[01:56] folks, launchpad is fucked [01:56] http://bazaar.launchpad.net/~ubuntu-core-dev/pulseaudio/ubuntu/annotate/head%3A/debian/patches/0091-dont-load-cork-music-on-phone.patch [01:57] whenever i try to access that i only get "Please try again" [01:57] i am asked to report that here [01:57] so here i got [01:57] s/got/go [01:58] http://bazaar.launchpad.net/~ubuntu-core-dev/pulseaudio/ubuntu/annotate/head%3A/debian/patches/0053-fix-sigsegv-module-bluetooth-device.patch [01:59] doesn't work either [02:00] mezcalero: works for me [02:02] yes, seems it unfucked itself now === oubiwann_ is now known as oubiwann [05:12] is the debian/changelog entry what controls which version of ubuntu is targeted by a package? [05:13] schmichael: yes [05:13] micahg: thanks. if i want to build for say jaunty & karmic, can i upload the same version of my package with just the changelog line switched? [05:14] you'll have to change the package version as well I believe [05:14] k, so i should just add a new changelog entry for each release i want to target? [05:14] that would work [05:14] it's also suggested to make the older releases a smaller version [05:15] like ~jaunty === mwhudson_ is now known as mwhudson [05:25] why isn't my ppa signed? [05:25] schmichael: what do you mean? [05:26] all PPAs are signed [05:26] https://launchpad.net/~schmichael/+archive/ppa [05:26] ^ does it just not show the sig info for my own ppa? [05:26] because i don't see sig info on that page like i do for other ppa pages [05:27] hmm [05:27] micahg: are you seeing sig info under Technical Details? [05:27] nope [05:27] didn't notice that before [05:28] huh, weird... [05:29] schmichael: did you just create it? [05:30] micahg: on friday [05:30] bung 452730 [05:30] bug 452730 [05:30] Launchpad bug 452730 in soyuz "lp_publish/ppa-generate-keys is broken atm" [Critical,In progress] https://launchpad.net/bugs/452730 [05:30] could be why [05:32] micahg: thanks === ripps_ is now known as ripps === jtv1 is now known as jtv === ahma_ is now known as ahma === geser_ is now known as geser [10:16] hello [10:16] is there a way to delete a ppa? [10:16] agateau: no, but we can disable them [10:17] bigjools: oh ok, is this something a ppa user can do? [10:17] agateau: no, admin only. You neeed to file a question on the soyuz project and we'll get it looked at. [10:18] bigjools: ok [10:18] thanks [10:18] np === agateau is now known as agateau|lunch === agateau|lunch is now known as agateau === mrevell is now known as mrevell-lunch === mrevell-lunch is now known as mrevell === thunderstruck is now known as gnomefreak === ursula is now known as Ursinha === thunderstruck is now known as gnomefreak [15:58] hi there again [15:58] need assistance with relase url glob pattern [15:59] how to set it? === rockstar changed the topic of #launchpad to: Read https://help.launchpad.net for help | Help contact: rockstar | Join https://launchpad.net/~launchpad-users | This channel is logged: http://irclogs.ubuntu.com/ | Launchpad is open source: see channel #launchpad-dev [15:59] src: http://php-fpm.org/downloads/0.6/archive/ [15:59] dest: https://launchpad.net/php-fpm/+download [16:00] dreamcat4, that's a very good question. [16:00] when i put Release url pattern = http://php-fpm.org/downloads/0.6/archive/* [16:01] then its creating 1 release per file [16:10] what i can't understand is how the release name is determined, for example: [16:11] filename: php-fpm-0.6-5.3.1.tar.gz => release name: "0.6-5.3.0 release" [16:11] filename: php-fpm-0.6-5.3.0-r103.tar.gz => release name: "0.6-5.3-r103 release" [16:12] so its seems like whatever the file glob pattern, or the project name it doesn't matter [16:13] it will be the first numerical digit encountered until the first dot '.' encountered after the last digit ? [16:13] sunzui ? [16:20] i'm going to submit a bug report (as feature improvement request) [16:25] sinzui, hi [16:26] hi rockstar, dreamcat4 [16:26] hi [16:26] I have uploaded a cryptographically signed package to a PPA, and the PPA has been signed (*after* the time of upload). There's a security warning when attempting to install the package (yes, I have remebered apt-get update after using apt-key to add the public key of the PPA). Is this because the PPA was not signed at the time when the package was uploaded? Should I remake the package to fix this problem, assuming this is the case? (I'd rather n [16:26] dreamcat4: The full filename is used to extract a debian compatible version [16:27] dreamcat4: versions == milestones, and their names are unique to the project not the series. [16:27] AskHL: wait a bit longer, there's a backlog of PPAs to be signed as the signing process was down for a while [16:28] dreamcat4: This is a nasty requirement to accommodate projects that are in a project group. [16:28] oh right [16:29] i just googled for "debian compatible version" and got nothing [16:30] bigjools, the PPA has been signed now though (wgrant helped me with this very issue last friday or so, and the signature finally appeared today). But doesn't this mean that everything should be correct now? Or am I waiting for the individual packages in the PPA to be sort of 'cross-signed' with the newly existing key? (I'm not a crypto expert) [16:30] anyway sinuzi, do you think its worthwhile to file a bug for extending this feature slightly ? [16:31] bigjools, to clarify, everything has a fingerprint in launchpad. [16:31] whereas before the problem was some things did not have one. [16:32] But now, in spite of the fact that everything appears to have a fingerprint, it's still not validating. But I can wait till later if there's a reason to expect this to change automatically [16:32] dreamcat4: We tried to fix this issue earlier this year. You are welcome to file bug since launchpad engineers believe this is a design flaw, but I do not think this something we can work on in the next year [16:33] dreamcat4: But I personally would love to see this nuisance fixed. [16:38] hey can i see the script that process these filenames / version number strings ? [16:38] sure [16:38] * sinzui looks === kfogel changed the topic of #launchpad to: Read https://help.launchpad.net for help | Help contact: kfogel | Join https://launchpad.net/~launchpad-users | This channel is logged: http://irclogs.ubuntu.com/ | Launchpad is open source: see channel #launchpad-dev [16:40] dreamcat4: http://bazaar.launchpad.net/~launchpad-pqm/launchpad/devel/annotate/head%3A/lib/lp/registry/scripts/productreleasefinder/finder.py [16:41] thanks [16:41] AskHL: Which is the PPA in question? [16:41] kfogel, how are you help contact? I thought I was help contact today. [16:41] maxb, https://launchpad.net/~askhl/+archive/ppa/ [16:41] rockstar: are you on CHR today? [16:42] kfogel, yessir. That's why I put my name in the topic. [16:42] rockstar: and if you are, update https://help.launchpad.net/HelpRotation please... [16:42] rockstar: ah, I think I know what's going on. [16:42] rockstar: were you aware of the new week-long CHR initiative? [16:42] kfogel, we have a Google calendar that shows help contact. [16:42] AskHL: There is no signature (Release.gpg file) here: http://ppa.launchpad.net/askhl/ppa/ubuntu/dists/jaunty/ [16:42] rockstar: the schedules have been completely redone; see my recent mail to canonical-launchpad@... [16:43] rockstar: we have a google calendar that needs updating then, and thanks, 'cause I didn't know about it. [16:43] kfogel, I saw talk of it, but I hadn't seen anyone pull the trigger. [16:43] maxb, oh. One moment... [16:43] rockstar: trigger was pulled. jml did the first week last week; I'm this week. [16:43] rockstar: which means you're off the hook this week, but if yuo'd like to add yourself for next week, that'd be great! (it's still uncovered) [16:43] kfogel, okay. It's all yours then. [16:43] rockstar: where is this google cal? I need to fix it. [16:44] AskHL: The problem is that Launchpad doesn't go back and sign PPAs that need it after it adds a signing key to them - they remain unsigned until some other change occurs to them, like uploading/deleting/copying a package [16:44] (And this confuses many people and really ought to be fixed.) [16:45] maxb, so apparently when the PPA *has* been signed, uploading should result in automatic signing - the problem is that the key wasn't there when the package was uploaded. Do I understand correctly? [16:46] So now I basically just copy the package. Although the destination package should be sort of the same as the present package. [16:46] AskHL: Yes, except not "when the PPA has been signed" - you mean "when the PPA has had a signing key generated for it" [16:47] maxb, ah. That was actually quite clarifying. [16:48] maxb and bigjools, thank for the help [16:48] thanks* [16:48] AskHL: If you like, you can copy maxb-dummy-package-to-trigger-ppa-key-generation from https://edge.launchpad.net/~mercurial-ppa/+archive/ppa/+copy-packages?field.name_filter=&field.status_filter=superseded&field.series_filter= [16:49] If you do, be sure to copy it into jaunty, not karmic [16:54] maxb, I suppose I should delete the original one and then copy back to it afterwards [16:54] I'm not sure that that's possible [16:54] or just copy on top of it [16:54] or that either [16:54] Hmmm okay. [16:54] let me test it [16:55] I copied things on top of each other previously, having edited the version number in the changesfile. That appeared to work [16:56] It overwrote the differently named package within the same series (so there would be only one jaunty package). (This was in a different PPA) [16:57] I don't understand what you mean [16:58] But testing on staging suggests that deleting and then copying might work. Be sure to select "Copy existing binaries" though [16:58] Uh, well. I had written the wrong version (something with 0ubuntu1) and found out that ppa1, ppa2, ... is the preferred form. So I changed the changesfile and rebuilt, and it replaced the previous one [17:00] Oh, so no "copying" involved [17:01] Not at that point. BUT: [17:01] I had previously used the copy feature to get karmic/intrepid/jaunty versions [17:01] and then I used copy on the newly renamed ones (ppa3), and that replaced the 0ubuntu1 ones. [17:02] https://launchpad.net/~pyg3t-dev-team/+archive/ppa/+packages <-- there used to be packages named something-0ubuntu1 on this list, and I didn't actively delete them [17:03] Which brings me to the next point, that they replaced the previous file within the same series. So if I copy the present python-ase (jaunty) to the same series, wouldn't that be some sort of conflict? [17:04] And there's no other way to copy to the jaunty series, AFAI can see [17:05] Sorry, all this might be confusing... [17:07] Ah, right - yes, new versions supersede existing ones within the same series [17:08] If you copy the present package to the same series, it will just tell you it's already there and do nothing. [17:08] maxb, right. That just happened. [17:09] So therefore, copy to *different* series, then delete the previous. Then copy back to jaunty [17:09] Rather inelegant solution though, but presumably it gets the job done [17:11] sinzui, it might avoid the debian version number if we replaced all dash '-' characters with underscore characters '_' [17:11] i will try this and see the result tomorrow [17:12] hmm [17:12] Yes that is correct [17:12] sinzui here is link which explains debian package version [17:12] http://manpages.ubuntu.com/manpages/jaunty/en/man5/deb-version.5.html [17:13] i don't know if this python script complies 100% (it may not) [17:14] sinzui, bye for now === salgado is now known as salgado-lunch [17:22] maxb, thank you! Copying an 'alien' package into the repository (precisely as you had said a long time ago) made it generate the sigfile correctly. I was just a bit slow to understand what was going on. === beuno is now known as beuno-lunch === rmcbride_ is now known as rmcbride === deryck is now known as deryck[lunch] === danilos is now known as danilos-afk === salgado-lunch is now known as salgado === deryck[lunch] is now known as deryck === jon is now known as Guest60630 === beuno-lunch is now known as beuno === matsubara is now known as matsubara-lunch === abentley is now known as abentley-lunch [18:51] i built some packages while there was a launchpad bug preventing ppa signing from occurring, do i need to re-upload my packages to get them signed? [18:53] schmichael, I had the same problem. Copy anything from some different PPA into the current PPA and it will make a sigfile. [18:53] (the copied package can then be deleted) [18:53] AskHL: my ppa has a sig, but when i try to install a package it says its untrusted [18:54] (must be copied to the same series - jaunty, karmic or whatever) [18:54] my ppa: https://launchpad.net/~schmichael/+archive/ppa/ [18:54] but 'apt-key list' shows the key [18:55] schmichael, yes. I had the same problem. If you go to a different PPA and copy a package from that into your own PPA with the right series, then it'll trigger creation of the key [18:55] err, trigger *signing* with the now available key [18:55] weird... [18:56] Yeah it is a bit weird. [18:57] http://ppa.launchpad.net/schmichael/ppa/ubuntu/dists/jaunty/ <-- when it works, a .gpg file should reside in this directory [18:58] (Release.gpg to be exact) [18:58] AskHL: thanks, working on it now [19:08] lfaraone: you're needing a bug import done? [19:16] rockstar, bug 320068 [19:16] Launchpad bug 320068 in launchpad-code "Allow setting merge proposal status when doing reviews" [High,Fix released] https://launchpad.net/bugs/320068 [19:17] rockstar, does that command work from the web UI? [19:17] jml, ah, no. I was under the impression that was about setting it through the email interface. [19:18] jml, could you re-open it with a comment that you want it through the web ui? [19:18] will do. :) [19:18] jml, thanks. It's still low hanging fruit, which is what I'm going through and squishing right now. I'm sure it won't take long to accomodate that. [19:19] rockstar, I've updated the title too. [19:19] jml, thank you sire. [19:19] Er, sir. You're no sire to me. :) [19:20] rockstar, glad to hear it! === abentley-lunch is now known as abentley === matsubara-lunch is now known as matsubara [19:37] what's the best place to get help on packaging? i'm trying to figure out how to properly handle a package that creates a new user [19:39] try #ubuntu-motu [19:40] cody-somerville: thanks, asking in there [19:52] https://launchpad.net/ubuntu/+milestone/ubuntu-9.10 [19:52] broken ? :s [20:17] Hi, I'm trying to support a friend who (don't ask me how, I don't know) is missing a lot of basic packages, like libc6, debconf, coreutils etc. The problem is that I can't find those packages with a standard ubuntu sources.list. So my question is, is there any repository in launchpad or elsewhere that contains those files? [20:18] alkisg, those absolutely are in the standard sources.list [20:18] apt-cache policy debconf [20:18] debconf: Installed: (none) Candidate: (none) Version table: [20:19] alkisg, either way, this is a broken ubuntu system, so you should ask in #ubuntu [20:19] beuno: thank you. [20:19] they *are* there by default, and are not especially available anywhere in Launchpad [20:20] Thanks, I don't see them in the mirrored repositories, so I assumed they'd be in launchpad somewhere [20:20] E.g. I can download debconf manually from https://launchpad.net/ubuntu/jaunty/i386/debconf/1.5.26ubuntu3 [20:20] yes, you can download all packages manually from Launchpad [20:21] but I'm sure that's not what you want to do ;) [20:21] Right :D [20:21] (I wonder how is he still able to boot his system!) [20:35] alkisg: debconf, libc6, and every other package on Ubuntu systems are all in the repository. [20:35] http://archive.ubuntu.com/ubuntu/pool/main/d/debconf/ [+] for instance. [20:35] soren: thanks, it was indeed a problem with his sources.list [20:36] If his system thinks they're not installed, his /var/lib/dpkg tree is likely busted. [20:37] Ugh, I think it'd be easier if he just reinstalled his system :) [20:39] alkisg: Perhaps. Without any more information, it's hard to tell. [20:39] alkisg: It may be as simple as a filesystem that didn't get mounted for some reason. [20:39] soren: any way to regenerate /var/lib/dpkg tree? [20:40] (sorry for abusing the channel) [20:40] alkisg: Depends on how badly (and how) it is broken. [20:40] soren dpkg -l dpkg => reports uninstalled [20:40] alkisg: Does /var/lib/dpkg even exist? [20:41] Yes, it does [20:41] Is it empty? [20:41] It has some directory structure, but no files [20:41] (available, info, statoverride etc) [20:42] no *.deb files, I meant [20:42] * alkisg compares the dir to his own.. [20:42] *deb files do not belong there. [20:44] Yes, sorry. It looks pretty much the same as mine, except for some a big status-bad file (1.5 M) [20:44] ...and the "status" file being too small (2K) [20:45] Heh, I copied the status-bad file over the "status" one, and all look ok now! :O [20:46] Thanks a lot soren :) [21:14] mdke: here? === huayra_ is now known as huayra === funkyHat is now known as funkyBass === funkyBass is now known as funkyHat [22:23] mwhudson: briefly [22:24] mdke: i commented on that bug report of yours about a corrupt branch [22:26] mwhudson: just read that, thanks. Any idea of the cause? [22:26] mdke: i didn't really try to think about that [22:27] mwhudson: I'm just wondering if it's a bzr bug, then it might be worth keeping the branch around. I'm obviously keen on fixing it however it can be done :) [22:27] looks like the pack-names file is mangled?? [22:28] mwhudson: I'm going to be no help diagnosing the cause. Can the branch be kept around somewhere for the bzr guys to look into? [22:29] oh maybe not [22:29] mdke: sure [22:31] mwhudson: ok, sounds good to me then [22:31] come on, lftp must have some version of 'cp -r'... [22:32] mwhudson: while you're here, I tried pushing a different branch for the same project to that project (to work around the branch breakage, I got the one from ~ubuntu-branches/ubuntu/gnome-user-docs, but it wouldn't let me push it to the gnome-user-docs project because it was in a different format to the "stacked" branch (gnome-user-docs/master). Can I work around that or must all branches in a project have the same format? [22:32] mdke: you can bzr push --unstacked [22:34] mwhudson: ok, thanks === flacoste is now known as flacoste_afk [22:40] mwhudson: thanks for the assistance [22:40] mdke: np [22:44] mdke: grr, my copy of the branch seems to be fine :/ === matsubara is now known as matsubara-afk [23:57] kfogel: yes, I was talking with deryck about it. Essentially, we want to test import-export functionality prior to a full migration.