[00:38] doko: so nevermind that, the old freecad is qt4-only and impossible to build against new libcoin, restoring the new version instead [00:46] :) [01:08] doko: gsequencer> why did it manage to run on amd64 without running out of memory? [02:27] -queuebot:#ubuntu-release- Unapproved: cloud-init (cosmic-proposed/main) [18.4-7-g4652b196-0ubuntu1 => 18.5-17-gd1a2fe73-0ubuntu1~18.10.1] (edubuntu, ubuntu-cloud, ubuntu-server) [02:32] -queuebot:#ubuntu-release- Unapproved: cloud-init (bionic-proposed/main) [18.4-0ubuntu1~18.04.1 => 18.5-17-gd1a2fe73-0ubuntu1~18.04.1] (edubuntu, ubuntu-cloud, ubuntu-server) [03:10] -queuebot:#ubuntu-release- Unapproved: cloud-init (xenial-proposed/main) [18.4-0ubuntu1~16.04.2 => 18.5-17-gd1a2fe73-0ubuntu1~16.04.1] (edubuntu, ubuntu-cloud, ubuntu-server) [08:25] vorlon: now https://bugs.launchpad.net/ubuntu/+source/gsequencer/+bug/1813456 [08:25] Ubuntu bug 1813456 in gsequencer (Ubuntu) "gsequencer fails autopkg tests on anything except amd64" [Undecided,New] [08:38] oh, why does journalctl have to be so incredibly slow [08:39] trying to see why node-husl results are missing [08:43] seems like we are having issues, autopkgtest exits after 2 seconds [08:45] um, might have been log reading bug [08:53] my understanding is that autopkgtest results might not be published correctly now for other packages either, because it seems to me like we stop fetching results of node-husl 404s [09:10] I think the solution is to re-run the test and put the results into the bucket where they are missing in [11:08] juliank: it's in the listing https://objectstorage.prodstack4-5.canonical.com/v1/AUTH_77e2ada1e7a84929a74ba3b87153c0ac/autopkgtest-disco/?format=plain&prefix=disco/s390x/n/node-husl/20190127_022342_96aef@ but when you retrieve that file it 404s [11:08] https://objectstorage.prodstack4-5.canonical.com/v1/AUTH_77e2ada1e7a84929a74ba3b87153c0ac/autopkgtest-disco/disco/s390x/n/node-husl/20190127_022342_96aef@/result.tar [11:09] Laney: yeah, it's a bit odd [11:09] Laney: there were no upload errors [11:10] sounds like swift did a bad thing to me [11:10] I retried it, afaik a newer result will stop the downloader complaining [11:45] +1 [12:53] vorlon, I was trying to heal nodejs/amd64 autopkgtestsuite... [12:53] can't we do anything about that ENOMEMORY error in autopkgtest? [13:00] old binaries left on amd64: nodejs-dev (from 10.15.0~dfsg-9) [13:00] is it ok to decuft it? [14:15] doko: gap-io is trying to make connections to the network [14:19] doko: (and doesn't respect http_proxy, apparently) [14:21] vorlon: are Ubuntu autopkgtests allowed to use the Internet? [14:21] jbicha: yes, but they have to honor http_proxy [14:22] but also, if your test depends on a network resource to exist, it might be a bad test [14:24] I was thinking about adapting https://salsa.debian.org/gnome-team/gnome-menus/blob/debian/master/debian/gitlab-ci.yml to work as an autopkgtest [14:25] since Ubuntu patches that file, a Salsa CI may not be sufficient (Ubuntu packaging for gnome-menus isn't in Salsa) [14:25] and it doesn't look like we can redistribute the DTD [14:26] (but I am checking with xdg) [14:26] can't redistribute the DTD> then it shouldn't be a dependency of anything in main? [14:27] the dtd isn't a dependency [14:28] including a test dependency :) [14:28] tests shouldn't regress when someone on the Internet changes their server [14:29] they shouldn't change their server because of how dtd is supposed to work (but I opened some bugs about that with freedesktop this week too ☹️ ) [14:29] jbicha: "change their server" might mean "domain goes away / is revoked" [14:30] I think that risk is acceptable for this particular kind of test: we can always modify or remove this autopkgtest [14:30] anyway, perfect is the enemy of the good etc, and having no tests is generally a worse bug than having tests that might fail in some hypothetical future [14:31] I broke Budgie this week so I want to try to make it harder for someone to make my mistake next time [14:44] LocutusOfBorg: we can move nodejs to a larger instance type if that's what's needed; but why does it run out of memory only on amd64 and not on other archs? Is it not running the full test suite on other archs? [16:04] hi all [16:04] Who doubts that gdb run out of memory as running functional tests of GSequencer? === OldManWi1ter is now known as OldManWinter [16:35] doko: please tell me [18:06] joelkraehemann: see above (looking for gap-io) [18:24] doko: I don't know how to show history of channel [18:26] doko: gap-io is trying to make connections to the network [18:26] --> Kaleo (~Kaleo@lec67-4-82-235-57-50.fbx.proxad.net) has joined #ubuntu-release [18:26] doko: (and doesn't respect http_proxy, apparently) [18:26] gap-io: did you see my email telling about ... [18:26] 16 * 20 * 30 * 131072 = 1258291200 [18:27] please can we track this in the bug report? [18:27] fine [18:27] ta [18:40] LocutusOfBorg: I removed tox from proposed. there is a reason we have auto-syncs off [19:16] vorlon, I would try to see if it makes some change, maybe the testsuite has more tests on amd64 wrt other archs? [19:16] btw, you did a no-change rebuild of tpm2-pk11, but it needs to be kicked out https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=913368 [19:16] Debian bug 913368 in src:tpm2-pk11 "tpm2-pk11 (build-)depends on cruft packages." [Serious,Open] [19:17] porting seems not trivial (sadly the tpm2-tss package in new version is really buggy too) [19:57] vorlon, Laney, cjwatson: please re-enable the auto sync. the entabgled transitions are done [19:58] the drawback is a mir binary in main which isn't approved by security [21:04] doko: autosync reenabled [21:08] Laney: armhf possibly a little overpowered now, it finished the last backlog before s390x did [21:10] LocutusOfBorg: tpm2-pk11 removal done [21:11] -queuebot:#ubuntu-release- New binary: ntopng [amd64] (disco-proposed/universe) [3.8+dfsg1-2ubuntu1] (no packageset) [21:12] -queuebot:#ubuntu-release- New: accepted ntopng [amd64] (disco-proposed) [3.8+dfsg1-2ubuntu1] [21:14] thanks [21:15] lets see it migrate on next run [21:17] also ndpi hopefully [21:19] vorlon, gnss-sdr is out of testing... can we have it removed so gnuradio migrates? [21:20] possibly, though first I'm reproducing the segfaults so I can see what's what [21:21] the maintainer's analysis is certainly wrong, 0.0.10-2 previously built and it only fails on reubild [21:25] smirk, r-cran-curl autopkgtest passed on ppc64el because the test failed to install libcurl from -proposed [21:27] but fixed in unstable, so merging [23:21] -queuebot:#ubuntu-release- New binary: debian-med [amd64] (disco-proposed/universe) [3.2] (no packageset) [23:22] -queuebot:#ubuntu-release- New: accepted debian-med [amd64] (disco-proposed) [3.2] [23:36] -queuebot:#ubuntu-release- New binary: radare2 [i386] (disco-proposed/universe) [3.2.1+dfsg-2] (no packageset) [23:36] -queuebot:#ubuntu-release- New binary: radare2 [amd64] (disco-proposed/universe) [3.2.1+dfsg-2] (no packageset) [23:37] -queuebot:#ubuntu-release- New binary: gatb-core [s390x] (disco-proposed/universe) [1.4.1+git20181225.44d5a44+dfsg-1] (no packageset) [23:37] -queuebot:#ubuntu-release- New binary: gatb-core [amd64] (disco-proposed/universe) [1.4.1+git20181225.44d5a44+dfsg-1] (no packageset) [23:39] -queuebot:#ubuntu-release- New binary: syslog-ng [amd64] (disco-proposed/universe) [3.19.1-2] (no packageset) [23:39] -queuebot:#ubuntu-release- New binary: xmlmarshaller [amd64] (disco-proposed/universe) [1.0.1-2] (no packageset) [23:40] -queuebot:#ubuntu-release- New binary: salmid [amd64] (disco-proposed/universe) [0.1.23-1] (no packageset) [23:41] -queuebot:#ubuntu-release- New binary: igor [amd64] (disco-proposed/universe) [1.3.0+dfsg-1] (no packageset) [23:41] -queuebot:#ubuntu-release- New binary: igor [i386] (disco-proposed/universe) [1.3.0+dfsg-1] (no packageset) [23:42] -queuebot:#ubuntu-release- New binary: gatb-core [i386] (disco-proposed/universe) [1.4.1+git20181225.44d5a44+dfsg-1] (no packageset) [23:43] -queuebot:#ubuntu-release- New binary: gatb-core [ppc64el] (disco-proposed/universe) [1.4.1+git20181225.44d5a44+dfsg-1] (no packageset) [23:44] -queuebot:#ubuntu-release- New binary: syslog-ng [i386] (disco-proposed/universe) [3.19.1-2] (no packageset) [23:46] -queuebot:#ubuntu-release- New binary: orthanc-wsi [i386] (disco-proposed/universe) [0.6-1] (no packageset) [23:46] -queuebot:#ubuntu-release- New binary: radare2 [s390x] (disco-proposed/universe) [3.2.1+dfsg-2] (no packageset) [23:48] -queuebot:#ubuntu-release- New binary: syslog-ng [s390x] (disco-proposed/universe) [3.19.1-2] (no packageset) [23:49] -queuebot:#ubuntu-release- New binary: orthanc-wsi [amd64] (disco-proposed/universe) [0.6-1] (no packageset) [23:51] -queuebot:#ubuntu-release- New binary: igor [s390x] (disco-proposed/universe) [1.3.0+dfsg-1] (no packageset) [23:52] -queuebot:#ubuntu-release- New binary: orthanc-wsi [s390x] (disco-proposed/universe) [0.6-1] (no packageset) [23:53] -queuebot:#ubuntu-release- New binary: radare2 [ppc64el] (disco-proposed/universe) [3.2.1+dfsg-2] (no packageset) [23:56] -queuebot:#ubuntu-release- New binary: igor [ppc64el] (disco-proposed/universe) [1.3.0+dfsg-1] (no packageset) [23:56] -queuebot:#ubuntu-release- New binary: syslog-ng [ppc64el] (disco-proposed/universe) [3.19.1-2] (no packageset) [23:59] -queuebot:#ubuntu-release- New binary: orthanc-wsi [ppc64el] (disco-proposed/universe) [0.6-1] (no packageset)