[00:05] <bdmurray> Is there a reason that its possible to report bugs about packages not in Ubuntu?  For example if you go to https://edge.launchpad.net/ubuntu/+source/automake1.11/ it is not available in Ubuntu but a PPA
[00:05] <bdmurray> However, if you click on Bugs then File a Bug you can report a bug about this non-ubuntu package
[00:09] <synic> a package isn't required to be in Ubuntu to make use of launchpad
[00:19] <wgrant> synic: It is to file bugs about it in Ubuntu.
[00:20] <bdmurray> Well it ends up raising an error if you actually go to file the bug
[00:20] <wgrant> Right.
[00:20] <wgrant> I filed a bug about that years ago.
[00:20] <wgrant> And it is still fixed.
[00:20] <synic> oh, heh
[00:20] <wgrant> (in that you can't actually file a bug)
[00:20] <wgrant> Odd that the button shows up, though.
[00:21] <bdmurray> well and the bug listing for Ubuntu shows up too
[00:21] <wgrant> Bug #157342
[02:54] <mhall1191> okay, my first attempt at both packaging and uploading to a PPA
[02:54] <mhall1191> I got the packaging working, and I think I dput okay
[02:54] <mhall1191> I get: Successfully uploaded packages.
[02:54] <mhall1191> Not running dinstall.
[02:55] <mhall1191> but still no package in https://launchpad.net/~mhall119/+archive/ppa
[03:01] <hggdh> mhall1191, it takes a bit for them to appear
[03:01] <mhall1191> oh, so I'm just being impatient?
[03:02] <wgrant> mhall1191: You need to wait up to 5 minutes.
[03:02] <wgrant> If it's not there after 5 minutes, and you don't have an email from Launchpad, you probably didn't sign the package properly.
[03:02] <mhall1191> so, it launchpad building the binary?
[03:02] <wgrant> You'll get an email before it starts doing that.
[03:02] <mhall1191> it says the signature is good
[03:02] <wgrant> The key is associated with your Launchpad account?
[03:03] <mhall1191> okay, got an email: Unable to find distroseries: unstable
[03:03] <mhall1191> wgrant: yes
[03:03] <wgrant> So, your changelog says 'unstable'. But that's not an Ubuntu release.
[03:04] <mhall1191> okay, must be from reading debian's documentation
[03:04] <mhall1191> what would it be for Ubuntu, "intrepid" or "jaunty"?
[03:05] <mhall1191> something like that?
[03:05] <wgrant> Exactly.
[03:06] <mhall1191> okay, rebuild and try again, thanks
[03:08] <mhall1191> dput -f to make it upload again?
[03:08] <wgrant> Yes.
[03:09] <mhall1191> okay
[03:09] <mhall1191> trying it again
[03:10] <mhall1191> qimo-usplash_1.0.0.dsc: Unknown section 'X11'
[03:10] <wgrant> x11, not X11.
[03:11] <mhall1191> thanks
[03:11] <mhall1191> noob mistakes
[03:11] <mhall1191> I appreciate the help though
[03:11] <wgrant> np
[03:16] <mhall1191> woo hoo, Accepted!
[03:16] <mhall1191> thanks again wgrant
[03:16] <wgrant> mhall1191: Excellent.
[03:16] <wgrant> Now, let's see if it builds...
[03:17] <mhall1191> https://launchpad.net/~mhall119/+archive/ppa shows 1 source package, 0 binary
[03:17] <wgrant> You've hit the build farm at a bad time.
[03:17] <wgrant> Right now there are lots of builds, and not many builders.
[03:18] <wgrant> If you expand the source package row and click on the builds, you'll get an estimate of the starting time.
[03:18] <mhall119|eeepc> sorry my laptop battery was dying
[03:19] <mhall119|eeepc> had to switch
[03:19] <wgrant> Ah.
[03:19] <wgrant> 12:17:48 < wgrant> You've hit the build farm at a bad time.
[03:19] <wgrant> 12:17:57 < wgrant> Right now there are lots of builds, and not many builders.
[03:19] <wgrant> 12:18:27 < wgrant> If you expand the source package row and click on the builds, you'll get an estimate of the starting time.
[03:22] <mhall119|eeepc> :(
[03:22] <mhall119|eeepc> build failed
[03:22] <wgrant> Indeed.
[03:22] <wgrant> Faily obvious fix.
[03:22] <wgrant> Build-Depend on whatever provides pngtousplash.
[03:24] <mhall119|eeepc> ah, okay
[03:24] <mhall119|eeepc> I'll have to find that
[03:26] <wgrant> mhall119|eeepc: You might want to look at https://wiki.ubuntu.com/PbuilderHowto. That will let you test the build-dependencies locally.
[03:27] <mhall119|eeepc> thanks
[05:07] <sanmarcos> does launchpad only support Bazaar as source control?
[05:10] <mwhudson> yes
[05:23] <mrooney> Anyone know if webhooks are a plan / current request for launchpad code hosting? So I can tell launchpad to POST to a given URL on commit like a post commit hook, triggering a build or something?
[05:27] <mwhudson> mrooney: we'd like something like that, but i don't think there's any decision made about what the mechanism would be
[05:27] <mwhudson> (maybe XMPP?)
[05:28] <mrooney> The general solution seems to be an HTTP POST because of its simplicity and how well understood HTTP already is
[05:28] <thumper> mrooney: file a bug on launchpad-code
[05:28] <mrooney> plus plenty of things already listen on http and can perform actions over them
[05:29] <mrooney> I haven't looked at github but I know it has that feature, I think it uses HTTP
[05:29] <thumper> mrooney: provide examples if you can
[05:29] <mrooney> thumper: okay. thanks!
[05:32] <micahg> I got an error on uploading to my ppa that 2 original source tarballs are different
[05:33] <micahg> if I delete the current package in tehre, will it solve the problem?
[05:34] <wgrant> micahg: Possibly, but that's a very bad idea.
[05:34] <wgrant> micahg: It might solve it after a day or two, but you shouldn't do that sort of thng.,
[05:34] <micahg> ugh
[05:34] <wgrant> Why would the orig.tar.gz differ?
[05:35] <micahg> idk
[05:36] <wgrant> By definition they should not differ.
[05:36] <wgrant> Or they are not the orig.tar.gz.
[05:36] <micahg> 1 know
[05:36] <micahg> I know
[05:36] <micahg> right
[05:36] <micahg> I think the one I uploaded originally is not the original
[05:36] <micahg> I got it directly from the source instead of from ubuntu
[05:37] <micahg> wgrant: here's what I got
[05:37] <micahg> http://pastebin.ubuntu.com/238145/
[05:37] <wgrant> If the Ubuntu one is different from upstream but has the same version, there is something wrong.
[05:38] <wgrant> (not unlikely, and it does happen, but it shouldn't)
[05:39] <micahg> so, what should I do now
[05:39] <micahg> I checked the help page and it says to regenerate the tarball
[05:39] <wgrant> Work out what the difference is
[05:39] <wgrant> If you really do need the new one, change the version number.
[05:40] <wgrant> (eg. add +repack1 if it's just a repack, or +dfsg1 if it's removing some non-free stuff)
[05:42] <micahg> I'm pulling in down now
[05:43] <micahg> is there a command that will spit out the differences between tarballs?
[05:44] <wgrant> I'd just unpack them and diff -Nur
[05:46] <micahg> ah
[05:46] <micahg> I found it
[05:46] <micahg> the one I built has -orig appended to the folder
[05:47] <wgrant> This is why we grab the tarball from upstream.
[05:47] <micahg> I did :)
[05:48] <micahg> ok
[05:48] <micahg> so what to do now?
[05:49] <wgrant> Who has the right tarball?
[05:49] <wgrant> Ubuntu or you?
[05:49] <wgrant> And which one is already in your PPA?
[05:50] <micahg> Ubuntu and mine
[05:50] <wgrant> you cannot both have a different right tarball.
[05:51] <wgrant> There is at most one correct tarball in existence, unless upstream is crazy.
[05:51] <micahg> Ubuntu has the right one and mine is in my ppa :)(
[05:51] <lifeless> micahg: if they are both right why are they different?
[05:51] <wgrant> Ah.
[05:51] <wgrant> I see.
[05:51] <micahg> lifeless: in essence both are correct, they contain the same files
[05:51] <wgrant> The tarball content is the same, except for that directory name?
[05:52] <micahg> yes
[05:52] <lifeless> micahg: then they are different; choose one and use it.
[05:52] <rafiu> ping
[05:52] <wgrant> rafiu: Failed.
[05:52] <rafiu> wgrant: send
[05:52] <wgrant> micahg: What is this new thing you are trying to upload with the Ubuntu tarball? I'd probably rebuild it to reference your version of the tarball instead.
[05:53] <micahg> hmmm
[05:53] <rafiu> anyone here familiar with the process for migrating to launchpad from trac once we've converted trac tickets to xml?
[05:53] <micahg> I think I'll just upload without the source :)
[05:53] <wgrant> micahg: Watch out for the hash in the .dsc.
[05:53] <micahg> ah
[05:53] <wgrant> rafiu: You need a Launchpad Bugs developer. They are in Europe, so are hopefully all asleep.
[05:54] <rafiu> asleep!?
[05:54] <wgrant> micahg: You might need to move the Ubuntu tarball out of the way, grab the one from your PPA, then debuild -S again.
[05:54] <rafiu> space invaders not supposed to sleep
[06:07] <micahg> wgrant: it accepted it this time :)
[06:07] <wgrant> micahg: Excellent.
[06:08] <micahg> although, I don't know if it'll even fix the problem I was having, but that's another story :)
[06:09] <mrooney> thumper: okay, I filed bug 407234, here's to hoping!
[06:09] <thumper> mrooney: do you think they should be configurable on a branch by branch basis?
[06:10] <mrooney> thumper: I'm up in the air, I think by project basis is a great simple, first step
[06:10] <mrooney> Plus if it passes the branch name, the listener can conditionally act if desired
[06:10] <thumper> mrooney: you don't really want it submitting for any branch associated with the project, surely
[06:10] <wgrant> By project doesn't make sense.
[06:11] <mrooney> neither makes sense for all purposes, project is the most simple and powerful
[06:11] <wgrant> Project is more complicated.
[06:11] <mrooney> why's that?
[06:11] <wgrant> If it's per-branch, there is no need for a special way to send the branch identifier.
[06:12] <thumper> mrooney: if you are a developer, I'd be happy to mentor this change in
[06:12] <mrooney> if I have a build server I may not want to register a new hook each time I branch a release. also I may want to do things on new branches made by others
[06:12] <mrooney> thumper: sure!
[06:12] <mrooney> wgrant: I think both might coexist in an ideal world?
[06:12] <mrooney> So you can have either or both
[06:13] <mrooney> thumper: I have yet to look at LP code but am moderately experienced with python and programming in general
[06:13] <thumper> mrooney: first up we'd have to get agreement on the right way to do it
[06:14] <thumper> mrooney: whether based on branch or project
[06:14] <thumper> mrooney: branch may be more flexible as it could work with +junk and package branches
[06:14] <thumper> mrooney: although we could move this to the bug report
[06:14] <mrooney> yeah, do you agree having both is not contradictory?
[06:15] <thumper> mrooney: having both is more work, and we would be able to get branch in more easily I believe
[06:15] <thumper> mrooney: although we'd have to work out who hits the hooks
[06:15] <thumper> mrooney: the scanner, or something else
[06:15] <mrooney> oh okay
[06:15] <thumper> we have infrastructure around branch jobs
[06:16] <wgrant> A job seems nicer, but overkill...
[06:16] <thumper> perhaps
[06:16] <thumper> however I'm EODing now
[06:16] <wgrant> But emails use it now, don't they?
[06:16] <thumper> wgrant: revision emails? yes
[06:17] <thumper> actually... the response time from the post leads me to think a job would be better
[06:17] <wgrant> Right.
[06:19] <mwhudson> should be a job
[06:19] <mwhudson> created by the scanner i guess
[06:20] <mrooney> well, the hookah library does it asynchronously so if that was used, a lot of the problems are taken care of like that, retrying with timeouts, et cetera
[06:20] <thumper> mwhudson: abreed
[06:20] <thumper> agreed
[06:20] <thumper> mrooney: there is a lot of infrastructure in LP that you'll soon learn about :)
[06:21] <mrooney> how fun :)
[06:21] <mrooney> if you think branch is easier to land, I'm fine with looking at that first
[06:21] <thumper> mrooney: I've updated the bug and offered mentorship (never done that before)
[06:21] <mrooney> hooray
[06:21] <thumper> mrooney: it is going to need a db patch
[06:21] <mrooney> yeah
[06:21] <thumper> mrooney: and that always takes more to get in
[06:21] <thumper> mrooney: but I can help with that too
[06:21] <mrooney> I think there will be some cool project wide hooks though that can be done
[06:22] <thumper> mrooney: first step, get the code
[06:22] <mrooney> a hook to call on a new branch creation for example, that would be neat
[06:22] <mrooney> okay, I'm throwing some more info in the description
[06:22] <mrooney> have you added anything that I'll stomp over?
[06:22]  * thumper is done for the day
[06:22] <thumper> mrooney: no
[06:23] <thumper> mrooney: I add comments, not updated the description
[06:23] <thumper> ciao
[06:23] <mrooney> good night, thanks for your help and suggestions, hopefully we can work together soon :)
[06:30] <wgrant> What's the replication lag threshold?
[06:31] <wgrant> And are there two replicas?
[06:31] <wgrant> I'm seeing some very inconsistent data here.
[06:31] <wgrant> Some of which makes no sense.
[06:32] <wgrant> Some of it is explained by replication issues (eg. the build page reporting that the build has not yet started, while the builder says otherwise), but other stuff isn't (eg. build start timestamps moving around as I refresh the page).
[06:33] <wgrant> See, now the build page says that it has been building for 7 seconds, when I first saw it hit the builder more than three minutes ago.
[06:35] <wgrant> Hmmm. Maybe the i386 buildd I hit is just broken and keeps dieing quickly and silently.
[06:35] <wgrant> Aha, the build has finally stuck.
[06:38] <mrooney> Man the automated translation export is so cool and exciting
[06:39] <mrooney> combined with automated import, Launchpad is basically an automated magical translation factory
[06:39] <wgrant> An open source one, too.
[06:40] <mrooney> yeah but things are more magical if you don't look at the source :)
[06:54] <mrooney> now it just needs to create a template automatically and I will be blown away
[06:55] <mrooney> it seems pretty doable, just run xgettext on the file pattern of the users choosing
[06:56] <wgrant> I'm not sure that template autogeneration is a good idea...
[06:56] <mrooney> oh? why's that?
[06:57] <wgrant> Good question.
[06:57] <wgrant> I can't actually see much wrong with it, except that it's always been done the other way.
[06:58] <wgrant> Which I presume there's a reason for.
[06:58] <mrooney> I think the reason is because there isn't yet a solution. Most repositories are passive things which sit there and hold your files.
[06:59] <mrooney> Until you have a system like Launchpad to do it for you, that's the only way to do it
[07:00] <mrooney> generate the template, find translators, give them template, collect translation files, integrate them. Those are the steps, basically
[07:00] <wgrant> LP would then automatically commit the template too?
[07:00] <mrooney> Launchad has now basically automated all but the first, right?
[07:00] <mrooney> Maybe.
[07:01] <mrooney> Do you even need the template in your code?
[07:01] <mrooney> It is autogenerated so there's no information loss to not have it. Couldn't LP store it internally wherever it currently does for Rosetta?
[07:02] <mrooney> From a programmer's perspective when you wrap your strings with gettext, you are saying, "I want translations for these strings". So I don't see why you couldn't have something smart enough to hide the stuff you don't care about, and just deliver translations
[07:27] <poolie> wgrant: i'm wonder what you think of bug 407189
[07:28] <wgrant> poolie: I wondered that myself.
[07:28] <wgrant> I'm not sure.
[07:28] <wgrant> It's hard.
[07:54] <noodles775> Hey poolie, thanks for the feedback re. ppa-index ideas. I was wondering whether you got to take a look at the new mockup? I tried to ensure that it deals with a number of your points?
[07:54] <poolie> thanks for listening
[07:54] <poolie> i hadn't clicked the new one yet but i can look now
[07:54] <noodles775> Thanks! https://dev.launchpad.net/VersionThreeDotO/Soyuz/PPAUI#Idea%202.%20Re-focusing%20the%20purpose
[07:56] <poolie> noodles775: it looks nice
[07:57] <mrooney> dear launchpad, I honor your achievements with this blog post: http://mrooney.blogspot.com/2009/07/launchpad-is-now-automatic-magical.html
[07:57] <poolie> I really like the top part, including
[07:57] <poolie> the "owned by %s"
[07:57] <poolie> mpt would point out that "read about installing" is a poor hyperlink text :)
[07:57] <noodles775> Yeah, kinda necessary when people can name their PPA "Official Mozilla Security PPA" if they like.
[07:58] <noodles775> Yes, I' just used what's on the current one... good point. I'll update it.
[07:58] <poolie> so i can see where you're going but i still have concerns about the "currently updates only ..."
[07:58] <noodles775> Yes, I'd done this version before your latest email...
[07:58] <poolie> firstly that it's just a point-in-time thing
[07:59] <poolie> and also that even for quite well focussed ppas there might just be quite a few packages
[07:59] <poolie> most of which the user doesn't care about or recognize
[07:59] <noodles775> what would be the problem when there were, say, 15 pkgs?
[07:59] <poolie> because packages are smaller than the end user's idea of applications
[08:00] <poolie> i guess basically i don't understand what this gives you beyond the package list down the bottom
[08:02] <noodles775> The aim was that it summarises information for a decision. Ie. if there were 15 pkgs, it would say "This PPA updates 15 packages on your system. You can browse those packages below and subscribe to be notified of new packages..blah"
[08:02] <poolie> sounds a bit like filler to be frank
[08:02] <noodles775> So I guess it is only ever summarising information that is present in the table.
[08:02] <noodles775> Yeah, ok. You're not the only one to question the duplication :)
[08:03] <poolie> how about instead: make it obvious that your can subscribe
[08:03] <poolie> make it obvious what the packages are
[08:03] <poolie> and use the space maybe for some stats like
[08:03] <poolie> 15 packgase
[08:03] <poolie> 23 uploads in the last month
[08:03] <poolie> 4 packages building and 3 waiting to build
[08:04] <poolie> oh, and i'll tell you what, the nice thing is that the build counts gives you a nice way to show developers what happened to their stuff
[08:04] <poolie> without necessarily having a portlet
[08:04] <noodles775> Great idea (the building stats will be present for owners/admins etc. already - see the note on the mockup).
[08:05] <noodles775> Hmm... were you against the build-summary-status only displaying for owners/admin etc.? (see note bottom left of mockup.)
[08:05] <poolie> because one of the issues you have to struggle with here is that you fire off with dput and then things become a bit invisible
[08:05] <poolie> i am against that actually
[08:05] <noodles775> Yep, as in you think it should always be visible?
[08:05] <poolie> i think generally, everyone should see the same page unless there's a good reason otherwise (like private data etc)
[08:06] <poolie> otherwise you get this "that's funny, why can't you see it" reaction
[08:06] <noodles775> Yes, good point.
[08:06] <poolie> mm
[08:06] <poolie> and it seems possible that there'll be an interaction like this
[08:06] <poolie> dev: ok i've uploaded a fix to the ppa, please try it
[08:06] <poolie> user: i can't see it
[08:07] <poolie> well, this will probably happen anyhow, but at least showing that stuff is building or waiting may help
[08:07] <noodles775> Why wouldn't they be able to see it? (ie. the latest uploads with their state?)
[08:07] <poolie> my mistake, they would see it there
[08:07] <noodles775> ah ok.
[08:08] <poolie> oh, the other big reason why people should see data they we don't expect them to use in this context -
[08:08] <poolie> it helps them learn what the system can do
[08:08] <noodles775> +1
[08:08] <poolie> people will probably see someone else's ppa before they create their own
[08:09] <noodles775> Great... thats excellent feedback, thanks! Anything else? (I'll try to do a v3 sometime later today).
[08:09] <noodles775> s/thats/that's
[08:09] <poolie> so the thing i meant by a timeline view is something like this - above or below the package list, have something a bit like a bug activity view
[08:09] <poolie> with items like
[08:10] <poolie> firefox-3.5-thoaeu-1234124-1234-124 by _Martin Pool_ at 2009-07-31 12:22 **waiting to build**
[08:10] <poolie>  * firefox now starts much faster
[08:10] <poolie>  * no more bugs
[08:11] <noodles775> OK - so people can see what changed in the latest upload, yeah, that is helpful info.
[08:11] <poolie> so then it's like an overall changelog for this ppa, not a timeline of builds as such, but a timeline of packages, including the current state of those uploads, whether they succeeded, failed or whatever
[08:11] <poolie> right, and so you get a sense of what kind of work's being done here
[08:12] <poolie> hm
[08:12] <poolie> i wonder if "You can update your system" is making too much of an assumption people are running karmic?
[08:12] <poolie> some people on Windows will find these pages through a google search :)
[08:12] <poolie> hilarity ensues :)
[08:13] <noodles775> heh, yes, we could make that "You can update your Ubuntu <distroseries> ..." when we autodetect...
[08:13] <poolie> more seriously people on jaunty will probably get stuck
[08:13] <noodles775> and present something else when it' pre-karmic that is more direct.
[08:13] <poolie> also remember people may be browsing from one system but ssh'd to another
[08:14] <poolie> so of course you need a way around autodetection
[08:14] <noodles775> Yes, but those people would surely know to look at the technical details (key/apt-get url), don't you think?
[08:27] <poolie> yes, as long as there's a drop-down to get the right system within that it's fine
[08:39] <sianis> hi
[08:39] <sianis> https://code.launchpad.net/~mvo - this page gives me timeout all day
[08:40] <wgrant> sianis: Known bug, and it's hopefully fixed. Try https://code.edge.launchpad.net/~mvo instead.
[08:40] <sianis> wgrant: thx
[08:44] <sianis> https://code.edge.launchpad.net/~mvo/ddtp-ubuntu/ddtp-ubuntu - why didn't launchpad put translations here?
[08:44] <wgrant> That's amusing.
[08:48] <wgrant> It could also do with having its whoami configured properly...
[09:08] <wgrant> sianis: The code seems pretty clearly right. I wonder if the right series is setup up for commits.
[09:35] <didrocks> dpm: to add new language available to be translated, even with the automatic launchpad import, do I have to add new blank .po file or can the user say "I want to translate in this language!" (https://translations.launchpad.net/quickly)
[09:37] <dpm> didrocks: if I'm not mistaken, you don't have to do anything, the user can easily create the new translation in the following way: depending on the language selected on his/her LP profile, only the translations for/his/her language will be shown in the +translate page, so even if the translation is empty, he/she should be able to create one with simply entering the first translation. jtv might be able to tell you more, but in fact, we can try this just no
[09:37] <dpm> wgrant, I'll create the Catalan translation :), give me a sec...
[09:38] <jtv> no need to create anything at all.
[09:38] <jtv> exactly what dpm said
[09:38] <didrocks> dpm: awesome, thanks for the clarification :)
[09:40] <didrocks> ok, it works. I added a new language to my profile and I can see it
[09:40] <didrocks> I just have now to see why it doesn't work this branch for automatic exporting translation: ~quickly/quickly/quickly-po
[09:41] <jtv> didrocks: what's not working?
[09:42] <dpm> didrocks: there you go, you should now have a Catalan .po file as well (it's not finished, I've just translated a few strings). Regarding automatic translation exports, jtv is the one to talk to
[09:42] <didrocks> dpm: yes, I can see it, thanks :)
[09:43] <didrocks> jtv: I try to add it to this page: https://translations.edge.launchpad.net/quickly/0.x/+link-translations-branch
[09:43] <didrocks> I just get "There is 1 error"
[09:44] <jtv> didrocks: did you pick the branch from the "Choose..." picker?
[09:44] <sianis> wgrant: mvo reviewd the settings, coldy you or somebody run an export again?
[09:44] <didrocks> jtv: I can't find it as well from the "Choose..." picker
[09:45] <didrocks> jtv: but the branch exists: https://edge.launchpad.net/~quickly/quickly/quickly-po
[09:46] <jtv> didrocks: team-owned branches don't quite work yet.  :(  Can you make yourself the owner of the branch, then set it as the translations branch, then make the team the owner again?
[09:47] <didrocks> jtv: this workaround works well. Thanks! Something to put in a FAQ ;)
[09:48] <jtv> didrocks: something to fix.  :(
[09:48] <didrocks> also :)
[09:48] <didrocks> (but in the meanwhile)
[09:48] <didrocks> jtv and dpm : thanks for your support
[09:48] <jtv> didrocks: no worries, hope you enjoy the feature.  Let me know if there's anything else.
[09:48] <dpm> didrocks: np, you're welcome :)
[09:49] <didrocks> jtv: I will! It seems to be a really cool feature (rosetta is the "unkown part of launchpad" where I didn't get until now)
[09:52] <jtv> didrocks: I've updated the blog post to link to the bug.
[13:20] <thekorn> hi, the graphs in the 'last modified' column of code overview pages like https://code.edge.launchpad.net/zeitgeist are all the same,
[13:20] <thekorn> is this an already known issue?
[13:22] <thekorn> nm, found it, it's bug 407006, did not know they are called sparklines ;)
[14:11] <wgrant> cprov: I ran into some strange build issues earlier this evening.
[14:12] <cprov> wgrant: can I see the buildlog ?
[14:12] <wgrant> cprov: Not an issue inside the build. It was very strange.
[14:12] <wgrant> Basically, an i386 got dispatched to aluminium several times.
[14:12] <wgrant> Each time, it sat building for 30ish seconds.
[14:12] <wgrant> Then restarted silentl.
[14:13] <cprov> wgrant: maybe because aluminium wasn't able to fetch the chroot+files in time?
[14:13] <wgrant> After about 6 minutes of trying, it finally caught and built fine.
[14:13] <cprov> wgrant: in those cases the UI lies for few seconds ...
[14:13] <RainCT> Hey
[14:14] <cprov> wgrant: were you able to view any relevant part of the buildlog tail (on the builder page) ?
[14:14] <wgrant> cprov: There wasn't any.
[14:14] <wgrant> cprov: I suspect the slave was still resuming.
[14:15] <RainCT> https://code.edge.launchpad.net/~thekorn/zeitgeist/flexible.engine.backend/+merge/9499 -I'm wondering, what does the "(community)" there mean?
[14:15] <cprov> wgrant: exactly, resuming or trying to fetch build files, then timed out
[14:15] <wgrant> cprov: So that's expected behaviour?
[14:15] <cprov> wgrant: they are blindly reset, in those circumstances
[14:15] <cprov> wgrant: not ideal, but yes, expected.
[14:15] <wgrant> cprov: OK, thanks.
[14:16] <cprov> wgrant: we should move the job to the next builder, I guess.
[14:16] <cprov> wgrant: it happens 'by accident' when the queue is full
[14:16] <wgrant> cprov: What do you mean?
[14:17] <wgrant> RainCT: It means something is broken, as it thinks you're not a member of the target's review team, when you clearly are.
[14:17] <wgrant> RainCT: I've seen this problem in a couple of other places.
[14:17] <cprov> wgrant: it seems that some builders don't cope with the timeouts we established for some periods
[14:18] <cprov> wgrant: after some time, they work normally, it's a race on resuming, I suspect.
[14:18] <wgrant> cprov: Right.
[14:18] <wgrant> So, it didn't cause any problems, but was really confusing while watching the build.
[14:19] <wgrant> Started 30 seconds ago... refresh... started 2 seconds ago...
[14:19] <cprov> wgrant: yes, it's very WTF, because the master keeps hammering a builder that is not working while the others are IDLE
[14:20] <RainCT> wgrant: ok then, just making sure you know about it :)
[14:20] <wgrant> RainCT: I'd file a bug, so Code people can look at it.
[14:29] <RainCT> ok, done (#407347), thx
[14:39] <ronny> whoever manages the bzr overlay for gentoo on lp - he failed at making it accessible to the rest of the world
[15:42] <NCommander> Can an LP admin bump the size of mp PPA
[15:43] <NCommander> *my
[15:48] <jpds> NCommander: It would probably be best to file a question against Launchpad so that they can track the request.
[15:49] <deryck> NCommander, yes, what jpds said.  File a question, and I can get it assigned for you.
[15:50] <NCommander> deryck, which project should I file it again
[15:50] <NCommander> *against
[15:52] <deryck> NCommander, https://answers.edge.launchpad.net/launchpad-project
[15:53] <NCommander> deryck, https://answers.edge.launchpad.net/soyuz/+question/78744
[15:59] <deryck> NCommander, it's assigned now, and the appropriate group pinged.
[16:25] <aboSamoor> Hi, I am trying to login to stackoverflow using launchpad open id, I did not use stack overflow for more than 6 months. It is not working it wants to create new user my launchpad open id !
[16:33] <aboSamoor> Hi, I am trying to login to stackoverflow using launchpad open id, I did not use stack overflow for more than 6 months. It is not working it wants to create new user my launchpad open id !
[16:38] <aboSamoor> can anyone help me ?
[16:44] <epsy> http://bazaar.launchpad.net/~armagetronad-dev/armagetronad/config_tournament_server/annotate/head%3A/ladle.cfg
[16:44] <epsy> what happened there?
[16:45] <aboSamoor> epsy, do you know who can I ask regarding open id support ?
[16:45] <deryck> aboSamoor, I haven't logged into stackoverflow before, but don't you have to associate your account on stack overflow with your LP account?
[16:45] <epsy> what are you trying to do?
[16:46] <epsy> ah, stackoverflow is a website?
[16:46] <deryck> aboSamoor, my guess is it wants to create a new stack overflow account because the site doesn't know which account your open id relates to.
[16:49] <aboSamoor> deryck, I created an account in SO using launchpad openid, what I can remember that my id was id.launchpad.net/~rmyeid and after a while it changed launchpad.net/~rmyeid and I can not access my account again !
[16:50] <deryck> aboSamoor, what is your user name on Launchpad?
[16:50] <aboSamoor> deryck, rmyeid
[16:52] <deryck> aboSamoor, and when you went to SO did you try to provide -- https://launchpad.net/~rmyeid -- as your open id login?
[16:54] <aboSamoor> deryck, I tried and it gives me that it is the first time so it will make a new user account !
[16:55] <deryck> aboSamoor, Stack Overflow tells you it is your first time and wants you to create a new Stack Overflow account?  Is that correct?
[16:56] <aboSamoor> deryck, yeah
[16:56] <deryck> aboSamoor, yeah, this is what I was saying earlier.  Stack Overflow doesn't associate your open id with any account that they have.  I'm not sure why that is, but it's on Stack Overflow's end, not Launchpad's.
[16:58] <aboSamoor> deryck, sorry, I am not that technical person what do you suggest to do ?
[16:59] <mhall119|work> help!
[17:00] <deryck> aboSamoor, I think you need to contact someone with Stack Overflow about the problem.
[17:00] <mhall119|work> okay, I uploaded a source package which finally (after some tweaking) was accepted, but the build failed because I lacked build-depends
[17:00] <mhall119|work> so, I added the build-depends, but now I can't get the package accepted again
[17:00] <mhall119|work> this is the email I get: http://paste.ubuntu.com/239489/
[17:01] <mhall119|work> do I need to change the version number and re-submit?
[17:02] <mhall119|work> my only changes were to debian/control
[17:02] <aboSamoor> deryck, thanks very much I will use IRC conversation to explain the problem to the SO guys :)
[17:12] <deryck> mhall119|work, does the info here help:  https://help.launchpad.net/Packaging/UploadErrors  ?
[17:15] <mb74> please can someone tell me how to make a kernel-package whic
[17:16] <mb74> which can be uploaded to launchpad?
[17:16] <mb74> how do i have to setup make-kpkg?
[17:18] <mhall119|work> deryck: nope
[17:18] <mhall119|work> deryck: I had it working previously, up to the point where it tried to build a binary
[17:18] <mhall119|work> I'm trying to fix the build now
[17:20] <mhall119|work> my ppa setup is right, my gpg keys are right
[17:20] <mhall119|work> it just doesn't like me uploading the package again
[17:20] <mhall119|work> changing the version in debian/changelog has allowed it
[17:20] <mhall119|work> do you have to change the version every time a build fails?
[17:23] <deryck> mhall119|work, sorry, soyuz isn't my area.  Maybe cprov could help explain?
[17:24] <maxb> mhall119|work: You can retry a failed build through the web interface. If you want to change the source in any way whatsoever, then yes you need to increment the version.
[17:24] <mhall119|work> maxb: is the debian/control file considred source?
[17:25] <maxb> Of course it is - it's part of the source package you upload
[17:25] <mhall119|work> ok
[17:25] <mhall119|work> so, what would be the point in retrynig a build through the web interface, if nothing has changed?
[17:26] <mhall119|work> presumably if it wouldn't build the first time, trying again without fixing anything will cause it to fail the second time
[17:31] <mhall119|work> are there any good tutorials covering how to handle meta-packages and non-code packages?
[17:32] <maxb> mhall119|work: A build may fail because of an issue with a package that it depends on. After that dependency is fixed, the build may be able to retry successfully
[17:32] <mhall119|work> ah, okay, that makes sense
[17:32] <mhall119|work> thanks for taking the time to explain all this to me maxb
[17:33] <maxb> mhall119|work: I'm not sure how you'd write a tutorial on meta-packages or non-code packages. They're really just packages that happen to be of certain kinds
[17:40] <mhall119|work> maxb: I meant, what do I do with debian/rules when nothing is being compiled?
[17:41] <mhall119|work> like, if I just have a GDM theme and other artowkr
[17:42] <maxb> mhall119|work: If you think about a debian/rules file, it does other things than compile the package. So you do all the rest, you just don't compile anything
[17:42] <maxb> There is https://wiki.ubuntu.com/MOTU/School/PackagingWithoutCompiling
[17:42] <mhall119|work> thanks
[17:43] <mhall119|work> ah, that looks like exactly what I'm looking for, thanks again maxb
[17:44] <mhall119|work> okay, I'm going to lunch while I wait for launchpad to compile my package
[17:45] <cprov> maxb: thanks for helping mhall119|work ... perfect explanation, as usual.
[17:47] <mhall119|work> yes, maxb deserves a raise for that
[17:47] <maxb> :-)
[17:47] <rafiu> any Launchpad Bugs devs around? need to migrate tickets from trac to LP (i have xml conversion ready)
[17:48] <mrevell> rafiu: gmb is the person to speak to but he's having some IRC issues, AFAIK
[17:49] <gmb> rafiu: Hi. As mrevell says my connection is a bit unstable at the moment. Normally I deal with bug imports but I'm going to be unavailable next week. Are you happy to wait until the week after that or would you rather get it done sooner?
[17:52] <mb74> nobody here who can tell me how to create a kernel source package for uploading to launchpad?
[17:53] <maxb> mb74: It's a fairly specialised topic
[17:54] <aboSamoor> what are the ubuntu websites that use openid ?
[17:54] <maxb> I assume from the way you're asking that "take the existing ubuntu source package and modify it" does not suit you
[17:54] <maxb> aboSamoor: several different wiki installations, REVU, ...
[17:55] <aboSamoor> maxb, I just discovered that ubuntuforums accept launchpad openid, I meant to ask if brainstorm, launchpad itself accept openid ?
[17:56] <maxb> launchpad itself provided openid identities for its registered users, but does not itself accept external openid identities
[17:57] <maxb> erm, I mean provides, not provided
[18:01] <SamB> yeah
[18:01] <SamB> I use mine to comment on Precocious ;-)
[18:02] <rafiu> gmb: if it's not going to put you out too much, the sooner the better
[19:04] <mhall119|work> just wanted to say thanks again to maxb, I got my first binary packages built and in my PPA
[20:33] <mhall119|work> is lpia the ARM arch?
[21:45] <geser> mhall119|work: no, lpia = low-power intel architecture, it's similar to i686
[21:45] <mhall119|work> oh, ok