=== davidm_ is now known as Guest33922 === davidm_ is now known as Guest89676 === Ursinha is now known as Ursinha-away === bilalakhtar_ is now known as cdbs [10:04] Is this one of those Tuesdays? === Ursinha-away is now known as Ursinha [10:06] persia: the last week was [10:07] persia: unfortunately, everyone except elky didn't think so [10:07] Ah, then next week will be. [10:08] i think we've missed enough we can probably agree to any tuesday and reset it from there. [10:09] Apparently our next scheduled meeting is 4th January. Shall we have one 15th February then, and try to get back with the program from there? [10:10] Heh almost enough for a quorum now? [10:11] When does more people get added to the board again? [10:11] head_victim, perhaps, but not in a way that is useful: we're clearly fail to attract candidates in a way that is useful. [10:31] * persia , in the absence of clear comment, updates the wiki page to claim a date === xaphir is now known as Guest99372 === Guest99372 is now known as xaphir === davidm_ is now known as Guest66558 === artir is now known as afk|artir === afk|artir is now known as artir === oubiwann is now known as oubiwann_ [14:57] mdz, pitti, cjwatson: meeting in a few minutes. can't find Keybuk or sabdfl, though. [14:57] o/ [14:58] here, more or less [14:58] kees, clan mailed saying sabdfl wouldn't make it [14:58] (t-b@) [14:59] mdz: ah, thanks. [14:59] kees, I tidied up the agenda based on pitti's minutes from last time [15:00] mdz: oh, excellent [15:01] (I've just moderated clan's mail) [15:01] #startmeeting [15:01] Meeting started at 09:01. The chair is kees. [15:01] Commands Available: [TOPIC], [IDEA], [ACTION], [AGREED], [LINK], [VOTE] [15:01] [TOPIC] action review [15:01] New Topic: action review [15:01] mdz - Set the release manager to ubuntu-release [15:02] kees, so I'm confused about this [15:02] the release manager for natty is already ubuntu-release [15:02] is that what was meant here, or did it mean something else? [15:03] I think that's what was meant; there were the notes from the rally TB [15:03] I think it wasn't when that action was taken [15:03] ok, so that's done then [15:03] ok, next [15:03] mdz - Review ubuntu-drivers membership and make sure it includes (only) people who need to manage blueprints [15:03] I haven't reviewed ubuntu-drivers [15:03] I think someone else should review ubuntu-release [15:03] ok, carried over [15:04] mdz - Review ubuntu-release membership [15:04] carried over [15:04] mdz - Set the owner/maintainer to techboard [15:04] done [15:04] the ubuntu-release membership looks correct to me [15:04] mdz - announce something to ubuntu-devel-announce [15:04] not done [15:04] bdmurray - file bug about bug supervisor not being able to set bug reporting guidelines for Ubuntu as a whole [15:04] I think this was done, but I can't find the bug currently [15:05] u-release> not sure about Iulian and Stefan (whether they still want to be involved), but otherwise it looks fine to me [15:05] [action] kees will follow up with bdmurray about his action [15:05] ACTION received: kees will follow up with bdmurray about his action [15:05] pitti - check with James Troup about current quality experience of pool.ntp.org [15:05] done, will present the details in the actual topic [15:05] okay [15:05] pitti - add TB as vendor contact point for http://groups.google.com/group/ntppool-vendors?pli=1 [15:05] saw that was done [15:05] done [15:06] okay, were there any other actions we needed to review that weren't already listed? [15:06] [topic] Default ntpd configuration https://bugs.launchpad.net/bugs/104525 [15:06] New Topic: Default ntpd configuration https://bugs.launchpad.net/bugs/104525 [15:07] Ubuntu bug 104525 in ntp (Ubuntu) "default ntp.conf should use pool.ntp.org servers" [Wishlist,Triaged] [15:07] where does this stand currently? [15:07] I talked to James about it [15:07] I brought it up, but I missed the last meeting [15:07] we reviewed that a couple of years ago, and then it was still not very reliable [15:08] but these days they have a pretty good probing system and network, and throw servers out of the rotation when they are off by .1 s or more (checked three times an hour) [15:08] so from a reliability POV James was okay with it now [15:08] his main concern is security [15:08] i. e. a server could send out bad times to all clients except to the one that is doing the probing [15:09] what's the worst case scenario? [15:09] so that you could infiltrate the network with several servers which send out the wrong time [15:09] that's quite a lot of effort, of course [15:09] the advantage is that it provides nice geolocation support, i. e. selects a server near you, and has a large round-robin network [15:10] it would be interesting to see whether they plan DNSSEC support, but then again ntp.u.c. doesn't have that either [15:10] pitti, quite a lot of effort, yes, and what is there to gain? [15:10] seems like DoS [15:10] I think we should remember that this is replacing a hardcoded list of server names [15:10] which we never maintain AFAIK [15:11] I have a hard time seeing how it wouldn't be an improvement [15:11] apt has rewind and staleness protections, so the worst-case with targetted time attacks would be a freeze attack convincing apt to never update. [15:11] kees, and most users probably inflict a much more successful attack on themselves by not installing updates ;-) [15:11] you could also envision running somethign like ebay on a machine like that, and tricking that into thinking that an auction is still happening when it shoudln't be any more, etc. [15:11] mdz: right [15:12] i. e. there are certaily interesting attacks, but it still by and large feels theoretical to me [15:12] (and it hasn't happened so far) [15:12] yup [15:12] pitti, on the plus side, if there are millions of clients using that pool, an attempt to subvert it would likely be noticed quickly [15:12] so in summary, from what I've heard from the admins there and after discussion with James, I'm happy about the proposal [15:12] anyway, this discussion has been had in the bug, see e.g. comment 6 [15:12] i. e. to have X.ubuntu.pool.ntp.org and ntp.u.c. in the default config [15:13] wfm [15:14] (replacing> in gnome-system-tools, I mean) [15:14] I'm also happy for us to go ahead at this point; especially security-sensitive sites are still free to use their own NTP servers, and they may well already do so [15:15] right. I'd agree as well. [15:15] cjwatson: indeed, the list in g-s-t is quite bad; we should IMHO replace that in any case, even if we'd not change (or revert) the ntpd.conf list) [15:15] so that there's only very little reason to change it at all [15:17] should we specifically vote on this? [15:18] I think we just did [15:18] we seem to have straw-poll consensus [15:18] okay, so the proposal would be to switch the default ntp servers to {0,1,2,3}.ubuntu.pool.ntp.org, yes? [15:19] kees: plus ntp.u.c. for ntpd.conf [15:19] (at least that was the proposal we discussed last time) [15:20] pitti: ntpd.conf would get {0,1,2,3}.ubuntu.pool.ntp.org and ntp.u.c? [15:20] that's how I understood it, anyway; perhaps just three from the pool [15:21] then if ntp.org would fail, we'd still have the ubuntu one [15:21] [vote] switch the default ntp servers to {0,1,2,3}.ubuntu.pool.ntp.org plus ntp.ubuntu.com [15:21] Please vote on: switch the default ntp servers to {0,1,2,3}.ubuntu.pool.ntp.org plus ntp.ubuntu.com. [15:21] Public votes can be registered by saying +1/-1/+0 in the channel, private votes by messaging the channel followed by +1/-1/+0 to MootBot [15:21] E.g. /msg MootBot +1 #ubuntu-meeting [15:21] +1 [15:21] +1 received from kees. 1 for, 0 against. 0 have abstained. Count is now 1 [15:21] +1 [15:21] +1 received from mdz. 2 for, 0 against. 0 have abstained. Count is now 2 [15:21] +1 [15:21] +1 received from pitti. 3 for, 0 against. 0 have abstained. Count is now 3 [15:22] I assume someone has checked that they are prepared to handle this volume of clients... [15:22] they seemed happy enough about it [15:22] +1 [15:22] +1 received from cjwatson. 4 for, 0 against. 0 have abstained. Count is now 4 [15:22] "they" -> ntp.org admins, haven't checked the servers obviously [15:22] [end] [15:23] er [15:23] * cjwatson eyes up https://bugs.launchpad.net/ubuntu/+source/ntp/+bug/715141 and wonders if pool.ntp.org does IPv6 ... [15:23] Ubuntu bug 715141 in ntp (Ubuntu) "ntp.ubuntu.com does not have AAAA record" [Undecided,New] [15:23] [endvote] [15:23] Final result is 4 for, 0 against. 0 abstained. Total: 4 [15:23] [etov]? :-) [15:23] $ host -t AAAA 2.ubuntu.pool.ntp.org [15:23] 2.ubuntu.pool.ntp.org has no AAAA record [15:24] http://lists.ntp.org/pipermail/pool/2010-October/005234.html etc. [15:24] LINK received: http://lists.ntp.org/pipermail/pool/2010-October/005234.html etc. [15:24] okay, that passed. who wants to implement it? [15:25] I'm happy to take g-s-t [15:25] jdstrand touched ntp last ;-) [15:25] want to coordinate a bit with upstream there [15:25] hey now [15:25] :) [15:26] but it's by and large sponsoring, the bug even has patches for both already [15:26] pitti: I think it's best for 1 person to do the changes, regardless of package. can you take both? [15:26] if the bug has everything in it, and the vote passed, I can do it now [15:26] kees: can do, I just don't know much about testing ntpd [15:26] as it happens, I am patch piloting as we speak [15:26] pitti: delegating to jdstrand works too :) [15:26] patch pilot! [15:27] jdstrand: I'll leave ntp to you then, and take g-s-t === bjf[afk] is now known as bjf [15:27] ack [15:27] [action] pitti to update g-s-t for new ntp pool [15:27] ACTION received: pitti to update g-s-t for new ntp pool [15:27] [action] jdstrand to update ntp for new ntp pool [15:27] ACTION received: jdstrand to update ntp for new ntp pool [15:27] jdstrand: to catch you up slightly, be sure that the to-be-sponsored ntp debdiff includes ntp.ubuntu.com as the final fall-back too. [15:28] ok [15:28] okay, that topic is done... [15:28] jdstrand: thanks! [15:28] [topic] the mailing list archive for anything we missed [15:28] New Topic: the mailing list archive for anything we missed [15:28] sure :) [15:29] I don't see anything from the mailing list. does anyone else see anything? [15:30] looking [15:30] is this utf8 thing resolved? [15:30] bug 666565 [15:30] Launchpad bug 666565 in Ubuntu Translations ""utf8" charmap in locale name is wrong" [High,Triaged] https://launchpad.net/bugs/666565 [15:31] looks clear apart from that [15:31] it doesn't seem like the TB needs to be involved in that bug? [15:31] can we unsub then? [15:32] done [15:32] [topic] Check up on community bugs [15:32] New Topic: Check up on community bugs [15:33] I see this: https://bugs.launchpad.net/ubuntu-community/+bug/374900 [15:33] Ubuntu bug 374900 in faac (Ubuntu) "Libfaac not LGPL" [High,Triaged] [15:33] this is from long ago, IIRC. [15:34] I took an action for that at one point, but my head is filled with despair any time I contemplate it [15:34] I hate legal stuff [15:34] hah [15:34] so I don't really know what to do, I could promise to do it but history suggests I may never get to it. Can I punt? === ian_brasil__ is now known as ian_brasil [15:35] let someone else worry about it for 2 weeks? [15:35] perhaps someone who isn't here today ;-) [15:36] looks like someone needs to get in touch with upstream [15:36] we can punt if that's the way to go; it looks pretty clearly like the encoder package need to be removed. [15:36] *needs [15:36] unless upstream agrees to fix it? [15:36] but I will add it to the agenda for next time and we can move forward. [15:37] sorry [15:38] [topic] other agenda items [15:38] New Topic: other agenda items [15:38] I'd like to bring up moving the TB meeting by an hour or two. Probably 1700UTC would be best for Keybuk, based on his last TB email about his current schedule. [15:39] I would like it moved away from 7am too, but I can usually make it. [15:39] 1700 is fine by me [15:39] still on tuesdays? [15:39] conflicts with desktop team meeting, so I'd need to sit in two in parallel [15:40] I have a tricky-to-move weekly meeting at that time [15:40] could do that time on mon or wed [15:40] I could do 1700 on Thu [15:41] I can't do Wed afternoon [15:41] Mon or Thu at that time would be OK [15:41] pitti, could you do mon? [15:41] (it's right after the 10.04.2/SRU/Natty interlock meeting, but I think that should be OK) [15:42] we have the SRU/10.04.2 meetings on Monday around that time [15:42] and after that I usually have dinner and then sports [15:43] so, while it would be very inconvenient, and I couldn't prepare for it, it'd be technically possible [15:43] I'd much prefer Thursday, though [15:43] I could do Thu but only if we go to alternate weeks [15:43] I have a 2-weekly meeting this week at that time [15:44] so we would need to change phase [15:44] phase doesn't matter much for me [15:44] due to DST, the earliest I can do Thu is 1800 [15:44] don't we usually move it an hour earlier during summer anyway? [15:45] pitti: no, we recently tied it to UTC [15:45] take this to the mailing list maybe? we will need input from the other members as well [15:45] ah [15:45] mdz: yeah, good idea. [15:45] 1800 UTC would be 2000 local in summer [15:45] [action] kees - bring up TB rescheduling on mailing list [15:45] ACTION received: kees - bring up TB rescheduling on mailing list [15:45] set up a doodle? [15:45] yeah [15:45] okay, any other topics? [15:46] I filed that bug asked about earlier [15:46] I'll leave Keybuk as chair for next time [15:46] its bug 703002 [15:46] bdmurray: ah, excellent. that was my memory too; do you have the bug # handy? [15:46] Launchpad bug 703002 in Launchpad itself "distributions have no +configure-bugtracker link" [Low,Triaged] https://launchpad.net/bugs/703002 [15:46] excellent, thanks [15:47] kees, I think it's sabdfl's turn to chair [15:47] mdz: oh, okay. it's been Keybuk's the last two, IIRC. [15:47] anyway, sure, sabdfl it is. [15:48] okay, that's a wrap then. [15:48] thanks everyone! [15:48] thanks everyone [15:48] #endmeeting [15:48] Meeting finished at 09:48. [15:53] o/ [15:55] hi [15:56] o/ [15:57] \o/ [15:57] o/ [15:58] \o [15:58] o/ [15:58] o/ [15:58] _o_ [15:58] ttx: feeling better? [15:59] JamesPage: slightly. I'm taking forever to recover [15:59] * smb \o [15:59] ttx, :( [15:59] JamesPage, Daviey: how was FOSDEM ? [15:59] ttx, good... tiring .. good... you were missed. [16:00] Daviey: not enough free beer ? [16:00] ttx: not much free beer :-( [16:00] o/ [16:00] ttx, maybe too much \;-0 [16:01] shall we get started? [16:01] yes [16:01] #startmeeting [16:01] Meeting started at 10:01. The chair is JamesPage. [16:01] Commands Available: [TOPIC], [IDEA], [ACTION], [AGREED], [LINK], [VOTE] [16:01] [TOPIC] Review ACTION points from previous meeting [16:01] New Topic: Review ACTION points from previous meeting [16:02] So there where no actions from last weeks meeting.... [16:02] \o/ [16:02] yuppie [16:02] [TOPIC] Natty Development [16:02] New Topic: Natty Development [16:02] Hows Alpha 2? [16:02] should we be iso testing [16:02] ? [16:02] not a good story for eucalyptus atm [16:03] still FTBFS? [16:03] think i solved that [16:03] but ftbfs in archive [16:03] and some dhcpd issues [16:04] sounds like fun [16:04] :/ [16:04] Regarding ISO testing... has everyone at least try installing A2? [16:04] fixable? [16:04] * robbiew needs to summarize A2 and put up the plan for A3 this week [16:05] JamesPage, yeah.. will be fixed this week.. [16:05] Not yet: the daily ISO testing looks OK at the moment (http://hudson.qa.ubuntu-uk.org/) [16:05] i somewhat plan on installing A2 today on netbook. [16:05] * Daviey needs to talk to upstream [16:06] So back to SpamapS original question - should we be ISO testing or do we need todo that closer to A3? [16:06] Generally A2 hasn't shown any kitten killers for me [16:07] (except euca) [16:07] virt-manager is quite broken right now in natty.. so that makes iso testing a little more of a pain for me. :-P [16:07] JamesPage: yes we should [16:07] SpamapS: TestDrive? [16:07] * SpamapS could do virtualbox I suppose [16:07] :P [16:07] RoAkSoAx: I'll look into it [16:07] Those with real hardware available... is pretty useful to use compared to pure virtualised [16:08] *cough* vmware *cough* ;) [16:08] i've seen some nasty kernel bugs on my desktop install... and waiting for some serverish ones to hurt. [16:08] Sounds like thats a specific action for this week [16:08] +1 [16:08] * SpamapS 's only real hardware available is an old apple G5 [16:09] you do know that A2 shipped, right ? [16:09] its kind of late to worry about dead kittens [16:09] no kidding [16:09] time to bury those and move on [16:09] [ACTION] ALL: ISO testing (including as much real hardware as possible) to smooth the way to A3! [16:09] ACTION received: ALL: ISO testing (including as much real hardware as possible) to smooth the way to A3! [16:10] in maverick we were getting hounded to do ISO tests.. this time I've heard nothing about iso testing. Just wondering if and when we should do iso tests. :-P [16:10] yeah.. but we need to start hammering it to find the kittens... not save them :) [16:10] SpamapS: I wonder who did the hounding [16:10] lol [16:10] what does hggdh do? [16:10] ttx: some uppity frenchman [16:10] SpamapS: the worst kind [16:10] ttx, you have been missed [16:11] Anything else for Natty dev at this point in time? [16:11] seriously...if you are on the server team and you have to be hounded to test your own iso at milestone...that's a bit concerning [16:11] i don't think hggdh has been able to test work loads... just installation. [16:11] well installations was the only thing that was tested for the longest time [16:11] on UEC? (sorry) [16:11] robbiew: +1. Even with a professional hounder, Hounding is no fun [16:12] hggdh, generally. [16:12] right [16:12] hounding != co-ordination [16:12] we are still prepping some automated tests -- based on LTP, and others [16:12] and, on Hudson (or Jenkins) we are running pretty much all of trhe ISO tests [16:12] hggdh: what about the powernap issue? [16:13] hggdh, maybe not now... but could you and JamesPage give us some deeper insight at some point? [16:13] robbiew: well the cloud doesn't use iso's ;) [16:13] Daviey: certainly [16:13] rocking [16:13] RoAkSoAx: I am starting to think the issue is not with powernap, but with euca itseld [16:13] SpamapS: yeah but the cloud tests are pretty extensive compared to the installation/work load tests [16:13] hggdh: ok ;) [16:14] SpamapS: actually we usually start the cloud (read euca, right now) tests with a preseeded ISO install [16:14] SpamapS: Yet.. [16:14] SpamapS: right, but it uses images...and it's ourproduct [16:14] I'm saying, it should be a high priority the week of the milestone, to test the images we put out...iso or cloud [16:15] soren, you are just trying to create more work for us! :) [16:15] Daviey: Quite the contrary. [16:15] Daviey: The biggest problem I had with automated iso testing was lack of resources. [16:15] Computing resources, that is. [16:15] ack [16:16] double-ack [16:16] If I could get a cloud to run the ISO for me, hooking up VNC to a screen scraping thing would be relatively easy. [16:16] Alright.. so the signal to iso test should be... the alpha freezes? [16:16] SpamapS: it's actually the candidate generation [16:16] I do not agree [16:16] we should run them daily [16:17] SpamapS: otherwise you don't have a page to record results on iso.qa.u.c [16:17] hggdh, i think we are talking manual [16:17] yes, talking about manual testing [16:17] in my mind.. automated complements manual [16:17] ttx: right, I don't get any emails when that happens. [16:17] I imagine most people tried the candidate iso before release, right? [16:17] despite being subscribed to all the iso tests I usually do [16:18] SpamapS: but you get the freeze notice right? [16:18] robbiew: yes [16:18] SpamapS, that is a good point... the current way of tracking the candidates is via watching -release...and that is a process which is broken [16:19] hmm [16:19] skaet: around? [16:19] * SpamapS gives in and just subscribes to every list on lists.ubuntu.com [16:19] so I can see about getting an email notification sent out when the isos are ready to test [16:20] SpamapS, that isn't enough :) [16:20] So what do we need in terms of notifications? [16:20] Subject: Candidate posted [16:21] Subject: Re-rolled... because XYZ. [16:21] I really dont think its that hard. [16:21] where? which list? [16:21] The week of a release, you need to be testing [16:21] we need to get through all the tests on iso tracker [16:21] you really should not need a reminder for release week [16:21] smoser, last cycle i wasted time testing a stale candidate.. [16:21] ack [16:21] well how do people know they've appeared now? [16:21] this needs better co-ordination. [16:21] #ubuntu-release [16:21] it's a great channel [16:22] I highly recommend it during the release [16:22] lol [16:22] heh [16:22] smoser: This is the first I've heard of this. The maverick cycle, a little man would pop up in my IRC window and say "test the isos now" ... [16:22] I'll take a todo on getting a notification sent out [16:22] reading scrollback seems to be a less clean process IMO. [16:22] [ACTION] robbiew to ensure notifications of release candidates (and re-rolls) sent out [16:22] ACTION received: robbiew to ensure notifications of release candidates (and re-rolls) sent out [16:23] OK so is the combination of watching #ubuntu-release and some extra notifications going todo the trick? [16:23] it better [16:23] Apologies for being naive above it. I just want to know when I should start focusing on testing the iso vs. other things. [16:23] we do have schedule [16:23] it's not like the release should creep up on anyone by surprise ;) [16:24] SpamapS understood, and I think you raise a good pointr [16:24] point [16:24] * JamesPage agrees with SpamapS [16:24] the release team should at least fire off a notice when ISOs are good to test [16:24] you should have it the back of your mind that we are releasing on this date and see what needs to be done [16:25] * kirkland wonders if Ubuntu should have a desktop indicator for that sort of thing ;-) [16:25] what we did during maverick is have a "release contact" in the team (was me) whose job it was to remind people to switch to ISO testing (and talked about it in weekly meetings). A bit inefficient, but worked [16:25] zul: +1 for that.. I have just now added iso testing to my calendar in line with the release calendar. [16:25] perhaps TestDrive could provide one [16:25] kirkland, byobu indicator! [16:25] if you have TestDrive installed, it puts a message in the indicator [16:25] +1 [16:25] yeah we could do that [16:25] I think we are missing the point of this meeting - surely we should be discussing activity over the next week in the 'Natty development' section? [16:25] A1/A2/A3/B1/RC/GA Candidate Available [16:26] Hah.. and of course.. the week before Alpha 3 is the ensemble sprint. :P [16:26] ttx, I had an advantage of seeing what was going on by being on the same TZ as you... i think that helped me. [16:26] JamesPage: true.. I think zul had something he wanted to bring up regarding Natty development. [16:27] i thought that was the next topic [16:27] if we are done with natty release stuff [16:27] zul: as our rep in the release meeting can you highlight any key activities over the next week just in case its slipped someones mind? [16:28] Indeed.. though before that was "events" ..which seems to have been wiped away. [16:28] JamesPage: i think i can do that [16:28] zul: excellent - I'll make is standing agenda [16:28] OK so moving on. [16:28] kirkland: +1 on the testdrive idea [16:28] [TOPIC] Ubuntu Server Team Events [16:28] New Topic: Ubuntu Server Team Events [16:29] JamesPage and I had a good weekend at foSdem ... report to follow [16:29]