[00:02] how do I proceed now. propose the branch for merge and let the sponsor subscribe sru or subscribe them myself? [00:32] Thanks RainCT. Somehow, I don't think I'll have much trouble with this package based on the ML feedback, but I'll definitely keep you in mind when I am ready [00:46] jtaylor: propose the branch for merge and subscribe ubuntu-sponsors [00:46] jtaylor: oops, no need to subscribe when proposing a merge [00:47] jtaylor: I'll be piloting in a bit, so I can run this through if it's ready [00:52] micahg: merge proposed [00:52] binary debdiff is just egg-file -> egg-folder with all the setuptools stuff in it [00:56] jtaylor: thanks === luciano_ is now known as virusuy === medberry is now known as med_out [07:04] good morning [09:50] micahg: I can't reproduce that foolscap problem in chroot, natty and oneiric, anything special you did? [10:08] hwo can I view patches to a package in fedora? anything like the PTS? [10:10] * Laney found pkgs.fedoraproject.org [10:11] Oh, is that useful? I always trawl the git. [10:13] it got me to the git in the end [11:34] Rhonda: is it known that packages.d.o's file search is broken for experimental? http://packages.debian.org/search?suite=experimental&mode=exactfilename&searchon=contents&keywords=csharp should return results, for example [11:40] hmmm [11:40] huhm, http://packages.debian.org/experimental/all/mono-csharp-shell/filelist [11:41] That's there since a month, right … [11:41] Laney: Can you file a bug about that, please? against www.debian.org [11:42] sure [11:42] Thanks. Not much time now to investigate further. [11:58] Hey, anyone up for reviewing http://revu.ubuntuwire.com/p/l2tp-ipsec-vpn [12:46] im getting this error when trying to upload changes file with dput: No host ppa:stlsaint/packagesources found in config [12:48] stlsaint: Is this from debians dput (does not support normal ppa syntax)? [12:54] arand: yes i am on debian [12:55] stlsaint: You can specify PPAs individually http://paste.debian.net/117487/ [12:55] Alternatively, port over the PPA argument magic from the ubuntu version into debian ;) [12:56] arand: is that the dput conf? [12:56] stlsaint: ~/.dput.cf yes [12:57] arand: aye gotcha [12:58] arand: is it suppose to be blank at first? [12:58] stlsaint: yes, /etc/dput.cf is read initially ~/.dput.cf overrides/adds, I assume.. [13:01] arand: so where you have ubuntu should i have debian? [13:01] arand: i tried running and it still gave same error [13:02] stlsaint: I'm also on debian, your's should be very similar, just replacing ppa names (redeclipse ppa unstable, in my case) [13:04] arand: i changed everything but still same error, same as yours except i change the name and ppa on all three stanzas [13:05] The title in [] is the one to use when dputting, eg, "dput ppa-unstable *.dsc In my case to upload to my ppa:arand/unstable" [13:07] arand: hrm one sec [13:08] arand: this is command im using: dput ppa:stlsaint/packagesources lxdm_0.3.0-0ubuntu5.1_i386.changes [13:09] stlsaint: You want to replace the "ppa:stlsaint/packagesources" with whatever to named the entry for that ppa (in my case, e.g. just "resvn") [13:12] arand: so i have a ppa on lp name packagesources, that should go in place of where you have resvn? [13:14] stlsaint: the thing within [] can be anything, but that is the name you will have to use when nivoking dput. [13:14] Gotta run, sorry [13:14] here is the commadn i use: dput ppa:stlsaint/packagesources lxdm_0.3.0-0ubuntu5.1_i386.changes [13:14] shucks [13:33] dang im slow [13:33] arand: i see what you were saying now [13:34] i fixed that but now im getting something about no signature on that file but signed the package at build [13:35] crap never mind on that either === Quintasan_ is now known as Quintasan [15:32] hi, I want to push to a ppa a fixed version of a package, tp_smapi. Is there any easy trick (debuild -something?) to help the infrastrcture to notice the diff? [15:33] any docu related to this case? [15:34] wrong channel? :) === EvilMTeck is now known as MTecknology [16:00] jtaylor: I built the package locally, could be I broke something, I'll push up to my SRU PPA to see if I get the same failure [16:02] micahg: wait I think I want to add a small change to the patch [16:02] the dependency on python-twisted is too strong, it only needs -core and -web [16:03] it just slipped in in the dh_python2 transition [16:07] or would that change also first have to be uploaded to debian? in this case its probably not worth it [16:08] jtaylor: changes must be made in the dev release first [16:10] then one can probably ignore it, it just installes a few extra packages [16:11] jtaylor: yeah, that's not worth fixing in an SRU [16:11] jtaylor: but Debian should get the fix if appropriate [16:12] yes going to commit it to the vcs soon [16:46] stlsaint: Sorry for rushing off, you got it working? [16:48] arand: well the command went thru but there is still nothing in my ppa [16:48] arand: though it said upload was successful [16:49] stlsaint: And the build queue is empty as well, and yo've received no confirmation by email? [16:50] Hmm, yea it appears to be completely empty... [16:52] stlsaint: The source (debiuld -S) is signed with one of the gpg keys that are specified on LP as well? [16:54] arand: yes [16:56] Then I'm not quite sure, I think then you'd have followed the same procedure as I normally do, and unless LP is in flux currently, I don't know what the issue would be .. :/ [16:57] arand: cool thanks [17:11] jtaylor: confirmed, I get the error on upgrading python-foolscap [17:12] I don't, what could be the cause? it upgrades in debian and a clean chroot and a oneiric vm [17:12] jtaylor: this is for the natty SRU [17:12] yes thats what I tried [17:13] also they are practially the same [17:13] no idea, I haven't come across this before [17:14] the only difference is a dropped breaks which was empty anyway [17:14] does the version in oneiric install an your natty machine? [17:15] * micahg can check [17:16] jtaylor: yep, that worked [17:16] oO [17:24] I don't get why it doesn't happen on my machine, but I guess I need to add a Break: python-foolscap (<< 0.6.1-3) [17:30] jtaylor: that version won't work for the SRU, I wonder what the difference is [17:33] lets see if I can get piuparts to run === ximion_ is now known as ximion [18:00] friggn piuparts, worked once with the debian version and now it fails on the same file with some weird unrelated error ... [18:11] micahg: as I can't reproduce the problem can you try adding that to the control: http://paste.ubuntu.com/610697/ [18:11] althout it should not be necessary ... [18:26] jtaylor: yep, in a bit, about to head out for lunch [18:43] Hello. [18:44] I was wondering if Bug 702316 is under review and if maybe there is anything else that I need to do for the patch to get applied. [18:44] Launchpad bug 702316 in clipit (Ubuntu Natty) "Generic Libindicate fallback support breaks applications (such as clipit) on non-Unity WMs/DEs" [Medium,In progress] https://launchpad.net/bugs/702316 === ximion_ is now known as ximion [18:49] evaluate: yeah, I have the patches, I just need to test build and upload [18:49] micahg, cool. [18:49] evaluate: you did everything right, the process has just been a little laggier than usual because of the recent release, UDS, etc. [18:49] evaluate: sorry, my piloting ran a little late, so I figured to wait until today to upload [18:49] but since I already looked at it, wanted to claim it [18:51] micahg, would enabling the indicator need a patch too? [18:52] evaluate: hmm, technically, probably should be since it's 2 different things [18:54] Thing is, I disabled it by default in the source package, since it generated a lot of problems for maverick/lucid, but I think it should work fine in natty, so it would be nice if it was enabled by default. [18:55] evaluate: right, and it seems that it was supposed to be enabled in the last upload but I don't think the merger noticed your source change [18:58] ok, so what would be the right thing to do to have it enabled in this case? [18:58] Should I submit another patch? Or edit the current patch? [18:59] evaluate: have a named patch to enable the indicator and a named patch to fix the crash [18:59] evaluate: if you want to fix it up, I can grab it over the weekend [19:00] Can I leave the current patch as it is and submit another one for enabling the indicator, or do you want me to also rename the current one? [19:00] evaluate: I thought the current one enables it [19:00] all in one patch that is [19:01] Ohh, yes it does. [19:01] Forgot that I also had that change in there. [19:01] oh, it's set to auto, not yes, but I think with the build flag we're good [19:02] Well, auto means that if the appindicator-0.1 package is present (which is set as a build-dep in ubuntu) it will default to yes. [19:03] ok then, so can I leave the patch as it is? [19:03] evaluate: yeah, are you going to push these changes into Debian as well? [19:05] Yes. I have sent the 1.3.13 version which includes these changes to my mentor, but it couldn't be uploaded due to some transitions going on in Debian and now I didn't bug him anymore about it since I want to roll out 1.4.0 soon anyway. [19:06] evaluate: k, I'll upload this to oneiric later today or over the weekend so we can get the SRU in [19:07] Good. Thank you very much! [19:07] evaluate: thank you for your work! [19:09] evaluate: actually, I might move the enable indicator piece out since it's not part of the upstream patch, but I can do that locally [19:10] Yeah, I enabled the indicator only in the 1.4 branch. [19:19] jtaylor: works now... [19:19] strange, it should not need that and why does the oneiric version work [19:20] jtaylor: no idea [19:21] jtaylor: if you want to have someone else test, I have the original in ppa:micahg/sru-test [19:30] ok from there I could reproduce it [19:31] jtaylor: cool :) [19:31] bdrung_: ping [19:32] jtaylor: so, do you want to discuss with some python gurus if the breaks/replaces is correct or if there's a better fix for this? [19:32] #debian-python says no replace needed, but I'll just check whats the difference to the package I built [19:33] jtaylor: k, well, you can show them the package in the PPA and the error and see if they change their minds :) [19:43] Quintasan: pong [19:43] bdrung_: sponsor-patch is broken I think [19:44] sponsor-patch --builder=pbuilder-dist 712534 --workdir kamoso [19:44] AssertionError: kamoso/kamoso_2.0-0ubuntu2.dsc does not exist. [19:44] ls kamoso/kamoso_2.0-0ubuntu2.dsc -> kamoso/kamoso_2.0-0ubuntu2.dsc [19:44] the file is there [19:44] Quintasan: that is fixed in trunk. can you file a bug report and add the needed stuff for a SRU? [19:45] bdrung_: well, I can, but later, uploading 24mb bug stacktrace now @_@ [19:45] Quintasan: thanks [19:47] tumbleweed: we need a online test infrastructure for these kind of bugs ^ [19:58] bdrung_: the version for SRU should be 0.123? [19:58] * Quintasan notes you might as well upload that to oneiric [19:59] Quintasan: 0.122.1 [20:07] Quintasan: http://paste.ubuntu.com/610750/ [20:08] oh [20:17] bdrung_: bug 785923 [20:17] Launchpad bug 785923 in ubuntu-dev-tools (Ubuntu Natty) "[SRU]ubuntu-dev-tools update" [Undecided,New] https://launchpad.net/bugs/785923 [20:17] I think I'm doing it right [20:18] bdrung_: Are you going to upload a newer version to oneiric soon? [20:19] Quintasan: yes [20:19] cool [20:28] Quintasan: uploaded to proposed [20:29] oh wait..what? so fast [20:29] :D [20:29] Quintasan: in rare cases :) [20:30] :) [20:36] broder: re bug #785854 - what's the reason for the broken pipe? [20:36] Launchpad bug 785854 in ubuntu-dev-tools (Ubuntu) "[sponsor-patch] (and probably others) don't reset SIGPIPE when shelling out" [Undecided,New] https://launchpad.net/bugs/785854 [20:37] bdrung_: what was the issue you say we need testing for? [20:37] oh that one [20:37] * tumbleweed sees the paste [20:38] tumbleweed: btw i am releasing 0.123 [20:38] tumbleweed: i give a few seconds to veto [20:39] cool [20:47] bdrung_: because diff got to the end of its input? there are a bunch of programs that expect SIGPIPE, not EPIPE [20:47] cjwatson walks through the basics at http://www.chiark.greenend.org.uk/ucgi/~cjwatson/blosxom/2009-07-02-python-sigpipe.html [20:48] because python changes the SIGPIPE handler, and signal handlers don't get reset on fork/exec, the program that expects SIGPIPE starts getting EPIPE instead, and gets confused [20:51] meh, we probably need to work around that in a bunch of places (or wrap subprocess, or use this subprocess32 backport) [20:52] tumbleweed: i may have a subprocess wrapper that resets a bunch of generally undesirable behavior [20:52] i need to dig it up [20:52] but why doesn't it always fail? [20:52] bdrung_: it'd only be an issue if the pipeline shuts down before the processes are done [20:52] i.e. if we stop being interested in output [20:53] bdrung_: there's some raciness involved. i don't remember the details [20:54] I guess it also breaks subprocess' childern [20:55] right, that would be what's happening here, i think [20:56] yeah [21:23] micahg: not even break works for me when installing from a ppa, but conflicts seems to work. I'll try to understand that and then probably go over debian [21:26] (although its not affected, I don't know why) [22:13] hi! i can't seem to be able to build avogadro with oneiric in a pbuilder, is there a problem with the gcc package ? [22:14] http://paste.ubuntu.com/610811/ << error in pbuilder [22:14] shadeslayer: update your pbuilder [22:14] afaik it's updated, but i'll check again [22:14] bleh ... outdated mirrir [22:14] *mirror [22:18] broder: didn't work [22:18] same issue [22:18] shadeslayer: all that error means is that the gcc packages you have installed don't match the version of the gcc packages in the archive [22:19] although it could be a publishing skew issue of some sort [22:19] okay, but i switched to the main archives [22:19] i know doko was in the middle of updating gcc for something, so maybe just wait a day [22:19] yeah looks like it [22:55] jtaylor: you should use breaks w/replaces [22:55] micahg: does not work [22:56] micahg: only conflicts [22:56] micahg: debian work fine, Im slowly thinking this is a dpkg bug [22:56] tries to remove a file from the new package instead of the old [22:56] jtaylor: breaks/replaces works for me with the package I built locally upgrading from the one in the repo [22:58] micahg: not for me, its kind of a heisenbug, as earlier I could not reproduce it to begin with [23:00] http://paste.ubuntu.com/610839/ [23:02] jtaylor: weird, upgrading from natty to oneiric version was fine, upgrading to version in my PPA + breaks/replaces was fine for me [23:02] yes my oneiric vm worked fine too [23:03] and according to debian policy it should work without the breaks too [23:03] (which it does in debian) [23:28] bdrung_: any reason for me not to upload the eclipse FTBFS fix over the weekend? === webjadmin is now known as JackyAlcine [23:38] micahg: no [23:39] micahg: can you check if 3.6.2 is affected too? (master-3.6 git branch) [23:39] bdrung_: k, thanks [23:39] i have only amd64 [23:39] (and there it builds) [23:42] bdrung_: idk if I'll be able to get to it, but I'll see === Guest45201 is now known as JackyAlcine