Bug 0 review pls

[Matt Darcy]

Dan Shearer wrote:
> On Sat, May 31, 2008 at 04:59:10PM +0100, Matt Darcy wrote:
>   
>> Dan Shearer wrote:
>>     
>>> I have put some text for Bug 0 up at 
>>> https://wiki.ubuntu.com/ServerTeam/Bug0
>>>       
>   :
>   
>> The aim of the text in the wiki page appears to (in my opinion) be aimed 
>> at small/medium intergration targets, which is of course a viable 
>> market, hence the focus on things like "gui tools".
>>     
>
> That's interesting (and quite possibly valid) but when I wrote it entire focus
> of my interest was on medium/large integration targets. It needs to be
> understood that in these sorts of corporate markets -- large banks, telcos,
> mil/aero, etc -- GUI tools are a precondition to even having a discussion. It
> is true that Microsoft is now working hard on producing commandline tools, but
> that's an extra. 
>
> Ah, reading further down, you're referring to GUI tools as pretty sugar. Yes,
> that would be a poor goal.  We need things that have the concept of repeating
> operations on groups of servers, tools like webmine don't even know multiple
> servers exist let alone how they might be related! Imagine something with the
> topographical understanding of a well-configured nagios and the detailed
> operation ability of webmin/$BETTER, and with authentication and fine-grained
> ACL control everywhere you browse/manage via LDAP/AD integration, and that's
> what I have in mind. I strongly dislike Microsoft Systems Management Server,
> but out of the box it is better than anything I can get from OSS with less than
> weeks of fiddling. To me GUI tools should build on underlying infrastructure
> for provisioning and configuration management, as well as status reporting and
> hands-on management for entire networks.
>
> As to the rest of your message, I thought there were really good points there
> and I'll reply in another thread about strategy for release cycle and
> maintenance. As a detail, you focussed on real hardware installs
> whereas very many server installs happen on virtual hardware these days. 
>
>   
>> There is also a lack of enterprise focus for me in Ubuntu server 
>> currenty, tools such as Redhats Satellite product which will allow easy 
>> mass managment of RHEL server and desktops are core to businesses 
>> picking this up.
>>     
>
> Has anyone seen a current summary of Linux management products? There are quite
> a few of them including
> http://www.oracle.com/technology/products/oem/omp_linux.html , and there's a
> 2006 article http://www.networkworld.com/newsletters/nsm/2006/0306nsm2.html . I
> don't feel up to date.
>
>   
>> I'd like to see some good focus and discussion on the points raised 
>> above, rather than trying to make "gui" tools for the server release.
>>     
>
> If we have a good provisioning, configuration management, hands-on management
> and monitoring system, but it hasn't got a GUI, medium and larger corporates
> just laugh. You can't even get in the door. I don't say Ubuntu Server
> has these things, but it does also have to have a GUI.
>
>   

Great responses all around, thanks for reading them through properly and 
taking the time to respond in the correct context.

I think we are discussing the same things in terms of gui, when you say 
enterprise managment tools, and I'm referencing thing like Redhat's 
Satellite server, we are on the same page, tools such as managing and 
interacting with LDAP/AD all core stuff that your getting no argument 
from me from. The "sugar gui" you mentioned is something I believe is 
being focused on too much (not from your mail - but from community 
discussion/support in general) as a response to Microsofts "home server" 
product, but as you rightly said thats not the intention of bug0.

The comments I made are from feedback from clients, and I only picked on 
Oracle as an easy example, not the defacto one, I'd be happy to pickup 
this discussion in more detail in a seperate thread perhaps.

Your points on virtual managment was only a valid point I didn't pick up 
on, and today I saw some mail discussion on tools such as Ovirt, which 
I'd love to see tools like this being looked at both as a tool to 
improve managment of a server system in the enterprise, and as an 
alternative way / tool that the current norm.

There is a lot of potential from what your saying, and something I feel 
has been lacking from the server product from an enterprise perspective.

I'll look forward to your comments on the release cycle, as I understand 
how much of a double edged sword that is.

It would be interesting to see some building on your bug0 text linked to 
specific discussion and plans to try to get some real targeted focus and 
consensus.

Again, great responses.

Matt