| [03:40] <teward> how long does it typically take for packages to be published to ppas after building? | |
| [04:07] <wgrant> teward: 5-10 minutes | |
| [08:16] <karni> Hi folks. I can't subscribe to a mailing list, because LP is timing out on me. Any pointers? Error ID: OOPS-d0ebcf5d853707cd9f623aa778b26324 | |
| [08:16] <ubot5> https://oops.canonical.com/?oopsid=OOPS-d0ebcf5d853707cd9f623aa778b26324 | |
| [08:36] <apachelogger> aloha, is there a way to find out if a package has been fully published through the API? | |
| [08:38] <geser> what is "fully published" for you? | |
| [08:40] <apachelogger> all binaries of all succesfully built builds of all published sources of all done uploads are published (available in the archive repo in question) | |
| [08:40] <apachelogger> I actually have everything but the binary part figured out | |
| [08:41] <geser> have you checked the status field of the binary_package_publishing_history object? | |
| [08:42] <apachelogger> yep | |
| [08:42] <apachelogger> I can actualy show you the entire gloriously crappy code, sec ^^ | |
| [08:43] <apachelogger> http://paste.ubuntu.com/8471508/ line 94 onwards is the intersting bits | |
| [08:44] <apachelogger> so, I thought I could use source_pub_history.getPublishedBinaries (line 156) but as it turns out that will not cover all potential binaries that will need publishing | |
| [08:45] <apachelogger> namely if i386 finished a bit after amd64 but not quite close enough the source_pub will be published, the build_pubs will be published and all (known) binaries will be published and programatically it is as though i386 never created any binaries at all | |
| [08:46] <apachelogger> related output from the pasted script when exactly this happened tonight: http://paste.ubuntu.com/8471521/ | |
| [08:51] <geser> hmm, is the output still wrong when you rerun the script for that upload now? | |
| [08:54] <apachelogger> geser: nope it's correct now | |
| [08:57] <apachelogger> it appears to me that there is a time gap between buildstate reaching 'successfully built' and its binaries appearing in getPublishedBinaries(), that's also where line 154 comes from. there is a chance that one can have everything indicate that the build is done but getPublishedBinaries() returns nothing yet | |
| [09:03] <wgrant> There is a gap, yes. The web UI uses getBuildSummariesForSourceIds. | |
| [09:07] <geser> apachelogger: could you use some wait after build.datebuilt before calling source.getPublishedBinaries()? (if wgrant call tell you how big the gap is) | |
| [09:09] <wgrant> There's no reliable gap | |
| [09:10] <wgrant> For very bad historical reasons, binaries only show up in getPublishedBinaries once the publisher starts. | |
| [09:10] <apachelogger> hm | |
| [09:11] * apachelogger takes a look at getBuildSummariesForSourceIds | |
| [09:11] <wgrant> getBuildSummariesForSourceIds takes the PackageUpload status into account. | |
| [09:11] <wgrant> It's what powers the right column on a PPA's +packages page. | |
| [09:11] <wgrant> Which shows when builds are in progress, or publishing, or all done. | |
| [09:16] <apachelogger> wgrant: that sounds like what I need | |
| [09:16] <apachelogger> geser, wgrant: thanks I'll give this a try :) | |
| [09:56] <florinel> Hi guys, I get a 403 response code when I try to download a package from a VPS Server hosted on Glesys (Swedish hosting company): wget http://ppa.launchpad.net/iacobs/redis/ubuntu/pool/main/r/redis/redis-server_2.8.14-0~ppa3_amd64.deb --2014-10-01 13:55:31-- http://ppa.launchpad.net/iacobs/redis/ubuntu/pool/main/r/redis/redis-server_2.8.14-0~ppa3_amd64.deb Resolving ppa.launchpad.net (ppa.launchpad.net)... ::ffff:108.166.117.240 | |
| [09:57] <florinel> is the host blacklisted or something? the provider does not filter any traffic | |
| [09:57] <florinel> or how can I proceed to find out what can I go from here | |
| [09:58] <wgrant> florinel: Have you tried using native IPv4? | |
| === karni is now known as karni-otp | |
| [09:59] <florinel> wgrant: good point, no, trying it now. | |
| [10:02] <florinel> wgrant: damn, that worked. Thank you. | |
| [10:02] <wgrant> florinel: I'm not quite sure why what you tried didn't work. Do you have some strange networking setup? | |
| [10:09] <florinel> wgrant: I am starting to look into it, why would it behave like this. | |
| === karni-otp is now known as karni | |
| === karni is now known as karni-lunch | |
| [13:07] <teward> wgrant: is it possible the builders last night were lagging with uploads? Had to wait 25 mins after building for it to publish | |
| === karni-lunch is now known as karni | |
| [14:00] <glev> Am I in the right place for a cloud-init question? | |
| [14:02] <glev> I wanted to know if fs_setup needs to be explicity set in the cloud.cfg modules list in order to declare it in my user-data | |
| [14:03] <dobey> you probably want to ask in #ubuntu-server perhaps? or #juju (if that's a channel)? | |
| [14:05] <glev> ok thanks, I will give that a try | |
| [14:17] <cjwatson> teward: there was an incident where a libvirt upgrade took out the networking on the virtualised builders, yes | |
| [14:17] <cjwatson> don't know whether that would have affected built -> published | |
| === mpt_ is now known as mpt | |
| === lan3y is now known as Laney | |
| === Pici` is now known as Pici | |
| [16:45] <teward> cjwatson: ok, thanks for the heads up on that, was that all fixed? | |
| [16:45] <cjwatson> so I'm told | |
| [17:01] <t1mp> hello | |
| [17:02] <t1mp> I'm getting "Sorry, there was a problem connecting to the Launchpad server." when trying to load https://code.launchpad.net/~tpeeters/ubuntu-ui-toolkit/20-sm/+merge/236750 | |
| [17:02] <t1mp> is there a known problem? | |
| [17:02] <t1mp> now I get: | |
| [17:02] <t1mp> 504 Gateway Time-out | |
| [17:02] <t1mp> The server didn't respond in time. | |
| [17:02] <gcollura> t1mp, I have the same error | |
| [17:03] <gcollura> everything else works though, just the merge interface seems broken | |
| [17:08] <allenap> Hello. Yes, I only call when I want something. I’m getting lots of timeouts looking at code. For example https://code.launchpad.net/~newell-jensen/maas/pxe-requests/+merge/236368 ends in “Please try again” consistently, for about 5-10 minutes now. | |
| [17:17] <allenap> cprov: Hi Celso, do you look after Launchpad these days? I’m having timeout issues, and so are a few others. ^ | |
| [17:19] <cjwatson> both those pages loaded first time for me FWIW | |
| [17:19] <cjwatson> I expect wgrant will want the OOPS ID from the timeout | |
| [17:20] <cprov> allenap: it worked for me too, temporary hiccup, maybe | |
| [17:21] <allenap> cprov, cjwatson: Yes, it’s working again for me too. Sorry for the noise. Unfortunately I didn’t write down the OOPS number :-/ | |
| [17:26] <cprov> allenap: it will show up in the reports, if it's working still. I will take a look later | |
| [17:26] <allenap> cprov: Ta, thank you. | |
| [19:11] <sethj> When I attempt to load a merge proposal at this link https://code.launchpad.net/~sethj/ubuntu-packaging-guide/lp1313053/+merge/236766 I get "Please try again Sorry, there was a problem connecting to the Launchpad server" and asking I let people know here. So here I am :) | |
| [19:11] <sethj> FWIW I tried reloading the page several times, but it always terminates at the same error. | |
| [21:29] <mibofra> hi guys, a thing. For the debs in ubuntu, on the launchpad like https://launchpad.net/ubuntu/+source/empathy, I find a debian.tar.xz tarball under the package details, and this tarball contains the debian directory that is used by the debian packaging tools to package the software into a deb package | |
| [21:29] <mibofra> now, I can't find a debian.tar.xz for linux :D | |
| [21:29] <mibofra> where I can find it ? | |
| [21:30] <mibofra> *can I | |
| [21:40] <mibofra> someone? | |
| [21:41] <mapreri> mibofra: looks like the linux package in ubuntu is not using the debian source format 3.0, but the 1.0, so you don't have a debian.tar.<compressionused>. instead the linux package in debian does, if you want it. | |
| [21:46] <mibofra> thanks mapreri | |
| [21:47] <mapreri> yw | |