[01:15] wgrant: oh ? [01:15] :( State: Chroot problem [01:16] lifeless: Ew where? [01:17] sandpaperfig [01:17] natty [01:18] A recipe build? [01:18] natty-cat-lpbuildd doesn't exist yet.. [01:19] So no recipes on natty until we get the new lp-buildd out. [01:19] Which is blocking on the Code bug being fixed. [01:19] Which probably needs a Julian. [01:19] Which I believe we will have tonight. [01:19] wgrant: yes [01:19] wgrant: anyhow, what do you mean by '12:27 < wgrant> Yeah, it looks like it has the same bug that r11812 fixed. [01:20] lifeless: There is a race in the fix for 669676. [01:20] oh, the 'submitted while starting up' thing ? [01:20] Yeah. [01:21] Bug #627608 [01:21] <_mup_> Bug #627608: Got a 401 on a fresh purchase

[01:23] wgrant: we are btw running 11858 now [01:27] lifeless: I thought you fixed the recipe feature flag? [01:29] lifeless: :( [01:29] No recipes yet. [01:31] thumper: [01:31] thumper: what do you mean? [01:31] lifeless: that the recipes will be visible on prod for a specified team [01:32] yes [01:32] once that code is deployed. [01:32] 11866 is the magic number to get qa'd and deployed [01:33] ok [01:34] (bzr log devel :P) [01:38] \o/ my +packages change seems to have worked. [01:38] https://launchpad.net/~yavdr/+archive/stable-vdr/+packages?start=0&batch=204 worked first time. [01:48] lifeless: Why does the deployment page sometimes show the revisions behind the one that is blocking rollout, and sometimes doesn't? [01:49] bad-commits show more [01:49] because they aren't a blocker *if* their rollback is present [01:49] unqaued stops the thing [01:49] its being altered to show more always [01:49] Excellent [01:50] oh OOPS-1773EA141 where art thou? [01:50] * StevenK notes qastaging wasn't reset over the weekend [01:53] lifeless: getBuildStatusSummaryForSourceIds hasn't had its timeout lifted? [01:53] hard_timeout pageid:Archive:EntryResource:getBuildSummariesForSourceIds 1 20000 [01:53] two possibilities [01:53] doesn't work on apis [01:53] its terrible [01:54] I'll check yesterdays oops [01:54] Hmm. It's not showing up in the debug footer's scopes. [01:54] in scopes {'pageid:BugTask:+create-question': False, 'pageid:Archive:+packages': True, 'pageid:Milestone:+index': False, 'pageid:Archive:+index': False, 'pageid:Person:+bugs': False, 'pageid:POFile:+translate': False} [01:55] that thing is a bit bustified [01:55] it shouldn't be showing all those for starters [01:58] SQL time: 19144 ms [01:58] Non-sql time: 1201 ms [01:58] Total time: 20345 ms [01:58] Statement Count: 155 [01:58] conclusion : its terrible. [01:58] https://api.launchpad.net/devel/%7Exorg-edgers/+archive/ppa [01:58] source_ids=%5B%221356771%22%2C+%221356770%22%2C+%221356769%22%2C+%221349686%22%2C+%221349688%22%2 ... 1164%22%2C+%221348857%22%2C+%221348858%22%2C+%221348870%22%5D&ws.op=getBuildSummariesForSourceIds [01:59] Yeah. [02:03] wgrant: hey [02:03] do you have a script that grabs attachments of private bugs, by chance? [02:04] lifeless: No. All I know that does that is apport. [02:04] lifeless: Going to try the shiny new librarian? [02:04] yes [02:04] all reports on qastaging were positive [02:04] Yay. [02:05] Erm, there is no DNS. [02:05] But OK. [02:05] thats interesting .. [02:06] It's there for staging and qastaging, but not production. [02:07] yes [02:07] folk are busy [02:07] its queued, or something [02:07] I've pinged on the rt [02:20] wgrant: the scopes that you see are only those evaluated in order to do the page [02:20] others below that in thepriority list are not show [02:20] *shown* [02:20] lifeless: Ah, so it queries them one at a time? [02:20] one query to get the rules [02:20] then one rule evaluated at a time [02:20] of the rules for the flag being requested [02:20] Ah. [02:22] many rules do not require db querying to evaluate [02:22] so far only team: does, in fact [02:22] wgrant: do you happen to know off the top of you head which template is used for binary builds? [02:22] thumper: Probably build-index.pt [02:22] Yes. [02:22] They weren't renamed when the class and interface were. [02:25] ta [02:26] I should probably fix that at some point, but there's so much overhead in getting stuff landed :( [02:29] thumper: do you think your [testfix] will really work or just hopeful? [02:29] wgrant: you'd like access to directly edit the source on the prod servers? >:) [02:29] spm: pls. [02:30] * spm tries to figure out how to express "no" in a polite, yet firm way..... [02:30] * wgrant pouts. [02:30] spm: FOAD [02:30] * spm avoids drinking, and hence avoids a replacement needing keyboard incident [02:31] lifeless: I was thinking it, i wasn't gunna say/type it [02:31] :) [02:31] spm: and here I was imagining you with a coffee [02:31] something about my cold dead corpse on the ground as well was needed [02:31] spm: That can always be arranged :) [02:31] Canberra is cold enough. [02:31] wgrant 1, spm 0 [02:31] OTOH [02:31] you'd have to go to canberra [02:31] True. [02:32] I'm not sure I'd risk it. [02:32] I could take you for a tour thru NPH! Watch your elected pollies in action! [02:32] I want a jack point in the NBN switch room [02:32] stuff the pollies [02:33] I've been in their DC. About 15 years ago, was bloddy impressively large then. [02:33] heh, had to get them to convert a 9" reel tape onto ... blah, some DEC VMS tape format whose name eludes me. [02:34] same physical format as DLC tape, I think it is. [02:34] oh [02:34] uhm [02:34] fail. same size as LTO!. [02:34] gah. DLT!!!. ffs. fail fail fail. [02:34] rofl [02:34] TK50, TK70 tape. [02:35] http://en.wikipedia.org/wiki/Digital_Linear_Tape <== *1984* format size. still in use today (MUCH higher capacity) [02:36] Even I've used SDLT. [02:36] you old man you [02:36] come to join us greybeards eh? [02:41] thumper: bug 671458. do you agree with it? all new mps will no longer have no reviewer set so is the behaviour described in the bug wrong? i can see both sides. [02:41] <_mup_> Bug #671458: Branch.addLandingTarget adds a review even when the caller requests there be none. [02:42] bac: I expect it to wokrk [02:42] * thumper needs to offload some kids [02:42] anyone want any? [02:42] thumper: thanks, but no. [02:58] sinzui: ping [02:58] sinzui: you seem to (perhaps) have a script marking bugs as fix released... I think if you do that it will need some improvements to be reliably in RFWTAD world [03:01] lifeless: I spoke about that to sinzui at UDS. [03:01] I think in the short term feeding it a list of bugs would be good [03:02] Well I'm specifically thinking of: [03:02] - incrementals [03:02] - not deployed to [03:02] cases [03:02] s/Fix (Committed|Released)/Fixed/g [03:03] wgrant: EWHINE [03:13] http://www.slideshare.net/padday/the-real-life-social-network-v2 [03:14] I saw that yesterday, got up to slide 40 or so, then gave up. [03:14] Is there anything interesting in the later bits? [03:14] not sure [03:14] The start is obvious but interesting [03:15] http://www.hideandseek.net/cant-play-wont-play/ is also entertaining [03:15] Wow that is a lot of Flash. [03:47] Yay, PQM doesn't want to eat my soul. === jtv is now known as jtv-eat [04:08] http://pastebin.ubuntu.com/527981/ some recipe stuff [04:11] sourcepackages being built by daily recipes that have successfully built === almaisan-away is now known as al-maisan [06:05] wgrant: its worth finishing the social network thing; its getting interesting around page 100 [07:02] lifeless: Hm, I may, then. [07:02] After I finish playing with this ARM netbook. [07:02] Don't you have study to do? :-) [07:03] Pfft. === jtv-eat is now known as jtv [07:51] bah, soyuz test fail [08:05] lifeless? [08:07] lifeless: hi [08:17] no danilo? [09:07] good morning [09:09] lifeless: still there? [09:12] Hi [09:25] hi henninge! [09:25] I thought you weren't here today? [09:25] jtv: Hi! [09:26] You thought wrong ... ;) I am just late. [09:27] Ah. I am late too, at least in terms of IRC login. [09:27] I think I have the causes now: id change on devpad, another server freeze of the kind I've been getting since the lucid upgrade, a crappy new router, and some temporary weirdness with my internet connection. [09:27] Not easy to debug so many variables. :-) [09:28] Which id changed on devpad? [09:33] jtv: that sounds hard :( [09:34] StevenK: the ssh pubkey [09:34] henninge: still not done tbh, but I've worked around the problems and that's something. [09:46] Project devel build (197): FAILURE in 3 hr 13 min: https://hudson.wedontsleep.org/job/devel/197/ [09:46] * Launchpad Patch Queue Manager: [r=allenap][ui=none][no-qa] log files are created in a logs [09:46] subdirectory instead of the root directory [09:46] * Launchpad Patch Queue Manager: [r=jelmer][ui=none][no-qa] Calculate a distroseries' architectures [09:46] and components directly from the DB, not via lucilleconfig. [09:46] * Launchpad Patch Queue Manager: [r=jml][ui=none][bug=671568] Remove the last imports from [09:46] canonical.launchpad.interfaces from lp.*. [09:46] wgrant: Ah, you finally managed to land your branch? [09:48] jelmer: Yes! [09:48] jelmer: PQM decided to jump out of testfix for a while. [09:49] Thanks for trying it a couple of times. [09:50] Now I just need to work out what to do about germanium's config's insanity, then the rest of the sequence can land, and lucilleconfig will evaporate. [09:50] A commit by gmb triggered a successful devel build this weekend, but the fix didn't make it into db-devel. So thumper did a manual merge this morning. [10:13] * bigjools is getting frustrated with people targeting bugs to "soyuz" just because they have the word "package" in the title [10:26] lifeless: ping === allenaptoo is now known as allenap [10:46] jtv: danilo_ is asking about you ... ;) [10:46] henninge: ah! [10:46] I got no answer from him earlier [10:46] hey jtv [10:47] jtv, you should have gotten my away message :) [10:47] hi danilo_! No, I got nothing. I couldn't log into the internal IRC. :-( === danilo_ is now known as danilos [10:47] But I rebooted my server so I can get in now. [10:47] jtv, ah, that's where my away message was :) [10:47] Figures. It was _not_ an easy day, IT-wise. [10:48] jtv, why, what happened? [10:51] bigjools: And so it begins :( [10:55] wgrant: ? [10:55] bigjools: My work queue is already non-empty, and I don't start for a month :P [10:55] wgrant: :D [10:56] dude, the queue is massive, it's just that the first one is written down ;) [10:56] Haha. [10:57] wgrant: stop laughing and start coding. Welcome on board. :-) === matsubara-afk is now known as matsubara [11:35] jml: Is the s/PPA/Software Archives/ part of an evil plan to introduce project-owned PPAs? [11:36] wgrant: not really. [11:36] wgrant: I mean, I have an evil plan to introduce project-owned PPAs [11:36] wgrant: but stage 1 of that plan is "introduce project-owned PPAs" [11:36] PPA is dead. Long live LSA. [11:36] bigjools: please don't replace PPA with another acronym. [11:37] almost LSD, which seems eerily appropriate. [11:37] Then the primary archive can be a Launchpad Ordinary Software Archive. [11:37] And then we have awesome acronyms. [11:37] jml: I am not going to type "Software Archives" in full every time, your request is going to be pretty futile I'm afraid [11:38] bigjools: what's wrong with just typing "archive" for short [11:38] it's not specific enough [11:38] I suspect ppa will live on in dev circles, frankly [11:38] me too. I'm not too worried about that. [11:39] but introducing another obscure acronym will be an utter fail [11:39] it would only be in dev circles, not in the UI :) [11:40] hmm. [11:40] what do I do with ArchivePurpose.PPA for example [11:40] we could just leave everything as ppa, internally [11:41] well, that's going to be an interesting question anyway, once you get project archives [11:41] indeed [11:41] ArchivePurpose is probably crack. [11:41] why? [11:41] prefer polymorphism to type checking! [11:41] I agree, I just wish PG would do that ;) [11:42] I'm not entirely a fan of the way PPAs are modeled at the moment. It feels like we should have PersonalPackageArchive which references an Archive, and Distribution which references an Archive, but I'm really not sure how it would all fit together. [11:42] I guess that's probably the case for most of the Soyuz model, though. [11:42] bigjools: separate problem. even if the table has a flag/enum row, the API doesn't necessarily need to have is_ppa() etc [11:43] it might need to for other reasons [11:43] yes, it was badly modelled in that regard [11:43] I have a first class honours in hindsight [11:44] Also, the partner archive sort of screws that idea up. [11:44] But we can hopefully kill that soon, right? [11:44] nope :( [11:44] !? [11:44] Commercial PPAs don't win? [11:44] not until lucid server is EoL [11:44] bigjools: We can't repurpose commercial-compat.sh to mangle a PPA into something that looks like partner? [11:45] we'll be doing commercial PPA^Wsoftware archives AND partner [11:45] :( [11:45] I think I'd prefer a mangling script to keeping this cruft around for five years. [11:45] totally - if it can be done, then I am +1 [11:46] I've not thought about it yet though [11:46] Sure. [11:53] hmm [11:53] I'm having trouble building from devel due to some conflicts in the vietnamese mailman po file [11:55] I wonder why. [11:57] hmm, not devel, but devel w/ c/l/interfaces/__init__ gutted [11:58] Morning, all. [11:59] deryck: hello [11:59] ahhh... deeper, earlier causes === Ursinha-afk is now known as Ursinha [13:00] Yippie, build fixed! [13:00] Project devel build (198): FIXED in 3 hr 14 min: https://hudson.wedontsleep.org/job/devel/198/ [13:00] * Launchpad Patch Queue Manager: [r=allenap][ui=none][no-qa] Unit-test translation permissions. [13:00] * Launchpad Patch Queue Manager: [r=thumper][ui=none][no-qa] Use patched version of zope.pagetemplate [13:00] 3.5.0 [13:02] 3h14 mins? is it by any chance 15.9265 seconds as well? [13:28] bigjools: at some point this week I'd like to talk about uploading to personal archives and the CoC. [13:29] jml: ah yes, we decided to fold the PPA ToS into the CoC [13:29] bigjools: esp wrt https://code.launchpad.net/~jml/launchpad/allow-upload-without-coc/+merge/40218 [13:29] or something with the right acronyms at least [13:31] bigjools: I was talking to sabdfl the other day, and he was fine w/ not requiring CoC for upload rights to PPAs. [13:32] jml: I still think we should make it part of the LP ToS, so it's just implicit [13:32] bigjools: make what part of the ToS? [13:33] the PPA ToS [13:33] bigjools: oh right. [13:33] bigjools: but that's a separate question from CoC [13:33] I probably mean the CoC :) [13:34] oh, there are separate PPA ToS ? [13:35] bigjools: you are either confused or being confusing. [13:36] "the thing you agree to when you sign up for a PPA" [13:36] bigjools: that's the terms of service. but there's also code in LP (see https://code.launchpad.net/~jml/launchpad/allow-upload-without-coc/+merge/40218) that prevents you from uploading unless you have signed the Ubuntu Code of Conduct [13:37] jml: right - what I mean is that we should amend the ToS to include anything we feel necessary from the CoC [13:37] there may be nothing of course, but we should check it over anyway [13:39] bigjools: I've just checked. There's nothing that should be specifically in the PPA terms of service. [13:39] ok great [13:39] bigjools: http://www.ubuntu.com/community/conduct ; https://help.launchpad.net/Legal (LP terms); https://help.launchpad.net/PPATermsofUse (PPA terms) [13:40] jml: I presume the PPA ToS will be implicit... ? [13:41] bigjools: the change I'm proposing doesn't touch the PPA ToS. You will still have to agree to them before creating a PPA. [13:41] bigjools: longer term, we should fold the PPA ToS into the Launchpad ToS. [13:42] but that's a discussion for another day. [13:42] ok === mrevell is now known as mrevell-lunch === adeuring1 is now known as adeuring === mrevell-lunch is now known as mrevell [14:47] gmb, hurrah! Tests pass and it's done! https://code.launchpad.net/~deryck/launchpad/rockstar-js-refresh/+merge/40329 [14:47] gmb, I'm going to land unreviewed since rockstar's already passed reviewed. So one more ec2 run, and it should land. [14:47] deryck: Let joy be unconfined. [14:47] indeed! [14:48] deryck: Cool. Many thanks for all the hard work :) [14:49] np! Glad to get it done. :-) === salgado is now known as salgado-lunch [15:28] what actually does the registering of adapters for top-level IEntrys? [15:29] I've got a branch that breaks it for IPillarNameSet by changing import order around (yay), and I'd like to see what stable is doing differently [15:30] jml: I think there's some zcml that includes canonical/interfaces/__init__.py [15:30] or summat like dat [15:36] something to do w/ lazr.restful.metazcml === matsubara is now known as matsubara-lunch [15:39] boo yah. [15:39] the webservice:register directive === beuno is now known as beuno-lunch [15:55] yeah that one [16:01] sinzui: call? [16:02] one moment I am working with a user [16:02] sinzui: np === matsubara-lunch is now known as matsubara === deryck is now known as deryck[lunch] === salgado-lunch is now known as salgado [16:45] jml, this hour has fallen apart for me. I just finished with a user, but I also just got a call to pick up my son from school. I wll not be available for 30 more minutes [16:45] sinzui: np. [16:45] sinzui: another day then. === beuno-lunch is now known as beuno === al-maisan is now known as almaisan-away [16:59] jelmer: hi [16:59] henninge: hi [16:59] morning y'all [16:59] lifeless: 'morning [16:59] Good early morning lifeless ;) [17:00] lifeless: This time you are actually up early! [17:00] jelmer: 6am, nominal. [17:00] :) [17:01] lifeless: You mentioned on bug 627608 that you did some setup on qastaging to test generate-ppa-htaccess [17:01] <_mup_> Bug #627608: Got a 401 on a fresh purchase

[17:01] oh right, you are on that little Island [17:01] henninge: yes, the little island that could [17:02] lifeless: What did you do exactly? Does qastaging do any publishing at the moment ? [17:02] jelmer: we did [17:02] jelmer: we setup a db user that was missing [17:02] and fixed the librarian config for qastaging [17:02] because of resource constraints, the previously announced 'ask a losa to run specific scripts' still applies. [17:03] lamont: what's the minimum RAM for an i386 virtual builder? Is it different for AMD64? [17:06] lifeless: So publishing doesn't happen by default? [17:06] jelmer: I don't know what 'publishing' stands for. Can't answer the question. [17:07] jelmer: if something you need in order to qa isn't present/happening on qastaging, please ask a losa to arrange it for you, for the patch you're qaing. [17:08] lifeless: I'm referring to the Soyuz publish-distro.py script which is used to publish PPA's. [17:08] lifeless: I'm wondering though where the results end up if that script ran, as private-ppa.qastaging.launchpad.net nor ppa.qastaging.launchpad.net appear to exist. [17:09] jelmer: ok, specifics. \o/ [17:09] jelmer: ask a losa though - its likely the first time we've qa'd something needing that. [17:10] lifeless: I figured the QA'ing of the generate-ppa-htaccess script had the same issue, as it would write to the same location. [17:10] jelmer: we asked the losa to look on disk for us. [17:11] lifeless: ahh, ok [17:12] having to ask a losa for that sort of thing will make QA'ing of Soyuz stuff quite a bit more painful though :-/ [17:12] jelmer: so there are two things [17:12] firstly, ask for the domains etc via rt, cc francis and me if you would [17:13] secondly when we get the qa sscripting server, we'll move all the background tasks for staging and qastaging to it [17:13] so the staging web ui's will be more representative of a deployed web ui [17:19] lifeless: Ok [17:19] lifeless: I'd like to file an rt about ppa.qastaging.launchpad and private-ppa.qastaging.net, is that ok ? [17:20] private-ppa.qastaging.launchpad.net [17:20] (although I do see qastaging.net is still free ;-) [17:20] you're going to have much pain running the publisher on staging [17:22] jelmer: yes [17:23] henninge: is it really week 1? [17:23] No, this is week 3. [17:23] lifeless: what makes you think that? [17:23] henninge: See the topic :-) === henninge changed the topic of #launchpad-dev to: Launchpad Development Channel | Week 3 of 10.11 | PQM is open | firefighting: - | https://dev.launchpad.net/ | Get the code: https://dev.launchpad.net/Getting [17:24] jelmer: what topic? [17:24] ;-P === lifeless changed the topic of #launchpad-dev to: Performance Tuesday | Launchpad Development Channel | Week 3 of 10.11 | PQM is open | firefighting: - | https://dev.launchpad.net/ | Get the code: https://dev.launchpad.net/Getting === deryck[lunch] is now known as deryck [17:30] good night folks [17:31] night [17:37] g'night all [17:42] night [17:45] henninge: so I'd like to go with my proposed mechanism [17:45] henninge: but its only a preference [17:46] henninge: I see the advantage of keeping the focus. [17:46] lifeless: ^ [17:46] lifeless: is this still up-to-date? [17:47] https://dev.launchpad.net/MergeWorkflow [17:47] That's what I used for my propsal. [17:47] I don't think its particularly stale. Let me see [17:47] ah, its stale. [17:47] Because you mentioned that we don't have production anymore [17:47] the big production circle is now reserved for private-patches only - there was a significant list thread about it [17:48] we're deploying straight from stable [17:48] the circle for production is not in the deployment path anymore its straight from stable to lpnet/edge once qa'd. [17:48] I probably saw that thread but the new model had not yet registered well with me so I probably could not follow. [17:49] Ursinha: hi, I think you did that lovely graphic ? [17:49] yes sir [17:49] Hi Ursinha ;) [17:49] hey henninge! :) === benji is now known as benji-lunch === leonardr is now known as leonardr-lunch [17:50] lifeless: so in that case your workflow makes more sense. [17:50] Ursinha: do you have the source still? we need to update it ... I'll update the text right after breakfast [17:50] I just wish we had a way to block developers less. [17:51] lifeless, yes, I do [17:51] have to find it, a minute, please [17:51] actually, updating the text on the page now [17:54] lifeless: you go for breakfast, I will go for dinner. ;-) But I'd like to continue this afterwards. I need a clear picture of the roll-out, obviously ... [17:55] obviously ;) [17:55] :) [17:59] I got a windmill test failure when trying to build re lp/registy/.../test_team_index.py is this fixed now? [18:00] jelmer: so, how are yo ugoing on that qa? [18:00] jelmer: it is the most important thing for the team right nw [18:05] flacoste: ping [18:06] flacoste: the 'release process db-devel->devel [18:06] Inbox [18:07] thread - you noted a small omission on my part about handling urgent fixes during the stabilisation period, but you didn't indicate any other approval/disapproval [18:13] lifeless, I've attached to the wiki page the latest version I have of the diagram [18:13] I think we just collidded :( [18:13] Ursinha: thanks [18:15] I've restored the end of the page [18:16] argh [18:17] silly question: I just attached the file, does that count as editing the page as if I was changing text? [18:18] Ursinha: I think so [18:18]