[00:00] oh, ok [00:00] I assume you have devscripts installed? [00:00] yes [00:00] in the source directory, type dch -i [00:00] THis will pop up an editor with a new changelog entry already added [00:00] Type in something like "Rebuild for Hardy" in the changelog [00:01] neat [00:01] :) [00:01] Then on the versioning string, you'll see something ending in -2ubuntu1, change that to -2~ppa1 [00:01] And make sure the series (which is right after the version says hardy) [00:01] ok [00:02] Save and exit [00:02] DO you know if this package will build out of the box? [00:02] (i.e., does it build correctly on hardy?) [00:02] oh, as a second note, make sure your username and email is right in the changelog (just open it with nano debian/changelog or some other editor) [00:02] yes it builds [00:03] (it needs to match what your GPG key says) [00:03] the version just ends with ubuntu1, not -2ubuntu1 [00:04] Just remove the ubuntu part of it ;-) [00:04] NCommander: got it [00:04] The reason you need the ~ppa1 at the end is because if the package is ever added to the Ubuntu repositories, you want upgrades to work sanely [00:05] so it's the latest changelog entry that debuild looks at when trying to sign? [00:05] Yup [00:05] So it should be *your GPG Key Name* <*your GPG key email*> [00:06] Once your done [00:06] run debuild -S; it will prompt for your GPG key to sign the package once your done [00:11] mtah, once that's done, simply upload the new changes to your PPA :-) [00:11] :) Launchpad is pretty impressive [00:12] and by that I mean awesome [00:12] but it's not approving of my trailer line in the changelog [00:13] \o\ [00:13] Launchpad is a combination of awesome and frustrating at times [00:16] NCommander: does this look alright? http://pastie.org/285535 [00:17] It can't linewrap [00:17] The final line has to be one line :-) [00:17] (which is why it likely rejected) [00:17] d'oh of course not [00:17] You need to run debuild -S -sa [00:18] (since the package isn't in ubuntu, the source isn't handy, so you need to include it) [00:25] NCommander: Success! So now I only have to wait until the "Pending" status is gone? [00:26] Its 20 minutes from upload before the source is available [00:26] And 20 minutes from when the binaries are built to when they are available [00:26] Why doesn't Launchpad's bug tracking system have a quick link to a list of bugs that I am involved in for a project? [00:27] I was expecting a tabbed system with bugs filed by me in one tab and those assigned to me or simply that I am subscribed to in another. [00:27] WillPittenger, it should. Feel free to file a bug requesting it, it's been talked about quite a few times, but I'm not sure if there's a bug open for it yet [00:27] you have that, just not per-project [00:28] but it would be very useful [00:28] Where? [00:28] https://bugs.launchpad.net/launchpad [00:28] NCommander: Ok. Thanks for all your help, really appreciated! :) [00:28] No. [00:28] No problem [00:28] Where is the list of bugs? [00:28] WillPittenger, go to your user page, click on the "Bugs: tab [00:28] "Bugs" [00:29] and, on your right side, you have a few filters [00:29] List assigned bugs, List commented bugs, etc [00:30] I haven't found it yet. [00:31] I also appear to have been unsuccessful changing my image. [00:34] Now I found it. I was looking at it like it was a project bug tab. [00:34] Confusing at best. I think the user name should be a drop down listing stuff like that. [00:35] Then change the CSS to keep users from thinking they are looking at a project page. [00:55] I'm pretty sure I know the answer: but is there any way to force a cvs import to store windows-style line endings in the bzr branch it creates? Until bzr itself gets some support for platform line endings, I'm guessing I'm kinda screwed though... === kiko__ is now known as kiko === jscinoz_ is now known as jscinoz === cjwatson_ is now known as cjwatson === jamesh_ is now known as jamesh [09:10] cprov & bigjools & kiko: I've got a hypothetical about Soyuz that needs to be asked. Any of you around? [09:10] NCommander: shoot [09:10] bigjools, is it possible for a package in a released distribution to be promoted to main [09:10] And not break everything to high heaven? [09:11] NCommander: I don't think that can happen in a released distro, no [09:11] bigjools, this is from a techological standpoint, right? (I'm just asking to see if I can get the archive admins to bump a package to main to resolve a bug in xfprint) [09:13] NCommander: the issue is that once a distroseries goes to release, the release pocket is frozen so no changes can be made in it [09:13] bigjools, could we do it via -updates? [09:14] bigjools, i.e., upload a new version to updates, and then bump that package to main? [09:14] theoretically, yes, but I'm not sure if the distro team can do that [09:14] you might want to catch up with them [09:14] bigjools, well, we needed to find out first if Soyuz could possibly do it [09:17] NCommander: I'm pretty sure it's ok as long as the release pocket is not involved. apt-get should cope, right? [09:17] If apt sees multiple versions of a package in multiple sources, whichever one is highest should one [09:17] indeed [09:18] s/one/win [09:18] whoops [09:18] Wrong shortcut key [09:18] I just have that weird feeling I can see things exploding in a giant fireball of fury [09:18] * bigjools also curses ctrl-q being next to ctrl-w [09:18] bigjools, can a promotion to main be done in a pocket without effecting the entire series? [09:20] * NCommander is just trying to preventively prevent the destruction of Launchpad :-) [09:22] NCommander: what do you mean by affecting the entire series? [09:22] Well [09:22] If you promote a package in RELEASE pocket, it also promotes it in UPDATES and BACKPORTS [09:22] I'm curious if that works in reverse [09:23] NCommander: mmm I don't know [09:23] Care to test on dogfood :-) [09:23] (I figure its better then testing it when/if this motion gets approved, and the archive self-destructs) [09:24] NCommander: ok pick me a package in a released distroseries [09:24] I'll override it on dogfood [09:24] hello [09:24] :-) [09:24] That's an easy one [09:24] well [09:25] bigjools, https://dogfood.launchpad.net/ubuntu/gutsy/+source/hello [09:25] oh [09:25] You need one from universe [09:25] d'oh [09:25] :) [09:25] xfprint? :) [09:25] a2ps [09:25] THat's what needs to be promoted anyway [09:26] Might as well see how badly the archive implodes [09:26] ok [09:26] (generally speaking, we probably should have a package in -updates to test what this realistically will look like) [09:26] * bigjools kicks the dog [09:26] * NCommander watches the dog kick back [09:27] it's low on memory so the first request in a while makes it grind when stuff is paged [09:27] Er, did you change a2ps? [09:27] not yet [09:27] There is no package in -updates [09:27] WHich is what we need to test [09:27] (that a change in the updates pocket won't break release pocket) [09:28] I can override universe/release to updates/main [09:28] And that will do the right thing? [09:28] let's see [09:28] (what we're expecting to do is upload a package to updates/universe, and then get it promoted there) [09:28] ok [09:29] I'd upload to dogfood, but I'm not a core developer ;.; [09:29] find one that fits the bill while I try this [09:29] Sure [09:29] * NCommander looks at gutsy-updates [09:30] NCommander: meh, I can't move pockets, d'oh [09:30] o_o; [09:31] so does this mean this plan wont' work? [09:31] NCommander: well let's try one in -updates [09:31] yeah, looking [09:32] Oh that's anonying [09:32] The dogfood snapshot is too old to have a bunch of these updates [09:32] yeah it's 2+ months old now [09:32] https://dogfood.launchpad.net/ubuntu/+source/avahi - bingo [09:33] oh wait [09:33] nm [09:33] lol [09:33] why is that in gutsy-updates/universe/Packages.gz o_o; [09:33] https://dogfood.launchpad.net/ubuntu/+source/basket - FINALLY [09:34] ok [09:34] * NCommander always wonders what the archive admin interface looks like [09:34] scripts :) [09:35] I thought there was some things that could be done by the web interface [09:35] Such as promotions/demotions [09:35] only the queues [09:35] ok check it out [09:35] I will re-publish [09:35] it will take a while [09:35] check what out? [09:35] the url you just pasted [09:35] Cool, it got promoted [09:36] now the question is does the backend glue do the right thing :-) [09:38] bigjools, dogfood is awesome; all my LP bugs now have backtraces attached :-) [09:38] NCommander: ahem :/ [09:38] Is that a bad thing? [09:39] config foobar, it should not happen, I'll be turning that off :) [09:39] bigjools, argh. kiko wasn't concerned about backtraces being public [09:39] oh well [09:40] So much for useful bug reports [09:40] really? [09:40] yeah [09:40] passwords and such are censored [09:40] So there isn't much of a security risk [09:40] well, don't rely on dogfood for bug reports please, it's rarely running the latest code [09:40] Well, I found the bug on production [09:40] (edge actually) [09:40] Then reproduced [09:40] and dogfoods code is new enough; it has black icons ;.; [09:41] when you cause an oops on production/edge, we can see the stack trace in our reports [09:41] even if you can't [09:41] so you just need to reference the oops number [09:41] right, webfacing tracebacks don't sound too useful [09:41] dogfood also often has highly experimental code on it [09:42] Ok, point taken ;-) [09:42] (this bug didn't generate an OOPS) [09:42] Two bugs actually, one was a DoS, and one was an annoyance [09:43] bigjools, so at least we know that Launchpad won't self-destruct if we promote in a pocket [09:43] NCommander: right, the publisher has finished and it was ok [09:43] cool [09:43] So you proved it was possible [09:44] Now I get to try and make my case to u-devel [09:44] \o/! [09:44] you're welcome [09:44] bigjools, well, if launchpad didn't do the right thing, picture how ugly it would be to fix [09:45] NCommander: no more so than the other crises we deal with :) [09:45] but thanks for asking about this [09:45] bigjools, I can picture it. Like a bad code upload causes the buildds to rm -r their chroots ;-) [09:46] that would be ok, the chroots are in the librarian :) [09:46] Ooooh [09:46] Maybe you can help me with this then [09:46] Can you find me the chroot for intrepid or hardy? We have a failure that only occurs on the actual buildds (doesn't happen in sbuild/pbuilder) [09:46] bigjools, cprov wasn't able to find it when I asked him :-/ [09:47] NCommander: I'll get back to you on that [09:47] If it can be done, it would be awesome. I know something similar happened back in gutsy and a chroot was made available via librarian === bigjools changed the topic of #launchpad to: https://launchpad.net/ | Channel logs: http://irclogs.ubuntu.com | bigjools is your help contact today [10:41] bigjools, I need help, I have no life :-) [10:41] How can I resolve? [10:42] * NCommander is shot [10:42] j/j [10:42] er jk [10:47] * bigjools furrows brow === bigjools-out is now known as bigjools === barry-away is now known as barry [12:35] bigjools: Any idea how I access a branch through launchpadlib? [12:36] wgrant: I don't, but I can find out [12:36] They're exposed in the API, but there's no obvious way to get to them. [12:36] bigjools: That's what I was hoping. Thanks. [12:40] wgrant: you could use Launchpad.load(), thinking about it [12:41] bigjools: My launchpadlib doesn't seem to know about that. Is it a new feature? [12:41] shouldn't be, wgrant [12:41] and project seems to have a branches_collection_link [12:42] bigjools: Ahh, I was trying to navigate to branches directly. I guess I could use that. Thanks. [12:42] np [12:43] wgrant: see "persistent references" in https://help.launchpad.net/API/launchpadlib for the load() call [12:45] bigjools: Still doesn't work, but going through the project should be fine. [12:45] okay [12:46] wgrant, is it worth filing a bug or asking a question about that I wonder [12:46] kiko: I suspect I'm using a very old version (the one in Intrepid). [12:47] wgrant, hmmm. can you confirm that? because if so we should definitely get it upgraded in Intrepid [12:47] 0.1~bzr7-0ubuntu1 [12:48] We're way past FF, but I'm sure an exception could be organised. [12:53] you should really try to update this version in intrepid, because it is missing all the nice features like load(), caching and support for hosted files [12:56] I'm going to work on that this week [12:57] if anyone wants to help that would be appreciated [13:02] james_w, on upgrading launchpadlib? [13:02] kiko: yup [13:03] james_w, awesome, I will tell francis and leonart to ping you. thanks for doing that! [13:03] no problem [13:04] I thought about asking to make sure there was nothing juicy landing in the next week we were going to miss out on [13:05] I was going to ask them exactly that :) [13:13] james_w, super, great news :) === mrevell is now known as mrevell-lunch [13:35] mrevell-lunch: Yay! CHR docs. === mrevell-lunch is now known as mrevell === lamont` is now known as lamont === stub1 is now known as stub [16:23] leonardr, ping [16:24] kiko, yo [16:25] leonardr, hey there, weekend okay? I was talking this morning with james_w who wanted to get in touch with you about upgrading launchpadlib in Intrepid -- can you give him suggestions or advice? [16:25] sure, i don't know anything about packaging but i can answer any questions he has about launchpadlib or the web service in particular [16:26] james_w, be sure to let us know [16:26] hey leonardr, how ya doing? [16:26] thanks kiko [16:27] hi james_w [16:28] leonardr: my first question is about whether current trunk is good to take. Is there anything juicy landing in the next week we would want? Any known major bugs with it? [16:29] there are known bugs but they're not going to be fixed in the next week [16:29] because i've got a deadline on another project [16:29] sure, but is there anything major that would make the current trunk worse than what we already have? [16:29] leonardr, how bad are the known bugs? (URLs help) [16:30] let me find the worst one [16:30] james_w: there are no regressions from what you already have [16:30] cool, that's good enough, as long as we aren't introducing known security holes or anything [16:32] the worst ones have to do with hosted files. bug 274815 and bug 277533 [16:32] Launchpad bug 274815 in launchpadlib "Cached binary files aren't retrieved from cache" [High,Triaged] https://launchpad.net/bugs/274815 [16:32] Launchpad bug 277533 in launchpadlib "Deleting a file from a release raises exceptions." [Undecided,New] https://launchpad.net/bugs/277533 [16:32] mrevell: Did you want to talk? [16:32] hey abentley, it's okay, don't worry [16:33] mrevell: Okie dokie. [16:33] leonardr: they look ok, we can always SRU for fixing important things. [16:34] leonardr: for future reference, how do you get the debugging information from the first bug? [16:36] james_w: set httplib2.debuglevel to 1 before creating any Launchpad objects [16:36] cool, thanks, that will come in useful [16:36] leonardr: I haven't seen any waddlib changes, have I missed any? [16:38] james_w: no, no wadllib changes for quite a while [16:38] cool [16:38] I'll request a freeze exception today [16:39] my last question would be about API stability. We've obviously had lots of problems with things like python-lp-bugs which do screen-scraping. Are future changes in launchpad likely to break launchpadlib in released versions, or is it far more robust against that sort of thing? [16:40] by it's nature it seems like it would be more robust, but I wondered if API stability was a launchpad goal. I realise this may be a post-beta thing. === salgado is now known as salgado-lunch === bac is now known as bac_afk === kiko is now known as kiko-fud [17:25] leonardr: I filed the freeze exception request at bug 279147, thanks for your time. [17:25] Launchpad bug 279147 in python-launchpadlib "[Freeze exception] Update to new bzr snapshot" [Undecided,New] https://launchpad.net/bugs/279147 [17:25] james_w, sure === bigjools changed the topic of #launchpad to: https://launchpad.net/ | Channel logs: http://irclogs.ubuntu.com === salgado-lunch is now known as salgado === matsubara is now known as matsubara-lunch [18:36] how may states are there in a bug lifecycle in launchpad's bug tracker? [18:40] kmouly, see http://news.launchpad.net/general/of-bugs-and-statuses [18:41] kiko-fud: thnx [18:41] YW === kiko-fud is now known as kiko === bac_afk is now known as bac === matsubara-lunch is now known as matsubara === matsubara is now known as matsubara-afk === matsubara-afk is now known as matsubara === dodobas_ is now known as dodobas [20:31] Is all well on lp? http://bazaar.launchpad.net/~jimw/mysql-server/bug-33812/diff/2673 -> "Sorry, there was a problem connecting to the Launchpad server." [20:32] CardinalFang, mysql is known to make loggerhead explode [20:32] mwhudson, mornin' [20:32] up yet? :) [20:32] beuno, Oh. Dang. [20:33] CardinalFang, I'd say, try a few more times, enough RAM may be freed up at some point [20:33] hi [20:33] it's a "known issue" [20:33] It certainly pushes bzr in general, too. [20:33] anyone from Mailing list approval? [20:33] beuno: hi [20:34] S-i-A, that would be Rinchen or barry (which I think is off sick today) [20:34] ok thanx [20:34] mwhudson, hi! is CardinalFang's timeout expected? [20:35] mm, loggerhead is using a lot of ram === nebi is now known as nebi|away [20:40] S-i-A, mailing list approval? let me look [20:40] Rinchen: https://launchpad.net/~linwx.net [20:40] beuno, actually any LP dev can do it now, unless it's ubuntu then we hand them off [20:41] CardinalFang: i restarted the code browser, please try again [20:41] S-i-A, done [20:41] Rinchen: thanx [20:44] Does anyone know why this doesnt work - bzr branch lp:~ubuntu-dev/mplayer/ubuntu - It just hangs and doesnt download anything... [20:53] stefanlsd, interesting [20:53] stefanlsd, i'm uploading something and it's taking a long time [20:53] Rinchen: maybe if its a big repo? [20:54] stefanlsd, let me try locally [20:54] stefanlsd, yea, it might take a while. It's a big repo. [20:54] rockstar: its been about 15 minutes, and i have 28k so far... [20:55] stefanlsd, yea, when I say a while, I mean a good long time. [20:55] If you want to make sure it's progressing, add -Dhpss after the bzr branch [20:56] stefanlsd, upgrading you version of bzr might also help. [20:56] stefanlsd, worked here but it's slow.. it's not been upgraded either [20:58] rockstar: im using 1.6.1... [20:58] maybe the bandwidth to LP just sucks from me [20:58] stefanlsd, I think 1.6.1 had the regression fixes in it, but I can't be sure. [21:02] This is interesting strace of bzr process. Its just repeating this... - http://pastebin.ubuntu.com/54761/ [21:03] looks like its working - just slow [21:04] stefanlsd, optionally, if you branch through http, it should be faster for the first time (after that, it's faster to use bzr+ssh, the default when you use lp: urls) [21:04] of course, if you've already start, it's probably best to leave it as-is [21:05] beuno: thanks. noted. will do http next time :) [21:05] this is http isn't it? [21:06] james_w: lp: [21:06] james_w, doesn't lp: default to bzr+ssh if you specified your launchpad-login? [21:06] yeah, but the strace looks like HTTP [21:06] traceroute from me to code.edge.launchpad.net if anyone is interested - http://pastebin.ubuntu.com/54762/ [21:06] recv(4, "HTTP/1.1 200 OK\r\nDate: Mon, 06 O"..., 16384, 0) = 640 [21:07] mwhudson, restarting it made it work. Thank you. [21:07] np [21:10] heh. still 28K. where is it saving this 'stuff'? [21:13] stefanlsd: I assume you are on intrepid? [21:13] james_w: yeah... [21:14] james_w: actually, it just started increasing... [21:14] stefanlsd: there may be a bugfix in this area not in Intrepid. I remember one, but I need to find some time to investigate and cherrypick it [21:15] james_w: and u were right, i actually am running bzr branch https://code.edge.launchpad.net/~ubuntu-dev/mplayer/ubuntu . run the lp one earlier... === salgado is now known as salgado-afk === kiko is now known as kiko-phone [22:00] hey guys, i was wondering, basically, this is my first time really doing some actual testing of ubuntu...and i was wondering if someone could direct me to somewhere that could tell me how to properly file a bug? [22:02] YaManicKill, launchpad.net/ubuntu/+filebug [22:02] YaManicKill, or http://help.launchpad.net/Bugs [22:04] Does launchpad offer any statistics [22:04] like number of times a package has been downloaded ? [22:04] for ppas [22:04] det, not at the moment, no [22:04] it is planned for the future [22:05] That would be really useful [22:05] Thanks [22:06] is it sad that i'm excited about the first bug i'm logging? [22:06] Filing bugs is great. [22:06] Community participation, hooray. [22:06] Open source, hooray. [22:06] nah, it's cool [22:08] man i'm a geek...but you are right...the whole feeling of, i can actually make a difference with the stuff i use on my computer - its great [22:10] anyway...i'm just bout to test if my video driver works...so gotta reboot...ciao [22:15] kiko-phone: so, I just uploaded a source package to a PPA, and it didn't install a depend that was listed in Build-Depends-Indep. Is this expected? [22:16] mtaylor, no... but that's not really something I'd expect us to not have noticed before [22:17] kiko-phone: me either :) [22:17] kiko-phone: http://launchpadlibrarian.net/18274662/buildlog_ubuntu-hardy-amd64.protobuf_2.0.2-1_FAILEDTOBUILD.txt.gz [22:18] make[1]: dh_pysupport: Command not found [22:18] mtaylor: Build-Depends-Indep are only installed for arch: all builds - ie. i386 [22:20] wgrant: ahh... so in this case it's an error in the packaging [22:20] It is. [22:21] Soyuz does give us some strange build failures, but that's not one of them. [22:21] :) [22:21] They're generally in strange, strange things like SCons. [22:22] SCons does not give me happy [22:22] Mhm. [22:22] is there a way we can ask for a Launchpad mirror of a CVS repo to use windows line endings in the repo? Otherwise, the mirror and the real CVS trunk are always different and never the same on my platform. [22:23] (problem is CVS checkout on windows has \r\n, but bzr branch has \n, so every file differs on every line. === mthaddon changed the topic of #launchpad to: https://launchpad.net/ | Channel logs: http://irclogs.ubuntu.com || bazaar.launchpad.net down from 22:00 UTC for up to 4 hours [22:49] hi, I am trying to push some stuff to my branch but I am getting an error [22:50] tretle, what error are you getting? [22:50] Unable to import paramiko (required for sftp support): No module named paramiko [22:51] tretle, install python-paramiko [22:51] * wgrant recommends using bzr+ssh, too. === kiko-phone is now known as kiko [22:54] tretle, bzr launchpad-login and then bzr push lp:... [22:55] well, bzr defaults to http if you don't have lp-login set, so he may be something manually === cjwatson_ is now known as cjwatson [23:08] now getting this bzr: ERROR: Connection closed: please check connectivity and permissions (and try -Dhpss if further diagnosis is required) [23:11] tretle: See /topic [23:12] ah [23:12] at least its not me doing something wrong, will try later.. thanks for the help === nebi|away is now known as nebi === Ng_ is now known as Ng