upstart socket activation

[Scott James Remnant]

On Mon, Mar 28, 2011 at 11:04 AM, Lennart Poettering
<mzhcfgneg at 0pointer.de> wrote:

> On Mon, 28.03.11 09:49, Scott James Remnant (scott at netsplit.com) wrote:
>
>> > Yes, it should.
>> >
>> > In fact we tried to get Scott into a discussion about a common API
>> > definition for this and other similar smaller interfaces for usage in
>> > daemons. Alas, Scott's only reply was that our approach was "too
>> > simple", and nothing more specific.
>> >
>> No, you have never tried to get me into a discussion about anything.
>
> Mind if I publish the mails I sent you and your responses?
>
Of course - most of them are archived on mailing lists anyway.

>> Every now and then you pop up and tell me that I should do things your
>> way, then when I try and engage with you, you disappear and after a
>> while, the cycle starts all over again.
>
> Oh, right. You engaged with me? I kinda felt talking to /dev/null when I
> sent you mails about these proposals...
>
How so? There were several iterations of patches, each of which varied
the approach based on feedback from everyone - especially including
you. You didn't give any further feedback after the first round.

>> Seems I said pretty much the exact same thing last year. And what
>> happened? I spent a fair amount of time attempting to get you to
>> discuss an interface (others: see the dbus list for December +
>> January, there are several RFC passes of the patches), you went
>> silent, then finally you followed up with:
>
> Dude, that's a completely different issue. don't mix bus and socket activation.
>
Our inability to work together is a fundamental issue, whether it be
about bus activation, socket activation, config formats, etc.

Clearly both of us feel like the other isn't trying.

If we don't address the fundamental issue, we're just going to keep
shouting at each other. And I don't want to do that.

>>         https://bugs.freedesktop.org/show_bug.cgi?id=34526#c1
>>
>>         NAK. I plan to add a very different interface for bus
>> activation in systemd
>>         now, so please do not merge this before this new scheme isn't clear.
>>
>> So after accusing me (again) of not being willing to discuss things
>> with you, I once again tried to engage with you - you didn't bother -
>> and then you NAK'd my attempts at working together because *you* were
>> going to design (on your own) a very different interface for systemd
>> *only*.
>
> That's just two weeks ago, and I was on vacation.
>
> Also, there's a big difference: for the socket activation API i pinged
> you at a time where the code was still in flux, where we could have
> agreed on something easily. Now with your D-Bus patches you are asking
> me to change stuff I already released, it took you a year or so to come
> back on this. Only naturally this will take longer time to
> do. Definitely longer than two weeks.
>
You went straight on vacation after NAKing a patch? And yet posted
dozens of mails to mailing lists, and even committed to D-Bus during
that vacation?

Sorry, not buying that ;-)

> You know, I can easily turn this around: the dbus hookup I prepped got
> zero response from you. Now you prepped your own dbus hookup and you
> already got waaaaaaay more response from me on it then I ever got from
> you. Sure, it wasn't necessarily postive responses, but hey, what do you
> expect if you play it like that?
>
Except I didn't prep "my own" dbus hookup, I attempted (in a few
different ways) to share as much of your hookup code as possible.

And at no point have I been at all hostile about your code being
there, where as you seem to be treating my patch as hostile in some
way?

> I have now commented on that bug, just for you.
>
You NAK'd the patch, why is your comment "just for [me]"?

> Also, I am not the dbus maintainer.
>
You are a dbus committer, as am I. There is no real "maintainer" of
dbus, just those of us that commit to it from time to time.

>> And again, I even responded on that bug to see whether you still
>> wanted to work together. And again, you have ignored it.
>
>> > If the opinions on cooperation on APIs on Canonical's side changed since
>> > then I am all ears. In fact I offered to come to UDS, if you guys want
>> > me to, and maybe we can discuss that there.
>> >
>> You aren't interested in cooperation, Lennart. You are only interested
>> in people doing things your way.
>>
>> I am not going to do things your way, but I am still interested in us
>> finding new ways to do things together, if you would like.
>
> Wow, you are special.
>
So now you're just going to be insulting?

In English, calling someone special is equivalent to calling someone
mentally retarded or disabled in some way. Perhaps this means
something different in German?

> Let me quote you:
>
> http://www.ustream.tv/recorded/11061470 at 1:05:14:
>
> ".. And why they backed off? Because there isn't really an advantage for
> us to use the systemd's API and config format. It's already different
> enough it wouldn't work the same. Compatibility is just merely a away
> ...  a different work for similarity and it would be surprisingly
> similar."
>
Interesting ... of course, if you start watching the video at 00:57:17
which is about the point that I begin talking about it, there are much
more interesting quotes from me:

  "if systemd already exists, why continue with upstart" - maybe this
wasn't your question, but that's the downside of having them relayed
to me via a third-person - it's the question I tried to answer

  "if systemd had existed four years ago, we'd have probably used it"

  "the problem is now, we've been doing this for four years, we've
built an entire operating system around a forward-booting system."

  "we've built entire products we go out and sell to our customers (as
Canonical) which are based on Upstart working the way it does"

  "we have derivatives like Chrome OS and like Palm OS" [sic: I mean
Web OS] "which rely on the way Upstart works, and are actually more
Upstart-native than Ubuntu is and use features of Upstart that Ubuntu
doesn't even go near yet and have added patches to Upstart"

  "can we switch to systemd now? it would be multiple years work"


And the comments I answer at 1:02:00 are also about why we can't just
replace upstart with systemd in Ubuntu as well.

Interesting that you say to start at 1:05:14 and not 1:04:45 ... then you miss

  "I remember us having an agreement that you weren't going to write
systemd and that you were going to contribute network-based
activiation to Upstart"

and jump straight into the quote you give above

  "And why they backed off? Because there isn't really an advantage
for us to use the systemd's API and config format. It's already
different enough it wouldn't work the same."

Now in context of answering a question about Ubuntu switching to
systemd, it should make perfect sense, no?  The "us" and "it" in my
answer is/are Ubuntu

> While what you said there doesn't appear very coherent this message
> appears clear to me: "I have no interest in sharing APIs with systemd".
>
It's not coherent because you cut off the context ... I think my
message there was very clear - "I have no interest in switching Ubuntu
from Upstart to systemd".

I even went to an effort to say, a couple of times, that I wouldn't
get in the way of someone else who *wanted* to do that.


What I've said, over and over again, including in that most recent
D-Bus thread is:

  "I am interested in sharing APIs with systemd, I have no interest in
sharing systemd's APIs"

In other words, I am very definitely willing to work with you to
create new APIs. I am definitely willing to work with you to adapt
*both of our* existing APIs so that they suit you both.

I am not willing to adapt all of your APIs, whether they fit, or not.

And so far none of your APIs fit.

Scott