20100817

Agenda

Minutes

Meeting Actions

  • SpamapS to submit rubygems change proposal as Important bug in Debian and CC ubuntu-devel
  • zul to nudge forward upstart script review process
  • jjohansen to review bug 493156
  • zul to review papercut status of bug 582963

Review ACTION points from previous meeting

  • everyone to pick 1 or 2 papercuts for the beta cycle
    • deferred until later point of order
  • zul to continue clean up the SRU assigne list by prodding people on status: DONE
  • SpamapS to finalize and send ruby gems proposal to ubuntu-devel
    • proposal finalized, action for coming week to submit as bug on debian rubygems package
  • ttx to send email about 10.04.1 release process to server team: DONE

Maverick development (jib)

Weekly Updates & Questions for the QA Team (hggdh)

  • (hggdh) UEC testing is progressing and build 1231 was pending build on a PPA before testing could commence on it.
  • (hggdh) QA is taking a look at bugs that are tagged regression-* and will be pushing server-team related bugs

Weekly Updates & Questions for the Kernel Team (jjohansen)

  • https://launchpad.net/bugs/574910 - high load for lucid on ec2 - is mostly phantom load, plan is to close that bug and open more specific bugs to target the issues that have caused so much piling on.

  • (Omahn) requested review of bug 493156 https://bugs.launchpad.net/ubuntu/+source/iotop/+bug/493156

  • (jjohansen) pv-ops kernel is still buggy, there is a chance the -ec2 kernel will have to be used instead. smoser doesn't want to see that happen, but acknowledges that it may be necessary.
  • (jjohansen) kernel tema will be making the "pv-on-hvm drivers" available for testing purposes in a PPA

Weekly Updates & Questions for the Community Team (kim0)

  • kim0 was unable to attend, so this topic was skipped

Weekly Updates & Questions for the Documentation Team (sommer)

  • (sommer) says things are going smoothly, and he will have time to work on some new content this week
  • (sommer) String Freeze is Sept 9. sommer has discussed dropping the ubuntu-serverguide package and forwarded that discussion to the ubuntu-server mailing list after the meeting.

Papercuts status (jib)

  • Two papercuts were left, and mathiz and zul stepped up to claim/evaluate them

Open Discussion

  • (mathiaz) 10.04.1 ISO testing had all server tests completed.
  • (andreserl) cluster stack is waiting on a MIR for pacemaker to be approved

Agree on next meeting date and time

Next meeting will be on Tuesday 2010-08-24 at 1800 UTC in #ubuntu-meeting

Log

[19:00] <SpamapS> #startmeeting
[19:00] <MootBot> Meeting started at 13:00. The chair is SpamapS.
[19:00] <MootBot> Commands Available: [TOPIC], [IDEA], [ACTION], [AGREED], [LINK], [VOTE]
[19:00] <SpamapS> [TOPIC] Review ACTION points from previous meeting
[19:00] <MootBot> New Topic:  Review ACTION points from previous meeting
[19:00] <SpamapS> everyone to pick 1 or 2 papercuts for the beta cycle
[19:01] <zul> hello
[19:01] <SpamapS> Anybody know that they failed to do that?
[19:01] <SpamapS> Oh and Hi
[19:01] <SpamapS> welcome to the server team meeting
[19:01] <SpamapS> The agenda is here: https://wiki.ubuntu.com/ServerTeam/Meeting
[19:01] <SpamapS> Apologies for jumping the gun.
[19:01] <jiboumans> SpamapS: there's still 2 up for grabs, i have an agenda item for it later
[19:02] <SpamapS> Alright, will defer that to jib later
[19:02] <SpamapS> * zul to continue clean up the SRU assigne list by prodding people on status
[19:02] <zul> done
[19:03] <SpamapS> excellent. I looked at it a few minutes ago, and the situation actually looks tenable now, though it seems we need to plow through some of the unassigned SRU's now.. ;)
[19:03] <SpamapS> * SpamapS to finalize and send ruby gems proposal to ubuntu-devel
[19:04] <SpamapS> I asked for some advice from a few DD's before sending it, and have decided on submitting it as an "Important" bug against rubygems in Debian, and CC'ing ubuntu-devel.
[19:04] <SpamapS> [ACTION} SpamapS to submit rubygems change proposal as Important bug in Debian and CC ubuntu-devel
[19:04] <MootBot> ACTION received: [ACTION} SpamapS to submit rubygems change proposal as Important bug in Debian and CC ubuntu-devel
[19:05] <SpamapS> * ttx to send email about 10.04.1 release process to server team
[19:05] <SpamapS> ttx isn't present, but I think I recall the email was sent, yes?
[19:05] <jiboumans> done
[19:05] <jiboumans> i updated the agenda to reflect that
[19:06] <SpamapS> Ahh, refreshing right before the meeting is a good idea. :)
[19:06] <SpamapS> Ok, moving on..
[19:06] <SpamapS> [TOPIC] Maverick development (jib)
[19:06] <MootBot> New Topic:  Maverick development (jib)
[19:06] <jiboumans> o/
[19:06]  * SpamapS tosses the ball to jiboumans 
[19:06] <jiboumans> So, this far into the cycle we should be around 40% completion
[19:07] <jiboumans> in most cases however, we're not and there's several high prio specs even at 0%
[19:07] <jiboumans> I listed them here: https://wiki.ubuntu.com/ServerTeam/Meeting
[19:07] <SpamapS> http://people.canonical.com/~pitti/workitems/maverick/canonical-server-ubuntu-10.10-beta.html
[19:07] <jiboumans> and there's 3 I wanted to specifically talk about
[19:07] <MootBot> LINK received:  http://people.canonical.com/~pitti/workitems/maverick/canonical-server-ubuntu-10.10-beta.html
[19:07] <jiboumans> https://blueprints.launchpad.net/ubuntu/+spec/server-maverick-uec-testing
[19:07] <jiboumans> hggdh, kirkland: do you have an update on this?
[19:08] <jiboumans> are there other things preventing us from knocking items off the list?
[19:08] <hggdh> jiboumans: Dustin just set a new revision -- 1231 to build that WIll Correct All Errors
[19:08] <hggdh> r1230 did not really get it done
[19:08] <kirkland> jiboumans: i just updated that spec, assigning a bunch of work items to me
[19:08] <hggdh> as soon as it lands I will start on it
[19:08] <kirkland> jiboumans: those will probably need reassigning, or dropping though
[19:08] <SpamapS> 1231 WCAE FTW
[19:09] <jiboumans> hggdh: that's great, but that's just 1 item on the list
[19:09] <hggdh> oh
[19:09] <kirkland> jiboumans: i'm testing the r1231 here now (shortly)
[19:09] <kirkland> jiboumans: on local hardware of smoser's
[19:09] <kirkland> one thing of note ...
[19:09] <jiboumans> kirkland: are any of the items on that list 'nice to have', or do we need to knock them all off?
[19:09] <kirkland> r1231, while only 1 top level revision, involves ~20 sub-revisions all merged together
[19:10] <kirkland> jiboumans: have not looked closely enough to say
[19:10] <jiboumans> kirkland: ok, let's take that offline; with beta freeze around the corner and no testing items completed yet, i'm a tad worried though
[19:10] <jiboumans> moving on: https://blueprints.launchpad.net/ubuntu/+spec/server-maverick-upstart-conversion
[19:10] <jiboumans> zul, the status says it's blocked on reviews. can you give us an update on that?
[19:11] <zul> jiboumans: i havent talked to anyone since last week...ill try to nudge it forward this week
[19:12] <jiboumans> zul: so we're still blocked on reviews then? if so, let's chekc in later this week and see if we need to escalate
[19:12] <zul> jiboumans: yes...sure
[19:12] <SpamapS> action?
[19:12] <jiboumans> action to have it dislodged by next meeting
[19:12] <jiboumans> (for zul)
[19:12] <jiboumans> hggdh: the other spec on your plate is https://blueprints.launchpad.net/ubuntu/+spec/server-maverick-qa-workflow
[19:13] <jiboumans> you mentioend before that depending on the uec testing load, that might not make it
[19:13] <jiboumans> hggdh: any updated thoughts on that?
[19:13] <hggdh> jiboumans: really did not have time to look at it
[19:13] <SpamapS> [ACTION] zul to nudge forward upstart script review process
[19:13] <MootBot> ACTION received:  zul to nudge forward upstart script review process
[19:13] <hggdh> many code drops on Euca, continuously
[19:13] <jiboumans> hggdh: does that mean it should be dropped from maverick?
[19:13] <hggdh> jiboumans: I think so
[19:14] <jiboumans> hggdh: go ahead and mark the work items 'postponed' thne please
[19:14] <jiboumans> a request to everyone else; your completion rate should be near 40% now. if you're not and you have any concerns about being able to complete your specs, please bring it up over mail or during our 1:1 this week
[19:15] <hggdh> jiboumans: doing it now
[19:15] <jiboumans> this week is about the last time we can reasonable adjust expectations, so get in touch
[19:15] <jiboumans> spamaps: that's it for me on this topic
[19:15] <SpamapS> Alright, moving on
[19:15] <SpamapS> [TOPIC] Weekly Updates & Questions for the QA Team (hggdh)
[19:15] <MootBot> New Topic:  Weekly Updates & Questions for the QA Team (hggdh)
[19:16] <kirkland> hggdh: do you think you'll get some test time on r1231 today/tomorrow?
[19:16] <jiboumans> hggdh: i'm quite curious to hear about the UEC testing progress
[19:16] <hggdh> kirkland: as soon as it is built I will start on the beast
[19:16] <kirkland> hggdh: https://edge.launchpad.net/ubuntu/+source/eucalyptus/2.0~bzr1231-0ubuntu1/+build/1923013
[19:16] <kirkland> hggdh: currently says it'll build in 49 minutes
[19:17]  * mathiaz cheers at hggdh for helping out in 10.04.1 UEC testing
[19:17]  * kirkland hugs mathiaz for his 10.04.1 work!
[19:17]  * SpamapS high fives everybody for 10.04.1
[19:17] <hggdh> jiboumans: I do hope this will be the one
[19:18] <SpamapS> Anything else for QA?
[19:18] <jiboumans> hggdh: of course. have we closed any open issues with the previous releases?
[19:18] <hggdh> jiboumans: very few. The critical issues are still open (or reopened)
[19:18] <jiboumans> hggdh: so it's fair to say we're roughly at the same stage as last week?
[19:19] <hggdh> jiboumans: yes, it is a fair statement
[19:19] <jiboumans> hggdh: ok, excellent thanks for the update
[19:19] <jiboumans> notihng more from me
[19:19]  * SpamapS signals to hggdh that jjohansen is open for a pass
[19:20] <SpamapS> anyone else?
[19:20] <hggdh> a heads up from QA
[19:20] <hggdh> we are starting to look into bugs that are tagged regression-*
[19:20] <hggdh> and I will start to nudge those related to the server team
[19:21] <hggdh> (that is, of course, having time ;-)
[19:21] <SpamapS> hggdh: I've got cycles specifically for bug fixing during this cycle.. so feel free to hunt me down before I get too wrapped up in random bugs.
[19:21] <hggdh> SpamapS: my hero!
[19:21]  * SpamapS blushes
[19:21] <hggdh> and I am done for now
[19:22] <SpamapS> Alright, much love for yo hggdh, moving on...
[19:22] <SpamapS> [TOPIC] Weekly Updates & Questions for the Kernel Team (jjohansen)
[19:22] <MootBot> New Topic:  Weekly Updates & Questions for the Kernel Team (jjohansen)
[19:22] <jjohansen> so still lots of bugs
[19:22] <SpamapS> jjohansen: there's this bug, in ec2, btw, where the load is high while idling. Had you heard?
[19:22] <jjohansen> Bug 574910 - is mostly phantom load
[19:22] <ubottu> Launchpad bug 574910 in linux-ec2 (Ubuntu) "High load averages on Lucid while idling" [Undecided,In progress] https://launchpad.net/bugs/574910
[19:23] <jjohansen> I am going to go with the test kernel we have up currently, and SRU it.  We will close the bug, and have people still experiencing issues open new bugs
[19:23] <jjohansen> basically there is a lot of piling on happening in that bug
[19:24] <jjohansen> And there are different issues.
[19:24] <Omahn> Would it be possible for someone from the kernel team to review this: https://bugs.launchpad.net/ubuntu/+source/iotop/+bug/493156
[19:24] <ubottu> Launchpad bug 493156 in iotop (Ubuntu) "Please enable CONFIG_TASK_DELAY_ACCT" [Undecided,New]
[19:24] <jjohansen> Omahn: I'll stick it on my todo list
[19:24] <Omahn> jjohansen: Thanks, having that option enabled in LTS would be very useful to us.
[19:25] <jiboumans> jjohansen: so is it fair to summarize that the load display is wrong, and there's not an actual high load?
[19:25] <jjohansen> no promises that it will make it through SRU though
[19:25] <SpamapS> [ACTION] jjohansen to review bug 493156
[19:25] <MootBot> ACTION received:  jjohansen to review bug 493156
[19:25] <ubottu> Launchpad bug 493156 in iotop (Ubuntu) "Please enable CONFIG_TASK_DELAY_ACCT" [Undecided,New] https://launchpad.net/bugs/493156
[19:26] <jjohansen> jiboumans: generally yes.  I have done a lot of testing and this has been the case
[19:26] <smoser> hooray!
[19:26] <jjohansen> this does not mean there are not other issues
[19:26] <jiboumans> jjohansen: that's not a bad outcome actually
[19:26] <jiboumans> of course
[19:26] <jjohansen> but we can get new clean bugs to deal with them
[19:26] <ulysses> o/
[19:26] <Dyth> o/
[19:27] <jjohansen> questions?
[19:27] <SpamapS> jjohansen: * pv-ops kernel status update
[19:27] <jjohansen> still buggy.
[19:27] <zul> hehe
[19:27] <jjohansen> Bug 606373 - didn't turn out to be related to either of the guessed bugs
[19:27] <ubottu> Launchpad bug 606373 in cloud-init (Ubuntu) "cloud-init output does not get to console when booted with pv-grub and ramdisk" [High,Confirmed] https://launchpad.net/bugs/606373
[19:28] <smoser> jjohansen you were going to discuss today with pgraner and others a cut-off date (of sorts) for falling back to -ec2
[19:28] <smoser> (a solution that nobody wants)
[19:28] <jjohansen> As things work with out an initrd, it should be easy to track down, it will just take some time
[19:29] <jjohansen> smoser: I will discuss, but there is no guarentee that the -EC2 kernel won't be worse.  In fact last time I had it up it was
[19:29] <smoser> right. but we need to have some plan in place. and you're skillful black magic has made -ec2 function in the past, so I have somewhat more confidence in it.
[19:30] <jjohansen> The majority of our current bugs look fix able.  The corruption bug 613083 is the worrying one
[19:30] <ubottu> Launchpad bug 613083 in linux (Ubuntu) "user-data is corrupted inside metadata service" [High,Confirmed] https://launchpad.net/bugs/613083
[19:30] <smoser> better the devil you know ...
[19:31] <jjohansen> smoser: your assuming that a maverick -EC2 kernel is the devil you know
[19:31] <smoser> agreed.
[19:31] <jjohansen> its not
[19:31] <smoser> oh, i know.
[19:31] <SpamapS> the only devil I know is zul + 3 red bulls
[19:31] <jjohansen> yikes, now that is frightening
[19:31] <zul> i still dont want to get involved with this ;)
[19:32] <SpamapS> Ok, anything else for jjohansen/the kernel team?
[19:32] <jjohansen> I have one other thing
[19:32] <jjohansen> We are going to make the pv-on-hvm drivers available in a ppa for testing purposes.
[19:32] <jjohansen> ..
[19:33] <SpamapS> Do you have the location of that PPA yet?
[19:33] <jjohansen> SpamapS: no I haven't set it up yet.
[19:34] <SpamapS> Alright, thanks jjohansen.. moving on to documentation..
[19:34] <SpamapS> [TOPIC] Weekly Updates & Questions for the Documentation Team (sommer)
[19:34] <MootBot> New Topic:  Weekly Updates & Questions for the Documentation Team (sommer)
[19:34]  * SpamapS lobs the ball toward sommer
[19:35] <sommer> hello, got a bunch of server guide bugs fixed this morning
[19:35] <SpamapS> sommer: ^5
[19:35] <sommer> so things are going pretty smooth, should have time to work on some new content this week as well
[19:36] <sommer> don't really have much else
[19:36] <sommer> at this time anyway
[19:36] <SpamapS> Alright, anybody have questions for sommer, other than "How do you find the time to do such an amazing thing for us?!"
[19:36] <mathiaz> sommer: what are the main release milestone documentation wise for maverick?
[19:37] <mathiaz> sommer: *milestones*
[19:37] <sommer> mathiaz: probably mainly bug fixes, but I'd like to get a puppet section in as well
[19:37] <sommer> I can update the blueprint for that
[19:38] <mathiaz> sommer: any relevant freezes coming up?
[19:38] <mathiaz> sommer: if so - when?
[19:38] <sommer> mathiaz: Sept 9 is String Freeze
[19:39] <sommer> mathiaz: but I've sent an email to mdke about dropping the ubuntu-serverguide package... so hopefully there will be some discussion on the doc list
[19:40] <sommer> mathiaz: if we only have HTML and PDF formats we could technically update whenever we'd like... but for translators we may want to do once a month updates or something
[19:41] <SpamapS> sommer: I'd be in support of that. Most people are probably reading the server guide on the web 99% of the time.
[19:41] <sommer> SpamapS: yeppers, I'll copy the server ml with any replies to that thread
[19:42] <SpamapS> Great! Anything else for documentation?
[19:42] <SpamapS> Ok, kim0 seems unresponsive, so I'll skip the community team update for now..
[19:43] <SpamapS> [TOPIC] Papercuts status (jib)
[19:43] <MootBot> New Topic:  Papercuts status (jib)
[19:43] <jiboumans> So most bugs are assigned, but there's 2 left for grabs
[19:43] <jiboumans> looking for volunteers for: bug 582963 & bug 601803
[19:43] <ubottu> Launchpad bug 582963 in apache2 (Ubuntu) "SSL pass phrase dialog can't read input" [High,Confirmed] https://launchpad.net/bugs/582963
[19:43] <ubottu> Launchpad bug 601803 in whois (Ubuntu) "when whois and mkpasswd are build locally they have .mo file conflicts" [Medium,Confirmed] https://launchpad.net/bugs/601803
[19:43]  * mathiaz took 601803
[19:44] <mathiaz> zul: is assigne to bug 582963
[19:44] <jiboumans> excellent
[19:44] <mathiaz> zul: seems like the best candidate?
[19:44] <mathiaz> zul: are you planning on working on the papercut?
[19:44] <zul> mathiaz: probably
[19:45] <zul> mathiaz: the ssl one is not exactly a papercut
[19:45] <SpamapS> How is it a High bug in apache2, but only medium in papercuts?
[19:45] <marianom> czajkowski: hi!.I already sent our application to the loco council email address
[19:45] <jiboumans> if it's not a papercut, let's remove it
[19:46] <SpamapS> marianom: Hi! This is the ubuntu server team meeting at the moment. Wrong window perhaps?
=== IdleOne is now known as NattyOne
[19:46] <czajkowski> marianom: great thre is a meeting on here now, and the LC doesn't meet for another hour
[19:47] <jiboumans> zul: action on you to figure out if that's papercut worthy
[19:47] <jiboumans> if it is, please take it. if not, let's remove it from the papercuts list
[19:47] <zul> jiboumans: yep
[19:47] <jiboumans> the last thing is a reminder:
=== NattyOne is now known as IdleOne
[19:47] <jiboumans> betafreeze is next week thursday
[19:47] <SpamapS> [ACTION] zul to review papercut status of bug 582963
[19:47] <MootBot> ACTION received:  zul to review papercut status of bug 582963
[19:47] <ubottu> Error: Could not parse data returned by Launchpad: The read operation timed out (https://launchpad.net/bugs/582963)
[19:47] <jiboumans> your papercut(s) should probably be fixed by then
[19:48] <jiboumans> so please keep that in mind
[19:48] <ubottu> Error: Could not parse data returned by Launchpad: The read operation timed out (https://launchpad.net/bugs/582963)
[19:48] <jiboumans> ok, that's all from me
[19:48] <SpamapS> Excellent thanks. Has anyone spoken with kim0 before I move on to open discussion?
[19:49] <SpamapS> I suspect he may be involved with some 10.04.1 revelry in Cairo. We'll move on then.
[19:49] <SpamapS> [TOPIC] Open Discussion
[19:49] <MootBot> New Topic:  Open Discussion
[19:49] <jiboumans> mathiaz: could you give us the 2 line summary of 10.04.1?
[19:50] <mathiaz> for server: all tests completed
[19:50] <mathiaz> we're good to go as far as -server isos are concerned
[19:50] <mathiaz> thanks hggdh for helping out with UEC testing
[19:50] <jiboumans> anything looking like it could be blocking / remain unresolved?
[19:50] <mathiaz> we're running into an issue with wubi though
[19:51] <SpamapS> mathiaz: wubi for server?
[19:51] <mathiaz> see #ubuntu-release to get your fix of real TV
[19:51] <andreserl> In the Cluster Stack, we still waiting for MIRs. We are hold up by Pacemaker. ivoks said he would review it and fix whatÅ› necessary to request MIR. However, it seems he hasn done it yet
[19:51] <mathiaz> SpamapS: nope - for ubuntu desktop
[19:51] <SpamapS> ah ok
[19:51] <hggdh> mathiaz: my pleasure
[19:51] <mathiaz> jiboumans: wubi is currently the blocker
[19:52] <jiboumans> mathiaz: excellent, thanks for the update
[19:52] <mathiaz> if anyone has a windows system, help in debugging/testing is welcome
[19:53] <mathiaz> hop in #ubuntu-release to help out
[19:53] <mathiaz> that's all on the 10.04.1 front
[19:54] <SpamapS> Alright, so that should be it then, going once.. going twice..
[19:55] <SpamapS> [TOPIC] Announce next meeting date and time
[19:55] <MootBot> New Topic:  Announce next meeting date and time
[19:55] <SpamapS> Tuesday 2010-08-24 at 1800 UTC
[19:55] <SpamapS> right here in #ubuntu-meeting
[19:55] <SpamapS> Thanks everyone!
[19:55] <jiboumans> thanks all
[19:55] <SpamapS> #endmeeting
[19:55] <MootBot> Meeting finished at 13:55.

MeetingLogs/Server/20100817 (last edited 2010-08-18 04:13:41 by 76-216-240-245)