/srv/irclogs.ubuntu.com/2012/02/01/#launchpad-dev.txt

wgrantwallyworld___: ppa:launchpad should work on precise now00:32
lifelessman once you start looking for dupes01:00
lifelessthere are tonnnnes01:00
lifeless(e.g. bug 132300)01:00
_mup_Bug #132300: mail for new bugs inconsistent with other bug mail (ignores mail-on-my-actions disabled, missing footers, duplicate mail vs structural subscriptions) <email> <lp-bugs> <notifications> <Launchpad itself:Triaged> < https://launchpad.net/bugs/132300 >01:00
wgrantlifeless: Well01:01
wgrantlifeless: They're not necessarily dupes01:01
wgrantThey would all be solved by making them consistent, but they're really separate bugs.01:02
lifelesswgrant: I think there is one that isn't necessarily a dupe01:03
lifelesswgrant: anyhow, we have limited facilities for saying 'X symptom', 'Y symptom' of a given root cause, so meh:)01:04
wgrantBah01:17
wgrantlxc has been apparmored01:17
wgrantAhhhh01:20
wgrantAnd that explains why my syslog has been all broken.01:20
wgrantrsyslog from a container sometimes takes hold of the same stream01:21
wgrantSo /var/log/kern.log only gets the occasional byte01:21
lifelessok so I need help finding a damn bug :)01:25
lifelessthere was one about bugs like bug 101:25
lifelesswhich get flashcrowded01:25
_mup_Bug #1: Microsoft has a majority market share <ubuntu> <Clubdistro:Confirmed> <Computer Science Ubuntu:Confirmed for compscibuntu-bugs> <LibreOffice Productivity Suite:New> <dylan.NET.Reflection:Invalid> <dylan.NET:Invalid> <EasyPeasy Overview:Invalid by ramvi> <Ichthux:Invalid by raphink> <JAK LINUX:Invalid> <LibreOffice:In Progress by bjoern-michaelsen> <Linux:New> <Linux Mint:In Progress> <The Linux OS Project:In Progress> <metacity:In Prog01:26
lifelessI cannot find01:26
wgrantBug #73122?01:26
_mup_Bug #73122: individual bug reports can become very noisy and confused when unrelated comments, tasks and other links are created <lp-bugs> <ubuntu-platform> <Launchpad itself:Triaged> < https://launchpad.net/bugs/73122 >01:26
lifelessaiie thanks01:27
wgrantHmm01:27
wgrantSo I guess by apparmoring lxc-start you actually end up apparmoring an entire userspace, which is probably unprecedented.01:27
wgrantNot surprising it doesn't work.01:27
lifelesswgrant: so it looks like it works but fails hard?01:28
wgrantMost stuff works01:28
wgrantBut eg. dpkg-divert sometimes doesn't.01:28
lifeless'nice01:29
lifeless'01:29
wgrantYeah, that's it.01:35
wgrantBreaks ephemeral entirely too.01:35
wgrantProbably because overlayfs doesn't like LSM much.01:35
lifelesswgrant: filed a bug ?01:58
wgrantNah, mailing gary about it.01:59
wgrantNeeds some investigation to see what's actually going on.01:59
wgrantAnd I've got two further non-feature things to do today02:00
lifelesswork? zomg02:00
wgrantI really hate Erlang.02:08
wgrantwallyworld_: Do you still have the rabbitmq-management-agent issue?02:13
wallyworld_wgrant: yes, not sure exactly what the root cause is, but the 4 or 5 rabbit packages don't install02:14
wgrantwallyworld_: sudo rm -r /usr/lib/erlang/lib/rabbit_management_agent02:15
wgrantThere's something wrong with part of the rabbitmq-management-agent package02:15
wallyworld_wgrant: and try reinstalling now?02:15
wgrantYes.02:15
wgrantAnd you can remove the libapt-pkg symlink as well02:15
* wallyworld_ fires up apt-get02:15
wgrantI've fixed the PPA up.02:15
wallyworld_and the debversion package too?02:16
wgrantThe fixed version is built.02:16
wallyworld_too slow at typing, yay, thanks. trying it now \o/02:16
wgrantSo just upgrade02:16
wgrantA fresh installation of launchpad-developer-dependencies works in precise now, apart from the rabbitmq issue02:17
wallyworld_wgrant: yes, it indeed works. thanks, that's awesome02:24
wgrantHuh02:25
wgrantrabbitmq-management installs lots of files that should be in rabbitmq-management-agent02:25
wgrantAhhh02:29
wgrantlynxman typoed it02:29
wgrant--- rabbitmq-management-2.6.1+hg20110927.orig/debian/install02:29
wgrant+++ rabbitmq-management-2.6.1+hg20110927/debian/install02:29
wgrant@@ -0,0 +1,2 @@02:29
wgrant+build/app/rabbitmq_management-2.6.1/* usr/lib/erlang/lib/rabbit_management_agent02:29
wgrant       85 /   69  MaloneApplication:+bugs02:36
wgrantWhy don't I believe that.02:36
wgrantlifeless: Any idea about those numbers?02:36
wgrantThey seem implausibly low.02:36
lifeless144 /    0  MaloneApplication:CollectionResource:#bugs02:36
lifelessmm02:37
lifeless2K total timeouts02:37
lifelesswgrant: '54 /  205    '02:37
lifelesswgrant: might be hiding something02:37
lifelesswgrant: the numbers seem consistent with the total timeout count02:38
StevenKwgrant: Do you mind QAing r14737?03:31
wgrantStevenK: Done03:32
wgrantlifeless: "may never" lolol03:34
* StevenK reaches for mawsonm03:36
StevenKs/m$//03:36
StevenKwgrant: Plan is to get r14738 deployed to NDT as soon as mawson behaves, and hopefully FDT it onto ppa/ftpmaster tonight03:37
wgrantCorrect.03:37
StevenKwgrant: I just came up with that plan, how can it be correct?03:38
wgrantIt is the right plan.03:38
wgrantThe only plan, even.03:38
lifelesshey03:40
* StevenK waits for mawson to finish WADLing03:40
lifelessdoes poppy need DB access any more ?03:40
wgrantNo.03:42
wgrantWell03:42
wgrantIn 5.5 hours it won't.03:42
wgrantBut it may still connect.03:42
lifelesscool, lets rip it out03:43
StevenKlifeless: Your bullet points should be seperate points03:44
StevenKlifeless: "or of bugs filed against launchpad-project "03:44
StevenKWhich doesn't make sense either03:44
lifeless * Increasing the number or difficulty of bugs filed against launchpad-project03:45
lifeless ?03:45
StevenKYes, which requires it and the previous point to be read together, and it scans awkwardly03:46
wgrantlifeless: Bug #246022: team admins/owners (and possibly members?) can contact all members (or the mailing list if one is configured). Everyone else can only contact the owner.03:48
_mup_Bug #246022: No way for team admins to contact all members of the team <contact-via-web> <feature> <lp-registry> <mailing-lists> <teams> <Launchpad itself:Triaged> < https://launchpad.net/bugs/246022 >03:48
lifelessStevenK: I'ma sking if that rephrasing is better03:48
lifelesswgrant: bah, I dropped the list condition from the summary03:49
StevenKlifeless: Oh, right. Sounds fine to me. There is a similar issue in the second group of bullets03:51
nigelbMornin'04:03
nigelbStevenK: For a minute there, I wondered why lifeless was skiing.04:04
StevenKHah04:04
StevenKnigelb: Looking forward to the first T20I? :_D04:04
nigelbNot really.04:05
nigelbI think they'll get crushed. :)04:05
StevenKwgrant: https://code.launchpad.net/~stevenk/launchpad/ttb-no-chroot/+merge/9101504:53
StevenKwgrant: Can haz review?05:26
wgrantStevenK: Is there a test for _getDistroArchSeries?05:29
StevenKIt just returns ubuntu.currentseries.nominatedarchindep, I suspect that is already well-tested05:30
StevenKlifeless: Didn't we kill oops prefixes?05:49
StevenKwgrant: Distracted much? :-P05:50
wgrantStevenK: Sure, but the method isn't tested any more.05:50
wgrantI would reinstate _getChroot and the test.05:51
wgrantThey're doing not much harm.05:51
wgrantOr is dispatchBuildToSlave fairly well tested?05:51
wgrantGiven what it does, it seems unlikely that its tests are stellar.05:52
StevenKwgrant: Why did you unassign yourself from bug 780835?06:00
_mup_Bug #780835: representation cache a pessimization <easy> <performance> <qa-ok> <tech-debt> <lazr.restful:Triaged> < https://launchpad.net/bugs/780835 >06:00
StevenKOh, since it probably needs to be ripped out of lazr.restful now that LP doesn't make use of it?06:01
wgrantStevenK: Right.06:07
wgrantIt was a lazr.restful task, which I didn't notice until after I'd landed the branch.06:07
* StevenK waits for the diff to update. :-(06:10
StevenKwgrant: The diff isn't updating, but http://pastebin.ubuntu.com/824788/ is a diff against submit:06:14
wgrantStevenK: Diff looks updated to me.06:16
wgrantNote that AJAX updating is presently broken.06:16
wgrantApprovalised06:16
StevenKI've been refreshing anyway06:16
nigelbWait. diff stuff is now queued?06:17
nigelbOr is it for launchpad team only for now.06:17
StevenKnigelb: Currently, you need to be a member of ~launchpad06:17
StevenKIt needs a bit of polish06:17
wgrantAnd it's been broken for a couple of days, but will hopefully be back next week.06:17
nigelbHrm, I suspect as much. But still, closer.06:18
StevenKnigelb: Jealous? :-P06:18
nigelbStevenK: You have *no* idea.06:18
nigelb4406:18
nigelbbah irssi06:18
=== almaisan-away is now known as al-maisan
=== al-maisan is now known as almaisan-away
lifelessStevenK: so oops_prefix -> reporter in the new oops stack08:07
lifelessStevenK: we have glue in LP to dynamically allocate them for scripts etc08:07
lifelessStevenK: but they are still used in appservers (but with little reason, we can discriminate via hostname...)08:07
=== almaisan-away is now known as al-maisan
wgrantlifeless: Can we migrate to one-config-per-user yet?08:33
lifelessI don't see why not08:45
lifelessI only ever wanted us to just limit the rate-of-change, for auditability08:46
adeuringgood morning09:03
danhgMorning09:16
stubwgrant: I see you rebuild postgresql-8.4-debversion for precise. Are you using the lucid packages atm. for the rest of things?10:16
wgrantstub: I grabbed slony (1) from oneiric in the primary archive, but had to rebuild debversion due to a libapt-pkg soname bump in precise.10:17
stubno signer listed for slony... don't know if that is a problem10:18
stubAre your PostgreSQL packages from precise?10:19
wgrantNah, the lack of signer is just because it was synced from Debian.10:19
stubJust wondering if I should leave things as they are or try to push the pg8.4 packages into the lp ppa.10:19
wgrantpostgres itself we're still using from the primary archive.10:20
wgrantWhich, as you say, will probably go away in less than a month.10:20
stubI'd like devs on PG 9.1 in the next week or two in any case10:20
stubok. I'll leave it as things are then and assume gary's missing PG packages are his problem.10:20
wgrantBut we can grab 8.4 from precise once it vanishes, if we haven't moved to 9.1 by then.10:20
wgrantHis problem was debversion and slony.10:20
wgrantWhich I fixed a few hours back.10:21
stuboic. thanks.10:21
wgrantI also diagnosed the postgresql-common issue, as you may have seen.10:21
wgrantUnrelated to postgres, as it turns out.10:21
stuband now I see your email :)10:21
stubyes, saw that thread earlier ta.10:21
=== matsubara-afk is now known as matsubara
stubHow is package syncing/package copyiHow is package syncing/package copying being done now? The sync_packages database user exists but I can't find anything in the code base that actually uses it.11:22
stubHow is package syncing/package copying being done now? The sync_packages database user exists but I can't find anything in the code base that actually uses it.11:22
wgrantstub: It's used by IPlainPackageCopyJobs11:24
wgrantWhich are run when IPlainPackageCopyJobSource is given to process-job-source.py11:24
wgrantWhich gets the dbuser from schema-lazr.conf11:24
stubWhich doesn't mention sync_packages, so must be using a different database user11:25
wgrantdbuser: sync_packages11:26
wgrantLine 195811:26
stubpebkac. Was running bzr grep in the wrong directory.11:27
wallyworld__nigelb: watching the T20? :-P11:31
nigelbwallyworld__: Not you too :P11:34
wallyworld__oh yeah!!!11:35
* wallyworld__ loves cricket11:35
wallyworld__especially when India lose :-P11:35
wallyworld__and another wicket falls!!11:36
nigelbfuuuuuuu11:36
bigjoolswallyworld__: haha :)11:36
wallyworld__yep :-)11:37
bigjoolsnigelb, wallyworld__: new Indian interweb domain: .ball11:39
wallyworld__lol!!!11:39
wallyworld__that is hilarious11:39
wallyworld__i almost spit out my wine11:39
wallyworld__all over the keyboard11:39
bigjoolsmy work here is done11:40
wallyworld__indeed it is11:40
nigelbbigjools: At least India didn't get bowled out for 72....11:40
wallyworld__ouch11:40
wallyworld__nigelb pwned bigjools11:40
bigjoolsnigelb: wanna talk about your tour to England last year? all those innings defeats? :)11:45
nigelbStill. 72. It happened even in T20 in a while.11:45
=== al-maisan is now known as almaisan-away
StevenKI was enjoying the '35 needed from 4 balls' counter12:19
=== salgado is now known as salgado-afk
=== matsubara is now known as matsubara-lunch
=== salgado-afk is now known as salgado
=== abentley changed the topic of #launchpad-dev to: https://dev.launchpad.net/ | On call reviewer: gmb, abentley | Firefighting: - | Critical bugtasks: 4*10^2
abentleyderyck: qastaging's database appears at least a month out of date from production.14:54
deryckflacoste, is this known ^^ ?14:55
flacostederyck: yes, it's a known problem14:58
flacostewe don't have a regular update of the qastaging db14:58
flacosteunlike staging which is reset every week-end14:58
deryckah ok14:58
abentleyflacoste: I see.  Thanks.14:58
deryckflacoste, do we schedule it on demand then?14:58
flacostederyck: tbh, i have no idea :-/14:59
flacostebut something like that yes14:59
deryckok, cool :)  thanks, flacoste14:59
flacostemthaddon might know14:59
mthaddonyep, it's on demand14:59
=== matsubara-lunch is now known as matsubara
rick_habentley: can you peek at https://code.launchpad.net/~rharding/launchpad/one_yui_instance/+merge/91111 when you get a sec?16:15
abentleyrick_h: sure thing.16:15
rick_habentley: ty16:16
abentleyrick_h: Why "LPJS"?  Wasn't it something else before?16:21
rick_habentley: it was LPS before but it was brought up that LPS stands for something and isn't as clear16:21
abentleyrick_h: r=me.16:21
rick_habentley: ty16:22
deryckactually, it was always meant to be LPJS.  But I can't type a J for some reason.  And my sed command was wrong. ;)16:23
deryckand no one caught it or complained until recently. :)16:23
sinzuijcsackett, do you have a few minutes to mumble?16:34
deryckabentley, I'm triaging bugs.  Bug 921213 looks like Bug 760735 to me.  Do you agree?17:05
_mup_Bug #921213: recipe does not build: make: Cannot open: Permission denied <Launchpad itself:Confirmed> < https://launchpad.net/bugs/921213 >17:05
_mup_Bug #760735: Daily recipe: failure because of Permission denied accessing build/patch directories <recipe> <soyuz-build> <Launchpad itself:Fix Released by jelmer> < https://launchpad.net/bugs/760735 >17:05
jelmerderyck: they are17:07
abentleyderyck: yes17:07
deryckabentley, jelmer thanks!17:08
deryckabentley, adeuring, rick_h -- I did my hour.  yay me! :)  I cleared project review, open bugs, and answered a couple questions.17:28
adeuringderyck: congrats!17:28
rick_hderyck: woot!17:28
abentleyderyck: cool17:29
jcsackettsinzui: i can mumble now, sorry i was on the phone when you pinged earlier.17:42
jcsackettsinzui: are you still free?17:42
sinzuiI am17:42
* jcsackett fires up mumble17:43
lifelessmorning18:13
mhall119can I check if a person is a canonical  employee or not through the launchpadlib api, using the anonymous login?19:36
james_wno19:36
mhall119ok, then I need to get an app authorized to access that19:55
dobeymhall119: authorized in what sense? you will only be able to see members of that team, if you are also a member of it, as it's a private team.19:59
mhall119dobey: I'm adding to the script that builds bug stats for unity20:00
mhall119I need to show the number MPs coming from community20:01
jelmermhall119: have you seen the contributors script in Launchpad?20:01
mhall119jelmer: no, does it involve zope?20:02
lifelessno, why would it?20:02
jelmermhall119: IIRC it doesn't; it just looks at bzr commits in the mainline rather than merge proposals though20:02
mhall119lifeless: because it's launchpad?20:02
mhall119where can I see this?20:03
jelmerutilities/community-contributions.py in lp:launchpad20:03
dobeymhall119: you can probably check that the owner of the source branch with is_valid_reviewer() on the target branch, in a proposal20:03
jelmermhall119: the result of that script is at http://dev.launchpad.net/Contributions20:04
dobeyjelmer: if it returns true, the owner is part of the unity team, if it's false, the owner is community :)20:04
mhall119jelmer: where can I see the script itself?20:04
jelmermhall119: it's linked from the wiki page20:04
mhall119dobey: what object has is_valid_reviewer?20:05
mhall119dobey: ok, I see <branch>.iPersonTrustedReviewer20:06
abentleyderyck: I'm OCR.  Could you possibly review https://code.launchpad.net/~abentley/launchpad/fix-index-download/+merge/91160 ?20:06
dobeymhall119: yeah, sorry, isPersonTrustedReviewer20:06
mhall119but that only tells me if someone is on the unity team, not if they're non-employees20:07
dobeymhall119: why does it matter if they're an employee or not?20:07
mhall119dobey: because that's the metric my manager wants20:07
deryckabentley, sure20:08
mhall119besides, I'm not even sure if everyone in the dx team would be isPersonTrustedReviewer on unity's trunk20:08
abentleyderyck: thanks.20:08
derycknp!20:08
dobeymhall119: possibly not, but not everyone on dx works on unity, either20:09
=== matsubara is now known as matsubara-afk
mhall119dobey: right, so I don't think jono wants DX members who don't work on unity counted in this metric20:10
dobey*shrug*20:11
deryckabentley, r=me20:31
abentleyderyck: thanks.20:32
deryckabentley, np!20:32
lifeless-> l'hopital20:39
lifelessbbiab20:39
mhall119dobey: ok, I've got it working, now to optimize it20:41
mhall119dobey: is there a way to pre-fetch all the people who can review lp:unity, and check against that for each MP against lp:unity20:41
dobeymhall119: you can get the default reviewer that's set for the branch, and branch owner, and compare against those (they might be different)20:45
mhall119dobey: right, but it'll always be the same team as reviewer20:46
mhall119so can I hit LP once to get all of the members of that team, and then check locally?20:46
dobeyyes; your script should check both in case they are different. if you've already got the branch object, it shouldn't be expensive to see if they are the same.20:47
dobey(to save having to fix the code later, if you decide to run the script against a project which does have different values set) :)20:48
mhall119how far back does <branch>.landing_candidates go?20:51
mhall119and is that the best way to get a historical list of merge proposals20:53
abentleyderyck, rick_h: Did my interrupt duties.  Project review, some Questions, spam deletion, and I updated https://wiki.canonical.com/Launchpad/FeedbackMonitoring20:59
deryckabentley, awesome, thanks.21:00
rick_hStevenK: ping when you're around to chat make/js build dir stuff21:31
StevenKrick_h: Hmmmm?22:06
=== abentley changed the topic of #launchpad-dev to: https://dev.launchpad.net/ | On call reviewer: - | Firefighting: - | Critical bugtasks: 4*10^2
StevenKabentley: If you tag a bug as 'qa-bad', could you also add 'bad-commit-<revno>' ?22:21
StevenKabentley: Not trying to pick on you, it just keeps happening. I think a mail to -dev is in order.22:22
abentleyStevenK: It seems absurd that I should need to.22:22
StevenKabentley: Why? Its needed for the qa-tagger if you rollback the branch22:25
abentleyStevenK: Because there's only one revno I could possibly be talking about.22:26
StevenKabentley: No, there could be more -- land, qa-bad, rollback, land, qa-bad ....22:26
StevenKIt has happened like that.22:26
abentleyStevenK: I'm saying there *aren't* more.22:27
abentleyStevenK: And at the time something is marked qa-bad, I don't think there's ambiguity either.22:28
deryck_Later on, everyone.22:41
wallyworld_rick_h: have you submitted your LPJS branch to ec2?23:02
lifelessStevenK: I agree, a mail to -dev is in order.23:29
lifelessStevenK: if the process is wrong / needs more automation, we can/should do that; but driving the tools inconsistently confuses folk and that consumes time and energy.23:30

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!