=== salgado-afk is now known as salgado === salgado is now known as salgado-brb === salgado-brb is now known as salgado === salgado is now known as salgado-lunch === salgado-lunch is now known as salgado [18:49] Rinchen: sigh. why is the last mootbot log for launchpad-meeting dated 10-jul? [18:49] That looks right to me [18:49] barry, it's manual update. I'll go poke the owner if he's around [18:50] Rinchen: thanks [19:00] #startmeeting [19:00] Welcome to this week's Launchpad development meeting. For the next 45 minutes or so, we'll be coordinating Launchpad development. [19:00] Meeting started at 13:02. The chair is Rinchen. [19:00] Commands Available: [TOPIC], [IDEA], [ACTION], [AGREED], [LINK], [VOTE] [19:00] [TOPIC] Roll Call [19:00] New Topic: Roll Call [19:00] me [19:00] me [19:00] me [19:00] me [19:00] you! [19:00] me [19:00] me [19:00] me [19:00] me [19:00] me [19:00] me [19:01] me [19:01] me [19:01] me [19:01] me [19:01] me [19:01] me [19:01] Rinchen: gmb apologises. he's traveling back oop north from london [19:01] Celso and Muharem send their apologies in an unpologetic kind of way [19:01] apologies from mrevell, statik, and Edwin [19:01] me [19:01] me [19:01] releases team is here [19:01] Foundations is here [19:02] Code is here [19:02] looks like bugs is here [19:02] me [19:02] mthaddon, ping [19:02] me [19:02] me [19:02] OSAs are here [19:02] and the 1 man soyuz team is here [19:02] super [19:03] [TOPIC] Agenda [19:03] New Topic: Agenda [19:03] * Next meeting [19:03] * Actions from last meeting [19:03] * Oops report (Matsubara) [19:03] * Critical Bugs (Rinchen) [19:03] * Bug tags [19:03] * Operations report (mthaddon/herb/spm) [19:03] * DBA report (stub) [19:03] * Sysadmin requests (Rinchen) [19:03] * New packages required (salgado) [19:03] * A top user-affecting issue (mrevell) [19:03] * Doc Team report (mrevell) [19:03] No doc team or user issues today though [19:03] even I'm here today! [19:03] [TOPIC] Next meeting [19:03] New Topic: Next meeting [19:03] OSCON is next week [19:04] Foundations/Commercialization sprint is next week [19:04] kiko and I won't be able to run the meeting next week. We have a session to be in [19:04] anyone want to run the mtg next week? [19:05] * kiko volunteers flacoste! [19:05] oh, he's at a sprint. [19:05] mpt? :) [19:05] kiko: well, if I host the meeting, you can be sure that all sprinters will attend :-) [19:05] sure, I don't mind [19:05] but mpt is such a cool host! [19:05] thanks mpt [19:06] and a great 2.0 UI hacker too it turns out! [19:06] so next week, same time, same channel [19:06] allenap, sinzui - I missed your items in the agenda but we'll do them. [19:06] [TOPIC] Actions from last meeting [19:06] New Topic: Actions from last meeting [19:06] Rinchen: I know you did [19:06] none [19:06] [TOPIC] Oops report (Matsubara) [19:06] New Topic: Oops report (Matsubara) [19:07] Today's oops report is about bugs 249543 [19:07] Launchpad bug 249543 in rosetta "Too many queries to render +products-with-translations page" [Undecided,New] https://launchpad.net/bugs/249543 [19:07] jtv: can you take that one for next week? [19:07] it's our highest query count, which means congratulations to everybody else who worked hard to fix the other pages that were hurting us [19:08] matsubara: OK [19:08] +10 kiko [19:08] yes, huge congrats to the team. current edge oops summary shows 3 hard timeouts and lpnet's only ~40. [19:08] and of course congrats to rosetta who managed to move to being an outlier when it comes to timeouts :) [19:08] storm n all! [19:09] anything else for matsubara? [19:09] Rinchen: no, I'm done. thanks everyone [19:09] kiko: 2 soft ones a day, I believe :) [19:09] and thank you jtv [19:09] matsubara, can we get a report with the slowest pages? [19:09] [AGREED] jtv to investigate bug 249543 [19:09] AGREED received: jtv to investigate bug 249543 [19:09] Launchpad bug 249543 in rosetta "Too many queries to render +products-with-translations page" [High,New] https://launchpad.net/bugs/249543 [19:09] thanks matsubara and jtv [19:09] matsubara, i.e. create a new section and finally get rid of the hard/soft timeout sections? [19:10] the section would have Time Page ID and N (7?) Sample OOPSes [19:10] and a page ID would only appear once [19:10] maybe 10 slowest pages [19:10] kiko: yes, I'll give it a shot on that next week the latest [19:10] cool [19:10] [TOPIC] Critical Bugs (Rinchen) [19:10] New Topic: Critical Bugs (Rinchen) [19:10] Bug 248691 and bug 196364 [19:10] Launchpad bug 248691 in launchpad "Many pages are mostly blank in Internet Explorer 7" [Critical,In progress] https://launchpad.net/bugs/248691 [19:10] Launchpad bug 196364 in rosetta ""Upload a file" link available when I don't have permission" [Critical,In progress] https://launchpad.net/bugs/196364 [19:10] matsubara, we should also add more OOPS samples to the time out count summary table we have there [19:11] mpt and jtv - how are you coming with these? [19:11] Rinchen, 248691 is already fixed, I just haven't got to the PQM message in my Inbox yet. [19:11] k, thx mpt [19:11] Rinchen: mine didn't quite make it in yesterday, so have to resubmit when PQM's open. (Problems fixed up already) [19:11] fantastic. [19:11] thanks gents [19:12] [TOPIC] Bug tags [19:12] New Topic: Bug tags [19:12] no new ones this week [19:12] [TOPIC] Operations report (mthaddon/herb/spm) [19:12] New Topic: Operations report (mthaddon/herb/spm) [19:12] Several times over the last week an app server has died and left a stale pidfile lying around. See Bug #247227. [19:12] On a couple of occasions in the last week we've had to restart codebrowse. Codebrowse wasn't responding and had grown > 1.5GB resident. [19:12] herb: Bug 247227 on http://launchpad.net/bugs/247227 is private [19:12] Yesterday we rolled out the new UI to production. Woohoo! [19:12] Bug #224623, high load on the DB server, crept up last week while the LOSAs were sprinting. This is at least the 5th time in the last 4 months. If there is anything else we can to do help isolate the problem, please let us know. [19:12] herb: Bug 224623 on http://launchpad.net/bugs/224623 is private [19:13] that's it from Tom, Steve and me unless there are any questions. [19:13] herb, is there any more detail/logs from the pidfile issue that can be added to the report? [19:14] Rinchen, it's repeatable in a dev environment, so I don't think too much else could be given [19:14] herb, does the appserver dying vary? [19:14] mthaddon, I didn't know that. Thanks. [19:14] Rinchen, kiko: per sinzui's comment [19:14] Rinchen: We pulled the access logs, launchpad log and nohup.out and there's nothing jumping out. [19:14] kiko: does it vary in what sense? [19:14] herb, i.e. not always the same instance/server? [19:15] mthaddon, sinzui's comment? [19:15] kiko: ah, yes. it does vary. different servers, different times. [19:15] mthaddon, did mwhudson or someone on code fix the new codebrowse up enough to get it deployed? [19:15] herb, gotcha. [19:15] kiko, on the bug report [19:15] I'm mistaken for everyone now. I have become the generic LP developer [19:15] kiko: we are running the new codebrowse [19:15] kiko, yes [19:16] thumper, hmmm, I was hoping that memory would be less of an issue with the new code [19:16] it is [19:16] but obviously not enough [19:17] kiko, thumper: as far as we can tell, it could be mysql browsing that kills it [19:17] hmm [19:17] hmm [19:17] or certain large branches... [19:17] mwhudson would know more, though [19:17] probably inefficiencies in the code [19:17] okay [19:18] should look into that, don't want us showing off for 2.0 and CB going legs up :) [19:18] will do [19:18] anything else for the OSAs? [19:19] thanks herb and mthaddon [19:19] [TOPIC] DBA report (stub) [19:19] New Topic: DBA report (stub) [19:19] no stub today [19:19] * Rinchen sighs. [19:19] [ACTION] Rinchen to email stub for DBA report [19:19] ACTION received: Rinchen to email stub for DBA report [19:19] Rinchen: religious holiday here. [19:20] jtv, last I checked stub wasn't religious. [19:20] kiko: but the country he lives in is. [19:20] [TOPIC] Sysadmin requests (Rinchen) [19:20] Is anyone blocked on an RT or have any that are becoming urgent? [19:20] New Topic: Sysadmin requests (Rinchen) [19:21] [TOPIC] New packages required (salgado) [19:21] New Topic: New packages required (salgado) [19:21] anything for me this week? [19:22] guess not [19:23] k [19:23] skipping mrevell's sections this week [19:23] [TOPIC] When is AccountStatus.SUSPENDED? How are bad users such as spammers locked out? - sinzui [19:23] New Topic: When is AccountStatus.SUSPENDED? How are bad users such as spammers locked out? - sinzui [19:23] I sent an email to the launchpad list for the LOSA and LP developers to look at today. I'm concerned that we have not clearly marked the accounts of bad users we want to suspend. [19:23] https://bugs.edge.launchpad.net/launchpad/+bug/246320. [19:23] sinzui: Error: This bug is private [19:23] I have a script that will update these accounts, and a replacement script to use when we want to suspend an account. [19:23] Rinchen: sorry, he mailed me asking to do the user-affecting section for hiom [19:23] him [19:23] There is a third script that I think is a starting point for identifying broken accounts that we need to fix. These accounts are in an consistent state. We need to decide what state they should be in. [19:24] matsubara, ok, we'll come back to that then. Thanks [19:24] sinzui, I currently believe having a clear delineation between deactivated and suspended is needed. [19:25] For normal folks who deactivate, we want them to be able to reactivate [19:25] Last week we saw about 3000 accounts in an inconsistent state [19:25] for bad folks, like spammers, we want them banned. [19:25] That's my interpretation at least. [19:26] mthaddon: provide me with the script that we have used. It is not good enough, but it did a very good job of clearly marking who we intended to SUSPEND. [19:27] I would like to get the SUSPENDED accounts really suspended this week [19:27] sinzui, was your intent at today's meeting to seek a +1 on the need for delineation? [19:27] i think he was looking for a +1 on running his scripts [19:27] Rinchen: last week was to understand the deliniation [19:28] This week I'm in the position to offer a plan to fix the accounts. [19:29] I am +1 on fixing the inconsistent data [19:29] me too. +1 [19:29] I also am happy that you've taken on the matter of clarifying this SUSPENDED stuff which looks half-baked [19:29] +1 to kiko's comment [19:29] fab [19:29] any objections? [19:30] sinzui: you have email. [19:30] ok, moving on [19:30] hate mail? [19:30] [TOPIC] Can we respond better to user questions raised in Answers? See the thread in launchpad@ and launchpad-users@. - allenap [19:30] New Topic: Can we respond better to user questions raised in Answers? See the thread in launchpad@ and launchpad-users@. - allenap [19:30] Hobbsee observed that 11 questions (in Answers) against Launchpad had expired because the LP staff hadn't addressed them. [19:30] allenap, I answered this publicly [19:30] sinzui: nope. just letting you know how we handle things currently. [19:30] releases team will start to do some of this [19:30] Rinchen: I know, I'm getting to that. [19:30] It struck me as strange that we weren't all answer contacts for Launchpad, so I wrote to the launchpad@ list. Some good responses, but it was all kind of summed up in Rinchen's reply to sinzui's statement that "the Release Team should be answer contacts". [19:30] Rinchen, allenap: the solution to this is to have support rotations. [19:30] Rinchen said: "... this is actually on my roadmap to begin when my second triage resource arrives and is trained up. So, a few weeks." [19:31] I also agree with kiko that we should at least try the support rotations [19:31] But I think intellectronica's idea of a weekly report (from Rinchen's triager?) was good. [19:31] we should not try [19:31] re: community help spec from Dec [19:31] we should DO a rotation [19:31] batching is nonsense, IMO [19:31] kiko: Channeling your inner yoda? [19:31] we should just make sure that there's somebody on-call at least once every 24h period [19:32] so that when I go on vacation [19:32] support requests still get answered! [19:32] abentley, I try. oops. I mean I do. [19:32] we could do it like we do with on call reviews [19:32] yes! [19:32] intellectronica, yeah, that's a similar vein to community help [19:33] https://launchpad.canonical.com/JoeyStanford/CommunityHelpRotation [19:33] ah ok, i'll need to catch up with that spec. joey thinks about everything [19:34] Rinchen: IIRC, there was some resistance to that... or is that new? [19:34] allenap, there was some resistance from stub, and I'm fine with him opting out. [19:34] :) [19:34] what kiko said [19:34] intellectronica had another good idea: an announcer on IRC. [19:34] it's still in draft which is why it's under my name [19:34] but it's been though through quite a bit [19:34] allenap, I think that technology alone won't fix this problem :) [19:35] kiko: I agree, but it can prevent the issue from being forgotten. [19:35] kiko: no, but it will raise awareness. people are more likely to fix the problem then [19:35] We used to have an announcer for bug reports [19:35] intellectronica, I disagree, because we already have users on #launchpad today that come in and ask questions which go unanswered. [19:36] so no, I don't think this is a problem that an announcer will fix. an announcer might make the on-call support person's task easier, though -- less polling. [19:36] kiko: But I don't get a *ping* when someone asks on IRC. [19:36] allenap, just make #launchpad always flash blue when a line with a question mark appears? :) [19:36] kiko: that's what i meant. announcing alone will not solve anything - there needs to be somewhere there waiting for the announcement eagerly [19:36] right [19:36] xchat could ping me when answer-bot sends a message. [19:37] aaaanyway. on-call support FTW. :) [19:38] +1 [19:38] cool. how do we go about setting that up? [19:38] I re-re-re-publish the spec.... [19:38] intellectronica: Do you mean the on-call or the announcer. I'd imagine that the on-call is still in Rinchen's court for now. [19:38] team leads discuss with teams [19:39] and I figure out a way to do scheduling [19:39] if we had one person take it per month... [19:39] everyone could choose the same day and seek coverage [19:39] We could also give this air time at the upcoming TL's meeting in August [19:39] Rinchen: i think it should be more along the lines of person per weekly shift, like reviews [19:40] anyway, i can respond to your spec, once i read it :) [19:40] intellectronica, it can be once a month for the whole team, actually -- we have a large team [19:40] Rinchen, I think we should JFDI [19:40] intellectronica, either way is fine. :-) Reply back. [19:40] another item back on my todo list :-) thanks allenap ;-) [19:40] On-call reviews work, we should have on-call answers using the same kind of rotor [19:40] Rinchen: Apologies ;) [19:41] it's all good [19:41] kiko: +1 on JFDI [19:41] sinzui: i also think so. we have a ready made process that works very well [19:41] we need to do this...which is why I have a spec for it :-) [19:41] intellectronica: agreed [19:42] Rinchen: place a schedule up on the wiki and let people place their name on it [19:42] [AGREED] JFDI on https://launchpad.canonical.com/JoeyStanford/CommunityHelpRotation [19:42] AGREED received: JFDI on https://launchpad.canonical.com/JoeyStanford/CommunityHelpRotation [19:42] Does anyone know who created the bug announcer? [19:42] allenap, yes I do [19:42] allenap, there are 2 of them actually [19:42] allenap: wasn't it mneptok? [19:42] yes plus seveas [19:42] * Rinchen wants to move on [19:43] Cool, thanks. [19:43] [TOPIC] A top user-affecting issue - matsubara [19:43] New Topic: A top user-affecting issue - matsubara [19:43] Some users reported permission issues while using Launchpad. More specifically a team ownership change reported by Hobbsee and a review request for a branch reported by Odd Bloke. I'll investigate those further and file bugs/contact the appropriate teams. [19:43] that's all [19:43] if anyone have any news regarding that, please let me know [19:43] I saw this problem myself [19:44] cool, we called this out on -code earlier as well [19:44] is this unrelated to the issue joey was telling me about, the team transitivity issue? [19:44] kiko, do you have a traceback? [19:44] matsubara: I asked Odd_Bloke for details but got no response. [19:44] of all people, you shouldn't have any permission problems kiko :-) [19:44] salgado, nope, but I'm sure I can try and reproduce on staging [19:45] kiko, that'd be great -- I couldn't reproduce it [19:45] request review is being looked at [19:45] we know what the problem is [19:45] and jml has a branch with fix [19:45] thumper, which branch is that? [19:45] cool. thanks thumper, abentley and jml [19:45] wooo [19:46] will request CP once finished and reviewed [19:46] Anything else before close the meeting? [19:46] Rinchen: not from me. [19:46] salgado: to fix the "request review" fubar [19:46] salgado: probably still on his laptop right now [19:47] ok then! [19:47] Thank you all for attending this week's Launchpad Developer Meeting. See the channel topic for the location of the logs. [19:47] #endmeeting [19:47] Meeting finished at 13:49. [19:47] thumper, ok, will chase him later [19:47] aha [19:47] ps the mootbot logs are not updating [19:47] Rinchen: i know :( [19:47] ps this was a fun meeting after all! [19:47] I'll be pasting in an IRC log until the bot is kicked [19:47] thanks Rinchen === salgado is now known as salgado-afk