| cjwatson | Laney: To succeed | 07:28 |
|---|---|---|
| Laney | grarg | 07:30 |
| Laney | Maybe jenkins is having a sad time anyway; can't see that those jobs are being run or get to the internal instance | 07:32 |
| cjwatson | We may have to force that one if nobody can be bothered to fix it - but the list of autopkgtests is sufficiently short that I think it's worth additional incentives to get the failures to zero routinely | 07:32 |
| cjwatson | Also possible that proposed-migration is wrong that they're running | 07:33 |
| Laney | I do see pango in https://jenkins.qa.ubuntu.com/view/Saucy/view/AutoPkgTest/portlet/dashboard_portlet_29/ | 07:34 |
| cjwatson | There's probably something wrong at my end. May look at it when I'm more awake ... | 07:35 |
| Laney | No need to spend weekend time on it really if it'll block on Libreoffice anyway | 07:37 |
| Laney | I'll just hit Sweetshark with it on Monday | 07:37 |
| * Laney goes to catch a train to Huntingdon | 07:37 | |
| Laney | ttfn | 07:38 |
| cjwatson | Huntingdon> I'm so sorry | 07:39 |
| === doko_ is now known as doko | ||
| Laney | points failure has me stuck at Grantham | 09:38 |
| Laney | not sure if that's better or worse ... | 09:39 |
| * Noskcaj is away: I'm either at school or soccer. or i just don't like you. | 09:45 | |
| Noskcaj | stupid xchat | 09:46 |
| * apw notes that the autopkgtest for linux is marked as running in proposed-migrations (autopkgtest for linux 3.9.0.7.8: RUNNING | 10:13 | |
| apw | ) any idea where i might find the said run to see when it started, how long it takes normally etc | 10:14 |
| cjwatson | I think it's probably a bug that it's listed as still running. | 10:15 |
| cjwatson | I was going to look at it at some point over the weekend. | 10:15 |
| cjwatson | https://jenkins.qa.ubuntu.com/view/Saucy/view/AutoPkgTest/ | 10:16 |
| * apw idly oneders if cjwatson is every offline :) | 10:16 | |
| apw | ever | 10:16 |
| apw | there is definatly something wrong with this keyboard, sigh | 10:16 |
| cjwatson | Hm, that doesn't show a recent run. | 10:16 |
| cjwatson | Has your autopkgtest ever passed? Doesn't really look like it. | 10:17 |
| apw | assuming it is saucy-adt-run, there was one 17 hours ago, that would be about the right time | 10:17 |
| cjwatson | Oh, back in Feb for raring apparently | 10:18 |
| * apw thought that their autopackage run was 'empty' and just triggered a kernel build | 10:18 | |
| cjwatson | apw: saucy-adt-linux. That run looks too early | 10:18 |
| * apw checks the config is even in the tree | 10:18 | |
| cjwatson | There's something wrong here but I think it needs jibel. How about I force it for now | 10:19 |
| apw | no great shakes either way i suspect | 10:20 |
| cjwatson | Do see if you can fix it though. Try running 'adt-run linux_blah.dsc --- adt-virt-schroot saucy-amd64' on a machine with a saucy-amd64 chroot | 10:20 |
| cjwatson | schroot that is | 10:20 |
| apw | cjwatson, yeah wil have a poke see what we added. i seem to remember it being empty | 10:21 |
| apw | cjwatson, though this saucy run looks about right. source published 21 hours ago, run was 17 hours ago | 10:21 |
| apw | (right time wise) | 10:21 |
| cjwatson | Maybe I can't do timezone maths | 10:22 |
| cjwatson | Or maybe I trusted the changelog date rather than the upload date :) | 10:23 |
| apw | oh, hmm, good point, maybe i am not taking them into account at all ;) | 10:23 |
| apw | though reading the jenkinsjob it does indeed seem it did nothing at all instead of testing right | 10:23 |
| Laney | https://jenkins.qa.ubuntu.com/view/Saucy/view/AutoPkgTest/job/saucy-adt-linux/39/ARCH=i386,label=adt/console | 10:24 |
| Laney | Does look like the right version | 10:24 |
| Laney | badpkg: Test Depends field contains an invalid dependency `' adt-run: erroneous package: Test Depends field contains an invalid dependency `' | 10:24 |
| * apw goes look at what we had in raring and what we carried forward, perhaps it got broken then | 10:24 | |
| apw | Laney, we do have an empty Depends: that was something pitti said we needed | 10:25 |
| apw | now what it might mean, erm, not sure | 10:25 |
| Laney | Well, see if you can reproduce and fix locally I guess. | 10:26 |
| Laney | Maybe the apec has something to say about it | 10:26 |
| apw | cjwatson, Laney, yep seems reasonable i'll take this away and poke it | 10:26 |
| Laney | Oho, looks like we are arriving in the promised land of Peterborough. See you. | 10:27 |
| Laney | (apec/spec) | 10:27 |
| apw | Laney, heh good luck with _that_ | 10:29 |
| jibel | apw, empty Depends field makes adt-run fail. The spec doesn't mention how it should behave in that case. | 11:29 |
| jibel | Strictly speaking, the meaning of an empty Depends field would be "do not install any package for the test" which is different from no Depends field which means "install package(s) generated by the source package" | 11:30 |
| jibel | but I never received an answer from upstream on this point. | 11:30 |
| jibel | apw, for linux adt test, if it is a rebuild test, you should probably have Depends: build-essential | 11:30 |
| jibel | instead of nothing | 11:30 |
| apw | jibel, ahh really, i can do that... | 11:43 |
| apw | jibel, do we do autopkgtest on anything other than development do you know ? | 11:51 |
| jibel | apw, we do autopkgtest on dev and latest stable. Precise and Quantal are kept only to run MaaS package tests. But going forward, with the increasing number of packages autopkgtest-able, I think it's helpful for SRUs to keep them running on all supported releases. | 19:44 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!