[05:16] wgrant, https://code.edge.launchpad.net/launchpad === fjlacoste is now known as flacoste [05:22] mwhudson: is it possible that your branch is stuck in PQM? [05:22] https://dev.launchpad.net/Getting [05:22] wgrant: ^ [05:22] flacoste: it's possible, i guess spm is the man who'd know [05:24] * thumper stretches in his new home [05:29] smell that? [05:29] smells like *freenode* [05:30] heh [05:30] :) [05:31] good night guys! === flacoste is now known as flacoste_afk [05:31] mwhudson: does loggerhead cause a test run? [05:31] thumper: no [05:31] good [05:31] it probably causes a make build though [05:31] right [05:33] all this excitement and wgrant isn't around :) [05:33] we should close it again [05:33] beuno: heh [05:33] Oh man, that's terrible... :) [05:34] haha [05:34] beuno: your branch listing page is now the top of my stack [05:35] thumper, awesomeness [05:36] spm: is pqm actually doing anything? [05:36] thumper: yes. ssh'ing [05:37] about 7 mins ago [05:37] hi BjornT [05:38] thumper: I had a stop patch in when I did tghe manual merge for flacoste earlier. forgot to remove it. my bad [05:38] spm: ta [05:38] hi thumper [05:38] BjornT: welcome to the brave new, public, world [05:39] hey guys :) [05:41] Glad I was up late tonight for the big moment. :) [05:42] deryck, :) [05:42] deryck, you're CDT, right? [05:42] rockstar, yup. [05:43] congrats LP devs! [05:43] LaserJock: thanks [05:43] LaserJock, thanks :) [05:43] deryck: i take it you won't be start working in 5 hours, like usual? :) [05:44] BjornT, no, I will actually. :) [05:44] my loggerhead branch landed at last [05:44] BjornT, there's always time for an afternoon nap, after work. :) [05:46] it wasn't clear from the announcement, is *all* of LP getting opened or is Soyuz (and Code hosting?) being held back? [05:46] it's all there. [05:46] LaserJock: have a look at the code :P [05:46] LaserJock, wysiwyg :) [05:46] flacoste_afk: any chance of getting this crucial fix into the rollout? https://code.edge.launchpad.net/~jtv/launchpad/bug-401835/+merge/9074 [05:47] flacoste_afk: very last-minute I know, but QA has been a huge bottleneck since May. [05:47] lifeless: well, https://dev.launchpad.net/Getting said it can take a couple hours to get the code, I was hoping for a shortcut :-) [05:48] LaserJock: it's the lot [05:49] LaserJock, http://bazaar.launchpad.net/~launchpad-pqm/launchpad/devel/changes [05:50] beuno, as if the servers weren't already going to take a beating, you link to loggerhead and watch it melt too. :) [05:50] https://bazaar.launchpad.net/~launchpad-pqm/launchpad/db-devel/files/head%3A/lib/lp/ [05:50] even better URL [05:50] :) [05:50] stress test! [05:51] * beuno is glad he worked on nice URLs for LH [05:52] grats everyone :) [05:53] radix, thanks :) [05:55] hi Hobbsee [05:55] hey thumper! [05:55] Congratulations all! [05:55] hey Hobbsee, early birthday pressie for you :-) [05:55] :) [05:56] hi :) [05:56] spm: indeed! [05:56] wow, just wow [05:56] Hobbsee: we look forward to your patches ;) [05:56] Congratulations everyone. [05:56] wuuu https://code.edge.launchpad.net/~jml/launchpad/branch-flow-utility [05:56] I can't believe this day has come :-) [05:56] my first public Launchpad branch [05:56] soo much code [05:56] LaserJock, yes. yes there is. [05:56] LaserJock, and so much history ;) [05:56] * mwhudson is very tempted to start drinking [05:57] mwhudson: me too [05:57] mwhudson: and since I'm going to be cooking dinner in about 15 minutes, I may well just pour a nice glass of pinot noir [05:57] it's 2am here, anything goes [05:57] thumper: mwhudson: me three, but um, someone needs to make sure servers don't melt :-) [05:57] mwhudson, _I_ am tempted to start drinking. :) [05:58] thumper: problem is, we really need to go shopping... i guess there's always pizza delivery :) [05:58] mwhudson, me too :) [05:58] is it generally going to require running a local instance of LP to work on bugs? [05:58] LaserJock: it certainly helps [05:59] LaserJock, the tools to do that aren't terrible. [05:59] cjwatson, hi :) [05:59] join the party :) [05:59] * cjwatson waves woozily [05:59] LaserJock, it's nothing like running the main LP servers. [05:59] spm: did flacoste_afk's branch land? [05:59] spm: I can't tell [05:59] rockstar: ok [05:59] spm: LP says nay [05:59] LaserJock: it is "make run" :) [05:59] LaserJock: if you've done the setup [05:59] thumper: which one? I manually merged one. should be a listening to Brazilian Girls something [06:00] I heard it was a party in here [06:00] Would this be accurate? [06:00] spm: the one that he submitted just before going to bed [06:00] ghindo, as close to an IRC party as you can get. [06:00] spm: it was in the queue just a few minutes ago [06:00] looking.... [06:00] kfogel, fwiw, I'm getting a lot of questions about whether or not soyuz & codehosting are included [06:01] g'day spm! got my feature working. :) [06:01] it wasn't very explicit was it? [06:01] * beuno blogged about it [06:02] jml: yeah, everybody was told that soyuz and codehosting weren't going to be included [06:02] LaserJock, right. [06:02] LaserJock: I'm about to send an announcement explaining that we chnaged our minds [06:02] they have been opened [06:03] I'm sure that will give Canonical some extra PR brownie points ;-) [06:03] soyuz is included? [06:03] jacob: yes [06:03] especially since for a lot of us soyuz is particularly interesting [06:03] kfogel, cool :) [06:03] mwhudson: wow, wasn't expecting that. cool. :) [06:03] LaserJock: let's hope it also gets us some patches, eh? :-) [06:04] extra present for me when going to DebConf, one fewer question to answer :-) [06:04] kfogel: possibly [06:04] cjwatson: exactly [06:04] cjwatson, :) [06:04] cjwatson, :D [06:04] I'll have to go through my LP bugs and see if there are any I dare trying to fix myself [06:05] spm: can you see why it didn't land? [06:05] thumper: running precommit hook: /bin/false [06:05] which tends to um fail [06:05] cjwatson: If you see Phil Kern there, the fact that Soyuz got opened my affect his gsoc project .... [06:05] from what I've heard and what little I've seen, I still don't think contributing is going to be a trivial exercise [06:05] spm: d'uh [06:05] spm: I need to land a testfix branch to get buildbot to rebuild [06:05] ScottK: what's his gsoc project? [06:05] thumper: hang on. there's a few in here. gimme a bit.... [06:06] spm: we usually do that during release-critical to prevent landings to devel and push all landings to db-devel to avoid the amount of time it takes to get through to db-stable [06:06] cjwatson: Rewriting the Debian build system. [06:06] sinzui, can you help me out with this? I've got some zcml that uses a path_expression that needs to be updated. The new version needs access to an element higher up in the path... can I do that? [06:06] mthaddon: yeah... bac was falling afould of it earlier. francis did know, so not sure what's up [06:07] thumper: can you resubmit flacoste's branch to db-devel? [06:08] ScottK: mm. I think Debian's pretty attached to its own system, I'd be surprised to see them switch to ... [06:08] ... LP1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 ... [06:08] mthaddon: I could manually merge devel into db-devel [06:08] ... 111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 ... [06:08] ... 11111111111111111111111111111111111111111111111111111111111111111111111111 [06:08] doh. sorry! [06:08] switch to LP, is what I meant to say [06:08] channel spam!! :) [06:08] cjwatson: Agreed, but it's always interesting to see how someone else solved the problem. [06:08] someone's overly excited :) [06:08] cjwatson: I expect there may be lesson's learned he can now take advantage of. [06:09] thumper: that'll happen anyway as a matter of course (devel -> stable -> db-devel) [06:09] ajmitch: stuck key repeat due to load :-( [06:09] mthaddon: well it won't [06:09] * ScottK goes to bed .... [06:09] * mwhudson opens a beer [06:09] mthaddon: as devel isn't building [06:09] mthaddon: which is what I'm trying to fix [06:10] thumper: oh, I see [06:10] "Launchpad is a Ruby on Rails blogging platform and CMS. It has categories and tags, RSS, a plugin architecture and theme architecture, and more. It's inspired by Wordpress." -- wtf [06:10] the name Launchpad has been taken on ohloh :( [06:10] LOL [06:10] hah [06:10] mthaddon: so... what do we do here? [06:10] thumper: I can temp open it up for you to kick it [06:10] but the url https://www.ohloh.net/p/launchpad is free. [06:10] mthaddon: that be good [06:11] thumper: you ready to do that now? [06:11] jml, trigger the import [06:11] they'll have fun :) [06:11] beuno, I have to come up with a name first. [06:11] wonder if they have an up-to-date version of bzr that can handle 2a [06:11] mthaddon: gimmie a minute [06:11] k [06:12] beuno, interesting question! [06:12] problems with branches already? I'm getting 404 on http://bazaar.launchpad.net/~launchpad-pqm/launchpad/devel/utilities/rocketfuel-setup [06:13] not a standard launchpad page saying that the page isn't found, just a boring "Not found" page [06:13] I think that the link is wrong [06:13] no rev on it [06:13] no? [06:14] hmm. ohloh gives 'https://code.launchpad.net/myproject/trunk' as an example branch url :\ [06:14] mwhudson, shouldn't there be a /files/? [06:14] https://dev.launchpad.net/Getting will need updated with a working link then [06:14] See you all back in a few hours. :) [06:14] ajmitch: you can edit the wiki apge :) [06:14] * deryck sleeps [06:14] mthaddon: ready now [06:14] lifeless: I need a working link first :) [06:14] getting that [06:14] ajmitch: bzr cat of that url really should work [06:14] ajmitch: that's not a http page - that's a bzr cat to the URL [06:14] thumper: go for it [06:15] http://bazaar.launchpad.net/%7Elaunchpad-pqm/launchpad/devel/annotate/head:/utilities/rocketfuel-setup [06:15] beuno, yeah, looks like it doesn't support 2a yet. [06:15] http://bazaar.launchpad.net/~launchpad-pqm/launchpad/devel/annotate/head:/utilities/rocketfuel-setup [06:15] ajmitch: ^ [06:15] lifeless: no [06:15] mwhudson: no? [06:15] no! don't send folks via loggerhead! [06:15] spm: ok, it was bzr cat that was failing for me, honest :) [06:15] mthaddon: ok, it's in the queue [06:15] lifeless: the instructions are for bzr cat [06:15] now, where did I put that diagram [06:15] mwhudson: ah [06:15] thumper: k, will wait til it's through and then turn it back off [06:16] I'm sure lh will survive:) [06:16] why it's not working for ajmitch i don't know [06:16] bzr: ERROR: Invalid http response for http://bazaar.launchpad.net/%7Elaunchpad-pqm/launchpad/devel/utilities/rocketfuel-setup/: Unable to handle http code 404: Not Found [06:16] is the real error I got [06:16] yeah, lh is holding up reasonably well [06:16] ajmitch: get rid of the trailing / [06:16] ajmitch: what version of bzr are you using? [06:16] 1.16.1 [06:16] ajmitch: although bzr may be adding that [06:16] spm, common! LH hasn't failed in what, 2 days? 3? [06:16] ajmitch: intercepting proxy? [06:16] bzr is [06:16] beuno: edge update/restart about 6 hours ago. I win. QED etc :-P [06:17] I'll unset http_proxy & see if it helps [06:17] ajmitch: I get the same [06:17] and I have no proxy [06:17] no difference [06:17] spm, that's just out of habit, I'm sure [06:17] * mwhudson tries with 1.16.1 [06:18] ajmitch: do you have pycurl installed? [06:18] LaserJock: I'm glad it's not just me going mad [06:18] mthaddon: I have to go and start cooking now, will check back in about 10 minutes [06:18] * mwhudson clutches at straws [06:18] ajmitch: well .... it could be both of us ;-) [06:18] so what's the actual command you're running that's failing? [06:18] mwhudson: I do [06:18] wgrant: ping [06:18] I think everyone has been underestimating the load we can handle :P [06:19] mthaddon: bzr cat http://bazaar.launchpad.net/~launchpad-pqm/launchpad/devel/utilities/rocketfuel-setup > rocketfuel-setup [06:19] mthaddon: bzr cat http://bazaar.launchpad.net/~launchpad-pqm/launchpad/devel/utilities/rocketfuel-setup > rocketfuel-setup [06:19] lifeless: yes -- habitual paranoia, but so far no problems [06:20] ajmitch: it works for me [06:20] me too... [06:20] takes its own sweet time (but I'll be spanked if I tell you better url's to use :P) [06:20] bzr 1.16.1 seems to be taking aaages [06:20] ajmitch: is it possible you have an intercepting proxy doing the wrong thing [06:20] it's not instantly failing if I try on another box at work, running jaunty [06:21] lifeless: http://bazaar.launchpad.net/~launchpad-pqm/launchpad/devel/annotate/head:/utilities/rocketfuel-setup > rocketfuel-setup isn't going to get you a very useful shell script [06:21] mwhudson: lp: [06:21] it's possible, but I'm not aware of any proxy in the way [06:21] it's failing rather quickly for me [06:21] ajmitch: do this please [06:21] on Jaunty with bzr 1.16.1 [06:21] LaserJock: you too [06:21] wget -S http://bazaar.launchpad.net/~launchpad-pqm/launchpad/devel/.bzr/repository/pack-names [06:21] ok, I get a failure on jaunty on a computer with a different ISP about branch version at least [06:22] lifeless: that works fine [06:22] 1.16 worked for me [06:22] lifeless: that grabbed a file [06:22] (and i do have an intercepting proxy, thanks voda!) [06:22] LaserJock: ajmitch: I want to see the headers you got [06:23] one second [06:23] http://paste.ubuntu.com/223265/ [06:24] ajmitch: thats from the failing machine? [06:24] yes [06:24] lifeless: you mean this: http://paste.ubuntu.com/223266/ ? [06:24] ok, you seem to be getting the right root node then [06:25] try running the command again with -Derror -Dtransport [06:25] lifeless: the wget or bzr command? [06:25] bzr [06:25] ajmitch, LaserJock: the script we're trying to get is here btw: http://pastebin.ubuntu.com/223269/ [06:25] I'm seeing bzr cat fail too [06:25] (in case you want to get the pull running in the background) [06:25] Congratulations Launchpad team. It's great to see this happen. :) [06:26] wget -S: http://paste.ubuntu.com/223270/ [06:26] http://paste.ubuntu.com/223271/ for the updated bzr cat output [06:26] -Derror -Dtransport shows a traceback: http://paste.ubuntu.com/223272/ [06:26] mwhudson: thanks [06:27] lifeless: http://paste.ubuntu.com/223273/ [06:27] I'm on Karmic with bzr 1.17 [06:27] err, or rather bzr 1.17~rc1-1 [06:27] ah [06:27] its bzr-svn FAIL [06:27] er ah eh [06:27] bzr-svn!! [06:27] run with --no-plugins [06:27] yeah [06:28] aha [06:28] mw care to get the karma for filing the bug? :) [06:28] i guess i'll edit the wiki page [06:28] ok, I'm getting some progress fetching stuff now :) [06:29] --no-plugins works [06:29] mwhudson: care to get the karma for filing the bug? :) [06:29] lifeless: looking now [06:29] lifeless: --no-plugins works for me as well [06:30] thumper: success [06:30] its open_containing failing is all [06:32] rs=me :( [06:32] jml: what? [06:32] oh [06:32] someone just landed a revision with that tag. [06:32] hi, is there any ppa where i can get a bazaar that allows me to download the LP codE? [06:32] lifeless: thanks for tracking that down [06:32] nxvl: ~bzr ppa? [06:32] ajmitch: sorry for not testing harder [06:33] mwhudson: https://edge.launchpad.net/~bzr/+archive/ppa that one? [06:33] nxvl: yeah [06:33] * mwhudson notices that 1.17 arrived there 17 minutes ago... [06:34] what's User sabdfl for in Host bazaar.launchpad.dev for the ssh config? [06:34] Heh. [06:35] convenience of manual testing of the ssh server; 'sabdfl' is one of the users in the sample data. [06:35] ah, i see [06:36] but all this /etc/hosts and ssh config munging is local, right? [06:36] jml: I just copied what flacoste_afk did :) [06:36] mthaddon: thanks, I'm EOD now [06:36] thumper: me too :) [06:37] LaserJock, yeah, that's local. [06:37] spiv, and that's obsoleted convenience -- ./utilities/make-lp-user is much nicer. [06:37] * spiv nods [06:38] mwhudson: want to copy the using lp codehosting to dev.launchpad.net? [06:38] lifeless: it's a bit odd, open_containing_tree_or_branch doesn't fail with my local apache [06:39] but surely does against bazaar.launchpad.net [06:39] thumper: ok, filing a bzr-svn bug first [06:39] mwhudson: ok [06:39] * thumper is really away now [06:39] mwhudson: fun [06:40] lifeless: https://bugs.edge.launchpad.net/bzr-svn/+bug/402063 anyway [06:40] Malone bug 402063 in bzr-svn "bzr-svn breaks open_containing_branch_or_tree against bazaar.launchpad.net" [Undecided,New] [06:41] I hadn't seen that doc before [06:42] jml: you mean the doc i wrote this morning? [06:42] mwhudson, yeah. :) [06:42] mwhudson, I saw the timestamp, but I'd never knew that such a doc was brewing [06:43] jml: thumper kept bugging me about it in standups [06:44] * jml is obviously a terrible listener. [06:44] I am here ... [06:44] ! [06:44] :) [06:45] jml: now at https://dev.launchpad.net/Code/HowToUseCodehostingLocally, any idea where to link to it from? [06:45] spm: Can you please clear the translation import queue on staging for me? If jtv does not object ... [06:45] * jtv does not object [06:45] spm: "delete from translationimportqueueentry" [06:45] henninge: sure. one sec. [06:46] mwhudson, probably from https://dev.launchpad.net/Code for now [06:46] spm: may take a while to replicate... [06:46] we're talking what, 150K entries? [06:46] mwhudson, I just added the system diagram there. [06:47] henninge: jtv: DELETE 129367 commit> [06:47] ? [06:47] spiv: please do. [06:47] spm: ^ [06:47] fail :-) [06:47] henninge: commited [06:47] * jtv plugs ears [06:47] * jtv squats under table [06:47] %§Ω#* auto-completion [06:48] jtv: all is well [06:48] jml: done [06:48] spm: thank you [06:48] henninge: and the one place where you really want auto-completion nobody ever wants to implement it. I keep asking for tab password completion, but do they listen? [06:48] mwhudson, woot. [06:49] jtv: Sounds like the next Windows feature for me! [06:49] henninge: well Windows doesn't have that problem. "Hi, you typed the wrong password 3 times. Would you like to create a new one?" [06:52] lifeless: had a bit of weirdness with my bzr branches if I build my TransformPreview but do not go on to commit anything... when I unlock the branch it says the lock was broken. [06:52] lifeless: as if the info file wasn't found for some extraneous reason (never moved into place unless I commit maybe?) [06:52] spm: can you please run the poimport script on staging for me. It is running from cron every 20 min but for speed I'd appreciate a manual run. [06:52] probably there is a transform thing you need to call to tell it you're not recording the result [06:53] lifeless: any idea who might know? [06:53] spm: "cronscripts/rosetta-poimport.py" just that [06:53] jtv: abentley; but - show me the backtrace [06:53] spm: btw is it really 20 minutes? That's fine, but istrm we had it set up for much fewer runs initially. [06:53] jtv: yah. every 20 [06:54] spm: that's great, thanks. [06:54] makes QA a lot easier. [06:54] henninge: running [06:54] lifeless: hang on, I'll reproduce. [06:54] henninge: done. I should have waited. [06:54] The problem! The problem! I'll reproduce _the problem_! [06:56] spm: thanks. why? [06:56] henninge: normally poimport runs for ages. finished in < 30 secs [06:56] spm: that's because we cleared the queue first! [06:57] spm: ages? Shouldn't—after 8 minutes (iirc) it stops taking on jobs and leaves the rest of the queue to the next script run. [06:58] so is the news going to hit slashdot, i wonder... [06:58] * maco hugs all [06:58] jtv: I may be talking based on old info - but I can find out if desirable? [06:58] mwhudson: and digg and LWN and ... [06:58] LaserJock: yeah [06:59] but i think the slashdot effect is still real, all these years on [06:59] spm: hmm... would be nice to know—yes please. [07:02] lifeless: https://pastebin.canonical.com/20127/ [07:03] jtv: pastebin.ubuntu.com these days, surely :) [07:03] jtv: brand new world [07:03] lifeless: oh, I wasn't aware of that one. I just type "paste" and let my browser figure out the rest. [07:04] sorry, paste.ubuntu.com [07:04] well, open :) [07:04] so, something is holding a lock on the branch [07:04] lifeless: I am. [07:05] oh actually [07:05] A write lock. And when I unlock it, boom. [07:05] that message says 'the lock was broken while open' [07:05] so perhaps something deleted the branch? [07:05] its in your teardown [07:05] is it possible you're causing 'rm -rf', 'branch.unlock()' to happen, in that order? [07:06] spm: I'd like to put staging into read-only mode for a few minutes. [07:06] henninge: ? why not just use the slave? [07:06] spiv: This is a patch that does that, can you apply it please and restart staging? https://pastebin.canonical.com/20128/ [07:06] already on it's way to digg: http://digg.com/linux_unix/Canonical_releases_complete_source_code_for_Launchpad [07:07] spm: No, the fix I am QA'ing checks that configuration directive. [07:07] spm: The bug is specifically about read-only mode. [07:08] bug 380852 [07:08] lifeless: it's the first teardown I do, afaik. I could try doing the unlock explicitly at an earlier stage; hang on. [07:08] bug #380852 [07:08] Malone bug 380852 in rosetta "translation page shows editable fields while in read only mode" [Medium,Fix committed] https://launchpad.net/bugs/380852 [07:08] ah! [07:09] spm: is the restart still such a pain with buildout? [07:10] lifeless: you're right, it happens in the teardown but not if I unlock before then. [07:10] lifeless: maybe useBzrBranches puts its cleanup in the wrong place? [07:11] henninge: patched and restarted [07:12] spm: great! [07:12] jtv: use addCleanup rather than tearDown [07:13] lifeless: already made that change in an as-yet unlanded branch, though it wasn't enough to fix the problem. [07:16] spm: thank you very much! Everything works as expected. [07:16] henninge: revert? [07:16] spm: Can you just revert or would you like a patch? [07:16] revert is fine [07:16] spm: ;-) [07:16] spm: please do [07:17] * henninge is blazing through the test plan here ... [07:17] henninge: is back [07:17] spm: cool, that was a great help! [07:18] just in the throws of setting up a local copy of LP here... \o/ [07:19] VK7HSE: I hope that's "throes," not "exception throws" :) [07:19] yeah something like that !!! ;) [07:20] <\sh> what did I read? LP is opensourced? Congrats to everybody involved...good job :) [07:21] I guess the word is out then [07:21] https://dev.launchpad.net/Getting === beuno changed the topic of #launchpad-dev to: Launchpad Development Channel | https://dev.launchpad.net/ | Please use #launchpad for support. | https://launchpad.net/~launchpad-dev | Get it: http://identi.ca/notice/6403823 === beuno changed the topic of #launchpad-dev to: Launchpad Development Channel | https://dev.launchpad.net/ | Please use #launchpad for support. | https://launchpad.net/~launchpad-dev | Get it: https://dev.launchpad.net/Getting [07:23] COPYNPASTE FAIL [07:23] should go to bed [07:25] failures in buildbot :( === al-maisan_ is now known as al-maisan [07:26] all seemingly to do with uncollectable garbage :( [07:26] mwhudson, actually no. [07:27] no? [07:27] mwhudson, one of them is to do with the move of bzr to buildout. [07:27] jml: oh "yay" ? [07:27] mwhudson, it ought to be fixed when my branch is finished running through ec2test. [07:27] jml: oh ok [07:28] except that's going to db-devel [07:28] hmm. [07:30] jtv: any clue how to best QA bug #388473 ? [07:30] Malone bug 388473 in rosetta "POTMsgSets on page not in ascending sequence order" [High,Fix committed] https://launchpad.net/bugs/388473 === lifeless changed the topic of #launchpad-dev to: Launchpad Development Channel | https://dev.launchpad.net/ | Please use #launchpad for support. | https://launchpad.net/~launchpad-dev | Get it: https://dev.launchpad.net/Getting | http://paste.ubuntu.com/ [07:31] henninge: in a case like this I'd just make sure that the page works properly, and let the oops reports show that the bug is gone. [07:31] jtv: cool, can do that ;) [07:32] mwhudson, and hey, that just finished with ec2test. I'll retarget the branch for devel & land appropriately [07:32] mwhudson, jml: yeah, somewhere we need to explain our four trunks thing, in the dev wiki... [07:32] congrats everybody! [07:32] thekorn: hey there! Thank you. Enjoy :-). [07:32] kfogel, yes. but not today :) [07:32] jml: indeed [07:36] how can I remove all changes introduced by rocketfuel-setup? [07:37] I didn't expect it to mess up my Apache+PHP setup [07:39] emilis_info, good question! [07:39] wow, and soyus and codehosting too, awesome [07:40] you must have a good PR manager, nice bluff ;) [07:40] emilis_info, there's isn't an automatic way of doing it, probably best to look at the script itself. [07:40] heh [07:40] thekorn: that's what I was thinking ;-) [07:41] I think you should recommend installing it on a separate / virtual machine, because it touches so many places it can break something for a lot of people [07:44] Morning [07:44] hi noodles775! [07:45] Hiya jtv, gee you've been around early and late lately? QA to finish? [07:45] jtv: even better, there is a link in the sequence bug that fails on production but succeeds on edge - QAed! [07:46] noodles775: yeah, QA's been a real bottleneck for the past few months. Finally got a full end-to-end QA on production though, so pretty cool. [07:46] noodles775: I've worked about 2.5 working days so far this week. :) [07:46] Sheesh [07:46] henninge: perfect. [07:47] henninge: that reminds me... I do _not_ plan to do much more today. I finished at 5 AM and got back to work at 11 AM, and the most crucial thing is done. [07:47] jtv: you are tough! [07:47] henninge: would you mind having a very quick standup soon and then doing without me? [07:47] henninge: I'm stupid. [07:47] jtv: what is the crucial thing? [07:47] jtv: I don't mind at all! [07:48] henninge: there was yet another bzr hiccup. Tickling yet another part of bzrlib that Launchpad hadn't touched yet. [07:49] oh, bzrlib [07:51] yes, when you reference a directory you need to look up or create an id for it. And first you need to do the same for its parents. [07:52] But the lookup doesn't work for directories you just created, only for ones that were already in the committed branch. [07:52] spm: how much longer will you be around? [07:52] So if you had n files in the same directory d, where d was not in the committed branch, I'd create d n times. [07:52] jtv: meep! [07:52] henninge: -22 minutes. I just haven't had a chance to sign off yet. :-) [07:53] lifeless: hey, this stuff is obvious to you, Here Be Dragons for us. :) [07:53] spm: ok, enjoy the evening, then. Thanks for your help! [07:53] jtv: well, you're using tree transforms, which I'm not actually familiar with [07:53] lifeless: see? Going Where No Man Has Gone Before. [07:54] spm: the script timings look good to me. === henninge is now known as henninge-brb [07:55] spm: that's good to know. === henninge-brb is now known as henninge [08:04] mwhudson: do you know if there's a general problem with imports at the moment, or have I just mis-reviewed some, like this one: [08:04] https://code.edge.launchpad.net/~vcs-imports/cairo-dock/trunk [08:04] There seem to be a lot of failures when I was CHR yesterday. [08:11] spm, PQM is now rejecting my branches with "Executing pre-commit hook /bin/false at Tue Jul 21 08:06:41 2009" [08:12] jml: I had that yesterday, but it was because I was trying to submit to devel rather than db-devel? [08:13] yeah. I'm deliberately submitting to devel. [08:16] congrats on the announcement!! [08:18] emmajane, thanks :) [08:19] \o/ well done [08:20] jtv: is that ok? kyleN: the only thing is that it'll probably have to wait until tomorrow, I've got a meeting in 30 min. [08:20] ok. [08:20] jml: I'll be sure to gloat on your behalf at oscon :) [08:21] oops, where did that come from ... ??? [08:21] :D [08:21] henninge: paste from old backlog? [08:21] jtv: probably. [08:21] jtv: what I meant to paste: https://answers.edge.launchpad.net/rosetta/+faq/619 [08:22] noodles775: that's just cscvs being lame abount reconnecting i think [08:23] * mwhudson AFK REALLY [08:23] OK, thanks mwhudson [08:23] mwhudson: bye [08:24] henninge: looks about right, apart from the tone maybe... what about it? [08:24] jtv: ok, i'll change it ... ;-) [08:25] jtv: you are talking about the en_GB remark, tright? [08:25] henninge: oh, you just wrote this? Great. Few suggestions... [08:25] That's the main one, yes. [08:25] congrats! [08:25] Also, don't say "iso," but "ISO 639." Also note that we prefer two-letter codes over three-letter codes if both are available. [08:25] a_c_m: thanks [08:26] henninge: also, there are two "levels" of exceptions to the country-code rule. I think it would be useful to distinguish them. [08:26] hopefully means it will get some love from the wider community and perhaps spawn something that can give github a run for its money :) [08:27] henninge: there are the exceptional _languages_ that really are meant to have separate codes: pt_BR, zh_CN, zh_TW. [08:27] henninge: then there are the exceptional _cases_ where adding a more specific country code may be justified, e.g. es_MX. [08:28] henninge: and finally there is English. English is assumed to be US English, and it's assumed to be the source language, so en_GB is justified as an exceptional "case" as above. [08:29] jtv: but en_GB is active whereas es_MX isn't. [08:30] henninge: yes, that's another level of distinction: some specialized languages are more desirable than others. [08:30] jtv: I think for the FAQ I'll just mention the first case and refer to the second as "Other special cases may be allowed on a pre-project basis." [08:31] henninge: as long as it's clear that the "default" country should definitely not be used, and there are a few exceptions where the country is always included. [08:31] henninge: I also wouldn't say there's no list of languages when you also point to the page where people can query the list. Just complement it with a pointer to the ISO-639 lists. [08:32] ok [08:32] henninge: finally, you may want to note that we don't support collective language codes. [08:32] henninge: e.g. "Indo-European (other)." No point in translating to that! [08:32] jtv: good point! [08:33] henninge: you did open a can of worms here: don't forget variants. :-) "Minimally supported for imports, but not in the UI." [08:34] henninge: and after that, we can have that quick call and I can bugger off. :) [08:34] jtv: right ... ;-) [08:35] henninge: just let me know when you're done. [08:38] g'night all. [08:38] happy hacking [08:39] 'night jml [08:44] jtv: I am done: https://answers.edge.launchpad.net/rosetta/+faq/619 [08:45] henninge: don't say "not the same as the language code"—that suggests that nb_NO and sv_SE are OK (and that ar_AR is not okay for the wrong reason) [08:46] jtv: ok, I also forgot the language group rule [08:46] henninge: say something like "the language's home country." [08:47] henninge: typo: excpetions [08:48] henninge: also, exceptions for "de_AT" are a very different matter from exceptions for "de_DE." I'd still keep the two separate: _never_ use de_DE, but _avoid_ de_AT. [08:48] * ddaa notices that the dev.launchpad.net wiki still points to SQLObject in the "External documentation" section [08:48] hasn't the code base switched to STORM a while ago? [08:48] ddaa: yes it did [08:49] ddaa: this is something for the Foundations folks (flacoste and team) [08:49] stub, you maybe? [08:49] ddaa: parts of the code still use SQLBase via Storms SQLBase support (as a transition) [08:49] jtv: I'll not mention those, I think that is covered by "please try without them first." [08:50] jtv: just drop the link already, before people start being confused and saying "Hey launchpad uses SQLObject, not STORM" [08:50] henninge: which ones? [08:50] jtv: _avoid_ using ... [08:50] ddaa: sorry, my head is exploding in slow-motion, not opening another tab right now! [08:51] henninge: if you insist... Got more: [08:53] The first bullet point needs punctuation. In fact I'd hoist it out of the bullet-points list and replace the leading line with "Launchpad uses ISO-639 language codes: two-letter ISO-639-1 codes where available (link) and three-letter ISO-639-3 codes for other languages (link). [08:54] henninge: if you don't like that, then insert a comma before the "or" and remove the full stop after "nds." [08:56] henninge: In the second bullet point, I'd say "do not include the country code except in a few exceptional cases." "Append a specific country code" doesn't sound like the sort of thing a non-programmer would suspect themselves of doing. [08:57] henninge: btw you can resolve the dichotomy between de_DE and de_AT by replacing "NOT it_IT, es_ES but also not fr_BE" with: "NOT it_IT, es_ES, but also avoid fr_BE" [08:58] hi guys, cheers on open sourcing Launchpad! great thing! [08:59] when i try to install it on Jaunty (32bit) i get the following error: bzr: ERROR: Unknown repository format: 'Bazaar repository format 2a (needs bzr 1.16 or later)\n' [08:59] any clues on what that can be? [08:59] beeman_nl: that you need to have bzr version 1.16 or later? [09:00] beeman_nl: https://dev.launchpad.net/Getting (you need a newer version of bzr) [09:00] ddaa: thanks, i fixed the link to point to storm instead [09:00] henninge: next, are you sure it's safe to say "because they are really considered separate languages"? We don't want to be showered in email about how different people's local dialects are different. In this case I'd only say that the country codes are needed for these languages, and not go into the why. [09:02] henninge: (I also wouldn't want to go into how Cantonese and Mandarin are completely different languages yet they get the exact same language codes, whereas Mandarin (Simplified) and Mandarin (Traditional) are the same language but get different codes :-) [09:02] LaserJock, jmarsden: excuse me, i did'nt read that well (it's early up here) [09:02] henninge: ahhh, the page is magically updating as I type this. :-) [09:03] * henninge edits and save like a madman [09:04] jtv: done it again [09:04] * jtv chuckles [09:04] jtv: I think we should leave it at that now. As you said, it is a can of worms. [09:05] henninge: sure. It's good to have this documented, so thanks for taking the trouble. [09:06] henninge: does that mean we can have the call? [09:06] jtv: call! [09:18] Hi guys [09:18] anyway to just get the latest revison from the branch instead of the entire history [09:18] ?? [09:19] use a checkout instead of a branch [09:19] so do i edit the rocketfuel scripts? [09:20] yeah, there's a bzr branch somewhere, just make it a bzr checkout [09:20] ok cool [09:20] wanted to check it out, but didn't want to branch the whole thing [09:20] at the moment it's just a quick test [09:20] it's only about 150M to branch [09:21] bzr currently isn't very optimised for getting just the current revision, unfortunately. [09:22] It shouldn't be too hard to fix, but it just hasn't bubbled to the top of the todo list yet. [09:22] that's cool [09:23] bzr: ERROR: Unknown repository format: 'Bazaar repository format 2a (needs bzr 1.16 or later)\n' [09:24] hmmm [09:24] I'm on 9.04 [09:24] i should be ok right? [09:24] cbx333: bzr --version ? [09:25] 1.16 is only about a month old; you need bzr from the bzr PPA I think. [09:25] ahh doh [09:25] ko [09:25] 1.16.1 has some essential fixes in it IIRC [09:26] Yeah, I meant 1.16.1. 1.17 is even better! [09:26] hehe [09:26] (and is already available in the ~bzr PPA) [09:28] cool [09:28] * cbx333 grabs [09:29] :) [09:34] * gmb rubs his eyes... [09:34] Why are all these people looking in at me? [09:35] Aah, that's it. We must be open source now. Bon. [09:35] Anyone tried installing launchpad on Debian? Anyone know if it will definitely not work? [09:37] * cbx333 pokes gmb [09:37] wakey wakey [09:38] an7: is there a debian package yet ? [09:39] proppy, no idea - was looking at doing it from bzr [09:40] an7: maybe there is a ppa for you to play with === proppy1 is now known as proppy [09:48] * silbs checks out the new digs [09:48] an7: AFAIK, no-one's ever tried installing it on debian. [09:48] an7: May eat your cat, etc... but give it a shot and let us know :) [09:48] well done, guys [09:49] silbs! [09:49] bigjools! [09:49] silbs, are you thinking of swapping the day job? You're in a dev channel now :) [09:50] <\sh> guys...I understand that all correctly, it comes even with the build infrastructure parts of LP (aka soyuz?) [09:50] bigjools: I hear this is the place to come if I want to contribute code to LP ;) [09:50] silbs: patches welcome! ;) [09:50] \sh: Yes. [09:50] \sh: yes, soyuz is open [09:51] \sh: if you have any questions about soyuz, I'm here to help. [09:52] <\sh> bigjools: I'll install it tomorrow on my buildserver in our DC...so I'll come and ask ;) [09:55] <\sh> bigjools: btw..which possibilities for chroots do you have? lvm+xen+sbuild or also other ways? [09:56] \sh: we only handle sbuild right now (well, the custom version of it we have in the tree) [10:04] beuno, the Launchpad front page is missing the sentence "All of it." in 48-pt font [10:06] beuno: and [10:06] is it already opensource ? i thought it was postponed. [10:07] * stub hands silbs his hat and goes out for a beer [10:07] tacone: It's open. It was never postponed, we just wanted to make sure that everything was working, so we didn't want to pin it on a specific date and then let everyone down. [10:07] k [10:08] congratulations. [10:08] thanks for the info, bye [10:12] * jtv afks [10:17] hi. so is there somewhere/somehow i can just get the code in stead of having to download and install the whole of lp? [10:18] walterl: I'm not sure I understand the question. If you bzr branch lp:launchpad you're just getting the code. If you're looking for a tarball, no, there isn't one available. === jtv is now known as jtv-afk [10:21] * silbs gives back stub 's hat, and takes his beer instead [10:22] * stub stabs silbs in the face [10:23] wow, talk about escalation! [10:23] Mine! Mine! [10:23] * bigjools grabs popcorn and watches [10:23] * silbs passes out beers to everyone ;) === noodles775 is now known as noodles775-afk [10:24] gmb: thanks :) [10:34] * cbx333 passes out victory slushies [10:49] mwhudson: hey [10:49] https://code.edge.launchpad.net/ [10:49] mouse over mplayer there [10:49] read the comment (128 days old) [10:49] then click on it [10:49] and look at the trunk [10:50] lifeless: that looks wrong [10:50] yes, it does [10:59] Morning, all. [11:01] mornign deryck [11:01] and morning, too [11:04] <\sh> bigjools: another question regarding soyuz and the underlaying infrastructure: you configure/install as well a full fletched archive infrastructure with it, or is it still self implementable (thinking about wanna-build infra etc.) [11:05] \sh: I don't understand your question [11:08] <\sh> bigjools: do you ship an automatical archive management with the rocketfull setup? or do the admin installs/configure the archive stuff him/herself? [11:09] \sh: it needs to be configured [11:11] * \sh 's testinstalling now @home...let's see where it will take me ;) [11:14] \sh: rf-setup gets the source for development ;P - deployment is a _whole nother problem_ [11:17] <\sh> lifeless: I just wondered if you bring all the stuff with the source tree, and it just needs to be configured to sane defaults for your own infra...(and/or setting up cron jobs etc.) and regarding the explanation on d.l.n/Getting : it starts to deploy at least a minimum ;) [11:18] <\sh> looks like that bazaar.launchpad.net is under heavy load ;) === henninge_ is now known as henninge [11:44] hi [11:45] hello [11:45] the first command from launchpad get the source howto stalls: [11:45] bzr --no-plugins cat http://bazaar.launchpad.net/~launchpad-pqm/launchpad/devel/utilities/rocketfuel-setup > rocketfuel-setup [11:45] do you know what is wrong ? [11:46] I think the servers/links are a little loaded, someone is looking at it to see if it can be improved [11:56] bigjools: lucille's db connection is being killed about every two hours (long running but idle transaction). [11:57] Last one 4 hours ago, so maybe it is ok now? [12:01] I think the LP server for the LP sources is a bit.... hammered at the moment :-) [12:01] Just a short: T H A N K Y O U ! This is *really* great news. Especially since soyuz and codehosting are included. This is a huge step forwards. I'm just hoping Canonical will make a brazillion with it's services. :-) Keep up the good work! [12:02] stub: that's not pleasant, how long has it been happening? [12:03] markust: great, I hope you can get involved with the code! [12:03] prusnak: it should be better now [12:03] First notification I have was from 14 hours ago. Then 11, 8, 7 and 4 hours ago. [12:03] Killing lucille(11230), 2009-07-21 06:51:54.161602+01:00, 2009-07-21 06:51:54.638500+01:00 [12:07] stub: do you know which query? [12:07] $ bzr --no-plugins cat http://bazaar.launchpad.net/~launchpad-pqm/launchpad/devel/utilities/rocketfuel-setup [12:07] bzr: ERROR: Invalid http response for http://bazaar.launchpad.net/%7Elaunchpad-pqm/launchpad/devel/.bzr/repository/packs/34bf81ec32ec262a521009fd975bfad1.pack: Missing the Content-Range header in a 206 range response [12:07] anyone seen that? [12:07] bigjools: There would have been no active query - that is why it was killed. The script will have been sitting around doing something non-db related for an hour. [12:08] an7: bazaar.lp.net is getting hammered and is hiccuping a bit at the moment. Someone's looking at it to see if matters can be improved [12:08] oh idle txn, ok [12:08] thanks gmb, will try later [12:08] The oops will be 'terminated by administrator' or similar [12:08] an7: IIRC that's generally caused by broken proxies [12:09] an7: especially dodgy intercepting proxies. If vila is around on #bzr he might be able to help debug it (assuming it isn't transient) [12:09] stub: can you see if there's an oops number available please? [12:09] bigjools: There is no oops at this end. === noodles775-afk is now known as noodles775 [12:10] spiv, I know there is a dodgy transparent proxy where i work - maybe that is causing it [12:10] bigjools: I don't even know what script was at the other end - just that it was connected as the lucille user :) [12:10] stub: bleh. I can't do a lot without an oops :( [12:10] stub: it's the publisher [12:11] bigjools: Yup. But now when you see the oops, you will know the trigger. [12:11] an7: That could be it.. the typical dodgy proxy error is to do with ranges and multipart separators [12:11] an7: ah, yep, in that case I'd say that's very likely to be the cause. [12:11] bigjools: We can increase that timeout if necessary, but it is better to turn the long running transaction into shorter transactions. [12:11] an7: I wouldn't try using bzr+ssh right now though, I think the server is too popular :-) [12:12] bigjools: lucille gets killed occasionally and nobody really notices - just that this time I saw a sequence of them so letting you know in case something is stuck. [12:12] stub: yes, but if it normally runs ok then we have another problem lurking [12:12] ok cheers [12:12] Just a thought - what about a tarball of the launchpad folder which you could spread across more redundant hosting as a "starter for 10" ? [12:13] You could even ship it with the trees removed and get a script to check them out [12:13] awilkins: no need; we're serving over http anyway [12:14] awilkins: we can do multiple backend stuff there; the problem was /not/ pure load [12:14] lifeless: Fairy snuff [12:14] what happened is that a cache with a bad failure mode (and heck, you can get the source and read it :P) failed [12:14] and once failed it won't recover. [12:14] we've bandaided it right now. [12:15] I'm excited, been waiting for this for a year :-) [12:15] :) [12:24] spiv: we bounced apache [12:24] spiv: its entirely possible it provoked a bunch of oddness [12:37] Hmm... rocketfuel-setup has bombed out... so I attempted to re run... now it's nagging me that it has "make: *** No rule to make target `install'. Stop. ERROR: Unable to install apache config appropriately" :( [12:39] so I didn't get it built! so is this a case of delete the created lp-branches and re run ? [12:41] yep that's done the trick! [12:44] VK7HSE: can you report a bug about that please, we should fix it. [12:46] bigjools: sorry for such a NOOB question, but against what? as in the place to do so, as I don't want to file in the wrong place... [12:46] VK7HSE: file a bug on the launchpad project, thanks! [12:46] ;) will do [12:59] <\sh> guys...should I really add "sabdfl" as username to my .ssh/config as rf-setup tells me to do ? [13:00] \sh it's just for access to the local instance [13:01] the sample data doesn't know about too many people [13:03] <\sh> bigjools: k :) [13:03] * wgrant wanders in. [13:04] So you did it on schedule after all. Very nice. [13:04] did you ever doubt us? :) [13:04] here's Bug #402187 for review... [13:04] Malone bug 402187 in launchpad "rocketfuel-setup can't resume if fails to complete" [Undecided,New] https://launchpad.net/bugs/402187 [13:05] VK7HSE: do you still have the screen scrollback from when it failed? [13:05] it would be useful to see exactly where it failed [13:06] nahh sorry out of the buffer... I wanted all that to post into the bug... but silly me re ran the script before thinking... [13:06] VK7HSE: can you remember if it had finished pulling the main trunk? [13:07] but it is whilst creating, Fetching revisions:Inserting stream [13:07] No I think this is where it bombed out I think my ISP had a hiccup that resulted in the fail connection... [13:07] yeah that's unfortunate [13:08] as lifeless said somewhere earlier, bzr and other DVCSes don't restart half-fetched branches :( [13:09] but so far by deleting the lp-branches directory all seem to be going fine... although the through put is woeful! 5-75 kbs [13:09] ouch [13:10] so it will take a while to grab it all! but hey it will be woth==rth the wait! [13:10] Gahh try, worth! [13:12] The fix would be to make the initial branch an empty one and to pull it [13:13] Or to just make the script delete the existing branch if it found it, because the revisions are already present, the overhead of remaking the branch would be minimal [13:13] * wgrant whines about it not running on Karmic. [13:13] Does it still need Python 2.4? [13:13] yes :( [13:14] * wgrant schroots. [13:14] but the good news is that we're working on that, or more to the point, barry is :) [13:14] Alas, can't post a merge branch without having pulled it first ;-| [13:14] <\sh> wgrant: what? LP is not working on karmic? [13:14] well, i wish! actually flacoste_lunch and gary are working on that. i'm just trying to noodge them along :) [13:15] \sh: It needs Ancient Python. [13:15] And the PPA isn't set up for Karmic. [13:15] barry: ah! I thought you were involved since your previous work on it [13:15] we're not far from python 2.5 or 2.6 though (hopefully the latter) [13:15] I knew there was no reason to be bleeding edge :-) [13:16] bigjools: i have several abandoned branches :) [13:16] we should write this up on the wiki ... [13:16] is anybody dealing with the tree breakage? this is not a very nice time to be in testfix mode [13:16] looks like many codehosting tests are failing [13:16] intellectronica: I didn't even know we were in testfix, was there an lpbot email? [13:17] bigjools: come to think about it, no, i haven't seen an email from the bot. but there's definitely failures. see the waterfall page [13:18] bigjools: you mean like a faq entry on the dev wiki? [13:19] barry: yes! and in the Getting page [13:19] Wow. Looks like just about every codehosting test ever has blow up... [13:20] Oh, no, just the AcceptanceTests. [13:20] barry: ah you added it! [13:21] bigjools: yep! you're editing the Getting page? [13:21] barry: yep. [13:21] * bigjools high-fives barry [13:21] * barry cancels his edit [13:21] :) [13:21] lol [13:22] bigjools: i was going to add a prerequisites section above "getting it" [13:22] barry: that's a good idea [13:27] hey guys, where can I just get the source code ? rather than an install. [13:28] I wonder how many people are attempting to get the new source ??? average now is 2kbs :( (looks like it will be a late/early morning!) [13:28] Mez: bzr get lp:launchpad, I suppose. [13:30] Mez: see https://dev.launchpad.net/Getting [13:30] * bigjools is shocked that wgrant hasn't downloaded it already [13:31] gmb: AFAICS, that only tells you howto do an install, not how to get a copy of the souce. [13:31] Mez: bzr [branch|checkout] lp:launchpad [13:32] bigjools: My Internet connection is unhappy with these dependencies. [13:32] And Codehosting isn't feeding me. [13:32] yeah, I set up LP on top of a JeOS image last week [13:33] the dependencies are a bigger download than LP itself [13:33] I decided I'd stop apt from installing Recommends. [13:33] Mez: In which case, you can just branch lp:launchpad, as wgrant suggested. [13:33] That cut the initial dependency installation from >300MB to 80MB. [13:33] wgrant: remind me how to do that .... [13:34] bigjools: Drop 'APT::Install-Recommends "false"; [13:34] ' in /etc/apt.conf.d/something [13:35] wgrant: you use karmic right? [13:35] bigjools: Yes. [13:35] bigjools: But I'm currently installing in a Jaunty chroot. [13:36] good morning everyone [13:36] As your PPA doesn't like Karmic much. [13:36] gmb: yes. I'd already got my answer from wgrant :D [13:36] my jaunty installation doesn't have that apt config [13:36] yeah, karmic isn't so supported yet [13:36] bigjools: Sorry, /etc/apt/apt.conf.d/something [13:37] wgrant: yeah I grepped that dir and there's no file with that apt config directive [13:38] bigjools: I think he means to add it to the file :D [13:39] bigjools: Oh, by 'drop in' I mean put in, drop drop from. [13:39] Sorry. [13:39] Gah. [13:39] s/drop drop/not drop/ [13:39] * Mez hugs wgrant [13:40] Currently downloading Launchpad at 0KB/s [13:40] ah [13:41] thanks wgrant [13:41] wgrant: do you know if there's a way of retroactively removing stuff only installed by Recommends ? [13:42] bigjools: A good question. I've wondered that myself, but never enough to actually go looking hard. [13:42] It should be possible, as aptitude and apt-get know what was installed explicitly. [13:42] And they can work out what isn't used any more. [13:42] I quite like the idea of shrinking my VM images [13:43] bigjools: One of the best ways I found of shrinking VM disks is to zero the empty space and repack the disk [13:43] bigjools: That works well for VirtualBox, anyway [13:43] I hate to think how slow this will be once it makes /. [13:44] no kidding [13:44] It's encouraging in a way ; it says that many many people are keen to put up their own Launchpad instances [13:45] (myself included) [13:45] awilkins: I will try that, hopefully it will work with images build by vmbuilder [13:45] Or is this obscene slowness because of that fix to make lp:launchpad resolve only to HTTP? [13:45] that should speed it up === flacoste_lunch is now known as flacoste [14:05] hello flacoste, I need an rc for https://code.edge.launchpad.net/~al-maisan/launchpad/apapi-fix/+merge/9089 [14:07] flacoste, I also have a branch I need rc for, when you have time. [14:09] al-maisan: looking [14:10] flacoste: thanks! [14:11] "This will take a while -- maybe a few hours to get everything, depending on your Net connection." Not wrong !!! 3kbs average !!! :P [14:14] * wgrant just broke the 100KB/s barrier for a second or two. [14:19] does anybody knows what is the issue with the devel tests failure? [14:21] flacoste: All I know is it was codehosting tests. I'm just pulling devel now to see what's going on... [14:22] flacoste: i'm on the blame list but it doesn't appear related to my change. i'm looking into it too. [14:23] bac: i'm probably also on the blame list [14:23] bac: There was a buildbot mail? I didn't see it... [14:24] i also didn't get a mail from the bot [14:25] wgrant: maybe if we cross streams it might come down quicker!!! :P [14:26] gmb, intellectronica, maybe it was sent to launchpad-dev@. I didn't get that either and just noticed I'm not subscribed to that list [14:27] salgado: I am, though... I think. [14:27] * gmb checks the lp-dev archive [14:27] Yep, I'm subscribed... [14:27] no, i'm subscribed to that [14:27] salgado: Nope. Didn't go there. [14:28] Oh, wait [14:28] bac and flacoste will get it because they're on the blame list [14:28] gmb: jml sent an email saying that he got it fixed [14:28] i don't understand why the list wasn't notified though [14:28] flacoste: "Bazaar upgrade branch being bounced, test failure in devel" or something else I haven't seen yet? [14:29] gmb: oh, of course [14:29] gmb: that one [14:29] Ah, right. [14:29] flacoste: maybe it's moderated because gary isn't subscribed? [14:29] intellectronica: to the launchpad-dev list? possibly [14:29] intellectronica: is there an automatic forward? [14:29] * gmb wonders why buildbot doesn't have its own email address, like PQM does... [14:30] otherwise, it should still be sent at the old list [14:30] gmb: interesting, the buildbot failure message only went to me, flacoste, and mwhudson [14:30] flacoste: i don't think so [14:30] Slashdotted a couple of minutes ago. [14:30] good morning! [14:30] but maybe it was changed in the bot [14:31] wgrant: oh oh [14:31] wgrant, ! [14:31] gmb: looking back to other failure messages it looks like the list gets one message and the blamers another [14:31] intellectronica: i don't think so [14:31] * wgrant is barely half-way through the branch :( [14:31] maybe we should create a torrent of trunk [14:32] barry, bac, Edwin, salgado: stand up in 2 minutes [14:32] intellectronica, interesting idea, whose key would sign it? [14:33] any losas around? [14:33] mars: why do you need to sign it again? the revisions are already signed by pqm [14:33] mars: losas still hang on the internal channel it seems === thekorn_ is now known as thekorn [14:33] intellectronica, just a knee-jerk reaction to distributing a tarball over bittorrent [14:34] i don't think there is a bzr torrent transport yet [14:35] flacoste: no, i just meant a torrent of a tarball of devel [14:35] I imagine that would be a whole lot faster and easier. [14:35] Just need to publish instructions on seeding the shared repo with it... [14:36] +1 for a torrent of a tarball [14:38] what's a good tracker for something like that? [14:41] It's a cool idea... is there any reason it'd be infeasible? [14:41] sinzui: cry [14:42] intellectronica, if we put the tarball on our own DC servers, we can all grab a copy directly at max speed and start seeding it - get say 6 seeds, instead of just say yourself :) [14:43] Woah. Speed. [14:43] wgrant, good speed, or bad speed? [14:43] Like, 200KB/s. 100x better than normal. [14:43] that would be kfogels' doing :) [14:44] Broken again :( [14:45] EdwinGrubbs:bug 401974 Create a milestone link icon looks broken [14:46] Malone bug 401974 in launchpad-registry "Create a milestone link icon looks broken" [High,Triaged] https://launchpad.net/bugs/401974 [14:50] wgrant, what broke? [14:50] mars: Speed dropped back to single digits. [14:50] It's now fluctuating every couple of minutes between fast and slow. [14:50] ok [14:51] Which is better than the first hour, where it was constantly slow. [14:51] really, interesting [14:51] intellectronica, ^ guess that is why a tarball of trunk sounds like a good idea? [14:52] mars: yup. and that's likely to increase as more people catch the news on /. [14:52] ok, testfix is being processed by PQM [14:53] yay [14:53] db-devel is green [14:53] trying to checkout rocketfuel setup .. gives bzr error invalid http response.. expected a boundary.. on a pack file [14:53] so we should be all reset very soon [14:53] I see that buildbot is still private. [14:54] hazmat, ok, abentley, ping can diagnose that? [14:54] abentley, trying to checkout rocketfuel setup .. gives bzr error invalid http response.. expected a boundary.. on a pack file [14:54] Ah, good, I see you already have people on /. accusing you of only doing this because of Chrome OS. [14:54] intellectronica, mars: If we can find a tracker then all we need to do is create a fresh branch of devel outside our repo, share that and write some instructions about seeding a fresh repo, right? [14:54] hazmat: what's your bzr version BTW? [14:54] 1.17 just upgraded [14:54] s/share that/tar and share that/ [14:55] hazmat: This is usually caused by misbehaving proxy servers. [14:55] gmb: yes, i'm doing that. any ideas for a preferred tracker? something decent and legit, preferably [14:55] intellectronica: No, I can't think of a legitimate one :) [14:55] abentley, quite possible.. i'm on a gov site.. trying from an external machine now.. [14:57] intellectronica, ask the ubuntu guys for a tracker [14:57] intellectronica, they must have something for the release images [14:58] mars: good point! [14:59] \m/ [14:59] :) [14:59] :) [14:59] Congrats to the now full-time Open Source devs :) [15:00] * al-maisan bows :) [15:00] greg-g: thank you [15:01] lmao @ but #393596 [15:01] bug *( [15:01] * Mez just wishes that he had working virtualisation on Karmic :D [15:01] al-maisan: thank you all for your hard work on LP over the years [15:01] Mez: No KVM? [15:01] :) [15:02] wgrant: I dont wanna put KVM on a desktop box. [15:02] (mainly cause I'm too lazy to use it!) [15:02] * Mez tries VMWare [15:02] Mez: Strange... [15:02] KVM is surely less of an effort/intrusion than VMWare. [15:02] It's not like Xen. [15:02] wgrant: my experience has been that KVM has been a bit b0rked. [15:03] (aka it decided that it wanted to set my system clock running at 1000 times faster than it should have. [15:03] Any time I tried to login, it timed out. [15:03] Huh [15:03] WFM [15:03] karmic 64bit ? [15:03] Yes. [15:04] Mez: AMD? [15:04] FATAL: Error inserting kvm_amd (/lib/modules/2.6.31-3-generic/kernel/arch/x86/kvm/kvm-amd.ko): Operation not supported [15:04] Ah. [15:04] Intel here. [15:04] I have a solution for that, one moment [15:06] Unable to complete install: 'Couldn't create storage volume 'Launchpad.img': 'cannot create path '/var/lib/libvirt/images/Launchpad.img': Permission denied'' [15:06] * gmb hand-cranks an ec2 instance [15:06] Damn this thing is slow. [15:06] Mez: I'm sure you can fix that one. [15:07] gmb: Better than production Codehosting. [15:08] hmm, seems I don't need kvm-amd [15:08] wgrant: Codehosting is still up, which given the kicking it's taking is quite impressive :) [15:08] wgrant, we're talking with our sysadmins now - hope to have an alternative download soon [15:09] gmb: Indeed. [15:09] mars: Great. [15:09] Though I've so far downloaded 170MB of the supposedly 150MB branch. [15:09] wgrant: awesomeness :D [15:09] wgrant: do a lightweight checkout [15:09] wgrant, is that the sourcecode size? And does it also have to grab the history? [15:10] Mez: options kvm_amd npt=0 [15:10] mars: It of course has to grab the history, but /Getting says the branch is around 150MB. I presume the bzr HTTP transport is just a little inefficient. [15:13] wgrant: It's ~150MB on disk when all's said and done. bzr also includes all the negotiation. Also, there's some inefficiency there at the moment for some reason. [15:13] gmb: Right. [15:14] wgrant: It's bug 402114, FTR [15:14] Malone bug 402114 in bzr "too many http requests with 2a fetches" [Critical,Triaged] https://launchpad.net/bugs/402114 [15:14] gmb: you were discussing the missing buildbot failure messages. was there any resolution? i see they aren't going to the old or new list. [15:14] gmb: Thanks. [15:15] bac: I don't remember there being any. flacoste might know [15:16] gmb, bac: emails are still sent to launchpad@lists.canonical.com [15:16] flacoste: Ah... So why didn't it turn up? [15:16] flacoste: but they went missing last night? [15:17] bac, gmb: i don't know yet [15:17] * gmb feels better about not knowing [15:18] flacoste, intellectronica, is rf-setup safe to run when you already have the lp-branches and lp-source-deps directory? [15:18] as in, it won' [15:18] 't try and fetch them again? [15:19] mars: it might do a pull [15:19] but that should be fine since everything will be there [15:19] flacoste, cool [15:19] wgrant, 150MB is the Launchpad repo. lp-sourcedeps has an extra half gig of stuff that comes from lp.net, I think [15:19] gmb: Have you spoken to mars or flacoste about bug 402160? [15:19] Malone bug 402160 in malone "Collapsible sections don't expand in IE8" [High,In progress] https://launchpad.net/bugs/402160 [15:20] nope [15:20] allenap: Not yet. [15:20] Mez: It is not a good idea to do a lightweight checkout if you don't have low-latency access to the branch (i.e. it's on your LAN or hard disk). [15:20] allenap: Go for it if you want to :) [15:20] * gmb is tied up with a bug import atm. [15:21] flacoste, mars: All collapsible areas in LP are broken in edge for IE[678]. The cause seems to be in setStyle (in YUI), where it tried to assign undefined to a style object. [15:21] flacoste, mars: I have a quick hackish fix for which there is a diff in the bug report. [15:22] flacoste, mars: It's a diff against YUI, so I thought I'd check in with you in case you had some better ideas. [15:22] allenap, I looked at your fix - have you traced the stack to see what is setting the style to undefined, and change it to ''? [15:22] allenap, it is either setStyle(undefined) or setStyle(null) that is doing it [15:23] mars: Can I trace the stack in IE? My blood pressure is already at about 120 bar. [15:23] barry, ^ didn't you encounter something similar in IE? [15:23] mars: i am a windows-free shop [15:23] * barry doesn't have ie [15:23] allenap, does the suite still pass with your fix? [15:23] mars: I haven't actually run it ;) [15:23] mars: ah, it helps if i read more scrollback [15:24] allenap, as a contingency, I'm +1 on patching YUI, will need flacoste's OK for landing it [15:24] flacoste: can I get an RC for https://code.edge.launchpad.net/~edwin-grubbs/launchpad/bug-401748-ie8-new-project/+merge/9083 BTW, there is also a lazr-js branch for this bug which will need an RC after it is reviewed. [15:24] allenap: yes, there is an issue with this. i will try to remember it [15:24] allenap, put an XXX in there :) [15:24] https://dev.launchpad.net/Testing seems incomplete, it only mentions running the testsuite on ec2. is there a way to run it locally? [15:24] a big one [15:25] make check? [15:25] mars: :) Yeah, I have done. I think gmb was going to file a bug against YUI, but for now I've XXXed it to the bug above. [15:25] been months since I ran that :) [15:25] mars: You mean in the LP tree? I don't think that'll even touch this code though. [15:26] allenap, you can patch the YUI in the tree, then run windmill. your fix should be in there. You can verify by looking in Firebug [15:26] to see the JS files pulled in [15:26] allenap, we bundle our own copy of YUI instead of pulling it from the Y! servers [15:29] jelmer: yes, the LP test suite can be run on your local system as well; "make check" will do it [15:29] mars: I've patched YUI in a lazr-js branch, so that's okay I think. I'll go and run Windmill. Is there a YUI test suite I can easily run too? [15:29] allenap, no, there isn't a suite for PR2 [15:30] al-maisan: Won't that take many many hours? [15:30] wgrant: Yes :) [15:30] mars: Okay. [15:30] Somebody might want to document this. [15:30] leonardr: is it true that with lazr.restful POSTs must be to named operations? [15:30] wgrant: it depends on your machine .. but, yes, it will take 2-3 hours [15:30] allenap, ok [15:31] wgrant: there are plans afoot to parallelize it and/or make it run faster in other ways.. [15:31] barry: yes, that's the only thing we use POST for [15:31] the other thing it makes sense to use POST for is constructors, but i haven't implemented that yet [15:31] leonardr: gotcha. the latter was what i'm looking for, but for now i'm just using a named op of 'new' [15:32] barry: that's a good name for now [15:32] hi stub! [15:32] leonardr: cool. glad it's on the radar [15:33] leonardr: also for factory functions, we always return u'' in the body, right? so if i wanted to dump the json that comes back from a request, i have to check that len() != 0 [15:33] wgrant: https://dev.launchpad.net/FasterTests [15:33] I wonder how hard it would be to replace the root-owned-and-running apache with a user-run twisted.web server for development purposes. [15:34] apache is used for some vhosting and path-mangling, I take it? is there anything more tricky than that? [15:34] barry: well, technically if it's sending you an empty string the response type shouldn't be application/json, but we probably do that [15:34] so, yeah [15:34] Darn, they slashdotted it [15:34] awilkins: kfogel slashdotted it ;-) [15:35] leonardr: ah, good point. i'll double check the type [15:35] but there's a slight possibility that we send you the quoted empty string, "" [15:35] leonardr: i think not, but i'll double check. thanks [15:35] radix, I don't believe there is much more that that, since on .dev requests are processed by the Zope server after routing [15:35] awilkins: hey, it's all part of the celebration ;) [15:35] radix, cool idea though [15:36] awilkins, we should have a copy of the source tree up on some faster servers in a few minutes. [15:36] mars: cool. [15:36] mars: Including the repo or just the sources? [15:36] direct HTTP download, as fast as an ubuntu release [15:37] herb, ^ ? [15:37] including the repo [15:37] You don't even need the sources to be in it if it has the repo.. you can just remove the tree and make downloaders check it out again [15:38] branching peaking at 1137KB/s \o/ [15:39] mars: The unittests in lazr-js all fail the same in my branch as in toolchain :) As in, not all unittests pass, which is unfortunate, but at least my change has not introduced any new ones. [15:40] allenap, in IE? or Firefox? [15:40] I should hope they pass in firefox :) [15:40] mars: In FF. [15:40] ah [15:40] allenap, not fun, but at least it is not a regression there [15:41] allenap, tried windmill yet? [15:41] mars: Yep, and the vast majority do pass. [15:41] ok, that is good [15:41] flacoste, ^ should we RC and push allenap's fix? [15:42] mars: No Windmill yet. I'm just bringing a couple more generators online before I can risk that. [15:42] flacoste, big day, some people would probably enjoy having it [15:42] allenap, :) [15:42] mars, flacoste: We need something, because otherwise IE users will be broken entirely tomorrow! [15:42] need to fix up "make jscheck" [15:43] allenap, mars iirc the proper invocation is foo.setStyle('height', null) [15:43] allenap, mars: merge proposal, please? [15:44] radix: apache is used for mod_rewrite and mod_ssl also [15:44] radix: two nice improvements would be to modify our setup so that the dev environment starts a local apache and postrgesl cluster [15:45] flacoste: Coming up. [15:45] flacoste: basically, it would be good if multiple instances could run on the same machine, i.e. two test runs. [15:45] flacoste: or a test run and launchpad.dev [15:46] EdwinGrubbs: what is this alert thing? [15:46] EdwinGrubbs: we should use Y.log, or something like that [15:46] not an alert [15:46] flacoste: ok [15:47] EdwinGrubbs: and can't you use a waits.forElement() or something like that instead of sleep? [15:48] abentley: agreed [15:49] flacoste: yes, I can do that. [15:53] allenap, mars The other option is that we make Y.lp.activate_collapsibles() defer to the old activateCollapsibles() function for IE users. [15:53] flacoste: ^^ [15:53] flacoste: hm, but is the mod_ssl really necessary for local development? [15:54] But I think a fix to lazr-js is preferable. [15:54] gmb: Or not run at all. [15:54] get your launchpad source code: http://people.canonical.com/~herb/ [15:54] allenap: Why not degrade only a little bit first :) [15:54] radix: well, it's good for regression testing [15:54] awilkins, wgrant, sourcecode is available at http://people.canonical.com/~herb/ [15:54] herb: That's with the tree, not just .bzr? (it looks far too big) [15:54] Many thnks [15:54] radix: spotting insecure warning and such other things [15:54] ah [15:54] gmb: Then we have two lots of code to kinda maintain. [15:55] I'm already 273MB through, so I'll see how it goes.. [15:55] True [15:55] wgrant: with the tree, yes === herb changed the topic of #launchpad-dev to: Launchpad Development Channel | https://dev.launchpad.net/ | Please use #launchpad for support. | https://launchpad.net/~launchpad-dev | Get it: https://dev.launchpad.net/Getting | http://people.canonical.com/~herb/ | http://paste.ubuntu.com/ [15:56] I guess we should update /Getting with instructions on the alternative sourcecode download, and how to use it? [15:58] leonardr: ping [15:59] herb, 625K/s - that's nice :) [15:59] mars: heh [15:59] ETA 6 minutes at that speed [16:00] barry, hi [16:00] leonardr: hi. are we trying to keep all the webservice/api doctests in lib/canonical/launchpad/pagetests/webservice/ ? [16:01] leonardr: i'm working on an api to expose message moderation and am wondering where to put the tests [16:01] barry: yes, tests of the launchpad web service in particular go into that directory [16:02] leonardr: cool thanks. btw, you know i'm taking over for rockstar on thursday? [16:02] barry: yes [16:02] cool [16:04] oh wow, herb, download speed is fluctuating [16:04] bursty [16:05] EdwinGrubbs: are you removing the bug count feature from the +series page? I can do it if you are not [16:05] mars: that's just me putting kinks in the hose at random times. :) [16:09] 9m 38s for the alternative source download [16:09] flacoste, mars: It seems that EdwinGrubbs found the root cause of that IE bug, and has fixed it in https://code.edge.launchpad.net/~allenap/lazr-js/set-style-fix-for-ie-bug-402160/+merge/9095 [16:09] How big is lp-source-dependencies? [16:09] wgrant: someone said a few hundred meg [16:09] (it actually downloads at a reasonably fairly constant speed) [16:09] EdwinGrubbs: I was chasing that same bug, but didn't find the cause, so I'm really glad you did :) [16:12] Blimey, loom is a required plugin? [16:12] awilkins: I never use loom [16:12] No.. I meant there's a link for it in the launchpad "bzrplugins" folder [16:13] it's optional [16:13] awilkins: yes, but you do not need to use it. I do not [16:13] Heh, why's it not in the optionalbzrplugins/ folder like git then :-) === noodles is now known as noodles775 [16:18] salgado: ready for our call? [16:18] sinzui, yes [16:19] flacoste, hi. I wants an r-c: https://code.edge.launchpad.net/~beuno/launchpad/open-the-world/+merge/9096 [16:19] sinzui: that would be great, since I'm still making changes to land my rc fixes. Do you want me to assign that bug and the sprite bug to you? [16:19] EdwinGrubbs: I think you should keep the sprite since you definitely can get an RC to land. [16:20] EdwinGrubbs: If my branch has issues, it can still be landed in the reroll [16:25] mars: any luck on updating /Getting to reflect the static files on people.c.c? [16:26] herb, haven't yet [16:27] flacoste, ping. [16:30] pkern@asterix:~/launchpad/lp-branches/devel$ make schema [16:30] Missing ./download-cache. [16:30] Developers: please run utilities/link-external-sourcecode. [16:30] That's somehow not helpful. [16:30] I did strange things I have to admit in order to avoid a re-checkout, but still... [16:30] sinzui, all translations-related templates have been moved out of lp.registry [16:31] \o/ [16:31] Hm, well, the real error is: [16:31] link-external-sourcecode: error: Parent branch not specified, and could not be discovered. [16:31] pkern: try "make" on its own [16:31] salgado: blueprints can move anytime next cycle [16:31] So I confused it too much. [16:31] pkern: ah, you need utilities/link-external-sourcecode [16:32] bigjools: What "parent" do I need to pass to that script? [16:32] is that not in the destructions? [16:32] pkern: a path to lp-sourcedeps [16:32] pkern: I take it you didn't run rf-setup to completion? :) [16:32] salgado: i just QAd bug 397148. first try timed-out but the next succeeded. marked it OK/OK [16:33] Malone bug 397148 in launchpad-registry "+adminpeoplemerge should be able to merge people without email addresses" [High,Fix committed] https://launchpad.net/bugs/397148 [16:33] bigjools: Indeed it didn't like me because it told me that my checkout and devel diverged and it needs merge. [16:33] bac, thanks! do you have the OOPS for the time out? [16:33] salgado: sorry, no [16:33] bigjools: I'll try again. [16:34] pkern: in your devel directory, you can try "bzr pull --overwrite" [16:34] bigjools: Sounds funny. [16:35] bigjools: Cool. [16:39] flacoste: can I get an rc for https://code.edge.launchpad.net/~edwin-grubbs/lazr-js/lazrjs-bug-401748-ie8-new-project/+merge/9081 [16:39] allenap: so EdwinGrubbs above fix is obsoleting what you wanted in? [16:40] hi deryck [16:42] beuno, EdwinGrubbs: done [16:42] thanks [16:42] Damn, I love my new 10Mbit/s connection [16:43] flacoste, hi, I'd like to get rc on: https://code.edge.launchpad.net/~deryck/launchpad/correctly-list-dupe-subscription-401779/+merge/9084 [16:43] flacoste, fixes a bug discovered during QA yesterday. [16:47] deryck: done [16:47] flacoste, awesome, thanks! [16:51] herb, please check https://dev.launchpad.net/Getting, "Alternative Sourcecode Download" [16:51] mars: looks good to me. [16:53] rockstar: chat? [16:56] I'm trying to do the whole rocketfuel-setup stage but primed with the revision data by branching herb's archive into lp-branches ; how long should that take? It seems to be stuck in an infinite loop at "Fetching revisions:Get stream source" [16:56] Is there a command to get just the last revision of launchpad? Instead of doing a bzr branch lp:launchpad which gets the massive history? [16:56] Ah, now it's saying "Insering missing keys" [16:56] Done [16:56] kripken__, you can either download the source from http://people.canonical.com/~herb/, or do bzr checkout lp:launchpad [16:56] kripken__: You could get the source tarball from the branch page, or get a stacked branch [16:57] "make schema" fails with: The found YUI root isn't valid: /home/[...]/launchpad/lp-branches/devel/lib/canonical/launchpad/icing/yui/3.0.0pr2/build [...] Traceback (most recent call last): [16:57] Or get herbs archive which has a full history [16:57] thanks guys [16:57] File "buildmailman.py", line 15, in ? [16:57] from canonical.config import config [16:57] File "/home/[...]/launchpad/lp-branches/devel/lib/canonical/config/__init__.py", line 19, in ? [16:57] import ZConfig [16:57] ImportError: No module named ZConfig [16:57] make: *** [compile] Error 1 [16:57] Any idea? [16:58] MarkusT, rocketfuel-setup ran? [16:58] Jup, no problems there [16:59] noodles775: cloak set [16:59] MarkusT: what does "make build" do? [16:59] noodles775: you may have to refresh with nickserv to see it [16:59] MarkusT: I believe "ImportError: No module named ZConfig" is due to you not having run utilities/link-external-sourcecode [17:00] yeah that's more likely [17:00] hmm [17:00] joey: thanks! [17:00] I'm now running setup again and will try to use your advise... just a moment.... [17:00] bigjools, al-maisan, that isn't in the /Getting instructions [17:00] should it be? [17:00] noodles775: you can /msg christel and thank her :-) I just own the cloak rights. :-) [17:01] mars: potentially, but you only need it if you shortcut the instructions in the first place [17:01] abentley, yeah. [17:01] bigjools, depends, I assume MarkusT was following the instructions [17:02] mars: Yes, i did :-) [17:02] mars: you typically run utilities/link-external-sourcecode after branching a development branch from (db-)devel .. [17:02] i.e. later in the process [17:02] well [17:03] you need it in trunk as well, so it can be built [17:03] MarkusT, was that error in a branch you just created, or devel/? [17:03] bigjools: I though rf-setup will take care of running utilities/link-external-sourcecode in the trunk .. [17:03] yes it does === salgado is now known as salgado-lunch [17:06] Ok, running Rocketfuel-setup again doesn't solve the problem. "Usage: link-external-sourcecode [options] [parent]", what do I need for parent or options? I'm not branching, just trying to get a first working system. Might it be useful to completely delete all files and start over again? [17:07] MarkusT, first, check the sourcecode/ directory in trunk/. Do you see a collection of links there? [17:07] Hmm, I'm getting all these "inconsistent details in skipped record" errors from bzr while it's branching source-deps [17:07] abentley, ^ ? [17:08] flacoste, fixed the text on /Getting - the source-deps aren't in the tarball, just the history [17:09] mars: This is because launchpad's trunk wasn't reconciled. I was in the process of reconciling it, when kfogel's announcement came. [17:09] flacoste, ^ [17:10] abentley, is that a problem? [17:10] mars: Where's trunk supposed to be? [17:10] awilkins: They seem to be coming from bzr-svn and bzr-git. [17:10] MarkusT, ~/launchpad/lp-branches/devel [17:10] And those two are taking absolutely ages longer than any of the others. [17:10] mars: Only have a Makefile there [17:11] MarkusT, ok, how about in ~/launchpad/lp-sourcedeps ? [17:11] mars: It's not a serious problem, but it does mean that people will see errors like that, and when they run bzr check. [17:11] sourcecode is empty, mars. ls [17:11] download-cache eggs sourcecode [17:12] What I did ; mkdir ~/launchpad ; cd ~/launchpad ; bzr init-repo --2a lp-branches ; bzr branch lp-branches/devel ; rm -rf lp-branches/devel ; ./rocketfuel-setup [17:13] mars: So should I start over? Seems the easiest way ..... [17:14] MarkusT: run rocketfuel-get and see what it does [17:14] Hmm, I wonder if the sourcedep pulling is so slow because it's upgrading to 2a. [17:14] Oops, stacktrace [17:14] When branching dulwich [17:15] SHA1KnitCorrupt [17:15] abentley, ^ ? [17:15] mars: Using saved parent location: http://bazaar.launchpad.net/~launchpad-pqm/launchpad/devel/ [17:15] awilkins, did it kill the download? [17:15] Does 2a even use Knits? [17:15] No revisions to pull. [17:15] Tree is up to date at revision 53. [17:15] Could not find /usr/local/bin/sourcedeps.conf [17:15] Updating sourcecode dependencies in rocketfuel: [17:15] mars: Yes, I think it did [17:15] NOTE: Each sourcedep may take a while; please be patient. [17:16] ALSO NOTE: You can ignore any warnings you see below about how [17:16] 'You have not informed bzr of your Launchpad ID ...' etc. [17:16] See https://bugs.edge.launchpad.net/bzr/+bug/401617 for why. [17:16] Malone bug 401617 in bzr "'bzr branch lp:foo' says "You have not informed bzr of your Launchpad ID ..." in anonymous pull (dup-of: 346677)" [Undecided,New] [17:16] Malone bug 346677 in bzr "confusing message about launchpad-login, even when doing readonly access to a public branch" [Medium,Confirmed] [17:16] Sourcedeps: /home/thielmann/launchpad/lp-sourcedeps/sourcecode [17:16] Checking /home/thielmann/launchpad/lp-sourcedeps/sourcecode/bzr-git [17:16] Couldn't find lp:~launchpad-pqm/bzr-git/devel/, skipping it. [...] [17:16] mars: I do not know why. [17:16] This goes on for all dependencies.... [17:16] abentley, ok, thanks, just checking :) === abentley is now known as abentley-lunch [17:17] Why are they all saying "Created new stacked branch referring to file:///home/adrian/launchpad/lp-branches/devel/." [17:17] What's devel got to do with the dependencies (in terms of stacking?) [17:18] MarkusT: hmmmm it might be because of what it says - "You have not informed bzr of your Launchpad ID" [17:18] bigjools, I think you can ignore those (like the message says :) [17:18] mars: I know, but he's getting them for sourcecode checkouts [17:18] mars: I'm just starting over .... See you an a few minutes ..... [17:19] MarkusT, ok [17:19] bigjools, that first line that MarkusT posted is odd: No revisions to pull. [17:19] Tree is up to date at revision 53. [17:19] Could not find /usr/local/bin/sourcedeps.conf [17:20] that's normal [17:20] ok [17:20] no revisions tp pull means his devel is up-to-date [17:20] bigjools, and the "Could not find /usr/local/bin/sourcedeps.conf" ? [17:20] I get that every time [17:21] ah, ok too then [17:21] I think it defaults to somewhere else [17:21] Oh. [17:21] rocketfuel-get is stupid. [17:21] the problem is the "Couldn't find lp:~launchpad-pqm/bzr-git/devel/, skipping it. [...]" [17:21] This explains why it's so horrifyingly slow. [17:21] I'm using Ubuntu 9.04, which has bzr 1.13, can I not clone the launchpad repo? [17:21] wgrant: patches accepted [17:21] MarkusT, you can save yourself the download of sourcecode when starting over if you save your ~launchpad/lp-branches directory [17:21] This part of rocketfuel-get makes no sense to me (the bit that's doing it) http://pastebin.ubuntu.com/223662/ [17:21] The detection code for SSO/ShipIt is broken, so it tries to branch everything as if it's meant to share history with the main tree. [17:21] YES. I have wanted to say that for 2 years. [17:22] MarkusT, sorry, I meant he download of the launchpad trunk [17:22] So it first branches devel to the dep's dir, then pull --overwrites the remote branch. [17:22] Causing an upgrade to 2a. [17:22] And a whole lot of copying. [17:22] Aha, all three of those comments were about the same thing [17:22] bigjools: Patches may come... [17:22] beuno: what ever happened to the front page badge idea for today's FOSS announcement? [17:22] wgrant: I am happy to be your sponsor [17:23] awilkins: The easy workaround is to preemptively create empty branches for each dependency. [17:23] That will speed things up. [17:23] And probably stop dulwich from breaking. [17:23] But let me fix rocketfuel-get properly.. [17:23] It's treating everything as optional and trying to stack them on trunk [17:23] mhall119|work, no, you will need to upgrade to Bazaar 1.16.1 at minimum, and even better, Bazaar 1.17 [17:24] I think it couldn't find sourcedeps.conf [17:24] awilkins: Not quite. [17:24] mars: any chance of that getting into the Jaunty repos? [17:24] mhall119|work, if you are on Ubuntu, you can use the Bazaar PPA to get them: https://launchpad.net/~bzr/+archive/ppa [17:24] awilkins: Oh, yes. [17:24] mhall119|work, the latest Bazaar, as a backport? [17:25] In fact, it said "can't find sourcedeps.conf on STDOUT" [17:25] I put the quote in the wrong place there [17:25] mhall119|work, I don't know, but the latest version will be in Karmic [17:25] mars: thanks [17:25] shit [17:26] there is a typo in rocketfuel-get :-( [17:26] Yes. [17:26] No $ before the optional. [17:26] right [17:26] That change was made just today, wasn't it? [17:26] yesterday [17:26] flacoste: cp rs=joey [17:26] flacoste, bigjools is sponsoring wgrant's fix, I believe? [17:26] joey, :) [17:26] mars: Requires signing of the contributors agreement and blah. [17:27] oh [17:27] I probably would have worked that out earlier if it wasn't 2:30. [17:27] * wgrant shall retry rocketfuel-get, then go to bed. [17:27] wgrant: Is it because $(dirname"$0") returns "." ? [17:27] flacoste, et al - I'm afk for a dr's appt. Ring my mobile if you should need me. [17:27] awilkins: No. [17:28] It's a nice missing $ in the 'if [ optional ]' [17:28] awilkins: 'optional' is of course always true. [17:28] $optional might not be. [17:28] wgrant: aha [17:29] herb: can you remove the launchpad tar ball from your home directory, please, we'll need to do another one [17:29] flacoste: what do you want it to contain? [17:29] :) [17:29] herb: rocketfuel-get is broken in it [17:29] flacoste: awesome. [17:29] flacoste: what revision does it need? [17:30] a yet unlanded one [17:30] i'm on it [17:30] ah-ha [17:30] ok [17:30] :-) [17:31] bzr taking ages on bzr-svn is also normal? [17:31] flacoste: they're gone [17:31] pkern: That's this bug. [17:31] pkern: There's a typo in devel/utilities/rocketfuel-get [17:32] wgrant: thanks [17:32] pkern: I just added the $, then reran lp-branches/devel/utilities/rocketfuel-get [17:32] It seems to be branching properly now. [17:32] Yes, I concur [17:33] So much faster without the 2a conversion... [17:33] sweet [17:33] No huge errors about "inconsistend details" too [17:34] flacoste, http://paste.ubuntu.com/223675/ ? [17:34] Hm, maybe one should join in on the development of this software... I love launchpad as it now and would gladly want to make it better [17:34] And hopefully no dulwich conversion crash. [17:34] mars: yes, that's what i'm submitting now [17:34] dulwich already branched OK here [17:34] Is there any formal process to become a developer or am I just to send in patches? [17:35] LordMetroid: The process for Bazaar is presently to submit a branch and merge proposal to launchpad ; I would expect that LP itself would probably have a similar process ? [17:35] LordMetroid, for now: https://dev.launchpad.net/PatchSubmission [17:35] joey, kfogel :) [17:36] A shame all those branches are conflicting formats [17:37] They could share a repo if they were all 2a [17:37] But there would be no benefit. [17:37] And it's nice that it doesn't tell me what' it's doing when looking at bzr-svn... [17:37] congratulations on making launchpad free software, guys! [17:40] samgee, thanks :) [17:41] * awilkins does `make schema && make run` [17:41] * wgrant is still two sourcedeps away. [17:42] Yay, it actually continues now. [17:42] My ISP might be caching the packs [17:42] They have a transparent proxy [17:42] btw, there's an error in https://dev.launchpad.net/Getting [17:43] awilkins: I have 320ms latency to LP. [17:43] wgrant: Ouch [17:43] awilkins: If you actually use http and not bzr+ssh... [17:43] it says the source tree is 150Mb, while it seems to be 150 MB [17:43] pkern: Didn't I see something above about reconfiguring the lp: responder to hand out http:// links for a while? === salgado-lunch is now known as salgado [17:44] Darn. "No module named cElementTree" [17:45] EdwinGrubbs: how goes your branches? [17:45] jtv-afk: all branch submission should go to db-devel [17:45] Need to add that to the list of packages that rocketfuel-setup installs for python 2.4 [17:45] awilkins: actually, it should come from the download-cache [17:46] flacoste, I don't have it in mine... [17:46] hmm [17:46] right [17:46] i thought it was part of elementtree [17:46] but probably not [17:47] i need to update the launchpad-developer-dependencies again :-( [17:47] flacoste, I think so. I'm fetching it from /usr/lib/python2.4 [17:48] At least this stuff is minor teething trouble [17:48] I've spent _weeks_ trying to get the build system working on a project I'm on, and it's frickin' Maven - it's supposed to be EASY [17:49] (this is a set of projects created by someone else - I'm not so slow as to have my own build system not working on me...) [17:50] uploaded a new l-d-d [17:50] flacoste: Just adding python-celementtree? [17:50] wgrant: yes [17:50] Is _pythonpath.py generated? [17:50] jtv-afk: stop submitting to devel! [17:51] awilkins: it is [17:51] flacoste: Cool, then it'll just pick up the change... which file is it, or is it one of the branches that rocketfuel-get fetches? [17:52] awilkins: it's a package in the ~launchpad PPA, give it a few minutes to be published [17:52] Righto [17:54] flacoste, any ideas what this means? http://buildbot.devpad.info/builders/db_lp/builds/550/steps/bzr/logs/stdio [17:55] beuno: trouble connecting to codehosting [17:55] herb: ping [17:55] flacoste: Does celementtree need an amd64 build as well [17:55] ah, seemed to obvious :) [17:56] sinzui: pong [17:56] awilkins: if you are on amd64, yes [17:56] flacoste, so how do I kick this off again? [17:56] herb: can you make me an admin on staging to test some team merge fixes? [17:56] beuno: i need another fix in, so i'll handle it [17:56] the PPA should handle the amd64 build as well, shouldn't it? [17:56] sinzui: I can [17:57] mars: Yes, it's just taking it's time, I think [17:57] mars: Yes, but there's a large backlog. [17:57] sinzui: done [17:57] Somebody should bump the build score. [17:57] eg. herb [17:57] flacoste, danke [17:58] The packages for launchpad-dependencies are only listing i386 builds as well [17:58] wgrant: build URL? [17:58] Although all of those buildds are going to be occupied by their current builds for at least an hour yet, I expect. [17:58] I can bump it [17:58] awilkins: actually, it doesn't need an amd64 build [17:58] it's an indep package [17:58] only deps [17:58] and i'm not building python-celementtree itself [17:58] that's part of universe [17:59] flacoste: Aaaah. Ok. [17:59] bigjools: https://edge.launchpad.net/~launchpad/+archive/ppa/+build/1127437 [17:59] There's a thing to fix in LP then :-) - misleading arch tags on package builds [17:59] wgrant: sorted [17:59] build start: 1 minute [18:00] * It appears your search path is unconfigured. [18:00] Have you read ? [18:00] a) nobody told me and b) even if I would try... [18:00] bigjools: It doesn't take into account currently building builds? [18:00] Heh, should be Build score : "one MEEELION points" [18:00] wgrant: yes, it does [18:01] * wgrant should have checked the code. [18:01] :) [18:01] pkern, looking now [18:01] pkern, on which page did you find that link? So I can re-link it with the new material [18:02] make[1]: Entering directory `/home/pkern/launchpad/lp-branches/devel/database/schema' [18:02] mars: There. [18:02] hmmm === barry-away is now known as barry` [18:02] ok, first I'll copy the page [18:03] I wonder why that search path is needed anyway. Storm not supporting schemas? [18:03] flacoste: have you gotten a chance to look at https://code.edge.launchpad.net/~edwin-grubbs/lazr-js/lazrjs-bug-401748-ie8-new-project/+merge/9081 === matsubara-afk is now known as matsubara [18:03] sinzui: their progressing pretty well, but I'm kinda derelict on my chr duties. [18:04] pkern, https://dev.launchpad.net/DatabaseSetup [18:04] EdwinGrubbs: I will free to start the CHR in 5 minutes [18:04] EdwinGrubbs: I'll start with projects and continue until you are available, which may not be at all [18:05] EdwinGrubbs: yes, it's approved already [18:05] pkern, but I don't guarantee that that page has an answer for you, I just brought it out into the light [18:05] So, I have my DB! [18:05] flacoste: ok, I didn't see a comment in the mp, and I wanted to make sure that you weren't confused by the two mps with the same bug number. [18:06] But I first had to create a postgres superuser for myself. [18:06] And change the search path. [18:06] * If this fails you need to run as the postgresql superuser [18:06] * eg. sudo -u postgres make create [18:06] There is no make create in there. [18:06] Hm, well there is, but in the subdir, heh ;-) [18:06] mars: It had, yes, thanks. [18:07] It's only adjusting the search_path it seems, I didn't touch hba. [18:07] EdwinGrubbs: no, there was both a lp and a lazr-js branch === barry` is now known as barry [18:07] * bigjools needs to run out, bb later [18:08] Ok, you need to touch hba to allow launchpad_main to connect. [18:09] That makes sense. [18:09] * gmb EODs. [18:10] * wgrant waits for his disk to catch up with postgres. [18:10] wgrant, are you having fun? [18:11] wgrant, there is a good reason why some of us still use hefty desktop workstations instead of laptops :) [18:11] * barry needs a new hefty desktop workstation :) [18:11] psycopg2.OperationalError: FATAL: database "launchpad_dev" does not exist [18:11] mpt: It's a bit late, and rocketfuel-setup is buggy, but otherwise it's not bad. [18:11] psycopg2.OperationalError: FATAL: database "launchpad_dev" does not exist [18:11] pkern: make schema didn't run properly? [18:11] Hrm. [18:11] Mine certainly created launchpad_dev. [18:12] wgrant: The problem is that now this stuff is owned by postgres I think. [18:12] * wgrant is still waiting for make schema to finish... [18:12] Well, let's try [18:12] Thanks for choosing AGPLv3. :-) [18:12] Oh great. [18:12] psycopg2.ProgrammingError: permission denied for relation pg_database [18:12] And as postgres [18:12] [Errno 13] Permission denied: '/home/pkern/launchpad/lp-branches/devel/eggs/test-easy-install-3674.pth' [18:13] *sigh* [18:13] pkern, according to database setup, it just needs your db user to be a superuser [18:13] ryanprior, :) [18:13] mars: It uses random own users to conncet. [18:13] pkern: Which are configured during make schema. [18:13] wgrant: So which one needs to be superuser? \-: [18:14] It started! [18:14] I haz a launchpad.dev. [18:14] pkern: The one matching your username. [18:14] pkern, "Your default PostgreSQL user has the same name as your login account (if it existed already, and it was not a superuser, drop it first)" [18:14] pkern, fwiw, stub is our database guru [18:14] Right. [18:14] mars: the future is SSD drive, postgres doesn't have any trouble catching up here :-) [18:15] "drop it first", heh... alter role $user superuser; [18:15] Are the usernames and passwords documented somewhere? I know a couple, but do I have to look through the sample data? [18:15] flacoste, nice :) [18:15] flacoste, and the future always will be SSD drives. [18:15] wgrant: they are in lib/canonical/launchpad/pagetests/README i think [18:16] flacoste: README.txt, but yes. Thanks. [18:18] Is there a root cert somewhere that I should tell Firefox to trust, or do I just whitelist the cert for each domain? [18:19] wgrant: you whitelist each domain [18:19] wgrant: i assume this means that you succeeded in running make run? [18:19] flacoste: Oh yes. [18:19] The webapp all works. [18:19] awesome!!! [18:19] Works here too. [18:20] EdwinGrubbs: branches must be landed on db-devel [18:20] Once I'd fixed rocketfuel-get, installed python-celementtree, createuser'd, added the postgres search_path and auth stuff, it all worked. [18:20] crap [18:20] What code is running on drescher for the archive? :-P [18:20] gmb, What does this mean? "ConjoinedBugTaskEditError: This task cannot be edited directly, it should be edited through its conjoined_master" [18:20] * pkern didn't need celementtree yet. o_O [18:20] Or maybe I just had it :-P [18:21] I presume there's some docs on the internal wiki about setting up a local Soyuz system. [18:21] The Codehosting ones have been made public. [18:21] * rockstar hates at DOM traversal [18:21] But not Soyuz. [18:21] rockstar: jQuery makes that so much nicer. [18:22] wgrant: actually, not really, Soyuz is that very mysterious secret guarded jealously by bigjools [18:22] and cprov [18:22] :-p [18:22] wgrant: seriously, i think nobody runs Soyuz locally [18:22] bigjools: soyuz, soyuz, soyuz! how! [18:22] wgrant: i know they usually do their QA on a dedicated server aka dogfood [18:23] flacoste: Right. But presumably they at least run things manually locally... [18:23] bigjools: you officially have a request to document how to test Soyuz stuff locally [18:23] * wgrant hears bigjools scream. [18:24] wgrant, that is a good list of roadblocks [18:24] * mars tosses it into a Tomboy note [18:25] wgrant, pkern wait 'til we ask you to review some soyuz code! [18:25] I think we have rf-get in process [18:25] python-celementree is fixed? [18:25] mars: The first has a branch, the second has been uploaded. [18:25] don't know what is up with the database setup stuff though [18:26] wgrant, the first also needs a CP [18:27] Is the default batchnavigator size lower on launchpad.dev? I notice team membership lists are only 5 per page, not 50. [18:27] wgrant, yes [18:28] see default_batch_size: 5 in configs/development/launchpad-lazr.conf [18:30] matsubara: Aha. [18:30] sinzui: can we delay our one-on-one call till tomorrow? [18:31] leonardr: all the pprint_*() methods in webservice.py require a json_body, but that seems tedious. would it make sense to have a method that calls .jsonBody() automatically? [18:32] EdwinGrubbs: sure [18:32] barry: well iirc there are a number of methods for using different http methods, get, post etc [18:32] if you added a jsonbody you'd expand that x2 [18:32] i guess you could just have a 'get' method, or an argument to the get method, that returned the body instead of the response object [18:34] leonardr: yeah, i see what you mean. what does webservice.get() return normally? [18:34] leonardr: i wonder if pprint_() could sniff its argument. ugly, but convenient [18:34] barry, it returns an httplib2 response object, or some kind of response object anyway [18:34] cody-somerville: A conjoined bugtask is the non-series task for a project which also has a task for its development series on the same bug - that dev series task is the conjoined master. While the dev focus task isn't Won't Fix, the non-series task cannot be altered. [18:36] leonardr: i.e. not a dict, right? looks like json_body is expected to be a dict (or maybe a duck type having a .items() method) [18:36] json_body will always be a dict in lasr_restful applications, at least for now [18:37] barry, i have a very obscure python question for you [18:37] sure thing [18:39] leonardr: i'm using a new irc arrangement, so i can't tell if you're pvtmsging me or not ;) [18:40] Which bit downloads the packaged dependencies? It's not picking up the cElementTree egg yet [18:40] awilkins, apt-get install launchpad-developer-dependencies [18:41] let me try one recipe and see if it works [18:41] wow, ok [18:41] barry: nm, the thing i tried worked for now [18:41] mars, herb: the db-devel branch now contains the rocketfuel-get fix, we could republish another tarball using db-devel [18:41] mars: Aha, I was fooled by not seeing it in synaptic... [18:42] leonardr: cool [18:42] * wgrant needs to sleep. [18:42] Thanks everyone. [18:44] flacoste: I'll take care of it as soon as I get back from lunch. [18:45] flacoste: what revision? [18:51] flacoste: I know this seems rediculous, but can I get an RC for this, since it seems silly not to fix it. https://code.edge.launchpad.net/~edwin-grubbs/launchpad/bug-401974-icon-positioning/+merge/9102 [18:52] Stupid question ; what's the root user on the default install? [18:53] awilkins: the default install of what? [18:54] Launchpad, as produced by getting a development set of Launchpad [18:54] There must presumably be a "god" user in the database [18:55] awilkins: when you run a local copy for development at http://launchpad.dev there are a number of users in the database [18:55] bac: I agree, I can see them. But to really explore the features you can't just register your own, you need an admin user, yes? [18:56] look at lib/canonical/launchpad/pagetests -- there is a README .txt and a REFERENCE.txt that list different test users [18:56] runyaga: Hi, Alan [18:56] Excellent, thanks [18:56] EdwinGrubbs, how it goes! good job re: cleaning up and releasing launchpad [18:56] bac: Maybe that should go in the FAQ? [18:56] awilkins: I think what you're thinking of is logging into an LP instance as a user that's a member of the ~admins team [18:56] really great [18:57] awilkins: yes, probably. [18:57] awilkins: although off hand I don't know which of the sample data users are in that team - I guess the README might show that [18:58] awilkins: "Foo Bar" is a an admin for launchpad.dev and the one we most frequently use when testing. [18:58] Just looking [18:58] Yes, there he is [18:58] in README [19:00] I don't think I've configured email sending correctly yet, my puny attempts at registering myself failed [19:11] barry: if it's one to one, the mailing list should be a property of the team [19:11] and it will show up as /~guadamen/mailinglist [19:11] that *should* work now [19:11] leonardr: it's always one-to-one... thanks! [19:12] leonardr: i think i just have to export that [19:13] congrets for the release! I knew this will come [19:14] sinzui, I'd love to understand why ie8 shows that stupid continue button in the middle of the page [19:14] pkern, wgrant, flacoste: right, we *rarely* run soyuz locally, and in many cases it's impossible because you need a build farm. you can run the publisher and the upload processor locally though. [19:14] Those bzr-1.17.tar.{gz,bz2} things at http://people.canonical.com/~herb/... They're not actually bzr, are they? > 160MB seems like a lot. [19:15] Ursinha: I will never be able to provide that answer. I so not have any IE browsers. [19:15] I'm guessing they're tared up launchpad trees, but the name suggests otherwise. [19:16] soren, they could be bzr branches of the bzr project [19:16] bigjools: How would I run e.g. the publisher, i.e. what's the wrapper to run scripts in the tree? [19:22] soren: it was bzr [19:22] freenode hickup? [19:23] flacoste: may I have your RC for this change to fix the timeouts on +series pages: https://pastebin.canonical.com/20166/ [19:23] herb: Oh, wow. [19:24] soren: It's bzr with the full history [19:24] soren: I had bzr branched from the bzr-1.17 branch and tarred it up. [19:24] soren: e.g. - includes a repository, not a just source === greg_g is now known as greg-g === ursula_ is now known as Ursinha [19:36] ok. new tarballs up with the fixed rocketfuel-setup [19:36] flacoste: ^ === matsubara is now known as matsubara-afk [19:49] sinzui, did you know about this: https://edge.launchpad.net/bzr/+download [19:50] empty series showing up [19:50] I have not seen that [19:50] beuno: I think we can fix that at the same time we link them [19:51] sinzui, should I assign it to wgrant? [19:51] * beuno giggles [19:51] He can fix his own bugs now... [19:51] ...except that I suspect I made most of the bugs he reported [19:52] shhhh [19:53] that was weird... [19:53] beuno: we are both karma whores. We have the most karma in the Launchpad Regisrty project. I enable him to report bugs, and he enables me to fix them [19:53] I just clicked a link on edge that sent me to code.launchpad.dev [19:53] the link seems fine now [19:57] kiko, :) [20:04] sinzui: was your branch reviewed? [20:05] flacoste: yes, by bac [20:05] sinzui: do you have the m-p? [20:05] flacoste: https://code.launchpad.net/~sinzui/launchpad/series-performance/+merge/9100 [20:06] * rockstar hunts for nourishment [20:07] Just a BIG THANK YOU !! [20:09] hi EdwinGrubbs, got a few minutes? [20:09] leonel, :) [20:09] bac: sure [20:13] EdwinGrubbs: did you see my msgs? [20:17] bac: ping [20:17] hi barry [20:22] So how will the launchpad wiki be opened? [20:23] pkern: it's opened already, dev.launchpad.net [20:23] Well done, it' [20:23] we haven't used the old one for a while now, edits to it have been closed for several months [20:23] flacoste: Well, there are many references to the old one. [20:24] Now I can re-gain access to launchpad code :-D [20:24] This right channel for launchpad crashes? [20:24] Like https://launchpad.canonical.com/LibrarianTransactions [20:24] pkern: yeah, it's been gardened properly [20:24] many of the pages have actually been moved, but not the interlinks [20:24] some pages have not been moved [20:24] but many of these are obsolete anyway [20:25] s/it's been/it's not been/ [20:25] \-: [20:28] Hah, scripts/ftpmaster-tools, oh well. [20:36] mthaddon, where can I find the loggerhead logs? [20:38] Hey herb [20:39] howdy [20:39] rockstar: looks like there's an in progress RT to fix that (missing firewall settings to allow rsyncing) [20:40] rockstar: although I think we're planning to move the machine into the LAN (currently in the DMZ) so it may need to wait til then [20:41] 1 [20:41] mthaddon, huh. Well, I'm getting an unhelpful "Internal Server Error" on a user's branch, and I'd like to see why. What's the process for finding out? [20:41] (typo) [20:41] rockstar: lemme check [20:42] mthaddon, I 'spect that mwhudson has super powers on the server itself, so it's not something that's been needed before. [20:42] rockstar: bzr error or apache error? [20:42] mthaddon, bzr error. [20:43] rockstar: so codehosting not loggerhead? [20:43] It's a loggerhead issue. When I get it in local loggerhead devel, I usually can see the output from the running server in the terminal. [20:43] mthaddon, no, it's a problem with loggerhead's server. [20:44] rockstar: ok, I see the debug.log - what should I be looking for? [20:48] mthaddon, something mentioning http://bazaar.launchpad.net/~richies/hypernucleus-server/trunk/annotate/head%3A/hypernucleusserver/config/ [20:49] Allright, I just removed all files and run the launchpad Setup again (according to the launchpad wiki). Still getting the same error, when running "make schema": [20:49] Traceback (most recent call last): [20:49] File "buildmailman.py", line 15, in ? [20:49] from canonical.config import config [20:49] File "/home/thielmann/launchpad/lp-branches/devel/lib/canonical/config/__init__.py", line 19, in ? [20:50] import ZConfig [20:50] rockstar: https://pastebin.canonical.com/20172/ [20:50] ImportError: No module named ZConfig [20:50] * barry runs [20:50] make: *** [compile] Error 1 [20:50] Any ideas? [20:50] Missing a python module. [20:50] MarkusT1, do you have the sourcedeps? [20:51] if so, did you run the utilities/link-external-sourcecode script? [20:52] Ursinha: Could you be more specific? I was just running rf-setup (two times now, including a complete wipe between). "mars" already told me to run utilities/link-externel-sourcecode, but I was unable to figure out what to use for "options" and "parents" (Usage: link-external-sourcecode [options] [parent]) [20:52] parent is the lp-sourcedeps dir [20:53] MarkusT1, do you see a sourcecode folder inside lp folder? [20:53] ls -l [20:53] total 12 [20:53] drwxr-xr-x 4 thielmann thielmann 4096 2009-07-21 19:35 download-cache [20:53] drwxr-xr-x 61 thielmann thielmann 4096 2009-07-21 19:50 eggs [20:53] drwxr-xr-x 2 thielmann thielmann 4096 2009-07-21 19:33 sourcecode [20:53] It's empty [20:53] MarkusT1, use the pastebin, please :) [20:53] it's empty because the links weren't created [20:53] So, how do I create the links? [20:53] using the script :) [20:54] Which script? [20:54] utilities/link-externel-sourcecode [20:54] *external [20:55] So just "utilities/link-external-sourcecode lp-sourcedeps" ? [20:55] cd into the devel dir, then run it as "utilities/link-external-sourcecode ../../lp-sourcedeps [20:55] MarkusT1, see, that's why I asked you if you have the sourcedeps [20:55] what bigjools said :) [20:56] Ursinha: there's zero output and I don't think it did anything at all? [20:56] that's normal [20:56] MarkusT1, if there's zero output it did nothing [20:57] bigjools, for me it shows the links [20:57] oh no it it's not :) [20:57] lol [20:57] So what do I do now? === matsubara-afk is now known as matsubara [20:58] MarkusT1, what do you have inside lp-sourcedeps [20:58] ? [20:58] As I said above: download-cache, eggs, sourcecode [20:58] what's in the sourcecode dir? [20:58] bigjools: Nothing [20:59] that's weird [20:59] ok, rocketfuel-setup didn't complete fully then, can you run rocketfuel-get [20:59] hm [20:59] yes [21:00] it's in utiltities [21:00] Got one question: Does Launchpad take care of version hirstory of source code or do I need to install subversion or something else? [21:00] utilities* [21:00] No1Viking: we use bzr for versioning the code [21:00] MarkusT1, please, run utilities/rocketfuel-get for it to finish downloading the dependencies [21:00] No1Viking: If you installed your own instance of Launchpad you already have Bazaar [21:01] I have a Hardy server here. Will it configure apache and everything automagically? [21:02] MarkusT1: You might also have to update your launchpad-developer-dependencies package [21:02] No1Viking: rocketfuel-setup will do, yes - be sure to read the comments as it's pretty invasive in terms of changes [21:02] No1Viking: Nope, you need Jaunty [21:02] Ursinha: Tried that before, doesn't work: http://pastebin.org/3454 [21:02] No1Viking: Patches gratefully accepted [21:02] * Ursinha looks [21:02] awilkins: I'd be surprised if it doesn't run on hardy... [21:03] mthaddon: Would you have to install the deps manually though? [21:03] awilkins: apt-get says: launchpad-developer-dependencies is already the newest version. [21:04] awilkins: Ah, sorry. Needed an update. [21:04] awilkins: I don't think so - there should be a meta-package for hardy as well [21:04] Will try that. [21:04] bigjools, do you know what's MarkusT1 problem doing the rf-get? [21:04] awilkins: It runs on hardy just fine - I ran it in a Hardy chroot for quite some time and the servers are running Hardy. [21:04] looking into it now [21:04] awilkins, at the site, https://dev.launchpad.net/Getting, it says the following: "Note that right now, Launchpad can only be built and run on Ubuntu (8.04 "Hardy" to 9.04 "Jaunty")." [21:04] Ok, fine [21:04] Ah, they updated that, I think [21:05] It said Jaunty a while back - pages are in flux :-0 [21:05] Or I'm just going mad [21:05] mthaddon: Interesting. I just got this from sucking down RF get: "Got a 200 response when asking for multiple ranges, does your server at bazaar.launchpad.net:80 support range requests?" [21:05] I hope so, cos I will try to install it tomorrow [21:06] joey: no idea I'm afraid [21:06] mthaddon: yeah, I think that might need to go to jml / thumper / mwhudson [21:06] morning [21:06] Last Q: Will it be possible to have all programming languages in it? [21:07] oh hi mwhudson [21:07] No1Viking: in what sense? You mean to host a project that's written in any language? definitely, yes [21:07] No1Viking: I don't understand the question. [21:07] No1Viking: Files are files. [21:07] MarkusT1: what happens if you run: bzr info lp:~launchpad-pqm/bzr-git/devel/ [21:08] mthaddon, thanks för the answer [21:08] sure [21:08] joey: proxy server in the way? [21:09] bigjools: none. I'm doing -Dhpss to see if I get anything interesting [21:09] joey: might be a transparent one [21:09] Good luck all. I'm happy with Ubuntu and I have it here in my house and we're getting it at work too. Thanks for your help all of ya! [21:09] bigjools: never stopped me before :-) [21:10] joey: the bzr guys were saying earlier you get that error with bad proxies, anyway [21:10] awilkins: updated launchpad-developer-dependencies, ran and link-external-sourcecode and rf-get again. Problem persists. [21:10] bigjools: http://pastebin.org/3457 [21:10] joey: what are you trying to do? [21:10] mwhudson: get rocketfuel-get the normal user way via the instructions on the wiki [21:11] joey: what command are you running? [21:11] flacoste: can you check MarkusT1's paste, http://pastebin.org/3454 [21:11] mwhudson: as posted on the wiki plus debug "bzr -Dhpss --no-plugins cat http://bazaar.launchpad.net/~launchpad-pqm/launchpad/devel/utilities/rocketfuel-setup > rocketfuel-setup" [21:11] I've no idea wtf is wrong [21:11] mwhudson: see https://dev.launchpad.net/Getting [21:11] joey: -Dhpss isn't going to help much for a http url [21:12] joey: you could try -Dhttp though [21:12] the reason I'm doing this, btw, is just to QA the wiki :-) [21:12] mwhudson: I'll give it go [21:12] MarkusT1: did you fix rocketfuel-get? [21:13] MarkusT1: there is a typo in the script [21:13] MarkusT1: at least some versions of it [21:13] Well, after the first install failed, I deleted everything and downloaded again. I didn't fix anything by myself. Where do I find information about this fix? [21:14] MarkusT1: you got the tar ball? or the branch directly? [21:14] I'm just following the wiki [21:15] flacoste: https://dev.launchpad.net/Getting [21:16] MarkusT1: using "Getting It" or "Alternative Sourcecode Download"? [21:16] MarkusT1: you are running rocketfuel-setup, right? [21:16] flacoste: Getting it. Yes, I'm running rocketfuel-setup. Two times now. (Well, in fact I ran it more often to check if it solves problem "by itself" :-)) [21:16] mwhudson: thinking of bigjools's comment, I also removed all my pre-existing locations info. I still get http://paste.ubuntu.com/223833/ with no "real" error [21:17] bzr 1.17 as well [21:18] MarkusT1: ok, hang on, the fix is being commited to devel as we speak [21:18] flacoste: Wow, great. What should I do once it's commited? [21:19] MarkusT1: once mthaddon gives us the go, you should just cd to ~/launchpad/lp-branches/devel and do a bzr pull [21:19] that should update utilities/rocketfuel-get [21:19] and you should be able to run rocketfuel-setup once again [21:19] (without deleting anything) [21:19] mwhudson: when I got to the URI in the wiki page, I get a not-found from codebrowse. http://bazaar.launchpad.net/~launchpad-pqm/launchpad/devel/utilities/rocketfuel-setup [21:20] still in progress, shouldn't be too long [21:20] joey: yes, that's expected [21:21] joey: can you paste a bit more of .bzr.log ? [21:22] flacoste: revno 8971 of devel [21:22] mthaddon: thanks [21:22] is bzr branch lp:launchpad supposed to take so long [21:22] ? [21:22] it's a pretty big branch, yeah [21:22] mhall119|work: yes, there is about 150M of data and it's really loaded now [21:22] wow [21:22] mwhudson: http://paste.ubuntu.com/223842/ [21:23] okay, well I'm at 90M now, so may as well finish [21:23] is that all code, or what? [21:23] joey: that's really strange [21:23] EdwinGrubbs: ping [21:24] sinzui: pong [21:24] mwhudson: just thought of something... I have some ssh config settings [21:24] mwhudson: let me remove the LP ones [21:24] EdwinGrubbs: have you seen any floating radio button when you tested with IE 8? https://answers.edge.launchpad.net/launchpad/+question/77504 [21:24] joey: this is over http, ssh really shouldn't matter... [21:25] mwhudson: yeah :-) [21:25] sinzui: that question doesn't mention radio buttons. It says "for an unknown raison a button", so it's probably what I fixed. [21:26] Which bug number is it so that I can link it to the question? [21:26] MarkusT1: can you try again? [21:26] mwhudson: I'm making you a new full Dhttp version of bzr.log [21:27] MarkusT1: so bzr pull followed by a rocketfuel-setup again [21:27] sinzui: I linked it and added a comment. [21:27] thanks [21:28] flacoste: mthaddon: You're really great, guys! Pulled the update and now rocketfuel-setup is actually branching the missing files. Will see if "make schema" run's this time. Thank you so much for updating the tools so fast! [21:28] EdwinGrubbs: feedback@ emails are the only thing I have not done. [21:28] well and I never answered any IRC questions [21:28] MarkusT1, :) [21:28] Ursinha: Thank you, too! :-) [21:28] sinzui: I did those already. There was only one email that wasn't one of your project processing emails. [21:28] MarkusT1, no problem :) [21:29] mwhudson, ping [21:29] mwhudson: full log http://paste.ubuntu.com/223845/ [21:29] sinzui: it really seems like feedback@ should use RT. [21:29] EdwinGrubbs: Then you wont have any CHR work untul next month [21:36] rockstar: hi [21:37] sinzui: thanks alot [21:37] mwhudson, do you have access to loggerhead logs? [21:38] rockstar: yes, but if it's the problem with annotate it's fixed in rocketfuel already [21:38] mwhudson, okay, so it'll be out in the rollout? [21:40] flacoste: was the fix to RF related to launchpad-developer-dependencies? [21:40] s/rf/rocketfuel-get [21:40] joey: no, that was a separate issue [21:40] flacoste: ok but it's know. cool [21:40] I'll work around it then [21:40] joey: what's the issue? [21:41] flacoste: I got "Package `launchpad-developer-dependencies' is not installed and no info is available." from rf-get but the pkg is there and I can install it by hand. [21:41] rockstar: loggerhead has edge-style rollouts now, around about now i think [21:41] joey: that's a new one [21:42] flacoste: I think it's not a blocker, just a confusing situation. http://paste.ubuntu.com/223857/ [21:43] joey: that's really odd [21:43] joey: are there any proxies lurking? [21:43] (i don't see any in the headers so they would have to be pretty evil ones) [21:44] mwhudson: that's what bigjools suggested. I'm direct connect to real-world and afaik there shouldn't be any [21:44] mwhudson: but it is possible something on my untangle box is getting involved [21:44] mwhudson, okay, great. [21:44] mwhudson: I have a webfilter :-) [21:44] * joey goes off and tries to whitelist launchpad [21:45] jml: wouldn't it be cooler if bzr-tools-grep did a case-insensitive grep? [21:46] joey: oh, well that's a little suspicious === _thumper_ is now known as thumper [21:46] morning [21:47] joey: can you run this command: [21:47] curl --silent --trace-ascii log.1 --range 10-20,20-30 http://bazaar.launchpad.net/~launchpad-pqm/launchpad/devel/.bzr/repository/indices/c4f5f44fe87847349a73c775afac3555.iix > log.2 [21:47] sinzui, on ie8 it shows all the boxes, as if there's no hide/show arrow [21:48] * Ursinha finally managed to make vmware work without borking everything else [21:48] mwhudson: done. log.1 is log, log.2 is bin [21:49] Ursinha: EdwinGrubbs ^ is IE 8 unsuable then? [21:49] mwhudson: [21:49] http://paste.ubuntu.com/223860/ [21:49] sinzui, no no, it's possible to use it because it apparently behaves as if js was disabled [21:49] let me take a screenshot [21:50] awww stupid me [21:50] looking in the wrong field :( fail [21:50] thumper, we have a call in 10, right? [21:50] rockstar: yes we do [21:50] well, but the answer wasn't wrong after all [21:50] joey: i think something must be dicking with your request en route :( [21:51] hi [21:51] barry: EdwinGrubbs: Ursinha: can either of you ensure that the the inline edit widget is not shown in IE since it does not work [21:51] thumper, okay, hadn't seen you around, just making sure everything was still koshery. [21:51] sinzui, it just doesn't work as expected, but you can type in the languages with text [21:51] Ursinha: what happens? [21:51] sinzui, it's not shown, yes [21:52] rockstar: I said "morning" :) [21:52] scuse me, but make schema return a error [21:52] rockstar: probably lost in the noise [21:52] utilities/shhh.py python2.4 bootstrap.py --ez_setup-source=ez_setup.py \ [21:52] --download-base=download-cache/dist --eggs=eggs [21:52] utilities/shhh.py make -C sourcecode build PYTHON=python2.4 \ [21:52] PYTHON_VERSION=2.4 LPCONFIG=development [21:52] make[2]: *** [download-cache] Erreur 1 [21:52] make[1]: *** [build] Erreur 2 [21:52] Ursinha: do you have a screen shot, or a vm i can (attempt to) attach to? [21:53] barry, let me check again [21:53] Misterniark: your error is way up higher than that, unfortunately [21:53] somebody have had this probleme [21:53] barry: sorry ? [21:54] Misterniark: sorry, i wasn't clear. ;) look higher up in your output to find the actual error [21:54] Misterniark: get the entire output and pastebin it [21:55] If any of the launchpad devs are here, I want to congratulate you on the announcement. I think this is as good as any of us could have hoped. [21:55] parolang: We're happy you liked it. Isn't that right kfogel? ;-) [21:56] joey: I'm technically away having food, but yes, absolutely right :-). [21:56] barry: thank for your help http://pastebin.com/m79581ae8 [21:57] joey: (actually, sending the rollout downtime announcement first -- food next :-) ) [21:57] mwhudson: I don't seem to get that except on that command. rf-setup seems to have run fine (despite two non-blocking issues) and rf-get is running at full speed. [21:57] mwhudson: heh, spoke too soon [21:58] barry, sinzui, okay, so here's what happens [21:58] barry, sinzui, you click in the exclamation icon, it redirects you to the +edit page [21:58] Misterniark: i'm sorry, but that's not enough to figure out what's going wrong :( one thing to try is "make SHHH= build" and capture all the output [21:59] Ursinha: so that could mean several things [21:59] well, hmm. it seems like the edit icon isn't hooked up to the javascript [22:00] though i don't know why that would be [22:00] herb: So did the lp canonical irc channel move here, or is it still live? [22:00] barry: thanks again http://pastebin.com/mc2a00bd [22:00] barry, oh, isn't hooked up to the javascript? [22:01] Ursinha: well, if it were, it would let you edit inline, so ie is doing something dumb here ;) [22:01] thumper: ping [22:01] Misterniark: did you run utilities/link-external-sourcecode? [22:01] abentley: pong [22:01] DoctorMO: this is the dev channel for launchpad. some of the canonical-specific operational issues will probably stay on the lp canonical irc channel. [22:01] thumper: chat? [22:02] barry, this is how ie has behaving for some time with our inline edits [22:02] abentley: just about to have a scheduled call with rockstar [22:02] thumper: Okay. [22:02] Ursinha: ah, so it's not a problem specifically with the inline language editor, but a more general problem with lazr-js? [22:03] barry: link-external-sourcecode : http://pastebin.com/m2975bdaf [22:03] barry, maybe. for instance: if you click in the icon to edit the status of a bug, it redirects you to +edit-status page [22:04] Misterniark: something's wrong with your checkout. did you run rocketfuel-setup, and did it succeed? [22:04] barry: sorry but i dont understand anythink [22:04] barry: yes ! [22:04] huh [22:04] weird [22:04] herb: Ah like the deployment and such, makes sense. [22:04] barry: an error on python -minimal at start [22:05] +editstatus, sorry [22:05] Ursinha: yep, that sounds like a more general lazr-js bug [22:05] barry: but ok after apt-get install -f [22:06] Misterniark: hmm. maybe you're missing some dependencies? [22:06] Misterniark: i haven't seen this problem before :( [22:06] herb: thanks :-) [22:06] DoctorMO: sure. you're welcome. [22:06] barry: argg [22:06] Misterniark: what is the "python -minimal" error? === salgado is now known as salgado-afk [22:07] herb: And well done on the release, I just got the code myself as I've been busy writing this sys-admin course. [22:08] Misterniark: do you have the python-dev package installed? you'll definitely need that [22:08] DoctorMO: I'm just an ops guy. :) Thank the other folks here for doing the hard work. [22:08] sinzui, barry, so, do you want me to file a bug about it? [22:08] herb: Well, you all get credit :-D [22:08] Ursinha: yes please [22:08] ok, a minute [22:08] http://pastebin.com/d4c5fb583 [22:09] Ursinha: yes, i think so. on lazr-js project [22:09] herb: Although if you don't feel like you've done enough community work ;-) I'm sure peer reviewing my course would be most welcome :-D [22:09] Misterniark: i have to hand you over to one of the bzr guys at this point. maybe thumper jml or mwhudson knows what that problem is [22:10] barry: thank you [22:10] barry: sorry for your time [22:11] uhh [22:11] that doesn't look right [22:11] * mwhudson tries for himself [22:11] Misterniark: no worries. i've been working for almost 14 hours now, so i think i'm done for the day ;) [22:11] DoctorMO: I'm not sure one can do enough community work. But the day job is keeping me busy lately. === barry is now known as barry-away [22:12] bye barry thank you [22:12] Just a coincidence that it's busy during this time. Anyway it won't be complete for a few more weeks. [22:13] OMG, it's actually working! Thanks again, folks! There was just one thing left: I needed to follow https://dev.launchpad.net/DatabaseSetup, since the "make schema" obviously didn't like my psql setup and my user was missing. But nevertheless, IT'S WORKING! Thanks again! :-) [22:14] jono: morning [22:14] MarkusT1: great! [22:14] jono: boy do we need you here :) [22:14] Misterniark: did you run rocketfuel-setup? or at least utilities/link-external-source? [22:14] thumper: community problems? [22:15] DoctorMO: well, jono is a funny guy, and we like fun here :) [22:15] thumper, hey, in a meeting now, back soon - happy to help :) [22:16] flacoste: yes , but it seems i have a probleme with python [22:16] Misterniark: what problem? [22:16] thumper: http://raisegrate.deviantart.com/art/Pirate-Penguin-Goes-Shopping-130282723 [22:16] i try to reinstall but apt say to me " no probleme buddy ! " [22:17] what did you try reinstalling? [22:17] flacoste: http://pastebin.com/d4c5fb583 [22:18] Misterniark: try going to ~/launchpad/lp-branches/devel and run bzr pull [22:19] then try again [22:19] oki i try thank you [22:20] flacoste: No revisions to pull. [22:21] Misterniark: what do you have in ~/launchpad/lp-sourcedeps/sourcecode? [22:21] flacoste: it seems like it is a bzr-git issue, but not sure [22:22] well, it might [22:22] but i want to make sure it's not an artefact of the rocketfuel-get screw-up [22:22] it treated all branches as an optional branch [22:22] and stacked them on devel [22:22] flacoste: yes i have [22:23] which of course would screw durwhich [22:23] Misterniark: i don't get that error, is there a possibility you have a "transparent" proxy between you and launchpad [22:23] ? [22:23] Misterniark: you mean, you have many directories in there? [22:23] mwhudson: i suspect it's an artefact from the rocketfuel-get error [22:23] flacoste: oh, i didn't see that [22:23] flacoste: notes and testes [22:24] Misterniark: ? [22:24] Misterniark: can you pastebin ls ~/launchpad/lp-sourcedeps/sourcecode [22:24] mwhudson : hello , proxy between me and launchpad ??? [22:24] Misterniark: yes [22:24] Misterniark: for instance, i have one thanks to my isp [22:24] flacoste: http://pastebin.com/d6d87f01 [22:25] (luckily they seem to have fixed it recently, and it doesn't break so often any more...) [22:25] Misterniark: you only have bzr-git i assume? [22:25] Misterniark: there are no other directories in ls ~/launchpad/lp-sourcedeps/sourcecode [22:26] Misterniark: is that right? in that case, simply delete bzr-git and try running rocketfuel-get again [22:29] flacoste: i have some error in output of rocketfuel-get [22:29] pastebin? [22:29] http://pastebin.com/d357e5db8 [22:30] it so big to have entier output in my termianl [22:30] mwhudson, bigjools - http://n1.netalyzr.icsi.berkeley.edu confirmed a transparent proxy somewhere on my line [22:30] joey: "transparent" heh heh heh [22:31] i get this error when running make schema: psql: FATAL: Ident authentication failed for user "bbo01" , i guess it's got something to do with Postgres authentication? [22:31] sinzui, barry-away, bug 402736 [22:31] Malone bug 402736 in lazr-js "Buttons seem not hooked up to js in IE8" [Undecided,New] https://launchpad.net/bugs/402736 [22:31] Misterniark: can you do this: cd /tmp; bzr get lp:~launchpad-pqm/dulwich/devel dulwich [22:31] i can give the complete output if needed [22:32] beeman_nl: you need to do things to your postgres install [22:32] sinzui, I thought we were disabling js magic on ie8, and that was on purpose [22:32] beeman_nl: have you seen https://dev.launchpad.net/DatabaseSetup ? [22:32] mwhudson: oke, thanks. it was not on the system before launchpad install, so it's probably not configured [22:33] mwhudson: nope :) i just ran the commands from https://dev.launchpad.net/Getting [22:33] beeman_nl: i guess we should mention something about this on there then! [22:33] i'll check the db setup page :) [22:34] mwhudson: and after ? [22:35] Misterniark: did it work? [22:36] Misterniark: if it completed successfully then something a bit strange is going on [22:37] same error [22:37] Misterniark: ah, ok [22:37] i start to zer0 tomorrow [22:37] mwhudson: it now continues, i ran "launchpad-database-setup $USER" from the utilities directory, so it's a good idea to mention it on that page :) [22:37] so stange [22:38] thank for helping [22:38] its night for me, so good night [22:38] good night Misterniark [22:39] beeman_nl: https://dev.launchpad.net/Getting?action=diff&rev2=18&rev1=16 [22:39] :) [22:39] Misterniark: np, hope it works better for you tomorrow [22:39] mwhudson: that's great, thanks :) [22:39] mwhudson: thank you [22:42] Ursinha: that is what I thought too. but I see there are IE 8 users reporting bugs. Since IE 8 was untested for that feature I asked you to test, I thought it best ti find out [22:42] sinzui, I see [22:43] Ursinha: EdwinGrubbs investigate 1 issues yesterday that turned out to be 3. I think that since IE is buggy, we cannot easily turn it off [22:43] hm [22:46] Might be a dump question, but is there a kind of superuser-login for the local branch? I can't find any references in ./config? [22:47] MarkusT1: do you mean you want to login as the test admin? [22:47] sinzui: Yes [22:47] MarkusT1: foo.bar@canonical.com:test [22:47] “Measuring programming progress by lines of code is like measuring aircraft-building progress by weight.” <- good quote [22:48] sinzui: Thanks! [22:48] from bill gates of all people?? [22:48] MarkusT1: I think it is in one or two README files in the doctest or pagetests directory [22:48] I once measured my wife in light years [22:49] I think I also gave her weight in liters [22:49] sinzui: please use admin@canonical.com:test [22:49] sinzui: using foo.bar is obscure [22:49] sinzui: which is why I added that "admin@canonical.com" email for Foo Bar [22:49] sinzui: Yeah, as I said: This might be a dump question. :-) I'm still struggling with all the files and directories :-) [22:49] thumper: I will. I'll give an rs to the person that fixes it [22:50] God, "dump". I better get some sleep.... [22:50] MarkusT1: I understand. The directly structure is better, but not good yet [22:51] ILaunchpadCelebrities... [22:51] sinzui, do you want me to test on ie6 and 7? [22:52] thought: shouldn't "Launchpad" be one of the featured projects on the launchpad.net homepage? :) [22:53] Ursinha: no, I am sure they are also broken. We need to be more certain that we are not giving IE an AJAX experience, or very certain it works for them [22:53] sinzui, okay [22:53] greg-g: It may be when we release in a day [22:53] sinzui: cool [22:54] lol [22:54] greg-g: that's awesome [22:54] joey: :) [22:54] figuted it was something to feature [22:54] I'm going to do that right now. :-) [22:55] ooh bug [22:57] beuno: I think this one is for you. https://bugs.edge.launchpad.net/launchpad/+bug/402747 [22:57] Malone bug 402747 in launchpad "The requested URL /@@popup-window was not found on this server." [Undecided,New] [22:57] Ursinha: matsubara - could one of you triage that for me pretty please. ^^ [22:57] flacoste: fyi ^^ [22:57] joey, sure [22:57] shouldn't be an issue anywhere 'cept there methinks [22:58] greg-g: your wish, my command. launchpad.net/ [22:59] joey: tight! [22:59] joey: nit-picking now: the 2 columns are now uneven by 2, thus could be exactly even. :) [22:59] greg-g: yeah, filing a bug about that :-) [23:00] greg-g: it's actually somewhat of a hack if you look at the code. [23:00] greg-g: when that has happened before we sometimes take off some of the older featured projects [23:00] joey: oh, wow, interesting. === awilkins is now known as nop === nop is now known as awilkins [23:03] greg-g: https://bugs.edge.launchpad.net/launchpad/+bug/402749 [23:03] Malone bug 402749 in launchpad "featured projects list columns are uneven" [Undecided,New] [23:04] kfogel: ^ I think that bug is your call [23:05] Has Rosetta (https://launchpad.net/rosetta) been released as part of Launchpad going GPL? [23:06] sinzui: yeah, I don't have access to the code atm (downloading) but I think there's a number there that tells us to split the display. [23:06] sinzui: I think *I* may even be responsible for it. [23:07] sinzui: 402749? I mean, I'm all for having an evenly divided list. Have no idea where that code is... [23:07] joey: The new design may have a fix for that problem. Well I tested the homepage with the new rules and they worked [23:07] oh cool sinzui [23:08] firefly2442: yes [23:08] kfogel: I believe someone with the right power needs to add or remove a project from one of the lists. I thought that power was conveyed to you [23:08] Morning. [23:08] sinzui: might well have been, not sure; can check [23:08] wgrant: morning! [23:08] kfogel: somewhere near lib/canonical/launchpad/templates/root-index.pt [23:09] *nod* [23:09] iirc [23:09] okay, really really food now :-) back later [23:09] sinzui: yeah it was unbalanced by one before. I added one just now and it's listing :-) [23:11] symmetry is overrated [23:11] * pkern is said that LP doesn't ship with a GPG key included. [23:11] *sad. Because GPG key import doesn't work out of the box. ;-) [23:11] pkern: But you have psql.... [23:12] thumper: bzr+ssh://bazaar.launchpad.net/~abentley/launchpad/notify-oops [23:15] wgrant: True [23:21] thumper, my internets - they eated themselfs [23:21] wgrant: How do I run something in that custom python2.4 env? [23:22] wgrant: I can hack sys.path in the script itself or supply a huge PYTHONPATH but both sounds wrong. [23:22] pkern, make harness === flacoste is now known as flacoste_afk [23:23] thumper, are you done, or can you call again? [23:23] bin/py, too. [23:24] * pkern chuckles at bin/py [23:24] Is LP going to support wiki any time soon? [23:24] I saw the Trac migration kit and thought "cool, wiki" [23:24] But it's just bugs :-( [23:25] barry-away, re bzr-tools -- I'm not sure, I've relied on the case sensitivity before [23:25] What kind of requirements does LP need to run decent? [23:25] MT-, check the topic :) [23:26] Ursinha: ? [23:26] MT-, https://dev.launchpad.net/Getting [23:28] Ursinha: that doesn't mention requirements though [23:29] awilkins: I believe it's been discussed, but not sure if there are firm plans [23:29] rockstar: we are done [23:30] thumper, okay. I missed like 10 minutes of that meeting though. Anything interesting? [23:30] rockstar: it ended a while ago [23:30] rockstar: you really didn't miss a lot [23:30] mthaddon: There's a bug ( #240067 ) but it's triaged [23:30] wgrant: It looks you need to put it into some keyring though. [23:30] mwhudson, okay. I assume just thumper waxing philosophical? :) [23:31] thumper, abentley, rockstar, mwhudson: sorry! [23:31] mthaddon: It's a "checklist feature" IMHO [23:31] yep [23:31] jml: morning [23:31] mthaddon: Not that it's a commercial product (although I can see a lucrative commercial support market) [23:31] jml, morning! [23:32] MT-, so I guess it's missing somehow in the docs, but you can file a bug against launchpad-documentation project mentioning that [23:32] jml: morning! [23:32] as your first contribution to lp now opensource :) [23:32] Dammit, now I can't tab complete jml's name so easily anymore. [23:32] sorry. [23:32] lol [23:32] Ursinha: :P - will do [23:32] thanks MT-! [23:33] pkern: Hm, indeed. What's broken about the web UI adding process? [23:33] rockstar, I've occasionally thought of assuming my old handle of mumak, but the rebranding would cost too much. [23:33] wgrant, good morning. [23:34] jml: Morning! [23:34] jml, mumak would work well for my tab completion needs at the moment. [23:35] rockstar: Sadly, when in his mumak guise, jml has a tendency to trample the forces of good. [23:35] Does Launchpad-FOSS allow us to create private branches? [23:35] gmb, well, I do that more subtly in this guise. [23:35] wgrant: It takes strange things about keyserver.ubuntu.com [23:36] *talkes. Oh crap, it seems to late but I wanted to continue. [23:37] gmb, I guess being a mumak would be farewell to subtlety [23:37] jml: Maybe you could just switch nicks whenever you're having a bad day, so we know not to make matters worse. [23:37] :) [23:38] "omg, quick, fetch the poncy elven archer" [23:38] MT-: if you were running your own instance you could be admin and decide who gets private branches, yes [23:38] jml: Exactly. I'm trying to think who might qualify for that role... [23:38] mwhudson: cool - I think my company could really benefit from this - as well as contribute back to it [23:39] gmb, stub has the hair for it :) [23:39] mthaddon != mwhudson :) [23:39] jml: Mind... boggles... [23:39] [IRunnableJob(job) for job in jobs] [23:40] jml: thank you for that image [23:40] MT-: however, I suspect the overhead of running it in house might be less than the cost of a license to host proprietary branches on launchpad.net [23:41] MT-: er, I mean, might be more... [23:41] mwhudson: I was looking for requirements to run it. bug 402762 [23:41] Malone bug 402762 in launchpad-documentation "Docs don't explain system requirements" [Undecided,New] https://launchpad.net/bugs/402762 [23:42] MT-: I don't know that you're really going to get much of an answer beyond "it depends" or "as much as you can throw at it" [23:42] MT-: please learn the difference between mthaddon and mwhudson ! :) [23:42] MT-: ??? many of the LP Dev's run it on their laptops. that sounds fine to me for a single user. [23:42] sorry... [23:43] spm, but we don't do anything _else_ while we're running it on our laptops. [23:43] I occasionally cook my thighs. [23:43] mwhudson: aren't all the launchpad developers just clones anyway? [23:43] ajmitch: Only if the experiment went tragically wrong. [23:43] would it freak out under 2GB RAM? or could it chug along? [23:44] * spm lol's at the continuing confusion between mwhudson and mthaddon. Michael - you need to grow a beard :-) [23:44] spm: I have one :( [23:44] MT-: that was at mwhudson :-) [23:44] ajmitch, close enough. we're all actually from the South Island of New Zealand, and thus spring from an extremely limited genetic pool :P [23:44] spm: Or was it noodles? [23:44] jml: now that is worrying [23:45] * mthaddon realises wgrant really does know everything... [23:45] I was wondering if I could play with it in a VM if I give the VM 1.5GB RAM and 20GB HD [23:45] Is there a way to browse the Launchpad sourcecode online instead of downloading the whole thing using Bazaar? [23:45] spm: i'm more working on "not shaving" than "growing a beard" [23:45] wgrant: does your brain ever hurt? [23:45] MT-: you can definitely play with it on a regular laptop, but as for running it as a "production" service, that'd be a different story [23:45] The only problem I had on this new laptop is postgres' disk-eating ability. [23:45] I think launchpad FOSS doesn't set up keyserver.internal or the keyserver in a sane way at all. [23:46] my regular laptop has 2GB RAM [23:46] I know some LP servers have over 100GB RAM [23:46] firefly2442: http://bazaar.launchpad.net/~launchpad-pqm/launchpad/db-devel/files [23:46] if you ahve a 32 bit install, 2 gigs should be plenty for playing around [23:46] MT-: that's similar to most LP devs I'm guessing, and they run a local LP instance for development/testing fine (the 2GB RAM, not the >100GB) [23:46] gmb: ty :) [23:47] MT-: I'm running it on 2GHz with 1.5 GB. It's not really fast with this setup (I would not want to use it on a regular basis), but I'm still running the debug environment. Not sure if that slows down that much. [23:47] good grief, 80 people in the channel. [23:47] My new T400 with 2GiB of RAM runs it fairly nicely. [23:47] Mapping keyserver.internal to keyserver.ubuntu.com works. Thankfully it then goes on to send mail to foo.bar@canonical.com (= [23:48] pkern: You altered /etc/hosts to point keyserver.internal at keyserver.ubuntu.com [23:48] can one turn off certain features to save on performance? [23:48] ... 's IP address [23:48] wgrant: Yes. [23:48] I can't test yet - I'm installing Vista in a VM :( [23:50] jono, hi [23:50] ooi, why does rocketfuel-setup request that you disable GSSAPIAuthentication ? [23:51] pkern: minor nitpick: launchpad FOSS is launchpad. we use the *same* code on the prod servers. [23:52] maxb: hm, that's probably not required any more [23:52] maxb: there was a bug that added 10s to the time to negotiate a connection with bazaar.launchpad.net [23:53] * MT- quiets excited self. You all rock immensely. [23:53] ;) [23:53] kiko: love you too :) [23:53] so can I safely ignore those errors like NoSuchRevision: KnitPackRepository('bzr+ssh://bazaar.launchpad.net/~launchpad/pygettextpo/trunk/.bzr/repository/') has no revision ('Arch-1:carlos.perello@canonical.com--2004%pygettextpo--devel--0--patch-5',) ? [23:53] (from running rocketfuel-get) [23:54] ajmitch: Are you using a fixed rocketfuel-get? [23:54] wgrant: no idea, I grabbed this yesterday & have re-run rocketfuel-get a couple of times [23:55] it appears to be up to date [23:56] I don't believe I got those errors. [23:57] it wouldn't surprise me if there was a proxy causing issues for me [23:57] ajmitch: it's probably ignorable [23:57] it's not like pygettextpo has changed in years [23:57] such as SHA1KnitCorrupt [23:57] ajmitch: if you have a proxy affecting your bzr+ssh connections, you have all kinds of problems [23:58] mwhudson: one would think so :) [23:58] So only launchpad is a hot_product at the moment? [23:58] aha, thanks. The expresses preference for apache worker mpm over prefork is also interesting [23:59] wgrant: yes