FAQ Search Today's Posts Mark Forums Read
» Video Reviews

» Linux Archive

Linux-archive is a website aiming to archive linux email lists and to make them easily accessible for linux users/developers.


» Sponsor

» Partners

» Sponsor


 
 
LinkBack Thread Tools
 
Old 06-02-2008, 11:20 PM
"Brett Alton"
 
Default Bug 0 review pls

On Mon, Jun 2, 2008 at 6:54 PM, Anders Häggström
<hagge.lists@intercorner.net> wrote:
> 2008/6/3 Daniel Robitaille <robitaille@gmail.com>:
>> On Sat, May 31, 2008 at 11:59 AM, Matt Darcy
>> <ubuntu.lists@projecthugo.co.uk> wrote:
>>
>>> 2.) LTS fixes and backports. There is no enough fixes, updates and
>>> upgrades to make LTS a viable long term (3 year) stratergy for business
>>> use. Too much focus is on "fix for next release" or "upgrade product for
>>> next release", which relates to point 1. If a server model is to be
>>> considered usable it needs to have regular fixes applied to that
>>> release, not pushed out into current +1. I understand why this is done
>>> as unless the bug is mission critical it makes more sense from an ubuntu
>>> standpoint to target the fix into the next release, as that release will
>>> have updates in and is less than 6 months away. This is not an option
>>> for a long term server audience. The 6.06 release was crippled on later
>>> edition dell servers due to the lack of back ports on the kernel for
>>> specific hardware controllers, if the LTS edition is to be truly LTS,
>>> then I'm afraid kernel updates/back-ports will need to be on the radar
>>> more, and things learnt from the non-LTS products need to be pushed back.
>>
>> Being the owner of a Dell server running 7.04 because that was the
>> only version of Ubuntu that could deal with its hardware when it was
>> set up in the spring of 2007, I totally agree with that paragraph.
>> Doing a double upgrade (7.04 to 7.10 to 8.04) on a production box is
>> not something I look forward in the next 6 months I have before
>> Feisty's support runs out. Given the choice last year, we would have
>> used an older LTS version instead of a currenty non-LTS version, but
>> we simply didn't have the option while continuing using Ubuntu.
>>
>> Daniel
>>
>> --
>> Daniel Robitaille
>> http://friendfeed.com/robitaille
>>
>> --
>> ubuntu-server mailing list
>> ubuntu-server@lists.ubuntu.com
>> https://lists.ubuntu.com/mailman/listinfo/ubuntu-server
>> More info: https://wiki.ubuntu.com/ServerTeam
>>
>
> I am an Ubuntu user (both desktop and server platforms) and I want to
> give my point of view. When I install a server with LTS I want to be
> able to update bugs found in the software for the whole LTS-period. If
> I install a new server (with fresh hardware) in the middle of an
> LTS-period I want to be able to use the last released LTS-release and
> upgrade drivers to support my new hardware that was not supported when
> the LTS-release was first released. I do not want to install a normal
> release just because the LTS-resease didn't support my hardware at the
> release time (when for example my hardware was not manufactured yet).
>
> For the GUI discussion I don't want the server depending on Xorg. I
> would rather see a good integrated webinterface with little overhead
> and nice ajax-applications for various tasks on the server.
>
> Thanks for your time!
> // Anders
>
> --
> ubuntu-server mailing list
> ubuntu-server@lists.ubuntu.com
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-server
> More info: https://wiki.ubuntu.com/ServerTeam
>

That makes a lot of sense. Almost all server users want to use the LTS
version as their needs for the absolute latest and greatest is not as
pressing as a desktop user's but the need for support is much higher.

Only releasing LTSes might be a bit of a stretch (6.06 -> 8.04 ->
10.04) but if the point releases had more meaning behind them, like
Anders said, supporting newer hardware, then releasing only LTSes
might make more sense.

This becomes more apparent when you realize that developers are
supporting up to four (!) releases at the same time. Currently
(Dapper, Feisty, Gutsy, Hardy). By the next LTS it will be five (!!)
releases (Dapper, Hardy, Intrepid+1, Intrepid+2, Intrepid+3 LTS) [see
http://www.markshuttleworth.com/wp-content/uploads/2008/05/ubuntu-release-cycle.png
and then remember that Dapper is supported until
2011-06]. If they only released LTSes with major point upgrades then
they would have to handle three releases at maximum.

That sounds easy to me.

I also got frustrated last summer when I had to rollout my new Dell
PowerEdge 2950 with a release (Feisty) that was only supported for 18
months because Dapper couldn't install properly on it. If 6.06.1 added
support for my new hardware then I would have been able to enjoy five
years of support, just like an IT manager would expect from their
server OS.

18 months just doesn't cut it.

--
Brett Alton
brett.jr.alton@gmail.com

Do you really need to print this email? Help preserve our environment!
--
ubuntu-server mailing list
ubuntu-server@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-server
More info: https://wiki.ubuntu.com/ServerTeam
 
Old 06-02-2008, 11:58 PM
"Aaron Kincer"
 
Default Bug 0 review pls

Focus on the core technology that makes Microsoft infrastructure take deep roots. They are (IMO):

1) Exchange
2) Active Directory
3) Sharepoint

Attacking these is not strictly a server issue. For many people, they couldn't care less about their operating system as long as they've got their MS Office/Outlook running. The Wine team is part of the solution, but for those devs who work for Codeweavers, they seem to reserve their enterprise efforts for getting Office only for the (partially) closed Crossover product.


For Exchange, the non-free version of Zimbra works great for Outlook connectivity.

For Sharepoint, the non-free layer on Plone by Enfold Systems is a great alternative except that one of the more attractive features Enfold offers (versioning) is only available on their Windows product. For those that are learned, I know versioning exists for Plone in general. I'm talking about specifically the Enfold products.


By the way, anybody noticing a trend yet for the enterprise-class versions of software that can fill a niche occupied by Microsoft?

The problem is that creating fully open source software to fill enterprise niches is non-trivial and the best model of it so far I've seen is Samba. The Samba team deserves some serious credit for the work they've done.


If only there were more billionaires putting their money towards open source software to pay lots of programmers to attack the problems, all of these pieces of the puzzle could be more quickly solved.

On the off chance that anybody is good friends with Warren Buffett, see if you can convince him to make significant investment into Yahoo* with strings attached so that they will open source all of Zimbra. That would really turbo charge this idea. Hey, it could happen.


Aaron Kincer


On Fri, May 30, 2008 at 8:11 PM, Dan Shearer <dan@shearer.org> wrote:

I have put some text for Bug 0 up at

https://wiki.ubuntu.com/ServerTeam/Bug0#preview . I didn't get into the

solutions we worked on at UDS, thinking this is what bug report followup

comments are for and the body was already too long.



Edit away people, but please don't try to turn it into a classic bug

report. This bug has a different purpose :-)



Regards,



--

Dan Shearer

dan@shearer.org



--

ubuntu-server mailing list

ubuntu-server@lists.ubuntu.com

https://lists.ubuntu.com/mailman/listinfo/ubuntu-server

More info: https://wiki.ubuntu.com/ServerTeam



--
ubuntu-server mailing list
ubuntu-server@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-server
More info: https://wiki.ubuntu.com/ServerTeam
 
Old 06-03-2008, 12:17 AM
Dan Shearer
 
Default Bug 0 review pls

On Mon, Jun 02, 2008 at 07:58:24PM -0400, Aaron Kincer wrote:
> Focus on the core technology that makes Microsoft infrastructure take deep
> roots. They are (IMO):
>
> 1) Exchange
> 2) Active Directory
> 3) Sharepoint
>
> Attacking these is not strictly a server issue. For many people, they
> couldn't care less about their operating system as long as they've got their
> MS Office/Outlook running. The Wine team is part of the solution

I got a bit lost here, can you explain how you see Wine helping deliver
solutions that don't come from Microsoft?

> For Exchange, the non-free version of Zimbra works great for Outlook
> connectivity.

Zimbra is fine for people who are able to modify their client Windows
machines. And that is something many, many large corporates refuse to
do, often for very good configuration management reasons. Until recently
this has been an impasse, because nothing else is a native Exchange
server except for Exchange. OpenChange is addressing that impasse.

> The problem is that creating fully open source software to fill enterprise
> niches is non-trivial and the best model of it so far I've seen is Samba.
> The Samba team deserves some serious credit for the work they've done.

With SAmba the barriers have been first and foremost non-technical.
Thanks to legal work in the EU the non-technical barriers have gone
away, and now Samba development has changed nature dramatically. For
Intrepid, there won't be a full Active Directory server. But there can
potentially be two kinds of proxies in Intrepid (CIFS and AD) that will
help stop companies making their internal borg bigger.

> On the off chance that anybody is good friends with Warren Buffett, see if
> you can convince him to make significant investment into Yahoo with strings
> attached so that they will open source all of Zimbra. That would really
> turbo charge this idea. Hey, it could happen.

Zimbra can only help in the rip-and-replace scenario, which is pretty
rare in large corporates. Zimbra can't be dropped in to a group of
Exchange servers, it can't natively interoperate with MAPI clients and
existing MAPI infrastructure including backup tols, it can't be a MAPI
client itself. In an Exchange infrastructure with thousands of mailboxes
you have to commit to replacing all of the servers *and* all of the
surrounding infrastructure.

I'm not against Zimbra, but I don't think it addresses the most common
use cases involving Exchange in corporates.

--
Dan Shearer
dan@shearer.org

--
ubuntu-server mailing list
ubuntu-server@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-server
More info: https://wiki.ubuntu.com/ServerTeam
 
Old 06-03-2008, 12:24 AM
Dan Shearer
 
Default Bug 0 review pls

On Tue, Jun 03, 2008 at 09:47:19AM +0930, Dan Shearer wrote:
> On Mon, Jun 02, 2008 at 07:58:24PM -0400, Aaron Kincer wrote:
> > Focus on the core technology that makes Microsoft infrastructure take deep
> > roots. They are (IMO):
> >
> > 1) Exchange
> > 2) Active Directory
> > 3) Sharepoint
> >
> > Attacking these is not strictly a server issue. For many people, they
> > couldn't care less about their operating system as long as they've got their
> > MS Office/Outlook running. The Wine team is part of the solution
>
> I got a bit lost here, can you explain how you see Wine helping deliver
> solutions that don't come from Microsoft?

Actually when I re-read it, I wasn't lost, just confused about the
implied goals :-) There's a very big difference between the use cases
that involve keeping the MS desktop untouched and moving away from MS
servers, and keeping the MS servers and moving the clients to
Linux+Crossover/Wine. What would the goals be where one is a substitute
approach for the other?

--
Dan Shearer
dan@shearer.org

--
ubuntu-server mailing list
ubuntu-server@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-server
More info: https://wiki.ubuntu.com/ServerTeam
 
Old 06-03-2008, 01:52 AM
Scott Kitterman
 
Default Bug 0 review pls

On Monday 02 June 2008 19:20, Brett Alton wrote:

> I also got frustrated last summer when I had to rollout my new Dell
> PowerEdge 2950 with a release (Feisty) that was only supported for 18
> months because Dapper couldn't install properly on it. If 6.06.1 added
> support for my new hardware then I would have been able to enjoy five
> years of support, just like an IT manager would expect from their
> server OS.

Did 6.06.2 solve the problem for you?

Scott K

--
ubuntu-server mailing list
ubuntu-server@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-server
More info: https://wiki.ubuntu.com/ServerTeam
 
Old 06-03-2008, 02:48 AM
"Aaron Kincer"
 
Default Bug 0 review pls

There are many ways to look at Wine, but I'll point out one obvious _potential_ use of the technology some day. There are many "server" applications that could be run on a Linux box with Wine. If you stop and digest what that suggests as a possibility, you can see that server side apps could be migrated as is.


Yes, Wine isn't really meant for that. I'm merely pointing out the tantalizing possibility. And I'm not talking about MS server products necessarily either. There are a whole host of other companies that just don't offer Linux versions of their products. I'm specifically thinking of legacy server apps that may one day be abandoned by their vendor or the vendor goes out of business.


By the way, I'm not sure what you mean by "modify their client Windows machines", but with Zimbra, you don't have to modify. Reconfigure, maybe. But you don't even have to touch a machine. You can create Outlook profiles and push those out with login scripts or via GPO (I believe).


Migrating to Exchange is then a snap as you can do what's called a split domain where you migrate users at whatever pace you wish. Heck, you could time it for when a user has a problem and you have to work on their machine so you can make sure the transition is smooth. Eliminate Exchange via attrition.


I'm not pimping Zimbra, I just don't think your assertion of complex barriers is accurate. Your assertion that you can't drop Zimbra in a group of Exchange servers is definitely not accurate. Unless you mean make it a member of a cluster. If that is what you mean then you are correct.


To put it simply, there is a clear and manageable migration path from Exchange to Zimbra if one is willing. But that wasn't the original topic of conversation as the full blown product is not a FOSS stack. I wish it were, but it isn't.


Aaron Kincer

On Mon, Jun 2, 2008 at 8:24 PM, Dan Shearer <dan@shearer.org> wrote:

On Tue, Jun 03, 2008 at 09:47:19AM +0930, Dan Shearer wrote:

> On Mon, Jun 02, 2008 at 07:58:24PM -0400, Aaron Kincer wrote:

> > Focus on the core technology that makes Microsoft infrastructure take deep

> > roots. They are (IMO):

> >

> > 1) Exchange

> > 2) Active Directory

> > 3) Sharepoint

> >

> > Attacking these is not strictly a server issue. For many people, they

> > couldn't care less about their operating system as long as they've got their

> > MS Office/Outlook running. The Wine team is part of the solution

>

> I got a bit lost here, can you explain how you see Wine helping deliver

> solutions that don't come from Microsoft?



Actually when I re-read it, I wasn't lost, just confused about the

implied goals :-) There's a very big difference between the use cases

that involve keeping the MS desktop untouched and moving away from MS

servers, and keeping the MS servers and moving the clients to

Linux+Crossover/Wine. What would the goals be where one is a substitute

approach for the other?



--

Dan Shearer

dan@shearer.org



--
ubuntu-server mailing list
ubuntu-server@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-server
More info: https://wiki.ubuntu.com/ServerTeam
 
Old 06-03-2008, 04:24 AM
"Dustin Kirkland"
 
Default Bug 0 review pls

On Mon, Jun 2, 2008 at 5:54 PM, Anders Häggström
<hagge.lists@intercorner.net> wrote:
> If I install a new server (with fresh hardware) in the middle of an
> LTS-period I want to be able to use the last released LTS-release and
> upgrade drivers to support my new hardware that was not supported when
> the LTS-release was first released. I do not want to install a normal
> release just because the LTS-resease didn't support my hardware at the
> release time (when for example my hardware was not manufactured yet).

As I understand it, LTS releases are allowed (intended?) to add
support new hardware in the dot-releases.

As you encounter hardware that does not work with a given release of
Ubuntu, by all means, please report it in Launchpad. It may well be
fixed in the current Ubuntu release first for development/test
purposes. Conditions allowing, however, those fixes can then be
proposed for -backports to the LTS release.

It's essential, though, that owners of the affect hardware work
through the appropriate Launchpad channels.

:-Dustin

--
ubuntu-server mailing list
ubuntu-server@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-server
More info: https://wiki.ubuntu.com/ServerTeam
 
Old 06-03-2008, 04:48 AM
"Dustin Kirkland"
 
Default Bug 0 review pls

On Mon, Jun 2, 2008 at 6:14 PM, Dan Shearer <dan@shearer.org> wrote:
> I was trying to catch people's attention mostly, wanting to focus on
> what the Hitchhiker's Guide to the Galaxy would define as an SEP Field -
> Someone Else's Problem and therefore invisible to everyone. From the
> feedback I got back at the time this much worked. Now to finesse it.

+1 for the HHGTTG reference ;-)

> My justification for that is that Linux was a success on the server
> before it was anywhere on the client, and the failure to compete on the
> server is an embarassment first and a challenge second.

Linux, perhaps. Ubuntu, not so. Bug #1 is about Ubuntu.

> As it happens there's a new set of opportunties and we aren't limited to
> just aping wherever Microsoft has gone... but refusing to acknowledge
> clear failings does not bode well for the bold new wave.

True--tons of new opportunities. Personally, I'm more interested in
new frontiers than replacing Microsoft infrastructure as such Active
Directory and Exchange servers. I think it would be great if Ubuntu
were the default choice for Web 2.0 servers, such as wiki's, gobby's,
pastebin's, streaming media, etc.

> As to numbers... I think even if this is Bug #7346492, if people take
> the content seriously it's going to be turning up in the mainstream
> press and maybe even in Microsoft PR.

2^2,079,460,347 perhaps? ... Strangely, this is also the telephone
number of an Islington flat where Arthur Dent went to a fancy dress
party, and met a very nice young woman whom he totally blew it with.
;-)

> [chopped advice on going through the lp process. I'll study it and
> probably just follow it as given. So far I have worked out that
> Blueprint == MRD in enterprise speak.]

Right, "Blueprints" are equivalent to "Line Items" in IBM-speak. It's
a loose suggestion of work that could be done. Note that the
"Drafter", "Assignee", and "Approver" can all be different people.
You can write a blueprint, approved by someone, and implemented by a
third person (or team). Perhaps someone from Canonical will work on
the item, or perhaps you or someone else in the community will run
with it.

> I get the feeling that if people just create content in launchpad it
> gets ignored unless there was buyin in the first place... is that fair?

True. It sounds like the UDS->Blueprint->Spec process has been
changed recently. Based on the sessions, discussions, and lessons
learned at UDS Prague, we're rapidly writing Blueprints and Specs for
the functionality we'd like to see make it into Intrepid.

See the Intrepid schedule, https://wiki.ubuntu.com/IntrepidReleaseSchedule

Specification (Blueprints) are due by June 5th (that's Thursday). The
powers that be will go through the submitted Blueprints and explicitly
bless some of them for Intrepid.

> Can you point me to anything online about these discussions, or should I
> start blueprinting?
...
> I'll need some help working through the process. Thanks for your advice.

If I understand the process correctly, Blueprinting (and the
corresponding Specs) should represent the results of discussions and
sessions at UDS. At least that's the intention. I'd say you can
start Blueprinting anything we discussed at UDS and you felt like you
got some support behind. Understand that not ever Blueprint can be
"approved" for the current release. Keep in touch with people in
#ubuntu-server and #ubuntu-devel to make sure you're on the right
track and not spinning your wheels.

:-Dustin

--
ubuntu-server mailing list
ubuntu-server@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-server
More info: https://wiki.ubuntu.com/ServerTeam
 
Old 06-03-2008, 08:05 AM
Serge van Ginderachter
 
Default Bug 0 review pls

----- "Brett Alton" <brett.jr.alton@gmail.com> wrote:

> > I am an Ubuntu user (both desktop and server platforms) and I want
> to
> > give my point of view. When I install a server with LTS I want to be
> > able to update bugs found in the software for the whole LTS-period.
> If
> > I install a new server (with fresh hardware) in the middle of an
> > LTS-period I want to be able to use the last released LTS-release
> and
> > upgrade drivers to support my new hardware that was not supported
> when
> > the LTS-release was first released. I do not want to install a
> normal
> > release just because the LTS-resease didn't support my hardware at
> the
> > release time (when for example my hardware was not manufactured
> yet).

> That makes a lot of sense. Almost all server users want to use the LTS
> version as their needs for the absolute latest and greatest is not as
> pressing as a desktop user's but the need for support is much higher.
>
> Only releasing LTSes might be a bit of a stretch (6.06 -> 8.04 ->
> 10.04) but if the point releases had more meaning behind them, like
> Anders said, supporting newer hardware, then releasing only LTSes
> might make more sense.
>
> This becomes more apparent when you realize that developers are
> supporting up to four (!) releases at the same time. Currently
> (Dapper, Feisty, Gutsy, Hardy). By the next LTS it will be five (!!)
> releases (Dapper, Hardy, Intrepid+1, Intrepid+2, Intrepid+3 LTS) [see
> http://www.markshuttleworth.com/wp-content/uploads/2008/05/ubuntu-release-cycle.png
> and then remember that Dapper is supported until
> 2011-06]. If they only released LTSes with major point upgrades then
> they would have to handle three releases at maximum.
>
> That sounds easy to me.

Let me throw in some post of a CentOS guy (Dag Wieers) which (rightfully, I think) points out some of the difficulties about supporting releases during a lot of years.

http://dag.wieers.com/blog/ubuntus-need-to-catch-a-wave

Now, I'm not sure what the point of this article exactly was, so I can't say I agree with the general tence of it, but let me repeat I just mention this article as extra info in this thread. He does put his finger on what is needed when one wants to support servers for enterprise environments.






Serge

Serge van Ginderachter http://www.vanginderachter.be/

Kreeg u een "odt" bestand en kan u deze niet openen? Zie http://ginsys.be/odf

--
ubuntu-server mailing list
ubuntu-server@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-server
More info: https://wiki.ubuntu.com/ServerTeam
 
Old 06-03-2008, 08:13 AM
Serge van Ginderachter
 
Default Bug 0 review pls

----- "Dan Shearer" <dan@shearer.org> wrote:
> > For Exchange, the non-free version of Zimbra works great for
> Outlook
> > connectivity.
>
> Zimbra is fine for people who are able to modify their client Windows
> machines. And that is something many, many large corporates refuse to
> do, often for very good configuration management reasons. Until
> recently
> this has been an impasse, because nothing else is a native Exchange
> server except for Exchange. OpenChange is addressing that impasse.
>
> > The problem is that creating fully open source software to fill
> enterprise
> > niches is non-trivial and the best model of it so far I've seen is
> Samba.
> > The Samba team deserves some serious credit for the work they've
> done.
>
> With SAmba the barriers have been first and foremost non-technical.
> Thanks to legal work in the EU the non-technical barriers have gone
> away, and now Samba development has changed nature dramatically. For
> Intrepid, there won't be a full Active Directory server. But there
> can
> potentially be two kinds of proxies in Intrepid (CIFS and AD) that
> will
> help stop companies making their internal borg bigger.
>
> > On the off chance that anybody is good friends with Warren Buffett,
> see if
> > you can convince him to make significant investment into Yahoo with
> strings
> > attached so that they will open source all of Zimbra. That would
> really
> > turbo charge this idea. Hey, it could happen.
>
> Zimbra can only help in the rip-and-replace scenario, which is pretty
> rare in large corporates. Zimbra can't be dropped in to a group of
> Exchange servers, it can't natively interoperate with MAPI clients
> and
> existing MAPI infrastructure including backup tols, it can't be a
> MAPI
> client itself. In an Exchange infrastructure with thousands of
> mailboxes
> you have to commit to replacing all of the servers *and* all of the
> surrounding infrastructure.
>
> I'm not against Zimbra, but I don't think it addresses the most
> common
> use cases involving Exchange in corporates.


I couldn't agree more with this. Whether we like ot or not, A lot of Microsoft technologies have become the (de facto) standard for long.

Every solution who tries to replace them with yet another solution, won't cut it far enoug. Whether that new soluion is proprietary or not, doesn't make any difference. Application compatibility is what it's all about, and Outlook is one of those killer apps people want to stick to. Zimbra's Outlook connector is good technology, but it remains an extra step.

Choices will only be viable when everybody speaks the same (MAPI) langue. The problem with standards is that there are too many to choose from.






Serge

Serge van Ginderachter http://www.vanginderachter.be/

Kreeg u een "odt" bestand en kan u deze niet openen? Zie http://ginsys.be/odf

--
ubuntu-server mailing list
ubuntu-server@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-server
More info: https://wiki.ubuntu.com/ServerTeam
 

Thread Tools




All times are GMT. The time now is 07:05 PM.

VBulletin, Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Content Relevant URLs by vBSEO ©2007, Crawlability, Inc.
Copyright ©2007 - 2008, www.linux-archive.org