? [10:54] <salgado> you could have used different pagetitles [10:54] <kiko-fud> that's only one of the benefits [10:54] <bradb> that's just a nasty bug in the pagetitle machinery [10:54] <bradb> that SteveA said he was going to fix yesterday [10:55] <bradb> in fact, we used to have a way to do it, but it got removed, according to SteveA [10:55] <kiko-fud> bradb, are you against re-adding the task word? [10:55] <kiko-fud> I'm okay with using something else, but... what? :) [10:55] <bradb> kiko-fud: not really. i worry about a normal person being able to work with that word though. but it's hard to say at this stage if it will be confusing or not. [10:56] <bradb> the word bug "sighting" has gone through my mind a few times, but, meh [10:56] <kiko-fud> bradb, I added a <span class="explain"> to the bug pag and tweaked it a bit [10:56] <kiko-fud> page [10:56] <kiko-fud> it might not be a full dud [10:56] <kiko-fud> bradb, if I commit this, would you be interested in pulling my branch and seeing what you think? [10:56] <bradb> kiko-fud: sure, but won't have time today, i don't think. is that ok? [10:57] <kiko-fud> you suck! [10:57] <bradb> (i.e. is monday ok?) [10:57] <bradb> ! [10:57] <bradb> how big is it? [10:57] <bradb> and what does it fix? [10:57] <kiko-fud> it's the same tree you looked at before, remember? [10:57] <bradb> oh, i didn't know you were talking about that branch [10:57] <kiko-fud> yeah [10:57] <kiko-fud> baby cakes [10:58] <bradb> sure, i could look at it [10:58] <kiko-fud> you rock! [10:58] <bradb> :P [10:58] <kiko-fud> let me diff, commit and mirror [10:58] <kiko-fud> I'll ping you [10:58] <bradb> sounds good [10:58] <kiko-fud> you're much better than PQM [10:58] <kiko-fud> I mean [10:58] <kiko-fud> like 100000x much better [10:58] <kiko-fud> like PQM if PQM was a naked waitress [10:58] <bradb> much better than pqm eh? wow, i'm *honoured*. :P [11:06] <salgado> bradb, oh, you have r=salgado with that changes, of course [11:06] <salgado> forgot to say in the email [11:07] <bradb> cool, thanks. just finishing up the one-bugmail-per-recipient patch. [11:12] <kiko> bradb, count-down to merge [11:12] <bradb> t-minus 30 mins... [11:12] <bradb> er, are your changes committed? [11:14] <jblack> Flight was canceled [11:14] <jblack> Somebody had the gal to schedule a hurricane during my flight days [11:15] <kiko> jblack, no joke? [11:15] <jblack> No joke [11:16] <kiko> when is it going to happen? [11:16] <jblack> THey booked me through a different airport tomorrow. [11:16] <jblack> But the eastern us is hosed, because of the remnants of that hurricane. [11:16] <kiko> that's foobed [11:16] <kiko> can you please tell me your rebooking date? [11:16] <jblack> Yeah. [11:16] <kiko> I need to cancel your pickup /now/ [11:17] <jblack> Saturday: Flight 4614 AVP (13:23) -> Atlanta (15:40) [11:17] <kiko> and from atlanta? [11:18] <jblack> Saturday: Flight 105 Atlanta (19:45) -> GRU (Sunday, 06:05) [11:18] <jblack> kiko: You should have already canceled my pickup. :) [11:18] <jblack> Remember? Fadm is picking me up and I was going to spend saturday night in Sao Paulo? [11:19] <kiko> ah. true. [11:19] <jblack> (fadm for fadmthiago. His first name is actually Thiago) [11:20] <kiko> yeah, thiago macieira [11:20] <kiko> his surname means appletree [11:20] <jblack> So, the good news is I should still get there on time. The bad news is that Eastern US flights are a mess, and there's a chance that there may be further problems. [11:20] <kiko> yeah [11:20] <jblack> (on time for work) [11:21] <kiko> hope for snow [11:21] <kiko> martin's safely here already [11:21] <jblack> Hope hope hope. Closing dulles & Atlanta (I imagine reagan too) .... <shudder> [11:21] <kiko> wow [11:21] <kiko> bradb, committed and mirrorred [11:22] <kiko> launchpad--devel--0--patch-84 [11:22] <kiko> or christian.reis@canonical.com--lozenge/launchpad--devel--0 [11:22] <kiko> pull it and smile [11:29] <bradb> heh, thansk [11:30] <bradb> kiko: so, just to be clear, will it be sufficient for me to look just at only launchpad--devel--0--patch-84? [11:31] <bradb> s/just // === terrex [~terrex@84-122-73-155.onocable.ono.com] has joined #launchpad [11:34] <kiko> bradb, nah, look at everything, it's good for you [11:36] <bradb> ok :) === kiko is finishing off a pretty neat lint script === bradb does some good ole ctrl-z on getting kiko's branch to delete a view gigs of cache hate [11:44] <jblack> Heh. I get a booby prize though. [11:44] <jblack> bradb: battlestar galactica tonight [11:46] <bradb> ? [11:47] <bradb> er, i meant "few" not "view"...anyway [11:58] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [trivial] Minor template fixes and import corrections; helps bug 1153 (patch-2108: christian.reis@canonical.com) [12:05] <kiko> wooo! [12:05] <kiko> pqm loves me

[12:07] lifeless: that would seem to address the problems i've been having with pqm. it might be useful if the terminate actions via the web UI included the output that would have been included in the failure email when pqm is bounced currently. [12:08] a little bit that shows me how long the current job has been running might be useful too [12:11] bradb: so, the former is non trivial - the output is in another processes memory buffer === Burgundavia [~corey@S0106000000cc07fc.gv.shawcable.net] has joined #launchpad [12:11] bradb: and not particularly relevant IMO. the submitter needs the error, not the admin/whoever [12:12] lifeless: what if the submitter isn't around anymore? [12:12] bradb: the second, would be doable by stating the file with the pid in it. [12:12] bradb: then they check their email the next day. [12:12] bradb: remember *they* are the person that has to correct their branch. [12:12] lifeless: if the problem is specific to their branch, yeah [12:12] bradb: and if its not they email the list [12:13] bradb: remember that pqm admins will have shell acess still. This isn't meant to substitute for that [12:13] its meant to provide a means for folk to not curse and swear when $admin isn't available. [12:14] right, i understand that [12:15] as a general principle, i just find it useful to get feedback from the system about something that went wrong, if possible. i can understand if it's non-trivial to implement though and, in that case, perhaps not worth the effort [12:16] there are also security considerations [12:16] its one thing to stop someone elses job [12:16] its another entirely to show $random the output from it [12:16] remember that pqm essentially runs arbitrary code. [12:17] sure, but these are trusted people that can access the web UI, right? [12:18] not mutually trusted no [12:18] for example, i don't consider the failure messages i get from pqm to be "for my eyes only" :) [12:19] I can trivially imagine uses of pqm where that would apply [12:19] lifeless: i'd be surprised if non-Launchpad people could kill launchpad jobs. likewise, i'd be surprised if non-baz people could kill baz jobs. [12:20] bradb: auth for pqm is via gpg. Unless the web auth uses 'please sign this via gpg and paste to me' to activate that red button, there will be almost no security on it [12:20] so yes - baz people will be able to kill lp jobs and vice verca, at least for a 1.0 release of that [12:20] oh :) [12:37] lifeless, so what did you fixed? [12:37] carlos: I emailed the list. I've disabled pomsgset [12:38] carlos: and anything that depends on it. [12:38] so you have disabled Rosetta [12:38] yup [12:38] there is no robots.txt I could find [12:38] so I couldn't try that solution [12:39] if we could identify the page (I've given you all the info I have) we could perhaps make it logged in users only or something [12:40] lifeless, that method is used by most +translations pages [12:40] but it was that way since long ago, so I don't understand why is it a problem now [12:40] carlos: like I said in my email, I think its a googlification change [12:41] what change was that? [12:42] carlos: I don't know. I'm guessing [12:42] lifeless, https://chinstrap.ubuntu.com/~dsilvers/paste/fileGAiq3d.html [12:42] lifeless, I got that error from PQM :-? [12:43] btw, it's the main hoary translations page [12:43] it took a lot to load using staging [12:43] right [12:43] but that page is not using that method at all [12:43] well [12:43] at least I don't see the info we get with that method [12:44] the query stub and I found is the culprit === carlos looks at the cod [12:44] carlos: no idea about that error === carlos resends the merge request [12:51] lifeless, I'm not sure the problem is that SQL query [12:52] I don't see anything that calls it from the URL I gave you [12:54] carlos: that query is the only symptom stub and I have identified so far. [12:55] carlos: please operate on the assumption that that is the problem, or tightly related to the problem. === lifeless goes for food [01:12] kiko, lifeless: The patch to fix that needs a db patch to store the value that the method returns [01:12] kiko, lifeless how will we handle that ? [01:12] it's a field addition to POFile [01:16] carlos: usual way. get it written, then review. [01:17] I will be in flight probably, but if its early enough, I'll review the db patch and apply. [01:17] once I'm in flight, it has to wait for stub to arrive, he arrives before me I think. [01:19] carlos: are you sure its not just giving anonymous users too much info ? or an object that does too much during __init__ ? [01:19] back in a bit [01:20] lifeless, I can give you the page that gave us problems [01:20] at least I think I found it [01:20] lifeless, if it's trivial for you to change it now.... [01:25] lifeless, It's late here, and I should go to sleep. At what time will you leave? (UTC) === cprov -> sleep === mdke [~matt@mdke.user] has joined #launchpad [02:07] any rosetta people still up? [02:08] clicking on any language here causes a system error, thus preventing any translation AFAIK https://launchpad.ubuntu.com/distros/ubuntu/hoary/+translations [02:10] mdke, we are having performance problems [02:10] and had to disable most of Rosetta pages [02:10] so malone and the other bits of launchpad are alive [02:11] I hope the situation will be better in a couple of hours with an alternative solution, until we get that fixed === carlos -> bed [02:11] good night [03:15] carlos: I leave home to start the trip at 1730 UTC [03:15] carlos: I think. something like that anyway. [03:15] 18 hours from now === vrsa [~vrsa@static-209-158-45-74.scr.east.verizon.net] has joined #launchpad === vrsa [~vrsa@static-209-158-45-74.scr.east.verizon.net] has left #launchpad [] [04:12] Merge to thelove@canonical.com/dists--bazaar--1.5: new build (patch-40) [04:12] Merge to thelove@canonical.com/hackerlab--devo--1.5: inline ar_size, and implement the upstream talloc alignment fix adjusted for hackerlabs needs (patch-8: robert.collins@canonical.com) === superted_ [~superted@213.167.101.222] has left #launchpad [] === asgeirf [~asgeirf@nat-pool-brisbane.redhat.com] has joined #launchpad === vrsa [vrsa@247.sub-70-216-148.myvzw.com] has joined #launchpad === lamont-away is now known as lamont [07:19] spiv: around? [07:25] jamesh: Yeah. [07:26] spiv: I was looking into how difficult it would be to set some SQL connection vars for webapp. What would the best place to do this be? [07:26] spiv: I saw there was some stuff in canonical/lp/sql.py, but that doesn't seem to be executed per-connection [07:27] would a wrapper round PsycopgAdapter be appropriate? [07:29] Hmm. [07:30] I think that's probably the best spot, yeah. [07:36] spiv: such an adapter would fit best in canonical/database? [07:43] Yeah, that's where all our other nasty database framework code is. === Burgundavia [~corey@S0106000000cc07fc.gv.shawcable.net] has joined #launchpad === BjornT [10183@82-135-221-189.ip.takas.lt] has joined #launchpad [08:14] good morning [08:15] spiv: how's it going with the distinct/order by fix? [08:19] BjornT: I'm about to come back to that, I'm sorting out twistd daemon running issues at the moment. [08:20] cool === Burgundavia [~corey@S0106000000cc07fc.gv.shawcable.net] has joined #launchpad === sabdfl [~mark@pc-n253.wlan.inet.fi] has joined #launchpad [09:41] morning all [09:42] hey [09:43] hey lifeless [09:43] anybody know if stub has tagged off rf for next weeks production yet? [09:48] he hasn [09:48] t [09:48] hes flying [09:49] we're currently running prod-25 [09:51] ok cool [10:22] Merge to thelove@canonical.com/bazaar-docs--devo--1.5: minimal update of locally-branching to fix typos (patch-1: robert.collins@canonical.com) [10:22] especially since that team is a sort of /win close [10:22] gah [10:22] sorry === mdke [~matt@mdke.user] has left #launchpad [] === carlos [~carlos@69.Red-80-33-181.pooles.rima-tde.net] has joined #launchpad [10:54] morning === _Raptor_ [~KillerApp@dsl-253-122.monet.no] has joined #launchpad [11:00] OSError: [Errno 17] File exists: '/var/tmp/fatsam.test' [11:00] lifeless, around? [11:01] lifeless, pqm is again failing... === SteveA [~steve@dhcp68.ams.attingo.nl] has joined #launchpad [11:02] rm rfed that dir [11:04] lifeless, thanks === carlos tries the merge again [11:08] lifeless, did you tried to protect with a login the pages I told you? [11:08] so Rosetta works again === terrex [~terrex@84-122-73-155.onocable.ono.com] has joined #launchpad [11:51] carlos: I don't know which pages you were telling me [11:52] carlos: nor (offhand) how to make them authenticated only. Give me a patch and I'll apply it [11:52] lifeless, don't worry then, It's faster if we just merge the final patch I'm preparing atm [12:12] carlos: I've put together a branch that lets you set the SQL statement_timeout from launchpad.conf [12:13] carlos: it might help in catching bad queries in the future [12:13] jamesh, thank you! [12:13] I think we should activate it on staging by default [12:13] yeah [12:13] before moving that into production [12:14] that will help a bit to prepare production [12:16] it might make sense to have a long timeout on production [12:17] carlos: I get a nice "ProgrammingError: ERROR: canceling query due to user request SELECT ..." exception with a traceback with the patch too [12:17] so we see the evil statement too [12:18] cool [12:18] jamesh, your rock! [12:18] jamesh, you rock! [12:18] :-P [12:19] I didn't have to do anything to get that info in the traceback -- that comes from postgres and psycopg [12:23] jamesh, anyway, that will save us some debugging work [12:23] :-) === gyaresu [~gyaresu@ppp111-83.lns1.hba1.internode.on.net] has joined #launchpad [12:41] lifeless, I have the patch [12:42] lifeless, around? [12:48] jamesh, do you have sometime to review the patch? [12:48] it should be easy [12:49] it touches many files but it's just due a latest_submission -> latestsubmission rename [12:49] isn't that against naming policy [12:49] latestSubmission surely, or latest_submission [12:49] latestsubmission is very hard to read [12:49] lifeless, it's a db field now [12:50] anf ? [12:50] I'm using the same name the db has [12:50] as mark wants [12:50] why not give the db field a good name then ? [12:50] I added a field POFile.latestsubmission [12:50] lifeless, all field in DB are that way [12:50] datelastupdate [12:50] for instance [12:51] I'm just following that naming schema, that's all [12:51] lifeless, btw, could you kill 21916 process at chinstrap? my changes stalled PQM [12:52] done === carlos waits for his new laptop next week to be back with Hoary and without all test problems with breezy.... [12:52] lifeless, thank you [12:54] carlos: I'm really hesitant about that db field name [12:54] lifeless, dude, I can use the old name latest_submission, but that's against all current fields we have in our database [12:56] lifeless, only contraints contain '_' chars in their name [12:56] lifeless, all fields follow that name schema (foobar vs. foo_bar) [12:56] carlos: I'm just looking for the spec on this [12:56] If you don't like them, I suppose you should request a global change [12:57] ok [12:58] oh, baz choose the worst moment to create a revision cache :-( [12:59] the mirror will take a while.... [01:00] carlos: I find myself passing --no-cached to baz archive-mirror most of the time these days [01:00] pep-8 is very clear on this, and we are pep8 conformant. [01:01] Kinnison: please don't, you'll make your merges a lot slower for chinstrap [01:01] lifeless, I see your point [01:01] but this is a db field unrelated to pep-8 [01:01] lifeless: I have mostly short branches and I let base-0 upload cached [01:01] lifeless: but I was in brazil and uploading cached revs was so painful it made me cry [01:01] lifeless, and Mark asked that the sqlobject fields should have the same name we use in the db [01:01] carlos: we are using this as a field name in python code : it is pep-8 related. [01:01] lifeless: I.E. circa an hour to upload a cachedrev [01:01] so we don't have problems to map those directly [01:02] Kinnison: ack. [01:02] lifeless, please, talk with Mark [01:02] lifeless, we had all Rosetta db fields following pep-8 and mark asked us to move all them to lowercase [01:02] carlos: I will. Problem is, I'm *backup DBA*. I'm not going to approve something I'm completely confident with. [01:02] carlos: pep-8 is lowercase for fields, with _ [01:03] lifeless, it's your choice, but take into account that then Rosetta will be fucked the whole weekend [01:03] carlos: dude, don't make like this is my faut [01:03] carlos: that really is not a good way to get what you want [01:03] lifeless, it's mark choice. I have the patch and I did it the way I was told to do it [01:03] I don't want to introduce an inconsistency with the rest of launchpad now [01:04] fair enough. [01:04] lifeless, I can open a bug about this if you are happier that way [01:04] what time does stub land [01:05] but rejecting this patch just because you don't agree with our current way to do it.... [01:05] lifeless, Monday [01:05] no [01:05] he lands way before that [01:05] hes been travelling for 18 hours laready [01:05] lifeless, he told me that will not be available until Monday === carlos checks the wiki [01:06] stub 15 Jul 17:17 [01:06] That's UTC-4 (I think) [01:07] 10 hours from now I think [01:07] yeah, something like that [01:10] plus the bus trip [01:11] 2 hours [01:11] can you mail me the 'baz diff rocketfuel@canonical.com/launchpad--devel--0' [01:11] or more... [01:11] that will give me a starting point [01:11] sure [01:12] I won't unnecessarily hold anything up - but to make changes in production, I'm wearing the channel-stub hat, and hes normally a nazi on style things - as I am when I am paying attention (I.e. I didn't write the code ;)). [01:12] I leave in 9 hours [01:13] lifeless, that's 100% normal and a good thing [01:13] so there are several things : [01:13] I won't roll out something just before I go : if it hoses, it could take the whole system down. [01:14] I won't roll out something that I have backup-dba-hat questions about [01:14] I'll happily rollout non db stop-gap measures. === _Raptor_ is now known as _Rappy_ [01:15] ok, so this patch will not be applied until stub's return, rigth? [01:15] once you mail me the diff, I'll give you an answer. [01:15] anyway, at least if it could reach rocketfuel, that should be enough, so we can check it on staging [01:15] but it needs a script run to migrate the data [01:15] well, cache the data more than migrate [01:16] I think the links to those pages should be made loggedin only or something for now [01:16] as we could roll that out right away [01:18] just a second... [01:20] lifeless, do you want it as a patch using rocketfuel or just a diff is enough? [01:20] it's a one line change [01:20] from zope.Public to launchpad.Person [01:20] what file [01:21] I'm logged into production - what file [01:21] --- orig/lib/canonical/launchpad/zcml/distroreleaselanguage.zcml [01:22] +++ mod/lib/canonical/launchpad/zcml/distroreleaselanguage.zcml [01:22] @@ -22,7 +22,7 @@ [01:22] for="canonical.launchpad.interfaces.IDistroReleaseLanguage" [01:22] - permission="zope.Public" [01:22] + permission="launchpad.AnyPerson" [01:22] class="canonical.launchpad.browser.DistroReleaseLanguageView"> [01:22] name="+index" [01:23] ok [01:23] reenabling rosetta [01:24] done [01:24] please check rosetta is ok and those pages are now protected [01:25] doing it atm. Thanks [01:25] lifeless, https://chinstrap.ubuntu.com/~dsilvers/paste/filewKixLt.html [01:26] that's the patch to solve it [01:26] the pages are protected now [01:28] right. codewise its trivial [01:28] the db field name though, I really am not up for oking - sorry [01:28] Merge to thelove@canonical.com/dists--bazaar--1.5: new build (patch-41) [01:29] Merge to thelove@canonical.com/hackerlab--devo--1.5: bugfix hashtree_fold to not segfault if the tree is empty (patch-9: robert.collins@canonical.com) === cprov [~cprov@200-171-140-32.dsl.telesp.net.br] has joined #launchpad [01:30] lifeless, ok, as I said I see your point, I don't agree with the rejection, but it's ok. Please, send an email to launchpad mailing list with your concerns [01:30] well, we have a workaround .. [01:30] morning all [01:30] so put your patch in the queue for normal approval (the dba queue and the normal review queue). [01:30] Stub can ok or not, and I'll raise this issue in brazil. [01:32] lifeless, ok [01:33] but please, don't forget to talk about it [01:33] I won't [01:33] lifeless, current solution has a problem if you visit the protected url without being authenticated [01:33] the +login redirects sends you to the wrong URL [01:34] hmmm and production is dead now [01:34] ? [01:34] I got a proxy error [01:35] looks fine [01:35] works fine for me [01:36] https://launchpad.ubuntu.com/distros/ubuntu/hoary/+lang/ca [01:36] try that URL [01:36] system error [01:37] theres and exception for you in /errors [01:37] 2005-07-15T12:36:47 ERROR SiteError https://launchpad.ubuntu.com/distros/ubuntu/hoary/+lang [01:37] Traceback (most recent call last): [01:37] File "/home/launchpad/dists/launchpad/utilities/../lib/zope/publisher/publish.py", line 135, in publish [01:37] object = request.traverse(object) [01:37] File "/home/launchpad/dists/launchpad/utilities/../lib/zope/publisher/browser.py", line 494, in traverse [01:37] ob = super(BrowserRequest, self).traverse(object) [01:37] File "/home/launchpad/dists/launchpad/utilities/../lib/zope/publisher/http.py", line 382, in traverse [01:37] ob = super(HTTPRequest, self).traverse(object) [01:37] File "/home/launchpad/dists/launchpad/utilities/../lib/zope/publisher/base.py", line 270, in traverse [01:37] subobject = publication.traverseName( [01:37] File "/home/launchpad/dists/launchpad/sourcecode/zope/src/zope/app/publication/publicationtraverse.py", line 56, in traverseName [01:37] ob2 = adapter.publishTraverse(request, nm) [01:37] File "/home/launchpad/dists/launchpad/lib/canonical/launchpad/webapp/metazcml.py", line 428, in publishTraverse [01:37] traversed_to = self._function(self.context, request, name) [01:37] File "/home/launchpad/dists/launchpad/lib/canonical/launchpad/browser/traversers.py", line 104, in traverse_distrorelease [01:37] lang = langset[langcode] [01:37] File "/home/launchpad/dists/launchpad/lib/canonical/launchpad/database/language.py", line 54, in __getitem__ [01:38] raise NotFoundError, code [01:38] NotFoundError: u'+index' [01:38] dude, use the nopaste for huge tracebacks [01:38] Kinnison: thats not huge. [01:38] It has lines which wrap three times in a normal terminal [01:38] Kinnison: eww, that must look fugly. === lifeless is glad he doesn't have a 'normal' terminal. [01:39] lifeless, that is from the problem I told you when you are not logged in and visit the problematic URL (+lang/ca) [01:40] ah === ..[topic/#launchpad:lifeless] : Rosetta is unwell, login before using! || Discussion with launchpad users and developers. || Includes Rosetta and Malone. || Register for your account on https://launchpad.ubuntu.com/+login [01:41] lifeless, :-D [01:41] hmm, those pages where not so slow last time I used them... === carlos still waits to get it loaded [01:42] well there are four copies of it loading now [01:42] it's normal that google or yahoo break launchpad... [01:42] 24298 | launchpad_prod | 25387 | 101 | launchpad | SELECT POSubmission.id, POSubmission.origin, POSubmission.person, POSubmission.pluralform, POSubmission.datecreated, POSubmission.potranslation, POSubmission.pomsgset, POSubmission.validationstatus FROM POMsgSet, POSubmission WHERE [01:42] POSubmiss | 2005-07-15 11:41:50.107465+00 [01:42] 24298 | launchpad_prod | 25388 | 101 | launchpad | SELECT POSubmission.id, POSubmission.origin, POSubmission.person, POSubmission.pluralform, POSubmission.datecreated, POSubmission.potranslation, POSubmission.pomsgset, POSubmission.validationstatus FROM POMsgSet, POSubmission WHERE [01:42] POSubmiss | 2005-07-15 11:41:52.882554+00 [01:42] 24298 | launchpad_prod | 25389 | 101 | launchpad | SELECT POSubmission.id, POSubmission.origin, POSubmission.person, POSubmission.pluralform, POSubmission.datecreated, POSubmission.potranslation, POSubmission.pomsgset, POSubmission.validationstatus FROM POMsgSet, POSubmission WHERE [01:42] that's why it's so slow... [01:42] Merge to thelove@canonical.com/dists--bazaar--1.5: new build (patch-42) [01:42] POSubmiss | 2005-07-15 11:41:50.120606+00 [01:42] 24298 | launchpad_prod | 25390 | 101 | launchpad | SELECT POSubmission.id, POSubmission.origin, POSubmission.person, POSubmission.pluralform, POSubmission.datecreated, POSubmission.potranslation, POSubmission.pomsgset, POSubmission.validationstatus FROM POMsgSet, POSubmission WHERE [01:42] POSubmiss | 2005-07-15 11:41:50.117912+00 [01:43] Merge to thelove@canonical.com/hackerlab--devo--1.5: inline ar_size (patch-10: robert.collins@canonical.com) === BjornT_ [10183@82-135-221-189.ip.takas.lt] has joined #launchpad [01:45] carlos: that proxy error is apache timing out the request [01:45] lifeless, so... [01:46] its just plain too slow on production, but at least it won't be hung by google or yahoo, and normal users have all the other rosetaa pages working. [01:47] lifeless, I get the proxy error always now... [01:48] those queries are still running [01:48] they will clear up soon I imagine [01:49] ok [01:58] Lunchtastic [01:58] ciao dudes [01:59] Kinnison, later [02:03] lifeless, I think I blocked pqm again as the mirror was not done, could you kill the poimport.py process again? if it's not still there, just kill current run, it's my merge [02:04] I think those pages are still a problem [02:04] perhaps we should make them admin only [02:06] lifeless, that will kill most Rosetta functionality, but it's better than disable Rosetta completely [02:06] so, go ahead, please [02:06] well [02:06] don't hit that page [02:06] and I won't [02:06] if its still up when I go to sleep [02:06] I'll leave it [02:06] otherwise, yeah will limit [02:07] lifeless, it's still down [02:07] I was not able to edit a malone bug [02:07] and that's about 20 minutes with production down already [02:08] carlos: its up now [02:08] carlos: I bounced it right before I said I think the page is an issue still [02:08] ok ;-) === salgado [~salgado@200-171-140-32.dsl.telesp.net.br] has joined #launchpad [02:26] Is Launchpad down? [02:29] morgs, it seems so [02:29] again... [02:29] Launchpad's been taking lessons from PQM :( [02:30] and unfortunately is doing quite good. [02:40] o/~ WHY DO LOVERS DIEEEEE o/~ [02:41] lifeless, protect the page with Admin, is my recom. === carlos -> lunch [02:41] morning kiko [02:41] lifeless, I think it's time to move to launchpad.Admin [02:41] kiko: so is Brazil the country much like Brazil the movie? [02:43] jamesh, no, it has nothing to do with the movie, actually! [02:44] you'll see shortly (I hope) [02:44] should I bring tim tams? [02:44] you know we don't have them here [02:44] could you bring a jar or two of vegemite? [02:44] I'm all out [02:44] I suppose so [02:45] tim tams are delish [02:46] hi kiko [02:46] how's it going with BBA? [02:46] how do you do BjornT_ [02:46] it hasn't gone much as you've seen [02:46] I see jamesh reviewed your patch though [02:46] I was about to comment on it [02:47] I'll do BBA after it, apologies for a rough yesterday [02:47] yes, so i'd like to have the spec reviewed soon ;) [02:50] it'd be nice if malone set In-Reply-To on mail it sent [02:51] either handle it properly, or fake it like bugzilla does, marking every message for a bug as being in reply to a particular message that doesn't exist [02:52] BjornT_, wait for my reply to fix stuff [02:52] jamesh, there's a bug on that [02:53] jamesh: i'll probably fix that one today === terrex is back (gone 00:56:06) [02:53] kiko: ok, i was about to reply to it, but i'll go out for lunch instead then === BjornT_ -> lunch === bradb [~bradb@modemcable082.64-130-66.mc.videotron.ca] has joined #launchpad [02:57] BjornT_, sent [03:03] jamesh: theres a bug on this [03:04] oh bah, should read all scroll back [03:05] carlos: done [03:06] carlos: [03:06] File "/home/launchpad/dists/launchpad/utilities/../lib/zope/app/rdb/__init__.py", line 283, in execute [03:06] return self.cursor.execute(operation) [03:06] ProgrammingError: ERROR: column potemplate.productrelease does not exist [03:06] 2005-07-15T14:06:24 ERROR SiteError https://launchpad.ubuntu.com/rosetta/products/wordpress/wordpress-1.5 [03:06] Traceback (most recent call last): [03:07] carlos: any ideas ? [03:08] lifeless: bug 1415 === debonzi [~debonzi@200-171-140-32.dsl.telesp.net.br] has joined #launchpad [03:14] lifeless, yeah, that bug is also know, it's related to the productrelease -> productseries change that was done last week [03:14] lifeless, I will look into it this weekend [03:15] carlos: I wonder if using sqlbuilder expressions more often would help here? [03:17] jamesh, no idea I hadn't time to debug it and I didn't change that code so I don't know exactly where the problem is, I just know that potemplate.productrelease is not valid anymore [03:18] morning all [03:18] carlos: how long has your merge request been in pqm's queue? [03:19] if it's still there... It's my fault [03:19] lifeless, please, kill it [03:22] dude, this is the third time [03:23] killed [03:23] lifeless, third time? [03:23] lifeless, I only sent it twice [03:24] it's the third time I ask you for it, but I thought you didn't kill it before... [03:24] yup, I did [03:24] :-? [03:24] lifeless, could be that pqm retried it ? [03:24] shouldn't have [03:25] may not have managed to kill it. lp test suite is quite ornerey sometimes [03:25] I only have two failure messages counting the one you just killed [03:25] lifeless, anyway, thank you. === carlos packs [03:25] lifeless, have a nice trip === terrex [~terrex@84-122-73-155.onocable.ono.com] has joined #launchpad [04:01] right, looks like rocketfuel is off limits until next week then [04:02] fucking pqm [04:02] kiko: whats up ? [04:03] is anyone looking into these librarian failures and whatever the "OSError: [Errno 17] File exists: '/var/tmp/fatsam.test'" stuff is? there's the huge moral-killing brick wall in front of rocketfuel right now [04:03] rm rf'd that dir again [04:03] I thought jamesh was going to look into it [04:03]