13:19:28 2016 From: vasilisc777 at gmail.com (Vasilisc) Date: Tue, 28 Jun 2016 16:19:28 +0300 Subject: syscall 142 In-Reply-To: <57724DE0.5000504@gmail.com> References: <57724201.6030805@gmail.com> <2D847C08-E14E-4E70-81C6-33F880E17665@canonical.com> <57724DE0.5000504@gmail.com> Message-ID: <57727960.1080702@gmail.com> 28.06.2016 13:13, Vasilisc пишет: > 28.06.2016 12:50, Zygmunt Krynicki пишет: >> >>> Wiadomość napisana przez Alekseenko Vasilii w >>> dniu 28.06.2016, o godz. 11:23: >>> >>> >>> Jun 28 12:15:22 vb kernel: [ 167.292913] audit: type=1326 >>> audit(1467105322.080:24): auid=4294967295 uid=1000 gid=1000 >>> ses=4294967295 pid=1805 comm="audioplayer-main" >>> exe="/snap/audioplayer/x1/usr/bin/audioplayer" sig=31 arch=c000003e >>> syscall=142 compat=0 ip=0x7f5aef3efca7 code=0x0 >>> >>> scmp_sys_resolver 142 >>> sched_setparam >>> >>> how to allow the audioplayer - sched_setparam? >> >> Currently this system call is not allowed. I’m not 100% sure but it >> looks like it is there to prevent processes from stalling the rest of >> the system. >> >> I think it would be really better if we didn’t kill the process but >> returned EPERM instead. DO you know if your audio player can handle >> that case? >> >> Best regards >> ZK >> > sched_setparam can be connected to threads? > http://linux.die.net/man/2/sched_setparam > Switch-off of threads can help? > I fixed this issue. Find and comment pthread_setschedprio in source code. -- Best regards, vasilisc From jamie at canonical.com Tue Jun 28 13:21:38 2016 From: jamie at canonical.com (Jamie Strandboge) Date: Tue, 28 Jun 2016 08:21:38 -0500 Subject: App denied access to /lib and crashes In-Reply-To: <577238DB.4030905@canonical.com> References: <577238DB.4030905@canonical.com> Message-ID: <1467120098.5164.76.camel@canonical.com> On Tue, 2016-06-28 at 10:44 +0200, Pawel Stolowski wrote: > Hi, > > I've been trying to create a snap package for Scid-vs-PC (and old-style  > TCL/TK based app) but have only been able to get it working in devmode. > In the strict mode it crashes in libtk8.6.so and the segfault appears  > right after a denied access to read "/lib", in dmesg which makes me  > think that tcl/tk doesn't handle such (unexpected) scenario very well. > > When running in the devmode I get: > [ 4039.752903] audit: type=1400 audit(1467102032.459:56):  > apparmor="ALLOWED" operation="open" profile="snap.scid-vs-pc.scidvspc"  > name="/lib/" pid=18523 comm="tkscid" requested_mask="r" denied_mask="r"  > fsuid=1000 ouid=0 > (and the app runs fine). > > I suspect that just making "/lib" readable to my snap would make that  > app happy, so a couple of questions: > - can I somehow expose "/lib" in read-only mode to my snap under  > "strict" confinement? > - or can I somehow simulate the presence of "/lib" (and let it be empty)? Can you file a bug and add the 'snapd-interface' tag? For now you can workaround this in strict mode by adding to /var/lib/snapd/apparmor/profiles/snap.scid-vs- pc.scidvspc:   /lib/ r, # trailing '/' is important then do: $ sudo apparmor_parser -r /var/lib/snapd/apparmor/profiles/snap.scid-vs- pc.scidvspc Then try again. Depending on what the program does, you might have to add '/usr/lib/', '/usr/local/lib/', etc. Please report all the accesses needed in the bug and I can get this fixed up. -- Jamie Strandboge | http://www.canonical.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: This is a digitally signed message part URL: From jamie at canonical.com Tue Jun 28 13:25:58 2016 From: jamie at canonical.com (Jamie Strandboge) Date: Tue, 28 Jun 2016 08:25:58 -0500 Subject: syscall 142 In-Reply-To: <57724201.6030805@gmail.com> References: <57724201.6030805@gmail.com> Message-ID: <1467120358.5164.79.camel@canonical.com> On Tue, 2016-06-28 at 12:23 +0300, Alekseenko Vasilii wrote: > Jun 28 12:15:22 vb kernel: [  167.292913] audit: type=1326  > audit(1467105322.080:24): auid=4294967295 uid=1000 gid=1000  > ses=4294967295 pid=1805 comm="audioplayer-main"  > exe="/snap/audioplayer/x1/usr/bin/audioplayer" sig=31 arch=c000003e  > syscall=142 compat=0 ip=0x7f5aef3efca7 code=0x0 > > scmp_sys_resolver 142 > sched_setparam > > how to allow the audioplayer - sched_setparam? > Can you file a bug and add the 'snapd-interface' tag? We'll need to investigate this but I think it may be possible to allow sched_setparam with the seccomp arg filtering branch for snap-confine that has landed in trunk and AIUI scheduled to land with snapd 2.0.10. -- Jamie Strandboge | http://www.canonical.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: This is a digitally signed message part URL: From jamie at canonical.com Tue Jun 28 13:27:25 2016 From: jamie at canonical.com (Jamie Strandboge) Date: Tue, 28 Jun 2016 08:27:25 -0500 Subject: syscall 142 In-Reply-To: <1467120358.5164.79.camel@canonical.com> References: <57724201.6030805@gmail.com> <1467120358.5164.79.camel@canonical.com> Message-ID: <1467120445.5164.80.camel@canonical.com> On Tue, 2016-06-28 at 08:25 -0500, Jamie Strandboge wrote: > On Tue, 2016-06-28 at 12:23 +0300, Alekseenko Vasilii wrote: > > > > Jun 28 12:15:22 vb kernel: [  167.292913] audit: type=1326  > > audit(1467105322.080:24): auid=4294967295 uid=1000 gid=1000  > > ses=4294967295 pid=1805 comm="audioplayer-main"  > > exe="/snap/audioplayer/x1/usr/bin/audioplayer" sig=31 arch=c000003e  > > syscall=142 compat=0 ip=0x7f5aef3efca7 code=0x0 > > > > scmp_sys_resolver 142 > > sched_setparam > > > > how to allow the audioplayer - sched_setparam? > > > Can you file a bug and add the 'snapd-interface' tag? We'll need to > investigate > this but I think it may be possible to allow sched_setparam with the seccomp > arg > filtering branch for snap-confine that has landed in trunk and AIUI scheduled > to > land with snapd 2.0.10. > I forgot to mention that you can temporarily work around this in strict mode by adding 'sched_setparam' to /var/lib/snapd/seccomp/profiles/snap.your.app. -- Jamie Strandboge | http://www.canonical.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: This is a digitally signed message part URL: From sergio.schvezov at canonical.com Tue Jun 28 16:22:36 2016 From: sergio.schvezov at canonical.com (Sergio Schvezov) Date: Tue, 28 Jun 2016 13:22:36 -0300 Subject: ANN: snapcraft has moved Message-ID: <5772A44C.9060706@ubuntu.com> I just wanted to announce that we have (finally) spent the couple of clicks required to move snapcraft over to snapcore to live with its related projects. The code can now be found on https://github.com/snapcore/snapcraft All PRs should have been transparently migrated. Trying to go to the old ubuntu-core repo on the webfront will correctly redirect. Keep in mind that bugs are still reported at the same location, https://bugs.launchpad.net/snapcraft/+filebug Cheers Sergio -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: OpenPGP digital signature URL: From jenny.murphy at episensor.com Tue Jun 28 20:11:48 2016 From: jenny.murphy at episensor.com (Jenny Murphy) Date: Tue, 28 Jun 2016 21:11:48 +0100 Subject: ANT plugin - location of jar files. In-Reply-To: <577144BC.70308@canonical.com> References: <5770E8D2.5080703@ubuntu.com> <577144BC.70308@canonical.com> Message-ID: Hi Sergio and Didier, Yes I will file a bug report. I think it is ok to have target as the default ( I am not sure what if any industry standard destination there is, maybe dist or build ? Unfortunately, no I am not python proficient :-(. Jenny On 27 June 2016 at 16:22, Sergio Schvezov wrote: > > > El 27/06/16 a las 05:50, Didier Roche escribió: > > Le 27/06/2016 10:24, Jenny Murphy a écrit : > >> Hi, > >> I am building a java application under snapcraft using the ant plugin. > >> My build.xml file puts the created jar files in a directory called > >> *lib*. The ant plugin complains about this because it was expecting to > >> find jar files in a directory called *jar*. > >> > >> Is it possible to tell the ant plugin to look elsewhere? > >> > >> I don't think the copy plugin is the solution because this errors at > >> the end of the ant part build step. > >> > >> Thanks in advance for any suggestions. > >> > >> > > > > Hey Jenny, > > I'm by no mean a java expert and it doesn't seem that "snapcraft help > > ant" shows up any options for achieving this. > > The line of the plugin itself seems guilty: "files = > > glob.glob(os.path.join(self.builddir, 'target', '*.jar'))" > > > > Before advising you to inheirt/create your own plugin, I'm CCing Sergio > > who might have another idea or may want to grow a plugin options for > this. > > Hi Jenny, I would welcome a bug report on > https://bugs.launchpad.net/snapcraft/+filebug about this, an industry > standard to follow by default would be good if `target` isn't the right > one. We can take it from there but if you know python, a PR on github > would be welcome as well :-) > > Cheers > Sergio > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > -- *Jenny Murphy* *EpiSensor, Georges Quay House, Georges Quay, Limerick, Ireland* jenny.murphy at episensor.com t | +353 (0) 61 512 511 w | http://www.episensor.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From vasilisc777 at gmail.com Wed Jun 29 08:57:25 2016 From: vasilisc777 at gmail.com (Vasilisc) Date: Wed, 29 Jun 2016 11:57:25 +0300 Subject: Please give me recommendations. Message-ID: <57738D75.5010508@gmail.com> Please give me recommendations. I packed gtk3 app. In snapcraft.xml -- parts: db: plugin: autotools source: deadbeef-0.7.2 configflags: - --prefix=/usr -- The application tries to get access to the files. /usr/share/NAMEAPP/pixmaps /usr/share/locale/CODE_COUNTRY/LC_MESSAGES/NAMEAPP.mo but the real path will be /snap/NAMEAPP/REVISION/usr/share/NAMEAPP/pixmaps /snap/NAMEAPP/REVISION/usr/share/locale/CODE_COUNTRY/LC_MESSAGES/NAMEAPP.mo If app uses absolute path, I have trouble. How to do right thing? -- Best regards, vasilisc From didrocks at ubuntu.com Wed Jun 29 09:36:39 2016 From: didrocks at ubuntu.com (Didier Roche) Date: Wed, 29 Jun 2016 11:36:39 +0200 Subject: Please give me recommendations. In-Reply-To: <57738D75.5010508@gmail.com> References: <57738D75.5010508@gmail.com> Message-ID: <577396A7.20301@ubuntu.com> Le 29/06/2016 10:57, Vasilisc a écrit : > Please give me recommendations. > I packed gtk3 app. In snapcraft.xml > -- > parts: > db: > plugin: autotools > source: deadbeef-0.7.2 > configflags: > - --prefix=/usr > -- > > The application tries to get access to the files. > /usr/share/NAMEAPP/pixmaps > /usr/share/locale/CODE_COUNTRY/LC_MESSAGES/NAMEAPP.mo > > but the real path will be > /snap/NAMEAPP/REVISION/usr/share/NAMEAPP/pixmaps > /snap/NAMEAPP/REVISION/usr/share/locale/CODE_COUNTRY/LC_MESSAGES/NAMEAPP.mo > > > If app uses absolute path, I have trouble. > How to do right thing? > You need to use some wrappers that help you redirect some of those calls to your snaps or HOME directory. Some options is to use a new shiny consolidated gtk3 cloud part! - prepend your apps: command with "gtk-launch - ensure that you add to it at least the [unity7, x11] plugs, (you may want gsettings and home plugs as well) - add to your part definition: after: [desktop.gtk3] Think about running "snapcraft update" if you have snapcraft 2.12 before running snapcraft. Note that the locale (transations) is still something that we need to figure out depending on the used technology. Cheers, Didier From pawel.stolowski at canonical.com Wed Jun 29 10:18:57 2016 From: pawel.stolowski at canonical.com (Pawel Stolowski) Date: Wed, 29 Jun 2016 12:18:57 +0200 Subject: Fonts issue (was: Re: App denied access to /lib and crashes) In-Reply-To: <1467120098.5164.76.camel@canonical.com> References: <577238DB.4030905@canonical.com> <1467120098.5164.76.camel@canonical.com> Message-ID: <5773A091.70307@canonical.com> Hi, On 28.06.2016 15:21, Jamie Strandboge wrote: > On Tue, 2016-06-28 at 10:44 +0200, Pawel Stolowski wrote: >> Hi, >> >> I've been trying to create a snap package for Scid-vs-PC (and old-style >> TCL/TK based app) but have only been able to get it working in devmode. >> In the strict mode it crashes in libtk8.6.so and the segfault appears >> right after a denied access to read "/lib", in dmesg which makes me >> think that tcl/tk doesn't handle such (unexpected) scenario very well. >> >> When running in the devmode I get: >> [ 4039.752903] audit: type=1400 audit(1467102032.459:56): >> apparmor="ALLOWED" operation="open" profile="snap.scid-vs-pc.scidvspc" >> name="/lib/" pid=18523 comm="tkscid" requested_mask="r" denied_mask="r" >> fsuid=1000 ouid=0 >> (and the app runs fine). >> >> I suspect that just making "/lib" readable to my snap would make that >> app happy, so a couple of questions: >> - can I somehow expose "/lib" in read-only mode to my snap under >> "strict" confinement? >> - or can I somehow simulate the presence of "/lib" (and let it be empty)? > Can you file a bug and add the 'snapd-interface' tag? For now you can workaround > this in strict mode by adding to /var/lib/snapd/apparmor/profiles/snap.scid-vs- > pc.scidvspc: > > /lib/ r, # trailing '/' is important > > then do: > $ sudo apparmor_parser -r /var/lib/snapd/apparmor/profiles/snap.scid-vs- > pc.scidvspc > > Then try again. Depending on what the program does, you might have to add > '/usr/lib/', '/usr/local/lib/', etc. Please report all the accesses needed in > the bug and I can get this fixed up. Ok, here you go: https://bugs.launchpad.net/snappy/+bug/1597259. I've only seen it with /lib so far. Now, it seems the real problem is not with /lib access, but it's font-related... After succesfully running the app in devmode yesterday, this morning after booting my PC it doesn't work anymore :O, looks like I was just lucky yesterday and the app is unstable due to a problem with fonts. I managed to run it with gdb inside the container and got this stacktrace on crash: #0 0x00007ffff759ee53 in TkpGetFontFamilies () from /snap/scid-vs-pc/x2/lib/libtk8.6.so #1 0x00007ffff74f3dc5 in Tk_FontObjCmd () from /snap/scid-vs-pc/x2/lib/libtk8.6.so #2 0x00007ffff7848e07 in TclNRRunCallbacks () from /snap/scid-vs-pc/x2/lib/libtcl8.6.so #3 0x00007ffff784ae75 in TclEvalEx () from /snap/scid-vs-pc/x2/lib/libtcl8.6.so #4 0x00007ffff7902908 in Tcl_FSEvalFileEx () from /snap/scid-vs-pc/x2/lib/libtcl8.6.so #5 0x00007ffff74fca9e in Tk_MainEx () from /snap/scid-vs-pc/x2/lib/libtk8.6.so #6 0x00000000004045aa in main () I'm trying to guess what font-related stuff may be missing. I've the following packages installed via stage-packages: - ttf-ubuntu-font-family - xfonts-base - xfonts-scalable - fontconfig-config Any idea what else to try? Thanks, Pawel From sergio.schvezov at canonical.com Wed Jun 29 17:50:02 2016 From: sergio.schvezov at canonical.com (Sergio Schvezov) Date: Wed, 29 Jun 2016 14:50:02 -0300 Subject: ANN: snapcraft 2.12 is no available Message-ID: <57740A4A.5010104@ubuntu.com> Hello snapcrafters! We are pleased to announce the release of version 2.12 of snapcraft: https://launchpad.net/snapcraft/+milestone/2.12 This is an exciting release with many new features and improvements. Among the features to highlight: - New parts ecosystem where remote parts are managed with some new snapcraft commands such as, update, search and define. - Many improvements to the parts parser backend. - More store interaction straight from snapcraft, `snapcraft register` is now available to register your snap name. - Subversion support as a source. - A gulp plugin - A qmake plugin - The possibility of loading from a .snapcraft.yaml (snapcraftyaml) for those wanting to hide this file from their trees. Among improvements: - rpath with $ORIGIN is properly taken care of now (no more /home/... paths added). - Error message improvements across the board. - cleanbuild now properly creates ephemeral containers. This release has seen many contributions from outside of the snapcraft core team, so we want to give a shout out to these folks, here's a team thank you for: - Simon Quigley - Bayard Randel - Evan Dandrea - Daniel Holbach - Rob Loach - Joe Talbott To consume the latest snapcraft release on Xenial Xerus (16.04), we suggest you install the snapcraft package from the Ubuntu Archives: sudo apt update sudo apt install snapcraft To get the source for this release check it out at https://github.com/ubuntu-core/snapcraft/releases/tag/2.12 A great place to collaborate and discuss features, bugs and ideas on snapcraft is snapcraft at lists.ubuntu.com mailing list or directly in the #snappy channel on irc.freenode.net. To file bugs, please go to https://bugs.launchpad.net/snapcraft/+filebug. Happy snapcrafting, - Sergio and the team -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: OpenPGP digital signature URL: From mark at ubuntu.com Wed Jun 29 18:41:19 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Wed, 29 Jun 2016 19:41:19 +0100 Subject: ANN: snapcraft 2.12 is no available In-Reply-To: <57740A4A.5010104@ubuntu.com> References: <57740A4A.5010104@ubuntu.com> Message-ID: <5774164F.4030609@ubuntu.com> \o/ "parts" are like "source libraries" for snaps. They let an upstream that provides a library describe that in a part, which anybody can reuse just by naming it. So, if you have a library that you want people to use in snaps, for example a library that lets a snap talk to a cloud API for data storage, then a part is what you want to publish :) In my snap I can then just define my own local parts, referring to the others are dependencies which will automatically be resolved and retrieved and build. We have noticed that in many cases, a part that builds a library from source results in a much smaller snap than installing that same library from primed packages. Mark -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 181 bytes Desc: OpenPGP digital signature URL: From evan.dandrea at canonical.com Wed Jun 29 18:44:22 2016 From: evan.dandrea at canonical.com (Evan Dandrea) Date: Wed, 29 Jun 2016 18:44:22 +0000 Subject: Fonts issue (was: Re: App denied access to /lib and crashes) In-Reply-To: <5773A091.70307@canonical.com> References: <577238DB.4030905@canonical.com> <1467120098.5164.76.camel@canonical.com> <5773A091.70307@canonical.com> Message-ID: On Wed, 29 Jun 2016 at 05:20 Pawel Stolowski wrote: > > I'm trying to guess what font-related stuff may be missing. I've the > following packages installed via stage-packages: > - ttf-ubuntu-font-family > - xfonts-base > - xfonts-scalable > - fontconfig-config > > Any idea what else to try? If Tcl/Tk is loading fonts via absolute paths, including some font packages in stage-packages unfortunately won't help. Any files in your snap will install relative to /snap/your_snap/current, not /. Only files in the OS snap can be installed to /. I ran into this font problem with Jenkins. As it also uses fontconfig, I provided a custom fonts.conf pointed at the needed fonts I put under the snap directory [1, 2]. If that doesn't work and you need to dig deeper, I've found it useful to run programs under ptrace and see what files they open (`strace -f -e file`). 1: https://github.com/evandandrea/jenkins-snap/blob/master/parts/plugins/x-jenkins.py#L12 2: https://github.com/evandandrea/jenkins-snap/blob/master/snapcraft.yaml#L20 -------------- next part -------------- An HTML attachment was scrubbed... URL: From david.planella at ubuntu.com Wed Jun 29 18:45:04 2016 From: david.planella at ubuntu.com (David Planella) Date: Wed, 29 Jun 2016 20:45:04 +0200 Subject: ANN: snapcraft 2.12 is no available In-Reply-To: <57740A4A.5010104@ubuntu.com> References: <57740A4A.5010104@ubuntu.com> Message-ID: Good work Snapcraft team and all contributors to this release! For those interested in a closer look at Sergio and the team's release highlights, do check out the article with video demos on the developer site too: https://developer.ubuntu.com/en/blog/2016/06/29/snapcraft-212 Cheers, David. On Wed, Jun 29, 2016 at 7:50 PM, Sergio Schvezov < sergio.schvezov at canonical.com> wrote: > Hello snapcrafters! > > We are pleased to announce the release of version 2.12 of snapcraft: > https://launchpad.net/snapcraft/+milestone/2.12 > > This is an exciting release with many new features and improvements. > Among the features to highlight: > > - New parts ecosystem where remote parts are managed with some new > snapcraft commands such as, update, search and define. > - Many improvements to the parts parser backend. > - More store interaction straight from snapcraft, `snapcraft register` > is now available to register your snap name. > - Subversion support as a source. > - A gulp plugin > - A qmake plugin > - The possibility of loading from a .snapcraft.yaml > (snapcraftyaml) for those wanting to hide this file from their > trees. > > Among improvements: > > - rpath with $ORIGIN is properly taken care of now (no more /home/... > paths added). > - Error message improvements across the board. > - cleanbuild now properly creates ephemeral containers. > > This release has seen many contributions from outside of the snapcraft > core team, so we want to give a shout out to these folks, here's a team > thank you for: > > - Simon Quigley > - Bayard Randel > - Evan Dandrea > - Daniel Holbach > - Rob Loach > - Joe Talbott > > To consume the latest snapcraft release on Xenial Xerus (16.04), we > suggest you install the snapcraft package from the Ubuntu Archives: > > sudo apt update > sudo apt install snapcraft > > To get the source for this release check it out at > https://github.com/ubuntu-core/snapcraft/releases/tag/2.12 > > A great place to collaborate and discuss features, bugs and ideas on > snapcraft is snapcraft at lists.ubuntu.com mailing list or directly in the > #snappy channel on irc.freenode.net. > > To file bugs, please go to https://bugs.launchpad.net/snapcraft/+filebug. > > Happy snapcrafting, > - Sergio and the team > > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/snapcraft > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From vasilisc777 at gmail.com Thu Jun 30 09:14:33 2016 From: vasilisc777 at gmail.com (Vasilisc) Date: Thu, 30 Jun 2016 12:14:33 +0300 Subject: font Message-ID: <5774E2F9.4030400@gmail.com> Developers have the plan to add the interface "fonts"? -- Best regards, vasilisc From didrocks at ubuntu.com Thu Jun 30 09:34:43 2016 From: didrocks at ubuntu.com (Didier Roche) Date: Thu, 30 Jun 2016 11:34:43 +0200 Subject: font In-Reply-To: <5774E2F9.4030400@gmail.com> References: <5774E2F9.4030400@gmail.com> Message-ID: <5774E7B3.1000109@ubuntu.com> Le 30/06/2016 11:14, Vasilisc a écrit : > Developers have the plan to add the interface "fonts"? > > Hey, It's definitively something that we discussed. Jamie would be able to shed some lights on where it is on the roadmap. Meanwhile, if you are using one of the gtk or qt launchers, you will notice that we do have some workarounds in place to have default ubuntu fonts copied inside the snap and picked up by the applications. Cheers, Didier From zygmunt.krynicki at canonical.com Thu Jun 30 09:36:57 2016 From: zygmunt.krynicki at canonical.com (Zygmunt Krynicki) Date: Thu, 30 Jun 2016 11:36:57 +0200 Subject: font In-Reply-To: <5774E7B3.1000109@ubuntu.com> References: <5774E2F9.4030400@gmail.com> <5774E7B3.1000109@ubuntu.com> Message-ID: <2FDAC07D-1B77-468B-A89A-6B255417DE16@canonical.com> > Wiadomość napisana przez Didier Roche w dniu 30.06.2016, o godz. 11:34: > > Le 30/06/2016 11:14, Vasilisc a écrit : >> Developers have the plan to add the interface "fonts”? All the heavy lifting for making fonts interface possible has landed. It will surely be a part of the 2.0.11 release next week. >> >> > > Hey, > > It's definitively something that we discussed. Jamie would be able to > shed some lights on where it is on the roadmap. > Meanwhile, if you are using one of the gtk or qt launchers, you will > notice that we do have some workarounds in place to have default ubuntu > fonts copied inside the snap and picked up by the applications. > > Cheers, > Didier > > -- > Snapcraft mailing list > Snapcraft at lists.ubuntu.com > Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/snapcraft From pawel.stolowski at canonical.com Thu Jun 30 10:05:20 2016 From: pawel.stolowski at canonical.com (Pawel Stolowski) Date: Thu, 30 Jun 2016 12:05:20 +0200 Subject: Fonts issue In-Reply-To: References: <577238DB.4030905@canonical.com> <1467120098.5164.76.camel@canonical.com> <5773A091.70307@canonical.com> Message-ID: <5774EEE0.7030100@canonical.com> Hi Evan, On 29.06.2016 20:44, Evan Dandrea wrote: > On Wed, 29 Jun 2016 at 05:20 Pawel Stolowski > > > wrote: > > > I'm trying to guess what font-related stuff may be missing. I've the > following packages installed via stage-packages: > - ttf-ubuntu-font-family > - xfonts-base > - xfonts-scalable > - fontconfig-config > > Any idea what else to try? > > > If Tcl/Tk is loading fonts via absolute paths, including some font > packages in stage-packages unfortunately won't help. Any files in your > snap will install relative to /snap/your_snap/current, not /. Only > files in the OS snap can be installed to /. I managed to get the app working by using desktop/gtk3 launcher temporarily. This launcher sets a bunch of environment variables, plus some gtk-related stuff (which I don't really need), so now it's just a matter of figuring out what's critical and pull it out into a custom launcher. And it works in strict mode now. > > I ran into this font problem with Jenkins. As it also uses fontconfig, > I provided a custom fonts.conf pointed at the needed fonts I put under > the snap directory [1, 2]. > > If that doesn't work and you need to dig deeper, I've found it useful > to run programs under ptrace and see what files they open (`strace -f > -e file`). > > 1: > https://github.com/evandandrea/jenkins-snap/blob/master/parts/plugins/x-jenkins.py#L12 > 2: > https://github.com/evandandrea/jenkins-snap/blob/master/snapcraft.yaml#L20 Thanks for the links, that's a useful trick. I hope it will soon be much easier when the fonts interface lands (as discussed in a parallel 'font' thread). Cheers, Pawel From daniel.holbach at canonical.com Thu Jun 30 15:34:55 2016 From: daniel.holbach at canonical.com (Daniel Holbach) Date: Thu, 30 Jun 2016 17:34:55 +0200 Subject: Snappy Playpen event next Tuesday Message-ID: <57753C1F.1080402@canonical.com> Hello everybody, Next week on Tuesday, 5th July, we want to have our next Snappy Playpen event. As always we are going to work together on snapping software for our repository [1] on github. Whatever app, service or piece of software you bring is welcome. The focus of last week was ironing out issues and documenting what we currently have. Some outcomes of this were: - documentation of some of our best practices [2], adding a FAQ [3] - a list of known issues [4] - consolidation in terms of using common parts (launchers, etc.) - some general cleanup We want to continue this work, but add a new side to this: upstreaming our work. It is great that we get snaps working, but it is much better if the upstream project in question can take over the ownership of snaps themselves. Having snapcraft.yaml in their source tree will make this a lot easier. To kick off this work, we started some documentation on how to best do that and track this effort [5]. You are all welcome to the event and we look forward to work together with you. Coordination is happening on #snappy on Freenode (IRC) and https://gitter.im/ubuntu/snappy-playpen (Gitter). We will make sure all our experts are around to help you if you have questions. WHAT: Snappy Playpen event WHEN: 5th July 2016 WHERE: #snappy on Freenode or https://gitter.im/ubuntu/snappy-playpen [1] https://github.com/ubuntu/snappy-playpen [2] https://github.com/ubuntu/snappy-playpen/wiki/Examples [3] https://developer.ubuntu.com/snappy/support/faq/ [4] https://github.com/ubuntu/snappy-playpen/wiki/Known-issues [5] https://github.com/ubuntu/snappy-playpen/wiki/Upstreaming Have a great day, Daniel From janssen at parc.com Thu Jun 30 16:05:57 2016 From: janssen at parc.com (Bill Janssen) Date: Thu, 30 Jun 2016 09:05:57 -0700 Subject: listing all the interfaces / sockets? Message-ID: <24070.1467302757@parc.com> I'm still trying to figure out how to access secure storage via the Secret Service API. Is there an interface for that? How would I list all the defined interfaces? How would I know what parts export which sockets? Bill From zygmunt.krynicki at canonical.com Thu Jun 30 16:12:12 2016 From: zygmunt.krynicki at canonical.com (Zygmunt Krynicki) Date: Thu, 30 Jun 2016 18:12:12 +0200 Subject: listing all the interfaces / sockets? In-Reply-To: <24070.1467302757@parc.com> References: <24070.1467302757@parc.com> Message-ID: > Wiadomość napisana przez Bill Janssen w dniu 30.06.2016, o godz. 18:05: > > I'm still trying to figure out how to access secure storage via the > Secret Service API. Is there an interface for that? I don’t think there is one yet. > How would I list all the defined interfaces? How would I know what > parts export which sockets? All of the interfaces as defined in snapd source code, in the interfaces/builtin directory. At runtime you can use REST API to query all the available plugs and slots (and their interface types and connections). This is documented in docs/rest.md. As for parts, I don’t know. Thanks ZK From mark at ubuntu.com Thu Jun 30 16:28:05 2016 From: mark at ubuntu.com (Mark Shuttleworth) Date: Thu, 30 Jun 2016 17:28:05 +0100 Subject: listing all the interfaces / sockets? In-Reply-To: References: <24070.1467302757@parc.com> Message-ID: <57754895.3090604@ubuntu.com> On 30/06/16 17:12, Zygmunt Krynicki wrote: > >> How would I list all the defined interfaces? How would I know what >> parts export which sockets? > 'snap interfaces' is a useful starting point. 'snap interfaces -a' might for example show you available but unused slots / plugs. Mark From janssen at parc.com Thu Jun 30 18:34:03 2016 From: janssen at parc.com (Bill Janssen) Date: Thu, 30 Jun 2016 11:34:03 -0700 Subject: listing all the interfaces / sockets? In-Reply-To: References: <24070.1467302757@parc.com> Message-ID: <28422.1467311643@parc.com> Zygmunt Krynicki wrote: > > > Wiadomość napisana przez Bill Janssen w dniu 30.06.2016, o godz. 18:05: > > > > I'm still trying to figure out how to access secure storage via the > > Secret Service API. Is there an interface for that? > > I don’t think there is one yet. Thanks. Is this something I could build for myself? If I did, how would it get onto my users' machines? What documentation should I look at to see how to do it? Bill > > How would I list all the defined interfaces? How would I know what > > parts export which sockets? > > All of the interfaces as defined in snapd source code, in the interfaces/builtin directory. > > At runtime you can use REST API to query all the available plugs and slots (and their interface types and connections). This is documented in docs/rest.md. > > As for parts, I don’t know. > > Thanks > ZK > From janssen at parc.com Thu Jun 30 18:38:33 2016 From: janssen at parc.com (Bill Janssen) Date: Thu, 30 Jun 2016 11:38:33 -0700 Subject: listing all the interfaces / sockets? In-Reply-To: <57754895.3090604@ubuntu.com> References: <24070.1467302757@parc.com> <57754895.3090604@ubuntu.com> Message-ID: <28575.1467311913@parc.com> Mark Shuttleworth wrote: > On 30/06/16 17:12, Zygmunt Krynicki wrote: > > > >> How would I list all the defined interfaces? How would I know what > >> parts export which sockets? > > > > 'snap interfaces' is a useful starting point. Thanks. I get this: $ snap interfaces Slot Plug :cups-control - :firewall-control - :gsettings - :home goodstuff :locale-control - :log-observe - :mount-observe - :network goodstuff :network-bind - :network-control - :network-manager - :network-observe - :opengl goodstuff :pulseaudio - :snapd-control - :system-observe - :timeserver-control - :timezone-control - :unity7 - :x11 - $ Looks like Zygmunt is right; I don't see anything that looks like secure-storage or secret-service. Hmmm, question: if I'm using :opengl, do I also need to use either :unity7 or :x11? What would be nice is a way to list the CORBA IDL (or some equivalent) for each interface. Bill > 'snap interfaces -a' might > for example show you available but unused slots / plugs. $ snap interfaces -a error: unknown flag `a' $ > > Mark From gustavo at niemeyer.net Thu Jun 30 20:34:02 2016 From: gustavo at niemeyer.net (Gustavo Niemeyer) Date: Thu, 30 Jun 2016 17:34:02 -0300 Subject: All MongoDB releases, everywhere Message-ID: Hello snapcrafters, Hot off the presses, all MongoDB releases in snap form: http://blog.labix.org/2016/06/30/all-mongodb-releases-everywhere Please drop me a note if you have any questions. gustavo @ http://niemeyer.net -------------- next part -------------- An HTML attachment was scrubbed... URL: