=== allquixotic_ is now known as allquixotic [08:30] Adri2000: copy-ppa-pkg.py, now with Debian support. [08:37] great, thanks [08:52] Adri2000: Note that the default mode of operation for it is to copy the binaries, too, but seeing as Launchpad doesn't have the binaries frm Debian, you need to pass --rebuild (or -r for short). [08:59] right [09:56] aloha [09:59] hiya [09:59] any problem with Loggerhead ? [09:59] I think it's not only loggerhead [10:00] http://paste.ubuntu.com/808371/ [10:00] I'm having troubles reporting a bug as well - it's stuck at "Please wait while bug data is processed" [10:00] for at least five minutes now [10:00] uh [10:00] http://identi.ca/launchpadstatus has no info [10:00] dholbach, yeah, i'm getting similar issues when trying to get branches too [10:00] diwic: Some background jobs are offline for a few minutes. [10:00] While we perform a DB upgrade. === chrisccoulson_ is now known as chrisccoulson [10:01] The codehosting stuff is unrelated, will investigate in a couple of minutes. [10:01] wgrant, so if I just hold out for a couple of more minutes it will eventually succeed, or should I start over and report a new bug? [10:01] ahh [10:01] wgrant, you are a hero [10:02] diwic: It will succeed. [10:02] wgrant, ok, thanks! [10:02] We just turn cronjobs off a few minutes before we rip the DB out from underneath everything :) [10:03] oh, actually it didn't succeed. It failed with the "Uh oh!" message. But will file a new bug later. [10:04] Refresh [10:04] That uhoh is the DB outage message from the 90 seconds of downtime. [10:04] It's back up now. [10:05] dholbach, eLBati: Is it working now? [10:06] yes [10:06] thanks [10:06] * dholbach hugs wgrant [10:07] wgrant, , yes thanks [10:10] wgrant, is the sponsor db change going in right now? [10:11] dholbach: The DB change was what was deployed just then, yeah. [10:11] awesomeness [10:11] I think. [10:11] Maybe not. [10:11] I think that's next in the queue, actually. [10:11] wgrant, do we have something that retroactively changes the old data? :) [10:11] ah, nevermind - it's not like I need it RIGHTNOW :) [10:11] it was just curiosity [10:11] Sorry, tonight was Ursinha's blueprint changes. [10:12] Tomorrow will be SPPH.sponsor. [10:12] go go Ursinha! :) === adeuring changed the topic of #launchpad to: Help contact: adeuring | https://launchpad.net | Launchpad is Open Source | This channel is logged === matsubara-afk is now known as matsubara === Quintasan_ is now known as Quintasan [12:48] Does Launchpad have a username squatting policy? === mdeslaur is now known as mdeslaur_ === mdeslaur_ is now known as mdeslaur === rick_h changed the topic of #launchpad to: Topic for #launchpad: Help contact: rick_h | https://launchpad.net | Launchpad is Open Source | This channel is logged [13:05] adeuring: got irc [13:05] rick_h: thanks! === luxigo_ is now known as luxigo [14:57] Hi! Can someone please help me fix an openid merged accounts problem? https://answers.launchpad.net/launchpad/+question/184522 [15:05] maploin: looking [15:15] rick_h, I've got IRC now. === deryck changed the topic of #launchpad to: Topic for #launchpad: Help contact: deryck | https://launchpad.net | Launchpad is Open Source | This channel is logged [15:16] maploin, I'm helping with this now, instead of rick_h. Checking with someone to see what we can do. [15:17] ok, thanks! [15:17] please tell me if there's anything else you need from me [15:22] maploin, I've assigned stub to the question. He's the engineer who can fix up the data for you. [15:23] deryck: ok, thanks [15:26] maploin: Should be fixed now. You might need to log in and out of various systems for the change to be noticed. [15:26] howdy -- is it possible to take a private ppa, and make it public? [15:27] kirkland: it's not possible [15:27] why didn't the retracer make this bug 878281 public? [15:27] Launchpad bug 878281 in modemmanager (Ubuntu) "modem-manager crashed with SIGSEGV in g_main_context_dispatch()" [Medium,Incomplete] https://launchpad.net/bugs/878281 [15:27] bigjools: even if i delete the ppa, and create a new one with the same name? [15:27] maploin, are you an openstack user? [15:28] kirkland: you can't create another with the same name [15:28] bigjools: okay [15:28] we need to fix that [15:29] sinzui: yes, that's where this problem started [15:29] deryck, rick_h, we discovered that openstacks offsite identity is not openid compliant and it assumes Lp is the openid proviser [15:29] provider [15:30] ah ok [15:30] maploin, stuart can fix the data to work with openstack, but I think openstack needs to recognise login.ubuntu.com provides the identity and look up the Lp user from there [15:31] ok [15:36] But that probably isn't your problem ;) [15:36] IIRC bugs have been filed; wgrant tracked down the source with an OpenStack user in Budapest. [15:37] stub, correct. I will ask wgrant to update the bug. === maco2_ is now known as maco === zyga_ is now known as zyga === deryck is now known as deryck[lunch] === zyga is now known as zyga-break [18:12] Anybody want to help tidy up a little bit of the launchpad database? https://answers.launchpad.net/launchpad/+question/183550 [18:13] hi, trying to register a new project, getting an error in its name, but LP doesn't give me advice on how to recover [18:13] The name 'u1.link' has been blocked by the Launchpad administrators. [18:14] seems to follow the rules, at least [18:14] possibly no . allowed? [18:14] that's what i thought but... [18:14] "At least one lowercase letter or number, followed by letters, numbers, dots, hyphens or pluses. " [18:14] *snerk* or .link is a keyword in their parser and it's barfing. thatd be funny [18:15] that's why i'm hoping some LP person can tell me what's up. :) [18:19] achiang, I'm pretty sure you can't use u1 === deryck[lunch] is now known as deryck [18:20] yeah, achiang, I think it's reserved name, like beuno says. hi beuno :) [18:20] * deryck is checking to confirm [18:20] o/ deryck [18:20] beuno: boo. :( there's a long tradition in open source of derivative/related projects incorporating names of underlying technology in their own names. [18:20] KMail [18:20] gBrainy [18:21] &c. [18:21] achiang, yeah, I don't think it's a problem, just that it's blocked by default [18:22] i'm not going to fight very hard for this, but i will note that a) the fact that U1 has an API encourages devs to build stuff on top of it and b) due to close relationship with ubuntu, it is likely these projects would use LP to host, vs github [18:25] achiang, I think it's something we routinely enable for people as it comes up, but we block by default due to branding. as I understand it, anyway. [18:25] I can't find where we store these blocked names, though. [18:26] sinzui or bac, could one of you perhaps help. Don't we block certain names from projects, i.e. u1 or ubuntu, and where would I find this info? [18:26] hello - how can I switch a ppa from public to private? [18:26] spena, you can't if you ever uploaded anything to the ppa. [18:27] deryck: we do, but the list only exists in the db [18:27] ah ha [18:27] debfx, I haven't uploaded anything yet. [18:27] debfx, it is a new ppa [18:27] debfx, it's not you, this is for deryck [18:27] deryck: the correc thing to do is create the project with a temporary name and ask a webops to rename it [18:27] deryck: i think you have a reasonable policy today; i'm only pointing out that i think this might be an increasingly common problem [18:28] gotcha. thanks, bac [18:28] deryck, I haven't uploaded any package yet [18:29] achiang, yeah, perhaps. can you create with a temp name and I'll ping someone to change it for you. [18:29] spena, ok. hold on a sec. [18:29] deryck: will do, thanks [18:31] achiang, yes, it's absolutely not a problem to use u1 in the name, it's just blocked by default to discourage abuse. Sorry about that. [18:34] rye: that is running. I will let you know when it completes [18:34] beuno: deryck: ok, thanks. i wonder how many other potential U1 developers never come into irc to ask these questions. just a thought [18:36] achiang, indeed. I will raise this issue. [18:36] spena, so we don't do private ppas without a commercial subscription. were you wanting to do that, or do you have a commercial subscription? [18:50] bac: deryck: NO about the project name! [18:51] sinzui, sorry, what? we can't change the project name for him? [18:52] bac, deryck, If the name is blocked for someone we know should be permitted to have the project name, we know we need to adjust https://launchpad.net/+nameblacklist [18:52] deryck, bac: look at the u1 rule, I think that should have the same team as ubuntuone [18:53] ah [18:53] sinzui, but achiang isn't in ubuntuone team. he would have to go through them for the name then. is that what we would expect? [18:53] thanks sinzui [18:53] deryck, bac: the rule is that the admin of a name is entitled to create the projects and teams of that name. Next week, they should also be able to set privacy settings too [18:53] ah ok [18:53] that makes sense. [18:54] this is not how you gain developer mindshare, fwiw [18:55] ~ubuntuone now admins the u1 namespace. [18:56] makes sense [18:59] achiang, I realize it's frustrating to have to jump through hoops for the name you want, but I hope you can understand the precautions taken here. [18:59] or at least the reason for the precaution is what I mean. [19:00] deryck: i am fine with it for me, because i am willing to jump through these hoops. my only concern is regarding the level of motivation for other people not familiar with this "process" [19:01] Is it just me or is launchpad being really slow right now? [19:01] achiang, deryck, We could set the admin of all canonical namespaces to ~canonical, but that means all staff need to know each groups rules. I doubt that is going to happen [19:03] sinzui: deryck: i think your reasoning is perfectly justified, and i'm not arguing against your precautions. i am simply raising the issue of unintended consequences. i think you get my point now, so i'll stop bugging you. :) [19:03] thanks [19:04] right, I do get it. it is something to consider, for sure. [19:04] abentley, I'll hand off to you now, sir. === deryck changed the topic of #launchpad to: Topic for #launchpad: Help contact: abentley | https://launchpad.net | Launchpad is Open Source | This channel is logged [19:13] sinzui: what stuff don't you know about on LP :) [19:14] I really do not know anything about Lp except milestones. My guesses about Lp's esoteric nature are often assumed to be knowledge. I am a fraud. [19:15] deryck, I have a commercial subscription, do you want me to give your the ppa url? === mwhudson_ is now known as mwhudson [19:52] hey guys are there some issues with the LP server [19:52] taking ages to load the login screen [19:53] eagles0513875: working ok here, local network issues? Logged out/back in without any delays [19:53] rick_h: no network at home is fine [19:54] now its loading but very very slow not doing anythign network intensive either [19:54] spena, sorry, on call. yes, please give me the url. and the project name you have the subscription for please. [19:54] havent logged in on the mac that im using before this is first time [19:54] rick_h: it's a bit slpw this evening [19:54] *slow [19:55] hmmm, it's loading pretty snappy for me, like for rick_h [19:56] yea, wondering if there's some pipe busted along the path. I've tried it from here on my mobile and from an offsite server elsewhere with <2s results [19:56] but both are US based [19:56] deryck, https://launchpad.net/~gazzang/+archive/development [19:56] deryck, project name is "Gazzang Team" [19:58] spena, done. [19:58] deryck, thank you. [19:58] soren: I've hit a bug in copy-ppa-pkg that happens when trying to sync a specific package from debian sid: the assert len(source_pkgs) == 1 at line 99 fails. I guess it's because launchpad knows about two different versions: https://launchpad.net/debian/+source/sivp [19:58] deryck, if I wanna create another ppa, I must tell your? or is there an option on my account to switch it to private? [19:59] rick_h: sry i dced from my znc but im in europe central europe and its taking ages over https [20:00] spena, yes, unfortunately, you have to request each time. [20:01] spena, you don't have to ask me personally. just ask here, or file a question against launchpad. [20:01] deryck, ok, thanks [20:01] np [20:02] Adri2000: What's the comand line you're using? [20:02] Adri2000: Oh, you're probably right. === yofel_ is now known as yofel [20:02] Adri2000: Is that because there's some stale ones due to not all arches building properly? [20:05] Adri2000: I guess I could let you specify a specific version rather than a source series and pocket. [20:07] im seriously having an issue here [20:10] soren: yes, though I don't understand why LP keeps the two versions. as you told me LP only imports source packages from debian, so it's not a problem of building correctly on all arches, is it? [20:14] Adri2000: No, but it may look to Launchpad as though Debian has two different versions in unstable. [20:15] sivp | 0.5.2-2 | http://ftp.fr.debian.org/debian/ unstable/main Sources [20:15] sivp | 0.5.3-3 | http://ftp.fr.debian.org/debian/ unstable/main Sources [20:15] indeed ^ [20:16] so, is there a way for copy-ppa-package to automatically select the most recent version? or otherwise to let the user specify the version, as you said [21:35] Adri2000: Sure. I'll get that done tomorrow. [21:41] what is this: https://launchpadlibrarian.net/90385693/buildlog.txt.gz [21:41] bzr dailydeb crash [21:52] that looks like an exciting bug [21:52] jtaylor: can you give a link to the recipe? [21:53] mwhudson: thttps://code.launchpad.net/~jtaylor/+recipe/ipython-daily [21:53] could it be a non ascii char in the changelog? [21:53] Ożarowski is in there [21:53] jtaylor: ah, i think 0.4 format recipes are broken [21:53] can you try changing it to # bzr-builder format 0.3 deb-version ... [21:54] k [21:54] changed and started a new build [21:54] 3 hours ... :( [21:55] nice :/ [21:58] would be nice if arch all packages could be built on other arches, amd64 queue is much shorter [22:00] yes === abentley changed the topic of #launchpad to: Topic for #launchpad: Help contact: - | https://launchpad.net | Launchpad is Open Source | This channel is logged [22:16] mwhudson: it apparently got past that point now, thx [22:16] jtaylor: yay [22:40] gna, it really sucks that lp builders forbids binding to localhost now, how is one supposed to test code that uses networking? [22:41] Hi -- a package I uploaded to my PPA failed, and I can't see the build log. It just says "No Such Resource" [22:44] spena: Are you using Chromium, and is the PPA private? [22:45] jtaylor: We don't forbid binding to localhost. What suggests that? [22:45] wgrant, yes, both of them [22:45] It's possible that name lookups don't work, however. [22:45] spena: Try Firefox instead. [22:45] spena: Chromium does some illegal reencoding of URLs. [22:46] wgrant: my (zmq based) packages bind to 127.0.0.1 and fail on lp builders but neither locally nor on debians builders [22:46] all with, cannot bind to port 127.0.0.1 [22:47] Do you have a link to the build log? [22:47] I also tried this make lp builder clone chroot script, it worked there too [22:47] wgrant, thanks, that was the reason. It worked on firefox [22:48] wgrant: see e.g. https://bugs.launchpad.net/ubuntu/+source/pyzmq/+bug/910757 [22:48] Launchpad bug 910757 in pyzmq (Ubuntu) "pyzmq ftbfs on amd64 and i386 (test failures)" [High,Confirmed] [22:48] spena: Great. It only affects packages with a ~ in the version. Chromium special-cases that and always reencodes it, in violation of just about everything :/ [22:48] wgrant: it runs on the arm and armhf builders, its just i386 and amd64 [22:49] powerpc also works [22:51] jtaylor: Does it fail on PPA builders too? [22:51] jtaylor: Can you convince it to not suppress the errno? [22:51] yes [22:52] and maybe [22:55] wgrant: 22 [22:56] assuming zmq maps the errno 1:1 which I'm not totally sure [22:56] EINVAL? Odd. [22:57] EINVAL The requested socket type is invalid. [22:57] from http://api.zeromq.org/2-1:zmq-socket [23:10] 15237 socket(PF_INET, SOCK_STREAM|SOCK_CLOEXEC, IPPROTO_TCP) = -1 EINVAL (Invalid argument) [23:12] elmo: thats the error you get on a builder? [23:13] jtaylor: that's an error I can see in an strace in a similar environment to the buildd [23:14] with pyzmq? [23:14] yes [23:14] isn't SOCK_CLOEXEC something new-ish? [23:14] the buildds are running a hardy kernel [23:15] http://paste.ubuntu.com/809158/ <-- my (obvious) test case [23:15] socket(2) says SOCK_CLOEXEC is >= 2.6.27 [23:16] thats interesting [23:16] * Fixed issue 273, O_CLOEXEC flag used in ip.cpp:192 is supported only... [23:16] so I need to upgrade the zmq library [23:17] elmo: thank you very much [23:17] jtaylor: no problem [23:22] are they going to be updated at some point? [23:23] jtaylor: I'd imagine so - but I'm not sure when, the platform folks tend to be pretty conservative when it comes to the build environment. I'd imagine as long as we need to build hardy, they'd want them on hardy [23:23] but ICBW [23:23] I want the testsuites to run, but I also want the flag enabled for users who run newer kernels :/ [23:24] jtaylor: hmm, yeah, that's sucky - I'm not honestly sure what to suggest [23:25] maybe make the test suite conditional on a modern kernel? it won't run on the PPAs, but at least it'll run when it can (e.g. presumably when folks build locally) [23:25] I'm surprised zmq can't/doesn't do the detection at run time though - but maybe that's naieve [23:28] Anything that does compile-time detection of kernel features is probably buggy. [23:29] I'll open a bug [23:47] Hello [23:48] I am wondering when will the staging code update will be done?