[00:04] Problem with Apache [00:04] I modified the Apache config as specified on the wiki [00:04] I then entered this into my web browser http://10.0.0.5 and I get the Apache It Works page [00:05] What can I do? [00:16] Hello Everone [00:16] *Everyone [00:16] hi weather15 [00:16] Does any one know what to do when you get the Apache it Works page when trying to access Launchpad for the first time? [00:17] OPen SOurce of course [00:18] Do I have to enable the Launch Pad SIte? [00:18] and Disbale the default site? [00:20] j [00:21] I disbale the default site [00:21] *disabled [00:21] and then I get the SSL cert error [00:21] But Then after accepting it:: The page isn't redirecting properly [00:22] What can I do about ^ [00:26] >> ^ [00:30] I odn't understand the question [00:30] perhaps you should start with a more regular development environment ? [00:31] Firstly, what are you trying to do? [00:32] I downloaded the Launch Pad Installer [00:32] Followed these instructions: https://dev.launchpad.net/Running [00:32] For what purpose? [00:32] Then These : https://dev.launchpad.net/Running/RemoteAccess [00:33] To Use Launch Pad Open Source [00:33] After starting using: make run [00:34] I tried to access LaunchPad via http://launchpad.dev [00:34] wouldn't the default virtual host config generally be matched first before the launchpad one? [00:35] Then Firefox throws this error: The page isn't redirecting properly [00:36] Any idea as to what to do? [00:37] I disabled the default Apache Site becuase I just got the It Works Page! when try to access Launch Pad [00:38] you probably do need to access it by name [00:38] Right. Launchpad uses virtual hosts extensively. [00:38] So you need to access it as 'launchpad.dev' [00:39] I'm using this: launchpad.dev [00:39] I did how ever need to set this in my Hosts file [00:39] as my DNS does not seem to resolve that name [00:41] thats expected [00:41] please note [00:41] that what you have downloaded is not an installer for launchpad [00:41] it is the launchpad source code - there are no installers [00:42] I understand that as I have followed the wiki instructions to get it to run [00:42] ok [00:44] \o/ milestones branch passed ec2 land [00:46] lifeless: thanks for the suggestion on the tests for my usage_enums branch; i was looking for a cleaner solution, and at midnight nothing was coming to me. :-P [00:49] I think I found the error [00:49] [Sun Aug 22 19:43:59 2010] [info] Initial (No.1) HTTPS request received for child 0 (server bazaar.launchpad.dev:443) [Sun Aug 22 19:43:59 2010] [debug] mod_deflate.c(615): [client 10.0.0.3] Zlib: Compressed 284 to 217 : URL / [Sun Aug 22 19:44:14 2010] [debug] ssl_engine_io.c(1892): OpenSSL: I/O error, 5 bytes expected to read on BIO#7fe8b8ba53c0 [mem: 7fe8b8be6c40] [Sun Aug 22 19:44:14 2010] [info] [client 10.0.0.3] (70007)Th [00:50] Just what can I do about it? [00:50] jcsackett_: my pleasure === jcsackett_ is now known as jcsackett [00:54] ?? ^ [00:59] weather15: I don't know [01:59] Hi there. Is there a way of linking directly to the Launchpad-hosted Changelog for a release? [02:00] I'd like to link to the Changelog text on https://launchpad.net/gcc-linaro/4.4/4.4-2010.07-0 [02:07] wgrant: around? [02:08] michaelh1: not AFAIK [02:08] OK. I'll submit a feature request. [02:08] a patch would be better :) [02:09] actually, more seriously [02:09] why? [02:10] michaelh1: ^ [02:10] lifeless: oh, we want to do a summary page in the Linaro wiki that lists the projects involved in the current release, a link off to the download, and a link off to the change log [02:17] lifeless: Sure. [02:17] I'm looking at [02:17] https://launchpad.net/ubuntu/+search?text=smplayer [02:17] timeouts [02:17] Ew DistroSeriesPackageCache [02:17] And DistroSourcePackageCache. [02:17] yes [02:17] Kill them with fire. [02:18] 8000ms queries [02:18] Ow. [02:18] what should be used [02:18] also I recall you had a query to establish the archive ids [02:18] to reproduce things on staging [02:18] I wanted to snarf that from you (or hell, just the ubuntu archive ids would do) [02:18] The magic number is 534. [02:18] That's parter. [02:18] primary is 1. [02:19] s/parter/partner/ [02:35] lifeless: I looked last week, and one of those caches is pretty useless. [02:35] I think DistroSeriesPackageCache. [02:35] But DistroSourcePackageCache might be slightly more performant than a straight query. [02:39] wgrant: https://bugs.edge.launchpad.net/soyuz/+bug/618372 [02:39] <_mup_> Bug #618372: Distribution:+search slow 50% of requests [02:40] its nice that the timeout predictor well, predicted [02:45] How bad is the plan? [02:45] dunno yet [02:46] but clearly pretty bad [02:46] Heh. [02:49] so to get the plan [02:49] I need values for distribution [02:49] ubuntu - 1 I'm guessing [02:49] releasestatus === Ursinha is now known as Ursinha-afk [02:50] releasestatus is an enum. [02:50] ubuntu might not be 1. [02:50] I'm not sure. [02:50] its 1 === Ursinha-afk is now known as Ursinha [02:58] grah [02:58] wgrant: lib/lp/registry/browser/distribution.py - I want a second opinion on line [02:58] 479 [02:58] I think its nuts; I don't see why it would work even. [02:59] Regardless, it's irrelevant now. [02:59] Storm is fixed. [02:59] Delete delete delete. [03:00] It may be that DRS has a workaround that's now gone since the original bug is fixed. [03:00] s/has/had/ [03:03] argh [03:03] has_exact_matches() [03:03] sigh [03:04] is there a tal construct for 'show this <> if iterates 1 or more times ? [03:06] Not directly, I don't think. [03:06] Anyway, I'm gone. [03:06] we need one [03:06] ciao [03:06] thanks [03:37] ✁☹ [03:37] damn simplistic freaking tests [03:38] * thumper taps fingers.... [03:42] * thumper hackerates [03:44] * thumper fixes \\o/ [03:45] ffs openid-apache-module-hate-hate-hate [03:48] poolie: so [03:49] poolie: I'd like you to send a hand-off mail to lp-dev, if you would, on flags [03:49] poolie: just a 'I'm not moving this forward for , folk that want to can look at , and I'd do next' or something [03:49] poolie: if you have an in-progress branch, snapshotting that and including its details would be awesome [03:50] yeah, good idea [03:50] i was still thinking of doing another increment on it today, but i will at least send that [03:50] i am a bit disappointed nobody else replied [03:50] i guess it needs some examples of actually being useful though [03:50] the lp dev list has an odd dynamic. [03:51] Not to worry; we will fix. [03:57] lifeless: you'll fix the dynamic? [03:58] * thumper runs when he looks at the time [04:32] thumper: https://code.edge.launchpad.net/~lifeless/launchpad/milestones/+merge/32855 [04:33] thumper: what does 'no longer in the source branch' mean? perhaps it should link to help.lp.net? [04:34] what it means is that the approved revision isn't in that branch [04:34] due to something like a push overwrite [04:34] ok [04:35] lifeless: is it right? [04:37] yes [04:42] is it because you reused the branch? [04:42] I had a bunch of cruft, which I collapsed into one commit [04:42] so reused - yes, but for that same mp. [04:42] history-edit [04:45] * thumper feels like a primal scream [04:45] clucking canonical_url fallout [04:45] and broken code [04:46] lifeless: Any luck? [04:55] thumper: :( [04:55] thumper: I thought c_u fix landed? [04:55] wgrant: plan is in the bug [04:55] lifeless: it has [04:56] thumper: untested code paths died as a result ? [04:56] lifeless: in my next branch, I've found lib/lp/testing/menu.py [04:56] which is wrong [04:56] so my fix for it breaks stuff [04:56] grr... [04:56] :( [04:56] didn't find it in the old branch as nothing that triggered the bug was being tested [04:58] spm: hi [04:58] spm: I'd like a kcachegrind please ;) [04:58] lifeless: err... [04:58] spm: ...rre? [04:59] of what where and probably more significantly, how :-) [04:59] spm: we did this before your sprint [04:59] ✁☹ [04:59] um, no we didn't. :-) [04:59] spm: you change the setting on staging, I hit up the url, you change it back [04:59] spm: yes, we did. [04:59] oh THAT. right. debug. [05:00] actually [05:00] nvm, I suspect that the underrepresentation of sql may be a factor [05:01] so I'm going to fix the darn sql [05:01] if its still naffed after that, we'll get a cleaner kcachegrind file [05:01] nod [05:01] sorry for the interrupt [05:01] wgrant: so - has the distribution search changed recently? [05:02] lifeless: Not since 3.0, AFAIK. [05:02] wgrant: if not, then its likely a transient DB issue, as its bad on edge and prod [05:02] and wasn't showing up as such before [05:02] so I've subscribed stub via the dba tag [05:02] however [05:02] the plan shows loop on loop on loop [05:02] for thousands of items [05:02] Yes, the plan makes me cry. [05:02] thats going to add up any which way [05:13] stub: I've tagged another bug dba [05:13] slow query, no code changes recently, wondering if you could see if its an operational issue / bad query design etc [05:13] stub: also, the PPR seems to have broken over the weekendish [05:24] jtv: Should suggestivepotemplate.potemplate be a foreign key reference to POTemplate (ON DELETE CASCADE I suspect as this is a cache) [05:26] jtv: Should suggestivepotemplate.potemplate be a foreign key reference to POTemplate (ON DELETE CASCADE I suspect as this is a cache) [05:53] thai internet doing well today, it seems [05:54] bouncy bouncy [06:07] mwhudson: Do you remember the discussion we had at least 3 weeks ago about my branch that added the beginnings of job support for InitialiseDistroSeries? [06:07] StevenK: vaguely [06:08] mwhudson: I removed job_type like you suggested, but now the test for InitialiseDistroSeriesJobDervied fails, since .create() doesn't raise an AttributeError [06:10] StevenK: err [06:14] * StevenK suspects he has lost mwhudson [07:36] Morning! [07:59] hi noodles775 [07:59] noodles775: I has a question for you [08:00] noodles775: lib/lp/registry/browser/distribution.py [08:00] line 479 [08:00] is that still relevant ? [08:07] * noodles775 looks [08:09] * noodles775 pulls a recent version of devel [08:13] lifeless: So the associated bug has been released, and we're using a version of storm that includes it. So I would say no. Remove it, and see (or I can do it if you're done for the day). [08:57] good morning === almaisan-away is now known as al-maisan === Guest73756 is now known as jelmer === jelmer is now known as Guest76162 [09:33] noodles775: I'm well done ;) [09:33] noodles775: I'd love it if you could; I found that code while starting to look at a perf issue [09:33] lifeless: np, I'll do it now. [09:33] thanks! [09:34] It'd be nice if we had a tool which would report XXXs referring to closed bugs. [09:35] that would be nice [09:35] it would make an awesome bzr precommit check too [09:35] or even better, a .... [09:35] that [09:36] night all; its only 8:30 but 5am starts really takes it out of you [09:36] Night lifeless. [09:37] and I want to be alert for perf tuesday! :) [09:37] Night lifeless. === Guest76162 is now known as jelmer === henninge_ is now known as henninge [12:03] Morning, all. [12:10] good morning deryck === jkakar_ is now known as jkakar === al-maisan is now known as almaisan-away === almaisan-away is now known as al-maisan === matsubara is now known as matsubara-lunch === Ursinha is now known as Ursinha-lunch === beuno is now known as beuno-lunch [17:46] is there a rockstar in the building? [17:47] * rockstar is a rockstar [17:47] deryck, how can I help you? === matsubara-lunch is now known as matsubara === Ursinha-lunch is now known as Ursinha [18:07] james_w: hi === benji is now known as benji-lunch === beuno-lunch is now known as beuno [18:32] bigjools, I've replied to your review. Megapologies about the delay. === al-maisan is now known as almaisan-away [18:48] sinzui, ping [18:49] hi deryck === benji-lunch is now known as benji [19:10] erm, this is a bit special. python-testtools is "Published" in jaunty in ~bzr/proposed, but doesn't occur in the Packages file === deryck is now known as deryck[lunch] [20:03] g'night all. [20:22] moin === almaisan-away is now known as al-maisan === lifeless changed the topic of #launchpad-dev to: hpad Development Channel | Week 1 of 10.09 | Performance Tuesday! | PQM is OPEN | firefighting: - | https://dev.launchpad.net/ | Get the code: https://dev.launchpad.net/Getting | On-call review in irc://irc.freenode.net/#launchpad-reviews === deryck[lunch] is now known as deryck [20:34] oops, forgot I marked myself away for lunch [20:34] bryceh, your branch looks good to me. r=me. [20:35] hi lifeless [20:36] deryck, great thanks [20:36] brianchidester, oh, wait. Did you test the new status conversions? [20:37] gah, tab expansion death [20:37] bryceh, oh, wait. Did you test the new status conversions? [20:38] if externalbugtracker-bugzilla.txt is not exhaustive, then no worries. If it was, then we should keep it so. [20:40] deryck, it isn't exhaustive [20:41] deryck, it just spot checks. I could add a few if you think it'd help [20:41] bryceh, nah, what you've added is fine then. [20:41] okie [20:44] deryck: hi [20:44] deryck: my bugtaskset change is in :) [20:44] excellent :-) === matsubara is now known as matsubara-afk [20:51] so [20:51] lets see whats timing out today [20:51] I suspect bug attachment api calls [20:51] deryck: has anyone claimed that yet [20:51] ? [20:51] lifeless, no, no one is on that yet. [20:52] deryck: also while you are here [20:52] have you seen https://bugs.edge.launchpad.net/malone/+bug/607935 [20:52] <_mup_> Bug #607935: timeout on bugtask:+index [20:52] 367 sql calls :) [20:53] lifeless, I have seen that in that I saw the few bugs you filed around the same day. And I just haven't got to prioritizing that yet. [20:53] deryck: it seems to be one-by-one retrieving people [20:53] I'm guessing its subscriptions or something [20:54] lifeless, ok, that's one of my spots I wanted to look at anyway next. [20:55] I'll be doing OOPS and timeout poking the rest of this cycle while the others finish up the subscriptions noise story. [20:55] \o/ [20:55] I'd love to help you [20:56] but we're in rather different tz's [20:56] :) [20:56] although, what time is it for you, now ? [20:56] well, work hours anyway. :-) [20:56] deryck, I see from the config that there is mention of a bug-heat user/script. I think it is gone. Is this right [20:56] it's only 3 PM for me. But that's EOD. I timeshift to match Europe better. [20:56] deryck: ah! [20:56] sinzui, yes, should be. [20:56] deryck, thanks [20:57] np [20:57] well, I'll try to be up @ 6 tomorrow and we should get 2 hours of overlap [20:57] Sure, that would be fun. [20:57] But I hate to ask you to get up early :-) [20:58] deryck: getting someone else to do perf stuff on tuesday is a pleasure, not a penalty! [20:58] heh, cool then [21:00] ok, so until tomorrow then.... [21:12] gary_poster: I saw your update of Foundations/Webservice [21:13] gary_poster: perhaps we should file a bug asking for a "%s.%s" % (__module__, self.__class__) to be gathered too ? [21:24] lifeless: I thought about that when we were first implementing. At the time I thought it would be unnecessary: it would make for a more unwieldy pageid, and we can always grep. I'm less sure now, but TBH I think the story of going from the named operation to the method is more annoying, at least potentially. [22:03] gary_poster: I'd like all the steps to be easy :) [22:03] * lifeless is lazy === al-maisan is now known as almaisan-away === Guest39999 is now known as jelmer [22:24] morning [22:28] hello === salgado is now known as salgado-afk [23:16] https://lp-oops.canonical.com/oops.py/?oopsid=1695L1219 should so not be a soft timeout [23:16] 45 seconds! [23:33] thumper: when you're around, a brief call would be nice [23:33] ack [23:33] is that 200 OK, or 101 please wait [23:34] 101 [23:34] kk [23:36] If I said object X is "a request timeline", what would you think it does ? [23:47] gary_poster: ping [23:47] gary_poster: does zope have something already existing for tastefully annotating request objects. [23:48] e.g. I have an object (say the db statement list we generate) that I want to associate with a request, such that I can, whenever I have the request, get this other object. [23:48] Its a little distateful to write to that object directly [23:49] and a weakrefkeydict is a little ergh to work with, though doable.