=== kiko is now known as kiko-zzz [00:36] New bug: #153762 in launchpad "python-medusa missing from developer dependencies" [Undecided,New] https://launchpad.net/bugs/153762 [00:41] New bug: #153763 in launchpad "we should never do bug redirects to disabled products" [Undecided,New] https://launchpad.net/bugs/153763 [00:43] hello, I was wondering if it is possible to take over the username of an inactive user? [01:02] guess it's a bit of a bad time :( [01:05] rob, you can make a request for it [01:05] https://answers.launchpad.net/launchpad [01:57] rob, which user. === RAOF_ is now known as RAOF [02:45] kiko-zzz, either rob or robert [02:48] rob, what's your user? [02:49] rob, both of them have validated email addresses, though [02:49] rob, you should contact them to see if they are willing to give up their names for you. if so, let me know. [02:49] zzz now for real === pitoow is now known as pitoow_zZz === rob is now known as robii === robii is now known as rob [04:15] Could I get some basic info about the tech behind launchpad? [04:17] Anyone here with knowledge of what is behind the scenes? [04:27] Rotund, what kind of thing in particular [04:28] it's twisted under the hood, right? [04:28] Including the web interface? [04:28] What templating engine is used? [04:29] well, it's zope [04:29] The web interface or the whole thing/ [04:30] https://help.launchpad.net/FAQ#head-34295746b9c12bbe42eee4a9bd5e2656306fd796 mentions some of the libraries its built on. [04:31] Interesting. So, what parts of Twisted? [04:34] Bits and pieces. The SFTP/bzr+ssh server at bazaar.launchpad.net, the "librarian" web server, an internal XML-RPC server, and some other internal systems. [04:36] Okay, what about the web interface? is that Zope or Nevow or something else? [04:38] It's Zope 3. (You can see that it's Zope by looking at the Server and X-Powered-By headers in the HTTP responses) [04:40] ahhh. I just looked at the HTML [04:40] * spiv -> lunch [04:40] excellent. just wanted a bird's eye view. thanks [05:04] someone mind confirming and looking at bug #153798 [05:04] Launchpad bug 153798 in launchpad "canonical partner repo packages showing as "in ubuntu"" [Undecided,Confirmed] https://launchpad.net/bugs/153798 [05:05] New bug: #153798 in launchpad "canonical partner repo packages showing as "in ubuntu"" [Undecided,Confirmed] https://launchpad.net/bugs/153798 [07:58] morning === poolie_ is now known as poolie === carlos_ is now known as carlos === Fujitsu is now known as Fujitsu_ === Seveaz is now known as Seveas === tonyy is now known as tonyyarusso === mwh_ is now known as mwh [12:20] hi all [12:21] on launchpad there is no trace of our mirror [12:22] it's uptodate since yesterday [12:22] i'm talking of ubuntu mirror [12:23] hi bostik [12:23] bostik, what is your mirror's name? [12:24] ubuntu.fastbull.org from italy [12:26] bostik, https://launchpad.net/ubuntu/+mirror/ubuntu.fastbull.org-release [12:27] bostik, looks like some files were missing when it was last checked (2h ago) === kiko-zzz is now known as kiko [12:32] salgado, last update 13.01 CET === mrevell is now known as mrevell-lunch === mrevell-lunch is now known as mrevell [13:50] New bug: #153884 in rosetta "Translation import queue needs better validation" [Undecided,New] https://launchpad.net/bugs/153884 [13:50] New bug: #153885 in soyuz "Refactor source and binary publishing presentation templates in a single macro" [Undecided,New] https://launchpad.net/bugs/153885 [14:05] New bug: #153891 in launchpad-bazaar "branch listings can be sorted by invisible quantities" [Undecided,New] https://launchpad.net/bugs/153891 [14:17] New bug: #153894 in launchpad-bazaar "two styles of sorting on one page branch listings probably confusing" [Undecided,New] https://launchpad.net/bugs/153894 [14:22] Hey, if i register a project as blueprint, must i program it all by myself is it possible that someone jumps in and helps [14:23] cause I think i have a good idea but im not sure i can complete it all by myself [14:25] New bug: #153901 in launchpad "OOPS registering attendance in a sprint with year before 1900" [Undecided,New] https://launchpad.net/bugs/153901 [14:30] hi warsocket [14:31] warsocket: If you register a blueprint in Launchpad, any other Launchpad user can become involved. You can even assign people to work on the blueprint. [14:52] cprov: is there any way from the LP UI (looking at build records, especially failed records) to say "please eliminate the records for any superseded source"? [14:54] lamont: no, they are eliminated only when they reach dispatcher (i.e, on-the-fly, not by batches) and since gutsy was released they won't be considered by dispatched and will remain in pending state (hint: it's a problem) [14:55] cprov: and manually deleting them will just be undone when the queue-builder fix hits production, eh? [14:55] my question wasn't really hppa/gutsy related... [14:56] when I pull up the list of failed-to-build packages, I don't really care about earlier versions: I only want to see things where the current version is FTBFS [14:57] lamont: no, failed build from old versions will remain as 'history', we do not provide a proper FTBFS list yet [14:57] cprov: right. they need to remain as history. I just want the option to filter records for superseded source. [14:57] so I'll file a wishlist bug [14:59] lamont: yup. === danilo_ is now known as danilos [15:00] new question. if the dispatcher strips any needs-build records and doesn't dispatch them for gutsy now, then I think I have another wishlist bug that says "please mark such records as failed, reason="stable release, no way (or some such)", rather than leaving the queue polluted with records that'll never go anywhere [15:01] Welcome to this week's Lauchpad development meeting! === pitoow_zZz is now known as pitoow_WAY === pitoow_WAY is now known as pitoow_AWAY [15:01] For the next 45 minutes or so, we'll be coordinating and reporting on Launchpad development. [15:01] which gets back to the base question of "will all those gutsy/hppa build records go away, or hang out" (if so, please make 'em all priority -1 in sql. kthxbye.) [15:01] Who is here today? [15:01] me [15:01] me [15:01] me [15:01] me [15:01] oops. meeting. [15:01] me [15:01] me [15:01] me [15:01] me [15:01] me === bac_afk is now known as babc [15:01] me [15:01] hi [15:01] me [15:01] me === babc is now known as bac [15:01] me [15:01] me [15:01] me, but I am ill so may not stay [15:02] me [15:02] me [15:02] meme [15:02] me [15:02] me [15:02] me [15:02] me [15:02] me [15:02] me [15:02] me [15:02] me [15:02] me [15:03] me (but may experience intermittent disconnection) [15:03] me? [15:03] me [15:04] apologies (according to the meeting page) [15:04] == Apologies == [15:04] * adeuring [15:04] * bac [15:04] * jamesh [15:04] * kiko [15:04] * ddaa [15:04] * stub [15:04] [15:04] obviously some of these are incorrect [15:04] that was lastweek [15:04] I'm not apologizing [15:04] that was last week [15:04] ergh [15:04] == Agenda == [15:04] * Roll call [15:04] * Agenda [15:04] * Next meeting [15:04] * Actions from last meeting [15:04] Did I forget to update it? [15:04] is it just a question of time when i can upload packages to my ppa using hardy as distro? [15:04] * Oops report (Matsubara) [15:04] * Critical Bugs (Rinchen) [15:04] * Bug tags [15:04] * Operations report (mthaddon) [15:04] * DBA report (stub) [15:05] * Sysadmin requests (Rinchen) [15:05] * A top user-affecting issue (mrevell) [15:05] ---- [15:05] * starred reviews - barry [15:05] * postgres shared memory - flacoste [15:05] (other items) [15:05] ---- [15:05] * Blockers [15:05] [15:05] * Next meeting [15:05] same time next week [15:05] Anyone know that they will be away? [15:06] I won't be here next week. [15:06] Schnitz, yes, a question of time. probably 1-2 weeks [15:06] * OOPS report : matsubara [15:06] me [15:06] Today's oops report is about bugs 153884, 153901 [15:06] Launchpad bug 153884 in rosetta "Translation import queue needs better validation" [Undecided,New] https://launchpad.net/bugs/153884 [15:06] Launchpad bug 153901 in launchpad "OOPS registering attendance in a sprint with year before 1900" [Undecided,New] https://launchpad.net/bugs/153901 [15:07] carlos, bug 153884 is the one we discussed yesterday. I couldn't trigger it in [15:07] the exact same way you did in production, but it's a start. Can you take that one? [15:07] SteveA: jtv, danilos and I will be in a sprint, although I guess we will be able to attend the meeting as usual [15:07] jsk, are you the right person to take 153901? [15:07] matsubara: sure [15:08] carlos: thanks. I have some other test cases that trigger a bug in that page. I'll add it to the bug report. [15:08] jsk, matsubara: isn't that already covered by work the jsk started on date formatting? [15:08] matsubara: ok, thank you. That form needs a lot of love so that's really helpful [15:08] intellectronica: I think it might be covered by bug 139413 [15:08] Bug 139413 on http://launchpad.net/bugs/139413 is private [15:09] carlos, it's very easy to cause that page to OOPS, btw. [15:09] intellectronica: there's also bug 134063 but that's not in progress... [15:09] Launchpad bug 134063 in launchpad "OOPS registering sprint or meeting using date format DD-M-YYYY" [High,Confirmed] https://launchpad.net/bugs/134063 [15:09] intellectronica: possibly - I was offline for a moment there. [15:10] ... I won't be here next week. (Sorry, just realized.) [15:10] jsk____: do you think you'll fix 139413 soon? [15:10] cool. [15:10] SteveA, move on. [15:11] kiko: why? [15:11] intellectronica, matsubara: this is primarily related to bug 139360 [15:11] Launchpad bug 139360 in launchpad "DateTimeParser fails silently for certain inputs" [High,Confirmed] https://launchpad.net/bugs/139360 [15:11] matsubara: are you done? [15:11] jsk____: ah so you need a upstream fixed [15:11] SteveA: not yet [15:11] s/fixed/fix/ [15:11] matsubara: yes this is related to an upstream bug. [15:11] SteveA, because this can be resolved privately. [15:11] the oops has been raised already [15:12] (and it's not a very common oops) [15:12] jsk____: ok then. thanks [15:12] SteveA: ok, back to you. [15:12] matsubara: welcome. [15:12] thanks matsubara [15:12] * Actions from last meeting [15:12] mrevell to report a bug about "ubuntero", etc. [15:12] Yes, that's done it is bug number [15:13] * # [15:13] * Critical Bugs (Rinchen) [15:13] # [15:13] Sorry [15:13] mislaid number [15:13] will report to list [15:13] Hi, Two bugs have been resolved and cherrypicked: Bug #147633 and Bug #148147. One bug remains on the list and that's Bug #150988. cprov, what's the current status on this? Please update the bug report too. SteveA, I've replied to you moments ago via email about bug #86185 [15:13] Bug 147633 on http://launchpad.net/bugs/147633 is private [15:13] Bug 148147 on http://launchpad.net/bugs/148147 is private [15:13] Launchpad bug 150988 in soyuz "queue-builder failing due duplicated entries in the DB" [Critical,In progress] https://launchpad.net/bugs/150988 - Assigned to Celso Providelo (cprov) [15:13] Launchpad bug 86185 in launchpad "Librarian should log OOPS reports" [Medium,Confirmed] https://launchpad.net/bugs/86185 [15:15] me (sorry, was distracted with a code review) [15:15] Rinchen: 1.1.10 code will cope with the problem pointed in 150988 [15:16] so we're targeting next week's roll-out for resolution then. [15:17] Rinchen: I didn't have time to work on the DB cleanup patch yet, but since we *can* live with the broken data, I think the bug should be demoted to 'high' [15:17] cprov, won't the broken data cause the constraint to not apply? [15:17] cprov, and isn't the broken data causing OOPSes too? [15:18] kiko: right, the broken data in gutsy-* will block the constraint [15:18] cprov, it needs fixing asap. [15:18] kiko: but I don't think they are causing any OOPSes [15:18] cprov, the "lastscore" oopses? [15:19] kiko: yes, if people hit 'retry' on the duplicated build :( [15:19] right. [15:20] so let's fix the damned builds. don't we have a script that does this already? [15:20] kiko, do you have some time later today to work this a bit further with cprov (in the interest of meeting time)? [15:20] kiko: partially, because we can't remove stuff published in gutsy [15:21] kiko: we have to tweak the SQL generated by the script manually [15:22] kiko: right, let's sort this together, but we need a DBA (postgres access to jubany). [15:22] Thanks cprov [15:23] SteveA, over to you. [15:23] thanks Rinchen [15:23] * Bug tags [15:23] there is one bug tag proposed today [15:23] lp-deps [15:23] bugs related to the launchpad*-dependencies packages in Ubuntu [15:24] however, the bug examples aren't linked [15:24] so I can't easily look them up === Fujitsu_ is now known as Fujitsu [15:24] why not something about lp-deps [15:24] sorry, lp-debs [15:24] I'm defering this until later in the meeting [15:24] or lp-packaging [15:24] kiko: please make the links [15:24] * Operations report (mthaddon) [15:24] since we might have other LP related packages in the future (for API stuff) [15:24] Issues with PostgreSQL and escalation which SteveA has discussed on the list [15:24] Working on codehosting staging setup and feeds staging setup [15:24] Lots of cricket work over the last week - also in the process of migrating cricket to a new server to take load off asuka [15:24] That's it from me unless there are any question [15:25] thanks mthaddon [15:25] * DBA report (stub) [15:25] At about 13:50 UTC either something chewed up way too much shared memory [15:25] resources on the Production DB or we ran out of locks. Restarting the [15:25] appservers and authserver appears to have fixed the issue, which indicates [15:25] it was either an appserver or the authserver chewing up the shared resources. [15:25] I'll increase the suggested tuning parameter next rollout, but we have no [15:25] way of knowing the actual cause or if we will ever see it again. Given the [15:25] codebase has been reasonably static as has the database there is a good [15:25] chance we will never see it again - just a one off glitch causing an [15:25] outage. [15:26] I don't know how long the outage was for or if it 'ramped up' to a [15:26] full on outage or switched from working to not working quickly. As far as [15:26] I am aware it was a single outage rather than an ongoing problem. [15:26] If you have approved branches with db patches land em. [15:26] I'm on leave tomorrow, back on Wednesday. I'll probably be around tomorrow [15:26] during the day. [15:26] stub: was the database restarted? [15:27] stub: was the shared memory max increased? [15:27] Not that I am aware of. And no mem changes have been made (they require a restart) [15:27] stub: please send a message to the launchpad list in future when we have any kind of database issue that causes failing pages [15:27] Or if you mean the linux max, I don't know. [15:27] DB wasn't restarted as far as I know either [15:28] * Sysadmin requests (Rinchen) [15:28] Hi! Is anyone blocked on an RT or have any that are becoming urgent? flacoste, did someone respond with the email information? [15:28] and yes, barry, I'll remind elmo today :-) [15:28] Rinchen: no [15:28] Rinchen: thanks! [15:28] Thanks flacoste. [15:29] Rinchen: #29134 [15:29] SteveA, you are pretty good at skipping /my/ topics, hmmm. [15:29] Who is the correct person to advise flacoste about how to access the incoming staging email pop? [15:29] kiko: what topic is that? [15:29] SteveA, the proposed bug tag. it was just a matter of outputting the bug numbers here and getting them linkified! [15:29] kiko: yep, so we'll come back to it later in the meeting. defered to later in this meeting, not skipped. [15:29] Rinchen, matsubara would know -- didn't he email people about this a while back? [15:30] flacoste, worst case I can have IS reset the pw on it and have them provide you that info. I would assume it was a config value. mthaddon ? [15:30] Rinchen, incoming mail is configured on staging - what access does flacoste need? [15:30] Rinchen, kiko, flacoste: I have access only to the outgoing mailbox [15:30] mthaddon: the credentials so that process-mail.py can access the POP mailbox where these email ends up [15:30] flacoste, it's already working and in place [15:30] incoming email that is [15:31] mthaddon: do they end up at the same place that the /outgoing/ ones go? [15:31] jtv, got it, thanks. [15:31] flacoste, there's a config file in the directory above where RF sits on staging that defines how to connect to the incoming mail box [15:31] mthaddon: ok, thanks! [15:32] Rinchen: case close [15:32] flacoste, mthaddon - thanks [15:32] jtv, elmo the good man he is, raised your ticket and assigned it earlier today so you can expect it soonish [15:32] SteveA, back to you [15:32] kiko: please make the examples links. that way, they're available and convenient for people to look at ahead of time, not at the last minute in this meeting. [15:32] thanks Rinchen [15:32] Rinchen, elmo: thanks [15:32] * A top user-affecting issue (mrevell) [15:32] In the "User deletion" thread on the launchpad-users list, Cesare Falco [15:32] raises bug 87326. [15:32] Launchpad bug 87326 in soyuz "Support PPA Renaming and Reassignment" [Wishlist,Confirmed] https://launchpad.net/bugs/87326 [15:32] SteveA, jesus christ, it's just 4 bugs requesting packaging stuff into lp-dependencies [15:33] When changing his Launchpad username, [15:33] Cesare's PPA remained under his [15:33] previous Launchpad username. [15:33] I'll send a mail to the launchpad-users list to kick of a discussion. [15:33] Thank you and back to you SteveA. [15:33] Apologies for odd formatting. [15:33] I thought we didn't allow people to change their Launchpad ID for precisely that reason. [15:33] mpt, we do allow them to change it. [15:33] we do allow people to change their id [15:34] Maybe I'm confusing that with project IDs [15:34] however, we have discussed the idea that there are certain things you do [15:34] that mean you can no longer change your id after doing those things [15:34] bug #141046 [15:34] Launchpad bug 141046 in launchpad "lp-dependencies missing ssh" [Undecided,Confirmed] https://launchpad.net/bugs/141046 - Assigned to Guilherme Salgado (salgado) [15:34] we need to think about how to warn people about this, and how all that would work [15:34] ok [15:35] I think some forum software or whatever allows people one name change shortly after joining [15:35] SteveA, or perhaps would should let them but with a warning. [15:35] I've seen that kind of thing before [15:35] * starred reviews - barry [15:35] I sent an email to launchpad@ about starred review status, please read and let [15:35] me know if you have any questions. In summary: if your branch is being [15:35] reviewed by a mentored reviwer (a.k.a. recruit) your review status will have a [15:35] star until it is approved by that reviewer's mentor. You are not allowed to [15:35] commit any branch that is still starred. [15:36] Once the mentor approves the review, your reviewer will remove the star and [15:36] then you can act on the status (e.g. commit if it's approved). Any status can [15:36] be starred, and when you respond to a needs-reply* branch, please keep the [15:36] star when you move it to needs-review*. [15:36] Rinchen: if we allow them to change a PPA directory will remain lost [15:36] done [15:37] thanks barry [15:37] * postgres shared memory - flacoste [15:37] Rinchen: I meant, a warning won't solve the issue, we should either deny changes or support them properly. [15:37] actually, that's my item [15:37] (If you aren't aware of it, go reat the 'Critical issue...' threads on [15:37] the LP mailing list.) [15:37] barry: btw, pendingreviews page needs to be updated to handle needs-mentoring status, or does it disappear with the * flag? [15:37] The DB is running out of shared memory causing transactions in many different [15:37] I just put flacoste's name on it because he's taking it forward :-) [15:37] parts of LP to fail. [15:37] We have two suspects: [15:37] - recent use introduction of usage of TEMP table in Translations code. [15:37] - increase usage of authserver and/or Librarian. [15:37] The more of think of it, the more I'm pretty sure it's the changes to use [15:38] SQLObject in the authserver that is the problem. [15:38] These changes were merged on 2007-07-3, so well after the last Ubuntu release. [15:38] Each Ubuntu release brings a spike in usage on the wiki, so it makes sense [15:38] that we are only seeing this now. [15:38] Temporary solution: revert the authserver to the old version not using [15:38] SQLObject. But since there were later changes to the authserver for Code [15:38] stuff, that may not be practical. [15:38] Permanent solution would be to fix bug #140817. [15:38] sorry, well, i prepared something :-) [15:38] Launchpad bug 140817 in launchpad "The librarian reconnects too often (and takes too long to do so)" [Medium,Confirmed] https://launchpad.net/bugs/140817 - Assigned to Francis J. Lacoste (flacoste) [15:38] carlos: bjornt is looking at issues related to pending-reviews and the new star flag [15:38] ok [15:39] flacoste: thanks [15:39] * Blockers [15:39] Foundations team: not blocked [15:39] SC: not blocked [15:39] Translations team: Not blocked. We wish. [15:40] Releases Team: Not Blocked. [15:40] Bugs team: not blocked [15:40] soyuz: not blocked [15:40] Code: not blocked [15:40] flacoste is my hero. [15:40] For today at least :-) [15:41] ok [15:41] and coming back to the bug tags... [15:42] for the proposed tag 'lpdeps', the heading and bugs are not linked, making it inconvenient for readers to see what bugs are refered to [15:42] even so, I think this is better done with a Product [15:42] as it's not launchpad code we're talking about, but packages for use by developers and installation on servers [15:42] so, I think we should use a launchpad-dependencies Project for this instead [15:43] kiko: you proposed the tag. comments? [15:43] +1 for the Project [15:44] SteveA, I'm fine with a separate project; it sounds more correct to me tbh. [15:44] cool [15:44] the only reason I proposed a tag was to be less controversial [15:44] but having it be a project will allow us to use Packaging links correctly too [15:44] great [15:44] since these are included in Ubuntu [15:44] my /only/ question is [15:45] why don't we just use the Ubuntu packages to manage these bugs? it appears that that is frowned upon [15:45] which ubuntu packages? [15:45] as Salgado commented in one of the bugs I looked at yesterday [15:45] launchpad*-dependencies? [15:45] those were in dapper iirc [15:45] by mistake [15:45] Collaborative Commerce: not blocked (sorry for the delay there) [15:45] isn't because they are published from another APT archive? [15:45] kiko, they don't really exist in Ubuntu [15:46] so, basically, they are not in ubuntu in any meaningful sense [15:46] we could put them in a PPA [15:46] they aren't? [15:46] okay let me shut up then. [15:46] right [15:46] they were in once, but it was a mistake [15:46] That may be related to the recently-reported bug about "partner" appearing as an Ubuntu pocket [15:46] a separate project is fine, I'll move the bugs across. [15:46] so they were removed again [15:46] cool [15:46] thanks! [15:46] so, 'lpdeps' as a bug tag is rejected in favour of using a Project [15:47] I'd like us to use PPA for these, if possible, in the future [15:47] dogfooding and all that [15:47] ok, that's all folks [15:47] thanks for being here and keeping the meeting productive [15:47] we can't report bugs against packages in PPAs yet? [15:47] MEETING ENDS [15:47] that would be a cool feature! [15:47] flacoste, we can't no. and that's a scary question. :) [15:47] flacoste, talk to sabdfl. kthxbye :) [15:47] each PPAs becoming a mini-distribution :-) [15:47] flacoste: Please, $DEITY, no. [15:48] kiko: I'm around if you want to discuss the pyspf upstream info you mentioned yesterday (or the day before). [15:48] i'm sure Mark would like the idea... [15:48] * mpt tries to imagine the navigation that would involve [15:48]