20100331

Agenda

Items we will be discussing:

Minutes

Beta2 status

ttx reminded everyone that Beta2 Freeze was around the corner.

There are a few beta2-targeted bugs:

Most of the Work Items targeted for beta2 are not impacted by the BetaFreeze, except for:

  • puppet/uec-ec2: integrate upstream patches for testsuite (mathiaz). Patches are available from upstream. However mathiaz may not have time to integrate them in time for BetaFreeze.

  • id-mgt-ref-env: package latest sssd (mathiaz): He may not have the time to package this bug-fix-only update before BetaFreeze.

ttx mentioned the addition of a work item covering libvirt compatibility testing. kirkland said he planned to work on libvirt with jdstrand a full day.

ScottK added that clamav reached RC2 for version 0.96. Testing is welcome and packages are available from the ubuntu-clamav PPA.

Weekly Updates & Questions for the QA Team

hggdh reported he had run the stress and scalability testing on the UEC testing network for Beta1. There are 2 topologies left for testing. Multi-network testing still requires a lot of manual steps. kirkland mentioned that he had the hardware to test multi-network but testing that topology was time consuming.

ttx asked about adding UEC topologies test cases to the QA tracker. hggdh replied he would add them today.

hggdh wondered whether the daily -server isos should be tested on a regular basis. ttx replied that pre/post milestones were the most important images to test. If the tests are fully automated and don't require a lot of hand-holding running them on the dailies could be useful.

Weekly Updates & Questions for the Kernel Team (jjohansen)

jjohansen reported that two bugs related to EC2 had been fixed in the latest kernel upload. Soft lockups are caused by a variety of things but they shouldn't happen near as often.

Server Papercuts status for beta2

ttx reported that 8 bugs were fixed while 2 were left open. ttx would like to discuss how successful the project was and if it should be done again, even on non-LTS cycles. He also confirmed that the project was now closed for this cycle and new nominations wouldn't be accepted anymore.

Bug zapping status (kirkland)

kirkland reported a successful UEC bug zap from last week. mathiaz mentioned he was running a MysQL Bug zap this week.

Community participation has been rather low for both the bug zapping and server paper cut projects. A session at the next UDS should be dedicated to improve Community participation.

Information about server survey (nijaba)

nijaba reported that the result of the Server survey were available to anyone in the community that might need it.

UDS-M topics

ScottK brought up the idea of having a session at UDS on what we wanted the next LTS to look like to have an idea of longer term stuff.

Agree on next meeting date and time

Next meeting will be on Wednesday, April 7th at 14:00 UTC in #ubuntu-meeting.

Log

[15:02] <ttx> #startmeeting
[15:02] <MootBot> Meeting started at 09:02. The chair is ttx.
[15:02] <MootBot> Commands Available: [TOPIC], [IDEA], [ACTION], [AGREED], [LINK], [VOTE]
[15:02] <ttx> Welcome to the Ubuntu Server team meeting
[15:02] <ttx> Agenda is: https://wiki.ubuntu.com/ServerTeam/Meeting
[15:02] <ttx> Today's scribe is: mathiaz !
[15:02] <ttx> [TOPIC] Review ACTION points from previous meeting
[15:02] <MootBot> New Topic:  Review ACTION points from previous meeting
[15:03] <ttx> Only one action, mathiaz to discuss with cjwatson about ubuntu-server upload team
[15:03] <mathiaz> ttx: not done
[15:03] <mathiaz> ttx: replace cjwatson with persia
[15:03] <ttx> mathiaz: maybe we shouldn't have it as an ACTION todo before the next meeting ?
[15:03] <mathiaz> ttx: ok
[15:04] <ttx> if you think you won't have time to complete it by next week, I don't see the point :)
[15:04] <ttx> mathiaz: Keep it warm in your GTD ?
[15:04] <mathiaz> ttx: it is
[15:04] <mathiaz> ttx: just lots other stuff there as well
[15:04] <ttx> [TOPIC] Beta2 status
[15:04] <MootBot> New Topic:  Beta2 status
[15:05] <ttx> So tomorrow, Beta2Freeze will be upon us
[15:05] <ttx> which means today is probably the last moment to get your non-exceptional fixes in
[15:05] <ttx> Starting tomorrow, they will require a BetaFreezeException.
[15:06] <ttx> Remaining beta2-targeted bugs:
[15:06] <ttx> bug 292971 (mathiaz)
[15:06] <ubottu> Launchpad bug 292971 in libnss-ldap "nscd leaking memory using libnss-ldap" [High,Triaged] https://launchpad.net/bugs/292971
[15:06] <mathiaz> ttx: that will probably require a BetaFreezeException
[15:06] <ttx> mathiaz: ack
[15:06] <mathiaz> ttx: as I don't plan to work on it before tomorrow
[15:07] <mathiaz> (tomorrow is the second MySQL bug zap day)
[15:07] <ttx> bug 493593 (zul)
[15:07] <ubottu> Launchpad bug 493593 in scgi "MIR for paste." [High,Fix released] https://launchpad.net/bugs/493593
[15:07] <mathiaz> arg - s/tomorrow/today/
[15:07] <ttx> and bug 408402 (zul)
[15:07] <ubottu> Launchpad bug 408402 in facter "Main Inclusion Report for facter." [Undecided,Fix released] https://launchpad.net/bugs/408402
[15:07] <ttx> Those are MIR bugs that still need to progress on some deps
[15:08] <ttx> A recent search turned out a few bugs milestoned to beta2 but not targeted to lucid
[15:08] <ttx> LP downtime prevented me from triaging them
=== jamie is now known as JamieBennett
[15:09] <ttx> I think at least one of them will fall into our plate:
[15:09] <ttx> bug 532733 (kirkland)
[15:09] <ubottu> Launchpad bug 532733 in qemu-kvm "apt/dpkg in qemu-system-arm hangs if a big task is installed" [High,Confirmed] https://launchpad.net/bugs/532733
[15:09] <kirkland> ttx: that probably won't be mine
[15:09] <kirkland> ttx: that'll be lool or ogra
[15:09] <ttx> ah
[15:09] <ttx> ok
[15:09] <kirkland> ttx: i'm happy to help them
[15:09] <kirkland> ttx: but it's very specific to their use case
[15:10] <ttx> ok, I'll ping you if there is anything left on that list that we need to care for before Freeze or Release
[15:10] <ttx> Looking at Beta2-targeted work items
[15:10] <ttx> on http://people.canonical.com/~pitti/workitems/canonical-server-ubuntu-10.04-beta-2.html
[15:10] <ttx> I think all the work items that needed to be completed before the Freeze were done
[15:11] <ttx> with maybe the exception of:
[15:11] <ttx> puppet/uec-ec2: integrate upstream patches for testsuite
[15:11] <ttx> id-mgt-ref-env: apockage latest sssd
[15:11] <mathiaz> ttx: I may have time to do that
[15:11] <mathiaz> ttx: (ie puppet patches)
[15:12] <mathiaz> ttx: upstream closed all the bugs yesterday stating there is a patch available
[15:12] <ttx> mathiaz: ah, good news
[15:12] <mathiaz> ttx: if I don't get it done by beta2 freeze, I'll postpone it to final
[15:12] <ttx> mathiaz: package sssd -> postponed ?
[15:12] <mathiaz> ttx: as fixing the test suite is not beta2 critical
[15:13] <mathiaz> ttx: sssd yes - If I don't have the time to do it
[15:13] <mathiaz> ttx: today
[15:13] <ttx> ok
[15:13] <ttx> Otherwise I think we are on track for Beta2, if you think you aren't, please shout now.
[15:13] <ttx> kirkland: note that I added a libvirt 0.7.7 compatibility test work item to the euca spec
[15:13] <ttx> to reflect the ongoing discussion with jdstrand
[15:14] <kirkland> ttx: i'm meeting up with jdstrand on thursday to work on libvirt
[15:14] <ttx> (can be done post-Freeze)
[15:14] <ttx> kirkland: cool !
[15:14] <kirkland> ttx: yeah, planning on spending a full day on libvirt
[15:14] <ScottK> New for Beta 2 we now have RC2 of clamav 0.96.  Unless something major comes up this will be their last RC before they release, so if anyone cares about clamav, now's a good time for testing (packages for earlier releases in the ubuntu-clamav PPA).
[15:14] <kirkland> ttx: if anyone else wants to help with libvirt testing this week, sync up with me and jdstrand ;-)
[15:15] <ttx> ScottK: ack
[15:15] <ttx> [TOPIC] Weekly Updates & Questions for the QA Team (hggdh)
[15:15] <MootBot> New Topic:  Weekly Updates & Questions for the QA Team (hggdh)
[15:15] <ttx> hggdh: hi !
[15:15] <hggdh> hi, life is good. A question
[15:16] <hggdh> stress and stability tests for UEC -- mathiaz's tests are enough?
[15:16] <hggdh> If so, I am done with B1
[15:16] <mathiaz> hggdh: do you mean the multi_config tests?
[15:16] <hggdh> yes
[15:16] <ttx> hggdh: I think so... unless kirkland or mathiaz have hidden tests up their sleeves
[15:16] <mathiaz> hggdh: ie starting 20 instances and make sure they all work?
[15:17] <hggdh> mathiaz: I ran with 20, 40, 100, 300
[15:17]  * kirkland high fives hggdh 
[15:17]  * hggdh blushes
[15:17] <mathiaz> hggdh: great - that should do it for stress/scalability tests
[15:17] <nealmcb> :)
[15:17] <kirkland> hggdh: if you got 300 instances running, you've been me :-)
[15:17] <kirkland> beaten
[15:17] <mathiaz> hggdh: have you gone through all the topologies?
[15:17] <hggdh> 2 to go through
[15:18] <hggdh> so I am (yet) not done
[15:18] <ttx> Should we plan to run the multi-network topology by hand ?
[15:18] <mathiaz> hggdh: multi-network being one of them?
[15:18] <mathiaz> ttx: for the time being yes
[15:18] <ttx> ok
[15:19] <hggdh> mathiaz: yes. multi-network is higjly hand-driven
[15:19] <mathiaz> ttx: I may be able to deploy it on the UEC test rig
[15:19] <ttx> kirkland: you managed to run a CC with 2 NICs on your setup ?
[15:19] <kirkland> ttx: i have not
[15:20] <ttx> kirkland: I suggest you go and buy a USB Gb NIC
[15:20] <ttx> if that's the blocker
[15:20] <kirkland> ttx: okay, so I should focus on testing multi-network at this point?
[15:20] <ttx> kirkland: I wouldn't say "focus", but having the ability to run that test would be good
[15:21] <ttx> kirkland: otherwise it sounds like I'm the only one to test that.
[15:21] <kirkland> ttx: well, it's a time-consuming effort, so if it's something i need to work on, it will need to be my "focus"
[15:21] <ttx> kirkland: libvirt has more priority
[15:21] <kirkland> ttx: i have the hardware to do so, I believe
[15:21] <kirkland> ttx: okay, thanks, that's what I needed
[15:21] <ttx> hggdh: I had a question for you about Beta2 ISO testcases. Any way we can preview what you plan to have for beta2, so that we can fix testcases and report issues ?
[15:22] <ttx> hggdh: in particular, I mentioned having tests for UEC install topologies in the ISo tracker for amd64
[15:22] <ttx> hggdh: is that done, any way to preview it ?
[15:23] <hggdh> ttx, no changes from what is already there, no time
[15:23] <ttx> hggdh: is the plan to change those before beta2 ?
[15:23] <hggdh> got my hands full figuring out how uec runs
[15:23] <ttx> hggdh: do you need our help ?
[15:24] <ttx> hggdh: could someone else working on the tracker (ara ?) help ?
[15:24] <hggdh> right now I am finally -- I think -- getting it
[15:24] <ttx> hggdh: if the cases are created, I can certainly document them.
[15:24] <hggdh> ttx will do them today. This is why I also asked if the stress tests were done for B1
[15:24] <ttx> ok
[15:25] <ttx> hggdh: keep me posted...
[15:25] <ttx> anything else for QA ?
[15:25] <hggdh> no, not right now
[15:25] <ttx> [TOPIC] Weekly Updates & Questions for the Kernel Team (jjohansen)
[15:25] <MootBot> New Topic:  Weekly Updates & Questions for the Kernel Team (jjohansen)
[15:25] <ttx> hggdh: thanks
[15:25] <ttx> jjohansen: hi
[15:25] <jjohansen> hi
[15:26] <jjohansen> fixes for Bug #540378, Bug #527208 have gone into the beta2 kernel
[15:26] <ubottu> Launchpad bug 540378 in linux-ec2 "BUG: soft lockup - CPU#1 stuck for 66s! [swapper:0]" [Medium,New] https://launchpad.net/bugs/540378
[15:26] <ubottu> Launchpad bug 527208 in linux-ec2 "ec2 instance fails boot, no console output on c1.xlarge" [High,Confirmed] https://launchpad.net/bugs/527208
[15:27] <ttx> jjohansen: great ! any reason why the bugs are not marked fixed yet ? You need to upload them ?
[15:27] <jjohansen> no promise that it will fix all soft lockups as they are caused by a variety of things but they shouldn't happen near as often
[15:27] <jjohansen> ttx: hrrm, I have to check into that, they were just sucked in yesterday
[15:28] <leoquant> 2/part
[15:28] <ttx> jjohansen: if they are in the pipe, I'm happy with it
[15:28] <ttx> anything else from/to kernel ?
[15:29] <jjohansen> I can't think of anything on my end
[15:29] <ttx> ok, thanks !
[15:29] <ttx> [TOPIC] Server Papercuts status for beta2
[15:29] <MootBot> New Topic:  Server Papercuts status for beta2
[15:29] <ttx> So we are nearing the end of the Server papercuts effort
[15:29] <ttx> as a papercut is not a valid BetaFreezeException motivation
[15:30] <ttx> Status at: https://launchpad.net/server-papercuts/+milestone/lucid-beta-2
[15:30] <ttx> We have 8 bugs fixed (the inprogress one will move to fixreleased as soon as LP unfreezes)
[15:30] <ttx> 1 invalid, 2 left open
[15:31] <ttx> feel free to try to fix the 2 last "confirmed" ones if you have the time
[15:31] <ttx> We'll discuss at UDS-M how successful it was and if we should do it again, even on non-LTS cycles
[15:32] <ttx> I think it failed to motivate new contrributors, but still had a positive effect overrall
[15:32] <ttx> Comments on that ?
[15:33] <hggdh> ttx: we should discuss on UDS-M how to motivate them contributors
[15:33] <kirkland> ttx: +1 for another discussion at UDS
[15:33] <nijaba> ttx: I take it we don't accept new papercuts now?
[15:33] <ttx> hggdh: yes. it's a more global "server community" discussion, I suspect
[15:33] <kirkland> ttx: we clearly need to investigate how to motivate our community
[15:33] <ttx> nijaba: we won't consider any new ones
[15:34] <nijaba> ttx: thanks for confirming
[15:34] <ttx> nijaba: for lucid
[15:34] <kirkland> ttx: i have a similar feeling about the bug-zapping effort
[15:34] <ttx> nijaba: nominating them could be useful if we decide to pursue the effort for M
[15:34] <ttx> good transition !
[15:34] <ttx> [TOPIC] Bugzapping status (kirkland)
[15:34] <MootBot> New Topic:  Bugzapping status (kirkland)
[15:34] <kirkland> ttx: we really smoked eucalyptus, euca2ools, cloud-utils, and cloud-init bugs last week
[15:35] <kirkland> ttx: was a really successful, focused week of bug fixing
[15:35] <ttx> so it was a "cloud" smoke
[15:35] <kirkland> ttx: yeah, like Lost
[15:35]  * kirkland is going to shave his head like John Locke
[15:35] <kirkland> :-)
[15:35] <ttx> kirkland: did you have external contributors to this one ?
[15:35] <kirkland> ttx: this week, as I said, I'm planning on spending at least all day Thursday on Libvirt
[15:35] <kirkland> ttx: no :-(
[15:36] <kirkland> ttx: but we did focus on euca* cloud* and we do have a better UEC because of the intense focus, I believe
[15:36] <ttx> mathiaz: did you get some external help on the bugday portion of the mysql bugzap ?
[15:36] <mathiaz> ttx: pedro and zul helped in triagging bugs
[15:37]  * ttx remembers having two external people helping for the samba bugday part of the samba bugzapping
[15:38] <ttx> kirkland: do you plan to run something for week 23 ?
[15:38] <ttx> (next week)
[15:38] <kirkland> ttx: i don't think so
[15:39] <kirkland> ttx: unless you or someone else has a suggestion
[15:39] <ttx> kirkland: freezes are not the best moment to do so anyway
[15:39] <kirkland> ttx: i just don't think it's generated the participation that we're hoping
[15:39] <kirkland> ttx: though it has stabilized the packages we need to be stable
[15:39] <ttx> ok.
[15:39] <kirkland> ttx: i think my time will be better spent crushing beta2 blockers
[15:39] <kirkland> ttx: okay by you?
[15:40] <ttx> kirkland: yes, and catching the overlooked critical bug in a sea of bug reports
[15:40] <ttx> [TOPIC] Weekly SRU review (mathiaz)
[15:40] <MootBot> New Topic:  Weekly SRU review (mathiaz)
[15:40] <ttx> https://wiki.ubuntu.com/ServerTeam/KnowledgeBase#SRU%20weekly%20review
[15:41] <mathiaz> http://qa.ubuntu.com/reports/ubuntu-server-team/fixedbugs.ubuntu-server.latest.html
[15:41] <MootBot> LINK received:  http://qa.ubuntu.com/reports/ubuntu-server-team/fixedbugs.ubuntu-server.latest.html
[15:41] <mathiaz> ^^ anything SRU worth on this list?
[15:42]  * ttx wonders why the openldap upgrade bugs don't show up in the list
[15:42] <mathiaz> ttx: the list is generated on Sunday night IIRC
[15:42] <ttx> ah
[15:42] <ttx> that will be next week topic, then :)
[15:43] <mathiaz> ttx: yes
[15:43] <ttx> nothing from me, then ;)
[15:43] <mathiaz> there are two bugs nominated for karmic:
[15:43] <mathiaz> bug 392190
[15:43] <ubottu> Launchpad bug 392190 in vm-builder "vmbuilder fails silently when providing invalid hostname" [Low,Confirmed] https://launchpad.net/bugs/392190
[15:43] <mathiaz> bug 502490
[15:43] <ubottu> Launchpad bug 502490 in vm-builder "ec2-init overwrites user/vmbuilder provided /etc/apt/sources.list" [Undecided,Confirmed] https://launchpad.net/bugs/502490
[15:44] <ttx> first one needs to be fixed in Lucid first
[15:44] <ttx> smoser: what's your opinion on the second one, any value in backporting as a karmic SRU ?
[15:45] <smoser> hmm....
[15:45] <smoser> i could go either way.
[15:45] <smoser> i dont think its terribly, terribly important.
[15:45] <smoser> as there are many work arounds, like using sources.list.d
[15:45] <mathiaz> ok - I'll decline both bugs then
[15:46] <ttx> mathiaz: you're done ?
[15:46] <mathiaz> yes
[15:46] <ttx> [TOPIC] Information about server survey (nijaba)
[15:46] <MootBot> New Topic:  Information about server survey (nijaba)
[15:46] <ttx> nijaba: o/
[15:46] <nijaba> ttx: that was for the meeting that was cancelled
[15:47] <ttx> and it doesn't make sense anymore ?
[15:47] <nijaba> since that I guess you've seen my email in the ml
[15:47] <ttx> ok :)
[15:47] <ttx> nijaba: so we rock more, right ?
[15:47] <nijaba> it was juste to annoucne that the full results are available to anyone in the community that might need it
[15:47] <nijaba> ttx: we rock HARD
[15:47] <nijaba> :)
[15:48] <ttx> ok, let's rock even harder
[15:49] <ttx> [TOPIC] Open Discussion
[15:49] <MootBot> New Topic:  Open Discussion
[15:49] <hggdh> ttx: question: should I keep on testing UEC on daily?
[15:49] <ttx> hggdh: if that can be done at a lesser cost, that would be great to spot regressions
[15:50] <ttx> hggdh: if it's fully automated and doesn't require your constant attention, I'd say yes
[15:50] <ScottK> ttx: I think we should have a session at UDS on what we want 12.04 to look like, so we have an idea of longer term stuff we need to start on now.
[15:50] <hggdh> I think *now* it is less cost, the most difficult was to understand the setup and UEC itself
[15:50] <ttx> ScottK: that's a very good point.
[15:50] <mathiaz> hggdh: note that daily are not always functional
[15:51] <ttx> Discuss anything that would span multiple releases, as well as what needs to be done for the next 6 months
[15:52] <ttx> hggdh: what needs to be done is the post-milestone validation and the pre-milestone candidate testing. Anything else (like testing all dailies) is bonus points
[15:52] <ttx> (what needs to be done is tracked in the spec)
[15:53] <ttx> ok, let's wrap up, if nobody else has any comment
[15:54] <ttx> Announce next meeting date and time : same time, same place, next week.
[15:54] <ttx> I gather DST makes this time slightly more convenient ?
[15:54] <Daviey> (less)
[15:54] <ttx> (equal)
[15:54] <kirkland> ttx: more!
[15:55] <mathiaz> more would even be better
[15:55] <ttx> mathiaz: sleepy head
[15:55] <Daviey> 15:00 UTC ++
[15:56] <ttx> Try to pay lots of beers to jib at UDS to make him change time
[15:56] <ttx> Daviey: ^
[15:56] <ttx> #endmeeting

MeetingLogs/Server/20100331 (last edited 2010-04-01 04:26:25 by mathiaz)