[06:00] <rozzin> Has Launchpad always required creation of an account before reporting bugs?
[06:02] <hyperair> yes.
[06:03] <rozzin> Hm.
[06:04] <rozzin> I could have sworn that I'd reported an Ubuntu bug without explicitly creating an account beforehand, back in 2008.
[06:04] <rozzin> Though, it wouldn't be the first time I'd just forgotten something.
[06:04] <hyperair> i created my launchpad account in 2006 when i first reported a bug iirc
[06:05] <hyperair> or maybe it was 2007
[06:05] <hyperair> 2007 it was.
[06:05] <rozzin> I do have some trouble convincing myself that I would have chosen the username I ended up with :)
[06:06] <hyperair> heh
[06:07] <wgrant> rozzin: Default usernames are generated from the email address you registered with.
[06:07] <wgrant> You can change your username later.
[06:07] <rozzin> Oh, but I guess you don't actually choose those, do you.
[06:07] <rozzin> I see.
[06:08] <rozzin> well, I guess it all fits together into a convincing story, then :)
[06:09] <rozzin> Thanks.
[06:57] <d1b> hi ah how much memory does one need 2 actually do a bzr branch of the stable launchpad repo?
[07:08] <lifeless> d1b: I'm not sure of the peak rss, but I'd say you want 500MB or maybe alittle more (especially if you are on 64bit)
[07:08] <d1b> lifeless: seems like a lot
[07:09] <lifeless> jam is working on reducing it
[07:09] <d1b> i couldn't find a git mirror :/
[07:09] <lifeless> but the repo has copies of debs and all sorts of stuff in its history
[07:10] <lifeless> and I dunno, 250K commits or something ?
[07:10] <d1b> ok doing a --lightweight this is silly
[07:12] <d1b> there isn't some tar ball / http download ...
[09:03] <maxb> d1b: I think a --lightweight may be sillier in the long run if yoou want to work with the project at all
[09:03] <maxb> d1b: Also, are you using the http or bzr+ssh protocol?
[09:09] <d1b> maxb: i just did a branch --lightweight
[09:10] <d1b> i don't care about re-syncing atm
[09:10] <d1b> just wanted to have a look through and a grep
[11:34] <yofel> hi, quite a few recipe builds are failing here with
[11:34] <yofel> dpkg-deb: error: control directory has bad permissions 700 (must be >=0755 and <=0775)
[11:35] <yofel> https://launchpad.net/~neon/+archive/ppa/+buildjob/2438371 for example
[11:35] <yofel> the debian folder is created by the nest command in the recipe, so that shouldn't be our fault, and it doesn't happen in all builds
[11:35] <maxb> lamont / wgrant : Is that that umask thing you referred to previously?
[11:44] <yofel> another error I'm getting is from https://launchpad.net/~neon/+archive/ppa/+buildjob/2437677
[11:44] <yofel> bzr: ERROR: No previous changelog to take the package name from, and --package not specified: debian/changelog didn't contain any parseable stanzas.
[11:44] <yofel> the changelog seems completely fine to me though, and VIM doesn't show any syntax error either http://bazaar.launchpad.net/~neon/project-neon/kiten-ubuntu/view/head:/changelog
[11:44] <yofel> and kiten failed to build every time so far with the same error
[11:48] <wgrant> maxb: Yes, thanks for the poke.
[11:49] <wgrant> yofel: Do you have links to other builds failing with permission errors?
[11:50] <yofel> probably, let me check the mails I got
[11:50] <wgrant> As for the changelog thing, I don't think the changelog is formatted properly.
[11:51] <wgrant> Let me find
[11:51] <wgrant> it.
[11:51] <wgrant> http://bazaar.launchpad.net/~neon/project-neon/kiten-ubuntu/view/head:/changelog has only one space between the email address and date.
[11:51] <wgrant> That could possibly be it.
[11:52] <yofel> aaah, someone should fix vims syntax highlighting I guess...
[11:52] <yofel> thanks
[11:52] <yofel> found one https://launchpad.net/~neon/+archive/ppa/+buildjob/2437179
[11:53] <wgrant> Maybe it's just that one builder.
[11:53] <yofel> https://launchpad.net/~neon/+archive/ppa/+buildjob/2437619
[11:53] <yofel> seems so
[11:53] <wgrant> I've set it to manual.
[11:53] <wgrant> lamont: I think thallium's lp-buildd needs upgrading.
[11:53] <yofel> thallium again https://launchpad.net/~neon/+archive/ppa/+buildjob/2437633
[11:54] <yofel> and 5 more failures, all on thallium
[11:54] <wgrant> If you see any non-thalliums, poke me and I'll kill them too.
[11:54] <yofel> sure
[13:28] <maxb> Gah
[13:29]  * maxb discovers yet another separate page on the lp wikis about reviewing code imports
[15:40] <rozzin> Hm.
[15:41] <rozzin> Isn't there some concern that requiring a launchpad account for bug-reporting actaully prevents some people from filing bugs?
[15:53]  * exarkun isn't concerned about that
[15:53] <exarkun> if you're concerned about it, then I guess there is some concern
[15:56] <rozzin> Well, yes--I'm concerned :)
[15:57] <ScottK> Personally, I'd rather make it harder.  We get more than enough useless bugs that don't help improve things.  They are a distraction from getting done what needs doing.
[15:57] <exarkun> I guess that's settled, then.
[16:02] <rozzin> ScottK: I don't have that problem, so far.
[16:03] <ScottK> Ubuntu certainly does as a whole.
[16:03] <rozzin> ScottK: If you do, congratulations may be in order :)
[16:03] <ScottK> Ubuntu is a very big project.
[16:03] <rozzin> ScottK: Yeah, I can see that for Ubuntu.
[16:04] <ScottK> I do maintain some small projects on LP and I seem to get bug reports OK.
[16:04] <exarkun> Indeed
[16:05] <exarkun> I don't have any projects on Launchpad which gets fewer bug reports than I am actually able to deal with
[16:05] <rozzin> A coworker of mine raised the issue to me in regards to SourceForge, after they had their compromise and reset all of the passwords;
[16:06] <rozzin> He said, "Well, I guess I'm done filing bug-reports for projects on SourceForge."
[16:08] <exarkun> You can't make everyone happy.
[16:08] <doctormo> I'm having trouble uploading to my ppa, it tells me that my changes file isn't signed. Even though using debuild -S I've never had a problem before.
[16:10] <ScottK> doctormo: Check and see if the package actually got uploaded or not.  IIRC someone else had that last night and it turned out the warning was in error and the package was accepted.
[16:11] <doctormo> It's a fresh ppa: https://launchpad.net/~doctormo/+archive/greeter
[16:11] <doctormo> I tried maybe an hour ago, so nothing so far.
[16:11] <rozzin> Then there's my user who decided that he'd rather just `file bugs' on IRC :\
[16:13] <doctormo> ScottK: Ah turns out I mistyped and did "ppa:~doctormo/greeter" with the extra ~, works now.
[16:13] <doctormo> Although the error was erroneous.
[16:17] <JanC> rozzin: I know projects that have a bug-reporting IRC bot...  ;)
[16:19] <JanC> IIRC that was a supybot plugin that inserted bugs into a trac bug tracker
[16:20] <JanC> (but it also required an authenticated user I suppose)
[16:22] <doctormo> JanC: You need my issue tracker, it's a mediated design between issues and bugs.
[17:16] <rozzin> If I wanted to `raise the bar and keep the stupids out' of my project, I'd be using git.
[17:21] <rozzin> Yow--yeah, this is basically the exact opposite of what I want: http://article.gmane.org/gmane.comp.bug-tracking.ditz.general/16
[17:24] <d1b> rozzin: wait what?
[17:29] <rozzin> d1b: "One of the big advantages of ditz is that it raises the bar for feature requests by requiring you to check out a repository and commit code."
[17:30] <d1b> rozzin: ok....
[17:31] <rozzin> "If you're not a developer comfortable with the VC system, I'm not interested in anything you have to say" is basically the opposite of what I want.
[17:32] <d1b> oh ok
[17:32] <rozzin> I'm going to tell myself that he's joking, so as to keep that from ruining my day.
[17:32] <rozzin> I'm not sure that he is, though.
[17:33] <d1b> why not let people email in patches
[17:33] <rozzin> er, "... is basically the opposite of what I want to tell my users."
[17:33] <d1b> if they dislike ur vc ssytem
[17:34] <rozzin> It's not a question of whether they dislike the VC system,
[17:36] <d1b> s/dislike/failsatvcs/
[17:36] <rozzin> because that's a second-tier barrier where an end user should never even have to deal with the first-tier one ;)
[17:37] <rozzin> Plenty of useful input comes with people who don't even have "preparing a patch" (or "writing code"!) as part of their skillset;
[17:38] <rozzin> e.g., because they spent their time working on a different skillset--like English grammer, or translation, or UI design, or whatever.
[17:39] <rozzin> Er, "English *grammar*"--see what I mean? :)
[17:39] <d1b> ha!
[17:39] <d1b> ./sleep &
[17:41] <rozzin> There was an interesting presentation I came across, actually,
[17:41] <rozzin> where the first 2 slides were something like:
[17:42] <rozzin> "Bazaar is version-control system for rocket scientists"
[17:42] <rozzin> "because rocket scientists have better things to think about than version control."
[17:46] <rozzin> JanC: Do people use bug-bots w/ Launchpad?
[17:48] <rozzin> JanC: It sounds like something that'd basically be circumventing site policy....
[17:59] <JanC> rozzin: I have seen no bots that can report bugs on LP
[21:22] <micahg> I seem to be having trouble uploading
[21:24] <micahg> http://pastebin.ubuntu.com/592321/
[21:29] <lifeless> micahg: what host
[21:29] <lifeless> micahg: actually, please file a bug
[21:30] <micahg> lifeless: ugh, it seems it was accepted already, but I'm getting this message, I think this is an old bug actually
[21:38] <syn-ack> Quick question.... you guys pushing an update or something to bugs.launchpad right now? I'm getting some weirdness here and I'd like to report it if you're not
[21:39] <lifeless> whats up
[21:39] <syn-ack> I'm getting all sorts of Time out errors
[21:39] <syn-ack> either trying to bring the homepage up or while searching for a project
[21:40] <syn-ack> heres the Error ID: OOPS-1926O1054
[21:43] <lifeless> homepage came up instantly for me
[21:43] <lifeless> whats the search url you had trouble on?
[21:43] <syn-ack> https://bugs.launchpad.net/ and searching for Ubuntu in general
[21:44] <lifeless> whats the *exact* url you had trouble on
[21:44] <syn-ack> I just told you the exact URL I had trouble on
[21:44] <syn-ack> seriously.
[21:45] <syn-ack> And now this one
[21:45] <syn-ack> https://bugs.launchpad.net/bugs/+bugs?field.searchtext=Ubuntu&search=Search+Bug+Reports&field.scope=all&field.scope.target=
[21:45] <lifeless> thats better, let me have a look see
[21:45] <lifeless> I'm syncing the error reports to see whats up with the one you quoted
[21:45] <syn-ack> lifeless, I wasn't trying to be difficult here. Sorry if it came off like that
[21:45] <lifeless> that takes about 10 minutes
[21:46] <lifeless> no worries
[21:46] <lifeless> so, the main sql query is timing out
[21:48] <lifeless> its a 7 secound count, sadly
[21:48] <lifeless> we'll get a detailed report later
[21:49] <lifeless> for now, you're probably best of clicking 'one project' putting 'ubuntu' in there, and then clicking search
[21:49] <syn-ack> k
[21:49] <lifeless> that isn't quite the same thing
[21:49] <lifeless> what I describe is 'all ubuntu bugs'
[21:49] <lifeless> what you searched for was 'all bugs with the word ubuntu in them'
[21:50] <syn-ack> ah
[21:51] <syn-ack> Just got a time out there too
[21:51] <syn-ack> https://bugs.launchpad.net/ubuntu/+bugs?field.searchtext=&search=Search+Bug+Reports&field.scope=project&field.scope.target=ubuntu
[21:51] <syn-ack> Error ID: OOPS-1926U1117
[21:52] <lifeless> we may have bloated indices again
[21:52] <lifeless> will get our dba to look at this when hes around later today
[21:53] <syn-ack> Good deal
[21:53] <lifeless> https://bugs.launchpad.net/ubuntu/+bugs
[21:53] <lifeless> works for me
[21:53] <syn-ack> do you guys cluster your DBs and this particular one that I'm hitting is fouled or what?:
[21:54] <lifeless> we have 3 db servers running LP
[21:54] <syn-ack> ah, yeah
[21:54] <lifeless> but the bug query logic is only partly optimised
[21:54] <lifeless> we're working on it
[21:54] <lifeless> its down from 17-18 seconds per page
[21:54] <syn-ack> ah
[21:54] <lifeless> 6 months ago
[21:54] <lifeless> still got a ways to go
[21:55] <lifeless> its pretty complex
[21:55] <lifeless> and ubuntu has 100000 open bugs
[21:55] <lifeless> which makes many indices a bit useless
[21:57] <syn-ack> yeah
[21:58] <lifeless> we also do this thing where we count all results
[21:58] <lifeless> rather than saying 'lots'
[21:58] <lifeless> so the actual search you have can return a single page pretty quickly
[21:58] <lifeless> but we calculate every possible row so show the count in the UI
[21:58] <syn-ack> Right
[21:59] <syn-ack> So I can understand the timeout then
[21:59] <lifeless> thats also something we're going to change :)
[22:10] <luks_> hi -- https://help.launchpad.net/Packaging/SourceBuilds/Recipes mentions that I can use "nest-part" in recipes, but when trying to do so I get this error "Error parsing recipe:3:1: Expecting 'merge', 'nest' or 'run', got 'nest-part'."
[22:10] <luks_> is there a way to use the "nest-part" command?
[22:13] <lifeless> sounds like we have an old bzr-builder library
[22:13] <lifeless> file a bug ?
[22:20] <luks_> ok, I wasn't sure if I'm just not missing something, thanks
[23:49] <poolie> maxb, i like your api breaks wiki page
[23:50] <poolie> though it's a shame it has to exist, of course