| *** EmilienM_ has joined #openstack-infra | 00:04 | |
| *** odyssey4me has quit IRC | 00:11 | |
| *** odyssey4me has joined #openstack-infra | 00:11 | |
| *** s-shiono has joined #openstack-infra | 00:14 | |
| *** ianychoi has quit IRC | 00:14 | |
| pabelanger | ianw: it does look like nb03.o.o is able to hit github.com for images: http://nb03.openstack.org/ | 00:16 |
|---|---|---|
| pabelanger | ianw: having issues caching etcd it looks like | 00:16 |
| pabelanger | we haven't built an image in 3 days it looks like | 00:16 |
| *** ianychoi has joined #openstack-infra | 00:17 | |
| scas | #openstack-chef got the drive-by vandalism while i was in the air, it seems | 00:19 |
| pabelanger | ianw: oh, maybe there isn't a etcd-v3.1.10-linux-arm64.tar.gz release | 00:20 |
| pabelanger | scas: yah, happened in a few channels | 00:20 |
| *** ykarel has joined #openstack-infra | 00:21 | |
| pabelanger | ianw: okay, I think the issue it is picking up the stable branches versions | 00:23 |
| *** markvoelker has joined #openstack-infra | 00:24 | |
| *** hongbin has joined #openstack-infra | 00:24 | |
| *** edmondsw has joined #openstack-infra | 00:25 | |
| *** edmondsw has quit IRC | 00:29 | |
| *** d0ugal has quit IRC | 00:35 | |
| *** jamesmcarthur has joined #openstack-infra | 00:39 | |
| *** agopi has quit IRC | 00:40 | |
| *** shu-mutow-AWAY is now known as shu-mutow | 00:40 | |
| openstackgerrit | Paul Belanger proposed openstack-infra/project-config master: Pause ubuntu-xenial-arm64 image builds https://review.openstack.org/558134 | 00:40 |
| pabelanger | ianw: clarkb: ^ we'll have to pause builds for arm64 while we patch devstack | 00:41 |
| *** markvoelker has quit IRC | 00:42 | |
| *** markvoelker has joined #openstack-infra | 00:42 | |
| *** d0ugal has joined #openstack-infra | 00:43 | |
| *** EmilienM has quit IRC | 00:48 | |
| *** EmilienM_ is now known as EmilienM | 00:48 | |
| *** EmilienM has joined #openstack-infra | 00:48 | |
| *** VW has joined #openstack-infra | 00:49 | |
| *** olaph has joined #openstack-infra | 00:50 | |
| *** olaph1 has quit IRC | 00:52 | |
| *** dizquierdo has quit IRC | 01:03 | |
| *** salv-orl_ has joined #openstack-infra | 01:24 | |
| *** salv-orlando has quit IRC | 01:26 | |
| *** d0ugal has quit IRC | 01:35 | |
| *** namnh has joined #openstack-infra | 01:37 | |
| *** spzala has joined #openstack-infra | 01:38 | |
| prometheanfire | pabelanger: anything preventing me from using the gentoo image? | 01:38 |
| *** david-lyle has quit IRC | 01:40 | |
| *** dklyle has joined #openstack-infra | 01:40 | |
| *** d0ugal has joined #openstack-infra | 01:43 | |
| *** anteaya has quit IRC | 01:45 | |
| *** jamesmcarthur has quit IRC | 01:45 | |
| *** caphrim007 has joined #openstack-infra | 01:46 | |
| *** caphrim007 has quit IRC | 01:46 | |
| *** germs has quit IRC | 01:47 | |
| *** anteaya has joined #openstack-infra | 01:47 | |
| *** spzala has quit IRC | 01:47 | |
| pabelanger | prometheanfire: we need a glean release first, then land your patch for nodepool.o.o. I think we can do that the beginning of the week | 01:49 |
| prometheanfire | ah, right | 01:53 |
| *** yamahata has joined #openstack-infra | 02:02 | |
| *** chandankumar has joined #openstack-infra | 02:02 | |
| *** iyamahat has joined #openstack-infra | 02:03 | |
| *** xinliang has joined #openstack-infra | 02:03 | |
| *** chkumar246 has quit IRC | 02:03 | |
| *** jamesmcarthur has joined #openstack-infra | 02:04 | |
| *** dklyle has quit IRC | 02:05 | |
| *** david-lyle has joined #openstack-infra | 02:05 | |
| dmsimard | Gentoo in the gate. That's pretty cool :) | 02:05 |
| *** chandankumar has quit IRC | 02:08 | |
| *** jamesmcarthur has quit IRC | 02:08 | |
| prometheanfire | :D | 02:11 |
| prometheanfire | it'll make it fairly easy to test new python versions | 02:11 |
| *** edmondsw has joined #openstack-infra | 02:13 | |
| openstackgerrit | Xinliang Liu proposed openstack-infra/project-config master: Add debian stretch nodes to arm64/amd64 clouds https://review.openstack.org/557902 | 02:17 |
| *** kiennt26 has joined #openstack-infra | 02:17 | |
| *** edmondsw has quit IRC | 02:17 | |
| *** chandankumar has joined #openstack-infra | 02:20 | |
| *** spzala has joined #openstack-infra | 02:23 | |
| *** caphrim007 has joined #openstack-infra | 02:24 | |
| *** dsariel has quit IRC | 02:26 | |
| *** spzala has quit IRC | 02:27 | |
| *** janki has joined #openstack-infra | 02:29 | |
| *** psachin` has joined #openstack-infra | 02:33 | |
| *** d0ugal has quit IRC | 02:37 | |
| *** gcb has joined #openstack-infra | 02:40 | |
| *** andreas_s has joined #openstack-infra | 02:42 | |
| *** jamesmcarthur has joined #openstack-infra | 02:43 | |
| *** d0ugal has joined #openstack-infra | 02:43 | |
| *** andreas_s has quit IRC | 02:47 | |
| *** ianychoi has quit IRC | 02:48 | |
| *** zhangfei has joined #openstack-infra | 02:50 | |
| *** ramishra has joined #openstack-infra | 02:52 | |
| *** kiennt26 has quit IRC | 02:54 | |
| *** janki has quit IRC | 02:57 | |
| *** ianychoi has joined #openstack-infra | 02:58 | |
| *** Jeffrey4l_ has joined #openstack-infra | 02:59 | |
| *** pgadiya has joined #openstack-infra | 03:08 | |
| *** spzala has joined #openstack-infra | 03:11 | |
| *** spzala has quit IRC | 03:11 | |
| *** ianychoi has quit IRC | 03:12 | |
| *** ianychoi has joined #openstack-infra | 03:13 | |
| *** Jeffrey4l_ has quit IRC | 03:22 | |
| *** l4yerffej_ has joined #openstack-infra | 03:22 | |
| *** janki has joined #openstack-infra | 03:22 | |
| *** janki has quit IRC | 03:23 | |
| *** l4yerffej_ has quit IRC | 03:23 | |
| *** janki has joined #openstack-infra | 03:23 | |
| *** l4yerffej has joined #openstack-infra | 03:23 | |
| *** yamamoto has joined #openstack-infra | 03:26 | |
| *** rwsu has joined #openstack-infra | 03:26 | |
| *** pgadiya has quit IRC | 03:29 | |
| *** zhongjun_ has quit IRC | 03:39 | |
| *** andreas_s has joined #openstack-infra | 03:41 | |
| *** pgadiya has joined #openstack-infra | 03:42 | |
| *** garyk has quit IRC | 03:44 | |
| *** ykarel has quit IRC | 03:46 | |
| *** andreas_s has quit IRC | 03:47 | |
| *** germs has joined #openstack-infra | 03:47 | |
| *** germs has quit IRC | 03:47 | |
| *** germs has joined #openstack-infra | 03:47 | |
| *** hongbin has quit IRC | 03:49 | |
| *** germs has quit IRC | 03:53 | |
| *** dhajare_ has joined #openstack-infra | 04:03 | |
| *** ykarel has joined #openstack-infra | 04:15 | |
| *** d0ugal has quit IRC | 04:23 | |
| *** sree has joined #openstack-infra | 04:26 | |
| *** psachin` has quit IRC | 04:27 | |
| *** psachin has joined #openstack-infra | 04:27 | |
| *** sree has quit IRC | 04:29 | |
| *** sree has joined #openstack-infra | 04:29 | |
| *** ianychoi has quit IRC | 04:35 | |
| *** ianychoi has joined #openstack-infra | 04:36 | |
| *** VW has quit IRC | 04:38 | |
| *** Sukhdev has joined #openstack-infra | 04:44 | |
| *** d0ugal has joined #openstack-infra | 04:45 | |
| *** links has joined #openstack-infra | 04:46 | |
| *** jchhatbar has joined #openstack-infra | 04:51 | |
| *** jchhatbar has quit IRC | 04:51 | |
| *** jchhatbar has joined #openstack-infra | 04:52 | |
| *** janki has quit IRC | 04:54 | |
| *** ianychoi has quit IRC | 05:00 | |
| *** ianychoi has joined #openstack-infra | 05:02 | |
| *** ykarel_ has joined #openstack-infra | 05:03 | |
| *** chkumar246 has joined #openstack-infra | 05:04 | |
| *** ykarel has quit IRC | 05:04 | |
| *** links has quit IRC | 05:04 | |
| *** chandankumar has quit IRC | 05:04 | |
| *** links has joined #openstack-infra | 05:05 | |
| *** ykarel__ has joined #openstack-infra | 05:05 | |
| *** ykarel_ has quit IRC | 05:08 | |
| *** chkumar246 has quit IRC | 05:08 | |
| *** ykarel__ is now known as ykarel | 05:08 | |
| *** ktsuyuzaki has joined #openstack-infra | 05:13 | |
| *** Sukhdev has quit IRC | 05:13 | |
| *** ktsuyuzaki is now known as kota__ | 05:14 | |
| *** kota_ has quit IRC | 05:16 | |
| *** kota__ is now known as kota_ | 05:16 | |
| *** eandersson has quit IRC | 05:16 | |
| *** eandersson has joined #openstack-infra | 05:17 | |
| *** e0ne has joined #openstack-infra | 05:20 | |
| *** chkumar246 has joined #openstack-infra | 05:21 | |
| *** iyamahat has quit IRC | 05:21 | |
| *** e0ne has quit IRC | 05:26 | |
| *** pgadiya has quit IRC | 05:30 | |
| *** pcichy has joined #openstack-infra | 05:35 | |
| *** jbadiapa has joined #openstack-infra | 05:35 | |
| *** jchhatbar has quit IRC | 05:43 | |
| *** jchhatbar has joined #openstack-infra | 05:43 | |
| *** d0ugal has quit IRC | 05:48 | |
| *** germs has joined #openstack-infra | 05:49 | |
| *** germs has quit IRC | 05:49 | |
| *** germs has joined #openstack-infra | 05:49 | |
| *** pgadiya has joined #openstack-infra | 05:51 | |
| *** germs has quit IRC | 05:53 | |
| *** d0ugal has joined #openstack-infra | 05:55 | |
| *** lpetrut has joined #openstack-infra | 06:09 | |
| openstackgerrit | vinay kumar muddu proposed openstack/diskimage-builder master: To check for the updates https://review.openstack.org/558163 | 06:12 |
| *** dsariel has joined #openstack-infra | 06:13 | |
| *** pcichy has quit IRC | 06:16 | |
| *** yamamoto_ has joined #openstack-infra | 06:18 | |
| *** dhajare_ has quit IRC | 06:20 | |
| *** yamamoto has quit IRC | 06:21 | |
| *** yamamoto has joined #openstack-infra | 06:29 | |
| *** chkumar246 is now known as chandankumar | 06:29 | |
| *** chandankumar has quit IRC | 06:29 | |
| *** chandankumar has joined #openstack-infra | 06:30 | |
| *** pgadiya has quit IRC | 06:31 | |
| *** yamamoto_ has quit IRC | 06:32 | |
| openstackgerrit | wangqi proposed openstack/diskimage-builder master: delete unused module https://review.openstack.org/558167 | 06:35 |
| *** pgadiya has joined #openstack-infra | 06:40 | |
| *** armaan has joined #openstack-infra | 06:44 | |
| *** dhajare_ has joined #openstack-infra | 06:53 | |
| *** HeOS has joined #openstack-infra | 06:55 | |
| *** chandankumar has quit IRC | 06:57 | |
| *** andreas_s has joined #openstack-infra | 07:01 | |
| *** imacdonn has quit IRC | 07:04 | |
| *** chandankumar has joined #openstack-infra | 07:04 | |
| *** imacdonn has joined #openstack-infra | 07:04 | |
| *** andreas_s has quit IRC | 07:05 | |
| *** alexchadin has joined #openstack-infra | 07:07 | |
| *** kjackal has joined #openstack-infra | 07:07 | |
| *** udesale has joined #openstack-infra | 07:08 | |
| *** tesseract has joined #openstack-infra | 07:12 | |
| *** jchhatba_ has joined #openstack-infra | 07:12 | |
| *** anteaya has quit IRC | 07:13 | |
| *** anteaya has joined #openstack-infra | 07:14 | |
| *** jchhatba_ has quit IRC | 07:15 | |
| *** jchhatbar has quit IRC | 07:15 | |
| *** jchhatba_ has joined #openstack-infra | 07:15 | |
| *** jchhatba_ has quit IRC | 07:17 | |
| *** annp has joined #openstack-infra | 07:19 | |
| *** xinliang_ has joined #openstack-infra | 07:27 | |
| *** e0ne has joined #openstack-infra | 07:31 | |
| *** kjackal has quit IRC | 07:35 | |
| *** kjackal has joined #openstack-infra | 07:35 | |
| *** amoralej|off is now known as amoralej | 07:40 | |
| *** andreas_s has joined #openstack-infra | 07:41 | |
| *** kjackal_ has joined #openstack-infra | 07:41 | |
| *** jtomasek has joined #openstack-infra | 07:42 | |
| *** kjackal has quit IRC | 07:42 | |
| *** bmace has joined #openstack-infra | 07:45 | |
| *** andreas_s has quit IRC | 07:46 | |
| *** andreas_s has joined #openstack-infra | 07:48 | |
| *** xinliang_ has quit IRC | 07:49 | |
| *** xinliang_ has joined #openstack-infra | 07:49 | |
| *** jtomasek has quit IRC | 07:50 | |
| *** xinliang_ has quit IRC | 07:51 | |
| *** xinliang has quit IRC | 07:51 | |
| *** jtomasek has joined #openstack-infra | 07:52 | |
| *** andreas_s has quit IRC | 07:53 | |
| *** xinliang has joined #openstack-infra | 07:54 | |
| *** xinliang has quit IRC | 07:54 | |
| *** xinliang has joined #openstack-infra | 07:54 | |
| *** xinliang_ has joined #openstack-infra | 07:57 | |
| *** xinliang_ has quit IRC | 07:57 | |
| *** garyk has joined #openstack-infra | 07:58 | |
| *** armaan has quit IRC | 07:59 | |
| *** xinliang3 has joined #openstack-infra | 07:59 | |
| *** zhurong has joined #openstack-infra | 08:02 | |
| *** gcb has quit IRC | 08:03 | |
| *** pcichy has joined #openstack-infra | 08:08 | |
| *** pgadiya has quit IRC | 08:11 | |
| *** dsariel has quit IRC | 08:14 | |
| *** armaan has joined #openstack-infra | 08:15 | |
| *** gcb has joined #openstack-infra | 08:17 | |
| *** jamesmcarthur has quit IRC | 08:19 | |
| *** pgadiya has joined #openstack-infra | 08:19 | |
| *** alexchadin has quit IRC | 08:22 | |
| *** alexchadin has joined #openstack-infra | 08:25 | |
| *** pcichy has quit IRC | 08:28 | |
| *** andreas_s has joined #openstack-infra | 08:36 | |
| *** d0ugal has quit IRC | 08:39 | |
| *** andreas_s has quit IRC | 08:42 | |
| *** shu-mutow is now known as shu-mutow-AWAY | 08:42 | |
| *** s-shiono has quit IRC | 08:44 | |
| *** jtomasek has quit IRC | 08:44 | |
| *** lihi has quit IRC | 08:49 | |
| *** lihi has joined #openstack-infra | 08:49 | |
| *** yamamoto has quit IRC | 08:51 | |
| *** ianychoi has quit IRC | 08:52 | |
| *** ianychoi has joined #openstack-infra | 08:54 | |
| *** numans has quit IRC | 08:57 | |
| *** numans has joined #openstack-infra | 08:58 | |
| *** yamamoto has joined #openstack-infra | 09:00 | |
| *** HeOS has quit IRC | 09:01 | |
| *** zhurong has quit IRC | 09:02 | |
| *** d0ugal has joined #openstack-infra | 09:03 | |
| *** bmace has quit IRC | 09:09 | |
| *** bmace has joined #openstack-infra | 09:10 | |
| *** kjackal_ has quit IRC | 09:11 | |
| *** kjackal has joined #openstack-infra | 09:11 | |
| *** aojea has joined #openstack-infra | 09:15 | |
| *** olaph1 has joined #openstack-infra | 09:17 | |
| openstackgerrit | Gaetan Semet proposed openstack-dev/pbr master: Add leading 0 on alpha release https://review.openstack.org/558181 | 09:18 |
| openstackgerrit | Gaetan Semet proposed openstack-dev/pbr master: Add leading 0 on alpha release in semver doc https://review.openstack.org/558181 | 09:18 |
| *** olaph has quit IRC | 09:18 | |
| *** HeOS has joined #openstack-infra | 09:20 | |
| *** ianychoi has quit IRC | 09:24 | |
| *** dsariel has joined #openstack-infra | 09:25 | |
| *** ianychoi has joined #openstack-infra | 09:25 | |
| *** pcichy has joined #openstack-infra | 09:25 | |
| *** olaph has joined #openstack-infra | 09:29 | |
| *** olaph1 has quit IRC | 09:32 | |
| *** sree_ has joined #openstack-infra | 09:35 | |
| *** sree_ is now known as Guest73591 | 09:35 | |
| *** alexchadin has quit IRC | 09:36 | |
| *** sree has quit IRC | 09:37 | |
| *** alexchadin has joined #openstack-infra | 09:38 | |
| *** armaan has quit IRC | 09:40 | |
| *** chandankumar has quit IRC | 09:40 | |
| *** armaan has joined #openstack-infra | 09:40 | |
| *** Guest73591 has quit IRC | 09:41 | |
| *** chandankumar has joined #openstack-infra | 09:41 | |
| *** andreas_s has joined #openstack-infra | 09:44 | |
| *** ianychoi has quit IRC | 09:44 | |
| *** sree has joined #openstack-infra | 09:44 | |
| *** ianychoi has joined #openstack-infra | 09:45 | |
| *** andreas_s has quit IRC | 09:48 | |
| *** sree has quit IRC | 09:48 | |
| *** alexchadin has quit IRC | 09:51 | |
| *** zhangfei has quit IRC | 09:52 | |
| *** rosmaita has joined #openstack-infra | 09:56 | |
| *** armaan has quit IRC | 09:56 | |
| *** armaan has joined #openstack-infra | 09:56 | |
| *** hwoarang has quit IRC | 09:59 | |
| *** kjackal has quit IRC | 10:01 | |
| *** zhangfei has joined #openstack-infra | 10:08 | |
| *** armaan has quit IRC | 10:15 | |
| *** ianychoi has quit IRC | 10:16 | |
| *** armaan has joined #openstack-infra | 10:16 | |
| *** ianychoi has joined #openstack-infra | 10:17 | |
| *** jlabarre has joined #openstack-infra | 10:19 | |
| sshnaidm | hi, where does zuul clones projects by default? in $WORKSPACE? | 10:25 |
| *** ianychoi has quit IRC | 10:31 | |
| *** nicolasbock has joined #openstack-infra | 10:32 | |
| *** garyk has quit IRC | 10:33 | |
| *** kjackal has joined #openstack-infra | 10:33 | |
| *** ianychoi has joined #openstack-infra | 10:38 | |
| mpeterson | sshnaidm: {{ ansible_user_dir }}/src/ | 10:38 |
| *** garyk has joined #openstack-infra | 10:38 | |
| *** namnh has quit IRC | 10:39 | |
| *** zhangfei has quit IRC | 10:40 | |
| *** andreas_s has joined #openstack-infra | 10:41 | |
| sshnaidm | mpeterson, does it have environment variable I can use in the job? | 10:41 |
| *** ianychoi has quit IRC | 10:43 | |
| *** andreas_s has quit IRC | 10:46 | |
| *** pbourke has quit IRC | 10:46 | |
| *** pbourke has joined #openstack-infra | 10:47 | |
| mpeterson | sshnaidm: you can use that in your job definition to pass it to bash scripts if you want... e.g. https://github.com/openstack/networking-odl/blob/master/playbooks/functional/pre.yaml#L19 | 10:47 |
| sshnaidm | mpeterson, thanks | 10:48 |
| *** ianychoi has joined #openstack-infra | 10:59 | |
| *** d0ugal has quit IRC | 11:00 | |
| *** ykarel_ has joined #openstack-infra | 11:03 | |
| *** chkumar246 has joined #openstack-infra | 11:04 | |
| *** dhajare__ has joined #openstack-infra | 11:04 | |
| *** alexchadin has joined #openstack-infra | 11:04 | |
| *** chandankumar has quit IRC | 11:04 | |
| *** links has quit IRC | 11:04 | |
| *** ykarel has quit IRC | 11:04 | |
| *** links has joined #openstack-infra | 11:05 | |
| *** ykarel__ has joined #openstack-infra | 11:05 | |
| *** dhajare_ has quit IRC | 11:08 | |
| *** chkumar246 has quit IRC | 11:08 | |
| *** ykarel_ has quit IRC | 11:09 | |
| *** dhajare__ has quit IRC | 11:09 | |
| *** hwoarang has joined #openstack-infra | 11:12 | |
| *** hwoarang has quit IRC | 11:13 | |
| *** hwoarang has joined #openstack-infra | 11:14 | |
| *** ykarel__ is now known as ykarel | 11:17 | |
| *** zhurong has joined #openstack-infra | 11:19 | |
| *** chkumar246 has joined #openstack-infra | 11:20 | |
| *** tpsilva has joined #openstack-infra | 11:21 | |
| *** dhajare__ has joined #openstack-infra | 11:21 | |
| *** xinliang3 has quit IRC | 11:25 | |
| *** ianychoi has quit IRC | 11:26 | |
| *** ianychoi has joined #openstack-infra | 11:27 | |
| *** d0ugal has joined #openstack-infra | 11:29 | |
| *** ldnunes has joined #openstack-infra | 11:38 | |
| *** evrardjp_ has joined #openstack-infra | 11:45 | |
| *** dprince has joined #openstack-infra | 11:53 | |
| *** rfolco has joined #openstack-infra | 11:54 | |
| *** rfolco is now known as rfolco|ruck | 11:59 | |
| *** katkapilatova has joined #openstack-infra | 11:59 | |
| *** lihi has quit IRC | 12:03 | |
| *** lbragstad has joined #openstack-infra | 12:03 | |
| *** oanson has quit IRC | 12:04 | |
| *** katkapilatova has quit IRC | 12:04 | |
| *** udesale has quit IRC | 12:04 | |
| *** udesale has joined #openstack-infra | 12:05 | |
| *** pgadiya has quit IRC | 12:11 | |
| *** weshay_mod is now known as weshay | 12:12 | |
| *** efoley has joined #openstack-infra | 12:13 | |
| *** jcoufal has joined #openstack-infra | 12:13 | |
| *** odyssey4me has quit IRC | 12:17 | |
| *** odyssey4me has joined #openstack-infra | 12:17 | |
| *** alexchadin has quit IRC | 12:20 | |
| *** udesale has quit IRC | 12:23 | |
| *** udesale has joined #openstack-infra | 12:23 | |
| *** pgadiya has joined #openstack-infra | 12:24 | |
| *** d0ugal has quit IRC | 12:25 | |
| *** d0ugal has joined #openstack-infra | 12:25 | |
| *** d0ugal has quit IRC | 12:25 | |
| *** d0ugal has joined #openstack-infra | 12:25 | |
| *** pgadiya has quit IRC | 12:28 | |
| *** efoley has quit IRC | 12:29 | |
| *** amoralej is now known as amoralej|lunch | 12:29 | |
| openstackgerrit | Andrey Kurilin proposed openstack-infra/project-config master: [rally] Remove several redundant jobs https://review.openstack.org/558204 | 12:29 |
| *** camunoz has joined #openstack-infra | 12:29 | |
| *** sshnaidm is now known as sshnaidm|bbl | 12:32 | |
| openstackgerrit | Andrey Kurilin proposed openstack-infra/openstack-zuul-jobs master: [rally] Remove unused jobs https://review.openstack.org/558205 | 12:33 |
| openstackgerrit | Andrey Kurilin proposed openstack-infra/project-config master: [rally] Remove several redundant jobs https://review.openstack.org/558204 | 12:33 |
| *** zhurong has quit IRC | 12:34 | |
| *** edmondsw has joined #openstack-infra | 12:43 | |
| *** gouthamr has joined #openstack-infra | 12:47 | |
| *** jamesmcarthur has joined #openstack-infra | 12:53 | |
| *** hwoarang has quit IRC | 12:53 | |
| *** hwoarang has joined #openstack-infra | 12:54 | |
| *** hwoarang has quit IRC | 12:54 | |
| *** eharney has joined #openstack-infra | 12:55 | |
| *** david-lyle has quit IRC | 12:58 | |
| *** david-lyle has joined #openstack-infra | 12:59 | |
| *** Guest88902 is now known as zeus | 12:59 | |
| *** zeus has quit IRC | 12:59 | |
| *** zeus has joined #openstack-infra | 12:59 | |
| smcginnis | Not sure if we really care, but some random person decided to start a "meeting" in #openstack-operators. | 13:02 |
| smcginnis | Looks to be part of the overall spam over the last day. | 13:02 |
| smcginnis | I just ended it now, but we can delete the logs if we want - http://eavesdrop.openstack.org/meetings/maintenance/2018/maintenance.2018-04-01-16.59.txt | 13:03 |
| *** david-lyle has quit IRC | 13:09 | |
| *** kgiusti has joined #openstack-infra | 13:10 | |
| *** sree has joined #openstack-infra | 13:12 | |
| openstackgerrit | Rishabh Dave proposed openstack-infra/openstack-zuul-jobs master: Remove the job manila-ui-dsvm to move it in in-tree https://review.openstack.org/557635 | 13:13 |
| openstackgerrit | Rishabh Dave proposed openstack-infra/project-config master: Remove the job manila-ui-dsvm to move it in in-tree https://review.openstack.org/557645 | 13:13 |
| *** eharney has quit IRC | 13:17 | |
| *** sree has quit IRC | 13:17 | |
| *** hwoarang has joined #openstack-infra | 13:18 | |
| *** jroll has quit IRC | 13:19 | |
| *** jroll has joined #openstack-infra | 13:20 | |
| *** olaph1 has joined #openstack-infra | 13:20 | |
| *** olaph has quit IRC | 13:21 | |
| *** sree has joined #openstack-infra | 13:21 | |
| andreykurilin_ | hi folks! Do we have a special ubuntu image with docker installed or should I install it manually? | 13:22 |
| *** evrardjp_ has quit IRC | 13:22 | |
| *** evrardjp_ has joined #openstack-infra | 13:24 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul master: Upgrade from angularjs (v1) to angular (v5) https://review.openstack.org/551989 | 13:25 |
| *** trown|outtypewww is now known as trown | 13:26 | |
| *** eharney has joined #openstack-infra | 13:31 | |
| *** felipemonteiro__ has joined #openstack-infra | 13:32 | |
| *** udesale has quit IRC | 13:35 | |
| *** udesale has joined #openstack-infra | 13:35 | |
| *** yamamoto has quit IRC | 13:37 | |
| *** yamamoto has joined #openstack-infra | 13:37 | |
| *** felipemonteiro_ has joined #openstack-infra | 13:39 | |
| *** felipemonteiro__ has quit IRC | 13:42 | |
| *** oanson has joined #openstack-infra | 13:45 | |
| *** sree has quit IRC | 13:46 | |
| *** amoralej|lunch is now known as amoralej | 13:47 | |
| *** sree has joined #openstack-infra | 13:48 | |
| *** links has quit IRC | 13:48 | |
| *** jamesmcarthur has quit IRC | 13:51 | |
| *** esberglu has joined #openstack-infra | 13:52 | |
| *** evrardjp has quit IRC | 13:52 | |
| *** gouthamr has quit IRC | 13:53 | |
| *** garyk has quit IRC | 13:53 | |
| *** gouthamr has joined #openstack-infra | 13:54 | |
| *** psachin has quit IRC | 13:55 | |
| *** sree has quit IRC | 13:58 | |
| *** caphrim007 has quit IRC | 13:59 | |
| *** gouthamr has quit IRC | 14:00 | |
| *** jamesmcarthur has joined #openstack-infra | 14:00 | |
| *** gouthamr has joined #openstack-infra | 14:00 | |
| *** sree has joined #openstack-infra | 14:01 | |
| *** sree has quit IRC | 14:01 | |
| *** sree has joined #openstack-infra | 14:03 | |
| *** hongbin has joined #openstack-infra | 14:04 | |
| *** lpetrut has quit IRC | 14:05 | |
| *** sree has quit IRC | 14:08 | |
| *** gouthamr has quit IRC | 14:09 | |
| *** gouthamr has joined #openstack-infra | 14:09 | |
| *** bobh_ has joined #openstack-infra | 14:11 | |
| *** dsariel has quit IRC | 14:12 | |
| *** sree has joined #openstack-infra | 14:12 | |
| *** gouthamr has quit IRC | 14:14 | |
| *** sshnaidm|bbl is now known as sshnaidm | 14:16 | |
| prometheanfire | looks like pip-10 is going to come out soon, so... hold onto your butts | 14:17 |
| *** gouthamr has joined #openstack-infra | 14:18 | |
| *** sree has quit IRC | 14:19 | |
| *** gouthamr has quit IRC | 14:19 | |
| *** chkumar246 has quit IRC | 14:20 | |
| *** chandankumar has joined #openstack-infra | 14:20 | |
| *** gouthamr has joined #openstack-infra | 14:21 | |
| *** jamesmca_ has joined #openstack-infra | 14:25 | |
| *** dsariel has joined #openstack-infra | 14:26 | |
| *** evrardjp_ is now known as evrardjp | 14:29 | |
| *** gouthamr has quit IRC | 14:34 | |
| *** gouthamr has joined #openstack-infra | 14:35 | |
| *** garyk has joined #openstack-infra | 14:35 | |
| clarkb | andreykurilin_: we don't have any special images. We try to make them as boring as possible then you can use tools like bindep to install what you need from there | 14:35 |
| clarkb | prometheanfire: ya fungi wrote mail to the dev list over the weekend about it. | 14:36 |
| clarkb | smcginnis: thanks, it looked like fungi had thought he got them all yesterday but if any were missed anyone can end them at this point then we can delete the logs | 14:36 |
| clarkb | dmsimard: when you get a chance what arguments did you use to launch-node.py for limestone's mirror? looks like we can only set a single mount path today so you probably did it with one volume then attached the other after? | 14:38 |
| clarkb | fungi: ^ this may be why two volumes have been used | 14:38 |
| *** eernst has joined #openstack-infra | 14:38 | |
| clarkb | Thinking about it I think the best way to do this replacement instance will be to not boot it with any volumes. Have it come up, stop apache, clean out /var/cache/apache2 if necessary then make two mounts out of a single pv and two lvs | 14:40 |
| *** david-lyle has joined #openstack-infra | 14:40 | |
| clarkb | I'm going to start booting a replacement with no volume with plan of attaching single 200GB cinder volume for ^ | 14:41 |
| *** tesseract has quit IRC | 14:42 | |
| *** tesseract-RH has joined #openstack-infra | 14:42 | |
| *** david-lyle is now known as dklyle | 14:44 | |
| *** dhill_ has joined #openstack-infra | 14:45 | |
| fungi | smcginnis: yep, i cleaned up about 20 of our channels they did that in yesterday and was watching for channel activity to trigger logging so i could spot more of them (i also just ended one for #openstack-watcher) | 14:46 |
| *** inc0 has joined #openstack-infra | 14:46 | |
| fungi | and i've been deleting the "meeting logs" as i go | 14:47 |
| smcginnis | fungi, clarkb: Thanks! | 14:47 |
| fungi | clarkb: that sounds like a fine plan | 14:48 |
| fungi | i'm only sort-of around today since i still have guests staying over through tonight | 14:48 |
| corvus | mtreinish: can you give https://review.openstack.org/535543 a look over? | 14:51 |
| corvus | fungi: ^ you too | 14:51 |
| fungi | sure thing | 14:51 |
| *** inc0 has quit IRC | 14:51 | |
| *** rkukura has joined #openstack-infra | 14:53 | |
| *** sai-ons is now known as sai1 | 14:55 | |
| *** spotz_ has joined #openstack-infra | 14:55 | |
| andreykurilin_ | clarkb: thanks | 14:56 |
| *** lpetrut has joined #openstack-infra | 14:57 | |
| *** spotz_ is now known as spotz | 14:59 | |
| *** VW has joined #openstack-infra | 15:00 | |
| clarkb | dmsimard: mordred fungi as a heads up the dns printing stuff for launch node did fail for me bringing up mirror02 in limestone, but thats fine I'll just manually add the records before I move the cname | 15:04 |
| clarkb | was there a fix for that? I think it had to do with the use of the clients and stuff? | 15:04 |
| pabelanger | clarkb: prometheanfire: started playing with USE_VENV for devstack, so far the issues I've found: https://review.openstack.org/552939/ | 15:06 |
| pabelanger | I need help from cinder to figure out why cinder-rootwrap isn't working properly with virtualenv | 15:06 |
| clarkb | pabelanger: that is all separate venvs though which is not what we want | 15:07 |
| pabelanger | clarkb: right, I figured I get that working first, then update it to dump everything into single virtualenv | 15:07 |
| pabelanger | since is it broken now | 15:07 |
| *** olaph1 is now known as olaph | 15:08 | |
| clarkb | well I think single virtualenv will be much simpler | 15:08 |
| mtreinish | corvus: ooh, mqtt. Sure I'll take a look | 15:08 |
| clarkb | since that is more similar to what we already had | 15:08 |
| pabelanger | clarkb: sure, it is easy to update to single virtualenv, can do that next patchset | 15:08 |
| prometheanfire | it'd also re-enforce co-installability (even if we already test that in reqs) | 15:09 |
| pabelanger | agree | 15:11 |
| pabelanger | clarkb: so we need to land https://review.openstack.org/558136/ and backport to fix nb03.o.o builds | 15:12 |
| pabelanger | going to find somebody to review | 15:12 |
| *** ramishra has quit IRC | 15:14 | |
| *** iyamahat has joined #openstack-infra | 15:16 | |
| *** e0ne has quit IRC | 15:16 | |
| *** dhajare__ has quit IRC | 15:21 | |
| *** gouthamr has quit IRC | 15:23 | |
| *** caphrim007 has joined #openstack-infra | 15:24 | |
| clarkb | pabelanger: apparently I do have approval rights on that, I'll take a look as soon as I manage to get new mirror in limestone going | 15:25 |
| *** gouthamr has joined #openstack-infra | 15:27 | |
| pabelanger | cool | 15:27 |
| *** garyk has quit IRC | 15:28 | |
| *** armaan has quit IRC | 15:28 | |
| *** armaan has joined #openstack-infra | 15:29 | |
| mtreinish | corvus: ok, reviewed | 15:30 |
| mtreinish | corvus: one thing I'm not sure about is what the hierarchy for topics we use is. It's hard to predict how people will want to filter events before it's available | 15:32 |
| *** jcoufal has quit IRC | 15:33 | |
| *** VW_ has joined #openstack-infra | 15:34 | |
| *** ianychoi has quit IRC | 15:34 | |
| clarkb | because I'm in the dns settings and noticing we still have old infracloud records hanging around I'm going to clean those up now | 15:34 |
| clarkb | mirror02.regionone.limestone.o.o is up now and mirror.regionone.limestone.o.o is cnamed to it | 15:34 |
| pabelanger | ++ | 15:35 |
| clarkb | maybe someone else can test ^ and then we can reenable limestone while I lcnea up these old infracloud dns records and mirror01 in limestone? | 15:35 |
| logan- | thanks clarkb, sorry for breaking the old one | 15:35 |
| clarkb | logan-: no problem | 15:35 |
| corvus | mtreinish: is the configuration of that zuul driver sufficient for us to be able to manage that? it appears set up to set a single topic per pipeline. so we could have all the pipelines report under "zuul" or we could have "zuul_gate", "zuul_check", etc... | 15:35 |
| *** jcoufal has joined #openstack-infra | 15:36 | |
| corvus | mtreinish: oh i just now see your inline comment :) | 15:36 |
| *** VW_ has quit IRC | 15:36 | |
| *** VW_ has joined #openstack-infra | 15:37 | |
| *** VW has quit IRC | 15:37 | |
| clarkb | ok infracloud mirror records cleaned up. Doing that I notice we have old pypi.foo records too which I'll lceanup once the old limestone mirror is cleaned up | 15:37 |
| *** VW_ has quit IRC | 15:37 | |
| *** VW has joined #openstack-infra | 15:38 | |
| clarkb | pabelanger: logan- is there a max-servers: 0 change to revert for limestone? | 15:39 |
| openstackgerrit | Logan V proposed openstack-infra/project-config master: Revert "Drop Limestone max-servers to 0" https://review.openstack.org/558224 | 15:40 |
| logan- | ^ | 15:40 |
| clarkb | logan-: +2 thanks | 15:40 |
| *** jcoufal has quit IRC | 15:40 | |
| clarkb | logan-: fwiw mirror01-afs does not appear to want to delete | 15:41 |
| clarkb | logan-: I'm not sure thats a major concern but I've asked it nicely twice and its gone back to available both times | 15:42 |
| logan- | thanks | 15:42 |
| logan- | will check on it in a few | 15:42 |
| *** VW has quit IRC | 15:42 | |
| *** bhavik1 has joined #openstack-infra | 15:43 | |
| pabelanger | +3 | 15:45 |
| *** VW has joined #openstack-infra | 15:46 | |
| *** hemna_ has joined #openstack-infra | 15:47 | |
| pabelanger | okay, booting review01.o.o today. Want to get the ML post out for new IP | 15:47 |
| *** armaan has quit IRC | 15:47 | |
| *** jcoufal has joined #openstack-infra | 15:48 | |
| *** gouthamr has quit IRC | 15:48 | |
| clarkb | infra-root can you give http://paste.openstack.org/show/718181/ a quick lookover? its a set of dns records that I think are safe to delete and one that I'd like input on | 15:49 |
| *** gouthamr has joined #openstack-infra | 15:49 | |
| pabelanger | clarkb: good, but also unsure about line 16 so will defer | 15:50 |
| *** felipemonteiro_ has quit IRC | 15:50 | |
| corvus | clarkb: lgtm. this looks pretty safe: http://codesearch.openstack.org/?q=pypi.openstack.org&i=nope&files=&repos= | 15:51 |
| fungi | heh, that was the first place i went too ;) | 15:51 |
| clarkb | turbo hipster would need updating but that is it I guess | 15:52 |
| clarkb | we ok with breaking turbo hipster? | 15:52 |
| *** bhavik1 has quit IRC | 15:52 | |
| openstackgerrit | Merged openstack-infra/project-config master: Revert "Drop Limestone max-servers to 0" https://review.openstack.org/558224 | 15:53 |
| clarkb | I'm going to go ahead and clean up the others to start | 15:53 |
| *** felipemonteiro has joined #openstack-infra | 15:55 | |
| *** patriciadomin has quit IRC | 15:56 | |
| fungi | thanks clarkb! | 15:56 |
| *** garyk has joined #openstack-infra | 15:56 | |
| fungi | maybe jhesketh will know whether turbo-hipster is still being run anywhere | 15:57 |
| *** felipemonteiro_ has joined #openstack-infra | 15:57 | |
| *** eernst has quit IRC | 15:57 | |
| *** patriciadomin has joined #openstack-infra | 15:58 | |
| *** ianychoi has joined #openstack-infra | 15:58 | |
| *** eernst has joined #openstack-infra | 15:58 | |
| fungi | clarkb: actually, looks like that cname's safe to delete too. that vhost no longer exists on static.o.o | 15:59 |
| fungi | it ends up redirecting to the docs site | 15:59 |
| clarkb | cool | 15:59 |
| *** aojea has quit IRC | 15:59 | |
| fungi | and the old docroot has been moved to /srv/static/old-pypi so can probably be cleared out as well at this point | 15:59 |
| fungi | last modified Jun 15 2014 | 16:00 |
| *** felipemonteiro has quit IRC | 16:01 | |
| openstackgerrit | Merged openstack-infra/zuul master: zuul autohold: allow operator to specify nodes TTL https://review.openstack.org/543403 | 16:01 |
| *** olaph has quit IRC | 16:01 | |
| *** olaph has joined #openstack-infra | 16:02 | |
| clarkb | I'm adding pypi.regionone.openstack.org to this list | 16:02 |
| clarkb | (found it in process of deleting things) | 16:03 |
| pabelanger | ack | 16:03 |
| dmsimard | clarkb: hey, I'm off today but yes, I attached things manually | 16:04 |
| clarkb | dmsimard: thanks for confirming, once I had woekn up I realized that must've been the process ued | 16:04 |
| dmsimard | Did you struggle with shade version for floating IP things ? | 16:04 |
| pabelanger | fungi: fyi: https://review.openstack.org/557902/ adds debian-stretch nodes for nodepool | 16:05 |
| pabelanger | https://review.openstack.org/557942/ shows them working | 16:05 |
| pabelanger | seems linaro would like to use them for arm64 testing | 16:05 |
| fungi | pabelanger: thanks!!! | 16:06 |
| clarkb | #status log Cleaned up old unused dns records per http://paste.openstack.org/show/718183/ we no longer use the pypi hostname for our pypi mirrors and some of the clouds don't exist anymore. | 16:06 |
| openstackstatus | clarkb: finished logging | 16:06 |
| clarkb | dmsimard: no, only the dns thing at the end so I just did dns manually | 16:06 |
| *** ianychoi has quit IRC | 16:06 | |
| *** armaan has joined #openstack-infra | 16:06 | |
| pabelanger | clarkb: logan-: did rdns get update for limestone mirror too? | 16:07 |
| dmsimard | Right, my patch was meant to fix that but I think mordred sent something better | 16:07 |
| *** ianychoi has joined #openstack-infra | 16:07 | |
| clarkb | pabelanger: I haven't done that, logan would have to do it on the cloud side I Think | 16:07 |
| clarkb | pabelanger: fwiw we don't typically have rdns for our mirror nodes because not all clouds support it | 16:07 |
| *** gouthamr has quit IRC | 16:07 | |
| pabelanger | clarkb: right, think logan- does it manually. But I think I was getting email warnings on the mirror for some reason about rdns | 16:08 |
| *** lpetrut has quit IRC | 16:08 | |
| *** gouthamr has joined #openstack-infra | 16:09 | |
| dmsimard | rdns is also one of those low hanging fruit when you want to improve email delivery (since our servers send emails about things like crons etc) | 16:10 |
| pabelanger | dmsimard: clarkb: hmm, dns.py is broken for me: http://paste.openstack.org/show/718184/ | 16:11 |
| pabelanger | just tried to launch review01.o.o | 16:11 |
| pabelanger | ah, I guess I need to update shade | 16:11 |
| clarkb | pabelanger: note that dns.py crashing didn't result in it deleting the server for me | 16:17 |
| clarkb | it just failed to print the helpful dns info | 16:17 |
| pabelanger | clarkb: yah, same. I manually deleted and tried to launch again (with shade updated) to confirm working | 16:17 |
| pabelanger | should know here shortly | 16:18 |
| clarkb | pabelanger: for the devstack etcd fix you ahven't backproted to the other branches yet ? | 16:18 |
| pabelanger | clarkb: no, wanted to make sure that patch was first good. can do that now if you'd like | 16:19 |
| *** VW has quit IRC | 16:19 | |
| clarkb | pabelanger: no I left a -1 on the first one | 16:19 |
| clarkb | pabelanger: so was checking if I had to apply it to others :) | 16:19 |
| pabelanger | kk | 16:20 |
| *** VW has joined #openstack-infra | 16:20 | |
| *** ykarel is now known as ykarel|away | 16:22 | |
| pabelanger | clarkb: updated | 16:22 |
| *** Sukhdev has joined #openstack-infra | 16:23 | |
| *** VW has quit IRC | 16:24 | |
| *** trown is now known as trown|lunch | 16:25 | |
| clarkb | pabelanger: +2 thanks | 16:25 |
| clarkb | corvus: might be willign to be second review on https://review.openstack.org/#/c/558136/2 to fix etcd caching | 16:25 |
| *** ykarel|away has quit IRC | 16:26 | |
| *** dprince has quit IRC | 16:26 | |
| *** jd_ has quit IRC | 16:26 | |
| clarkb | logan-: pabelanger 10 limestone nodes in use now | 16:27 |
| *** VW has joined #openstack-infra | 16:28 | |
| pabelanger | yay | 16:28 |
| clarkb | http://zuul.openstack.org/stream.html?uuid=f78c93e727174f7aa0a36e36df603520&logfile=console.log is one of them | 16:28 |
| *** felipemonteiro_ has quit IRC | 16:29 | |
| logan- | excellent | 16:29 |
| *** felipemonteiro__ has joined #openstack-infra | 16:29 | |
| logan- | i will update rdns shortly | 16:29 |
| pabelanger | clarkb: dmsimard: dns.py still failed after update shade :( | 16:29 |
| dmsimard | pabelanger: you need a patch | 16:29 |
| logan- | also happy to give infra access to the portal account for this deploy, which would allow rdns edits etc over time | 16:29 |
| dmsimard | pabelanger: oh wait maybe not, I didn't know https://review.openstack.org/#/c/555433/ had landed | 16:31 |
| *** esberglu has quit IRC | 16:31 | |
| *** esberglu has joined #openstack-infra | 16:31 | |
| *** HeOS has quit IRC | 16:31 | |
| openstackgerrit | Merged openstack-infra/nodepool master: Add debian-stretch to nodepool-functional-py35-debian-src https://review.openstack.org/557942 | 16:32 |
| *** gouthamr has quit IRC | 16:33 | |
| *** yamamoto has quit IRC | 16:33 | |
| pabelanger | dmsimard: think I see the issue, testing | 16:34 |
| *** armaan has quit IRC | 16:35 | |
| openstackgerrit | Merged openstack-infra/nodepool master: Enable AFS mirrors for ubuntu-bionic testing https://review.openstack.org/558015 | 16:37 |
| pabelanger | clarkb: prometheanfire: are we also interested in tagging glean today to pickup fixes for gentoo? That would be needed if we want to start offering images in nodepool | 16:38 |
| prometheanfire | clarkb: it'd be apprecieated :D | 16:42 |
| *** yamamoto has joined #openstack-infra | 16:42 | |
| *** eharney has quit IRC | 16:43 | |
| openstackgerrit | Paul Belanger proposed openstack-infra/project-config master: Use AFS mirrors to build ubuntu-bionic DIB https://review.openstack.org/558237 | 16:43 |
| pabelanger | Shrews: clarkb: fungi: update to nodepool.o.o to also use bionic AFS mirrors when building images | 16:44 |
| openstackgerrit | Paul Belanger proposed openstack-infra/project-config master: Use AFS mirrors to build ubuntu-bionic DIB https://review.openstack.org/558237 | 16:45 |
| openstackgerrit | Paul Belanger proposed openstack-infra/system-config master: Fix syntax error with dns.py print_dns() https://review.openstack.org/558238 | 16:47 |
| pabelanger | clarkb: dmsimard: ^ fixes dns.py for me | 16:47 |
| openstackgerrit | Merged openstack/python-jenkins master: modify version spelling errors https://review.openstack.org/534598 | 16:47 |
| pabelanger | clarkb: dmsimard: actually, I lied. I just noticed we no longer print href info for rdns. Checking why now | 16:49 |
| *** udesale has quit IRC | 16:50 | |
| *** david-lyle has joined #openstack-infra | 16:51 | |
| *** dklyle has quit IRC | 16:51 | |
| openstackgerrit | Clark Boylan proposed openstack-infra/zuul master: Report to all reporters even if one fails https://review.openstack.org/557859 | 16:53 |
| clarkb | pabelanger: there was a change mordred wrote to fix this. Unsure if that change merged and was broken or if it is still unmerged and needs to get in to fix this | 16:54 |
| pabelanger | clarkb: k, let me find that and tes | 16:54 |
| clarkb | pabelanger: the problem is/was that iterating over the list of things doesn't have the href data so instead we just show the specific server we want and get the href that way | 16:54 |
| pabelanger | test* | 16:54 |
| *** eharney has joined #openstack-infra | 16:56 | |
| *** yamamoto has quit IRC | 16:59 | |
| *** olaph1 has joined #openstack-infra | 17:02 | |
| *** trown|lunch is now known as trown | 17:03 | |
| *** olaph has quit IRC | 17:03 | |
| *** dhajare__ has joined #openstack-infra | 17:05 | |
| *** david-lyle has quit IRC | 17:06 | |
| *** david-lyle has joined #openstack-infra | 17:06 | |
| *** david-lyle has quit IRC | 17:07 | |
| *** david-lyle has joined #openstack-infra | 17:07 | |
| openstackgerrit | Merged openstack-infra/nodepool master: Reduce logging in _cleanupCurrentProviderUploads function https://review.openstack.org/557791 | 17:08 |
| *** david-lyle has quit IRC | 17:13 | |
| pabelanger | clarkb: okay, I can't find fixes from mordred, so assuming not fixed. It seems cloud.get_server(server.id) doesn't return any href info by default, trying to see how we get that | 17:13 |
| clarkb | pabelanger: I think tahts the fix from mordred so likely that the fix doesn't fix it | 17:14 |
| clarkb | pabelanger: maybe we need to tell it to give us all the infos back | 17:14 |
| pabelanger | detailed=True maybe? | 17:14 |
| clarkb | pabelanger: maybe? I'd have to go raed some shade source | 17:14 |
| pabelanger | yah, doing that now | 17:16 |
| *** sfeole has joined #openstack-infra | 17:16 | |
| *** _alastor_ has joined #openstack-infra | 17:17 | |
| *** david-lyle has joined #openstack-infra | 17:18 | |
| *** felipemonteiro_ has joined #openstack-infra | 17:18 | |
| *** lpetrut has joined #openstack-infra | 17:19 | |
| *** felipemonteiro__ has quit IRC | 17:22 | |
| *** david-lyle has quit IRC | 17:23 | |
| *** lpetrut has quit IRC | 17:26 | |
| clarkb | pabelanger: https://review.openstack.org/#/c/555433/3/launch/dns.py just confirming that is the fix that merged whcih I think is still broken | 17:26 |
| clarkb | (as you've found) | 17:26 |
| beisner | hi all, anyone else having small fits of mayhem after pip released 10.0.0b2 ( ! ) https://pypi.python.org/pypi/pip/ | 17:26 |
| *** lpetrut has joined #openstack-infra | 17:26 | |
| clarkb | beisner: its know to not work with openstack tools like devstack currently | 17:27 |
| clarkb | beisner: however pip won't install prerelease releases by default so you should be largely protected until the releease happens | 17:27 |
| clarkb | at which point if we haven't addressed the concerns yet we'll have to quickly | 17:27 |
| beisner | that's weird. we're getting 10.0.0b2 in all of our virtualenvs right now, clarkb | 17:27 |
| pabelanger | clarkb: yah, I'm think we might need to revert that until we figure out hrefs. I can't seem to get that info from latest shade, but I am assuming mordred knows more | 17:28 |
| clarkb | pabelanger: well reverting it doesnt' work either beacuse shade can't do that old method or something | 17:28 |
| clarkb | pabelanger: reading it you may want bare=True, detailed=True | 17:28 |
| clarkb | beisner: how are you installing pip? | 17:29 |
| beisner | clarkb: ex: `tox -e pep8` | 17:29 |
| clarkb | beisner: that should update to latest release of pip but not pull in prereleases unless you are using a virtualenv install that is also a prerelease | 17:30 |
| clarkb | beisner: I would check your virtualenv install version | 17:30 |
| clarkb | beisner: creating a local virtualenv I get 9.0.3 not 10.0.0b2 | 17:31 |
| pabelanger | clarkb: http://paste.openstack.org/raw/718192/ | 17:31 |
| clarkb | as I expect | 17:31 |
| pabelanger | but don't see the information | 17:31 |
| beisner | clarkb: ditto. for the life of me i can't see why i'm getting 10.0.0b2 when the venv is built with tox. | 17:32 |
| *** dsariel has quit IRC | 17:32 | |
| clarkb | pabelanger: does openstack server show foo.openstack.org include the info? If so I wonder what osc is doing differently | 17:32 |
| clarkb | beisner: I'm gonna check one of our test runs too | 17:32 |
| *** caphrim007_ has joined #openstack-infra | 17:32 | |
| clarkb | we don't actually appear to log the version of pip when running tox | 17:33 |
| clarkb | beisner: also do you have logs you can share for how this is breaking? things in a virtualenv should work regardless. Its only interaction with system pacakges that we know are potentially going to break | 17:34 |
| pabelanger | clarkb: all I see right now: http://paste.openstack.org/show/718195/ | 17:34 |
| beisner | clarkb: --allow-unverified went away, which we apparently use for pkg. | 17:34 |
| beisner | + for one pkg, unfortunately | 17:35 |
| *** neiloy has joined #openstack-infra | 17:35 | |
| clarkb | beisner: oh openstack should've stopped doing that a while ago | 17:35 |
| beisner | so, if using 10.0.0b2, it breaks on that. | 17:36 |
| clarkb | I think we use global config option to trust non https sources and then all packages are directly hosted not spidered | 17:36 |
| *** caphrim007 has quit IRC | 17:36 | |
| *** dhajare__ has quit IRC | 17:36 | |
| clarkb | pabelanger: ya I wonder if that is rax specific and you have to go asking for it specifically | 17:37 |
| beisner | clarkb: can you point me at tox config example that does that? | 17:38 |
| openstackgerrit | Paul Belanger proposed openstack-infra/nodepool master: Test growroot in boot tests https://review.openstack.org/555103 | 17:39 |
| clarkb | beisner: we don't do it in tox, we do it in the job itself, http://logs.openstack.org/32/551832/1/gate/openstack-tox-py35/f78c93e/ara-report/file/4c8dc876-8029-435c-9d26-b16866b50d39/#line-15 | 17:40 |
| pabelanger | clarkb: possible, I'd have to see what mordred says | 17:40 |
| clarkb | beisner: https://git.openstack.org/cgit/openstack-infra/zuul-jobs/tree/roles/configure-mirrors/templates/etc/pip.conf.j2 is the template used. The trusted-index is how we explicitly allow certain locations | 17:41 |
| *** felipemonteiro_ has quit IRC | 17:41 | |
| *** felipemonteiro_ has joined #openstack-infra | 17:41 | |
| clarkb | beisner: why are you using unverifiable packages? | 17:41 |
| clarkb | beisner: we worked really hard to get rid of all of those from openstack proper. | 17:42 |
| clarkb | either by having them host directly on pypi or by changing deps | 17:42 |
| beisner | i'm disabling to find out, clarkb :-) | 17:42 |
| *** e0ne has joined #openstack-infra | 17:44 | |
| beisner | clarkb: haha it's python-apt | 17:45 |
| clarkb | beisner: probably because it is hosted on launchpad? | 17:45 |
| clarkb | for something like that it is probably reasonable to have them move to pypi especially if you can show them that new pip will break their users otherwise | 17:46 |
| *** Sukhdev has quit IRC | 17:46 | |
| clarkb | yup its on launchpad | 17:47 |
| clarkb | https://pypi.python.org/pypi/python-apt/ the download url is the problem | 17:47 |
| *** dprince has joined #openstack-infra | 17:48 | |
| clarkb | pabelanger: reading novaclient code it explicitly removes the links and addresses when printing to the user | 17:48 |
| clarkb | pabelanger: I half wonder if shade/osc are approximating that behavior but I can't find that on the sahde side | 17:49 |
| beisner | clarkb: still a mystery why with vanilla tox usage, i'm getting 10.0.0v2 | 17:49 |
| beisner | b2, even | 17:49 |
| beisner | clarkb: agree that this and all other similar authors will and should get ahead of this. appreciate your pointers here. | 17:50 |
| *** caphrim007_ has quit IRC | 17:50 | |
| *** caphrim007 has joined #openstack-infra | 17:51 | |
| *** jamesmca_ has quit IRC | 17:51 | |
| *** lpetrut has quit IRC | 17:52 | |
| clarkb | pabelanger: do you have a stack trace? the thing I ran into was different and likely due to having too old of a shade | 17:52 |
| clarkb | pabelanger: but reading shade more there are two different ways it can get a server and one way (use_direct_get) doesn't appear to do the extra server detail listing. But the default method should | 17:53 |
| *** dprince has quit IRC | 17:54 | |
| *** lpetrut has joined #openstack-infra | 17:54 | |
| *** eharney has quit IRC | 17:55 | |
| clarkb | pabelanger: aha! I think I found it. shade._normalize.Normalizer._normalize_server does not seem to do anything with the links argument | 17:59 |
| *** yamamoto has joined #openstack-infra | 18:00 | |
| clarkb | pabelanger: I'm going to have a shade patch in a few that I think you can tes with | 18:04 |
| clarkb | I just have to figure out what the expected munch behavior is here :/ | 18:04 |
| *** yamamoto has quit IRC | 18:06 | |
| *** dprince has joined #openstack-infra | 18:10 | |
| *** eharney has joined #openstack-infra | 18:10 | |
| *** hemna_ has quit IRC | 18:11 | |
| mordred | clarkb: sorry - reading scrollback | 18:12 |
| mordred | pabelanger, clarkb: link hrefs are deleted from results from shade | 18:12 |
| clarkb | mordred: sort of | 18:13 |
| clarkb | mordred: they aren't incldued | 18:13 |
| clarkb | mordred: https://review.openstack.org/558247 I think fixes this | 18:13 |
| mordred | so ... | 18:14 |
| mordred | before we go doing that | 18:14 |
| mordred | the patch above does not use the shade layer to get the hrefs | 18:14 |
| mordred | it uses the sdk layer which already preserves them | 18:14 |
| mordred | so I'd like to figure out what is actually wrong | 18:14 |
| clarkb | mordred: reading the code I don't think that is true | 18:14 |
| clarkb | mordred: it goes through normalize_server which is what I've fixed above | 18:14 |
| mordred | https://review.openstack.org/#/c/555433/3/launch/dns.py | 18:14 |
| clarkb | (though I've not yet gotten a traceback from pabelanger so not 100% sure) | 18:14 |
| mordred | look at line 81 - 'import openstack' then at line 38 https://review.openstack.org/#/c/555433/3/launch/dns.py | 18:15 |
| mordred | line 38 uses the sdk layer connection.compute.get_server( which does not go through the shade normalize layer | 18:15 |
| mordred | so if that's not working, there is a bug that we should fix | 18:16 |
| *** e0ne has quit IRC | 18:16 | |
| mordred | "It seems cloud.get_server(server.id) doesn't return any href info by default" <-- *that* is correct - but that's not what launch/dns.py is doing | 18:17 |
| mordred | (at least not with that patch applied) | 18:17 |
| *** armaan has joined #openstack-infra | 18:18 | |
| clarkb | ya a traceback would be helpful | 18:19 |
| clarkb | my issue with it was unrelated and due to old shade breaking under new fix | 18:19 |
| clarkb | so I don't have the traceback to go off of | 18:19 |
| *** dprince has quit IRC | 18:19 | |
| mordred | nod | 18:20 |
| *** tosky has joined #openstack-infra | 18:20 | |
| mordred | clarkb: also, fwiw, your patch won't work - but I left comments inline | 18:20 |
| clarkb | mordred: fwiw reading the sdk side of things it isn't anywhere clear how the compute connection is made via connect() | 18:20 |
| clarkb | its definitely using the shade openstacloud stuff as a mixin | 18:21 |
| mordred | clarkb: connect() just makes a Connection. That Connection has all of the shade methods, as well as an attribute corresponding to each service named after its official service-type | 18:21 |
| clarkb | (which is why I'm guessing its broken) but not making the jump immediately | 18:21 |
| *** jamesmca_ has joined #openstack-infra | 18:22 | |
| mordred | clarkb: those attributes are either just a raw keystoneauth1.adapter.Adapter (a requests Session mounted on a particular service endpoint) or they are a Proxy object ... in thise case, conn.compute can be found in openstack/compute/_proxy.py | 18:22 |
| mordred | clarkb: would it be helpful if I tried to use launch.py to do something? | 18:24 |
| clarkb | mordred: maybe. I think shade and sdk are being too magical here and its bioting you. This is going through normalize because the only thing that defines get_server() is shade | 18:25 |
| clarkb | mordred: your compute proxy methods do not implement this method | 18:25 |
| mordred | that's just not accurate at all | 18:25 |
| mordred | or, rather, it's 100% defintely not going through shade normalize | 18:26 |
| *** sree has joined #openstack-infra | 18:26 | |
| clarkb | ok there is no other get_server() | 18:26 |
| mordred | it's possible that get_server is not defined | 18:26 |
| clarkb | so its possible its getting garbage back | 18:26 |
| clarkb | either get_server() is being called and going through normalize or its not being called at all and that is the error | 18:26 |
| mordred | there are no possible ways in which get_server is being called and is going through normalize if that patch is applied and not reverted | 18:27 |
| *** armaan has quit IRC | 18:27 | |
| mordred | buI **really** want to see a traceback or something | 18:27 |
| mordred | because theoretical arguments are not useful :) | 18:28 |
| *** armaan has joined #openstack-infra | 18:28 | |
| mordred | clarkb: mind if I try booting a test server? | 18:29 |
| mordred | clarkb: and is there a region in particular that we were having issues? | 18:29 |
| clarkb | mordred: no its any cloud/region I think | 18:29 |
| clarkb | mordred: you shouldn't need to boot a test server just run dns.py against it | 18:29 |
| clarkb | er against an existing one | 18:29 |
| mordred | kk | 18:29 |
| *** sree has quit IRC | 18:30 | |
| clarkb | mordred: reading the proxy api _get_resource() is likely what you want as the call but its private | 18:31 |
| clarkb | looking to see if that is aliased anywhere | 18:31 |
| *** tosky has quit IRC | 18:31 | |
| *** rpioso has quit IRC | 18:32 | |
| *** dprince has joined #openstack-infra | 18:32 | |
| clarkb | ya I don't think there is a vanilla get Server object via openstacksdk that is public | 18:32 |
| *** rpioso has joined #openstack-infra | 18:32 | |
| jlvillal | gerritbot review request: https://review.openstack.org/#/c/545469/ Some cleanup/refactoring and adding unit tests. Has one +2 Thank you. | 18:32 |
| mordred | clarkb: is there a reason why we have old copies of software in /root/launch-env ? | 18:33 |
| clarkb | mordred: I do not know. I always run it out of my homedir | 18:33 |
| clarkb | mordred: could be to make the rax dns client work though | 18:33 |
| *** tosky has joined #openstack-infra | 18:34 | |
| clarkb | mordred: my current reading on this is openstacksdk lacks a get server method we can use that is public to replcae the current get_server() call taht doesn't exist. And if we switch to using shade we won't get the links attribute without my change. | 18:34 |
| clarkb | So either one or both of those things will need to be addressed then dns.py can be updated | 18:35 |
| mordred | clarkb: the script works fine | 18:36 |
| mordred | clarkb: it just requires a new version of shade being actually installed | 18:36 |
| clarkb | mordred: what get_server is being called? | 18:36 |
| mordred | http://paste.openstack.org/show/718202/ | 18:36 |
| clarkb | mordred: there does not exist a get_server outside of shade from what I can tell | 18:36 |
| *** yamamoto has joined #openstack-infra | 18:37 | |
| clarkb | there is a _get and a _get_resource and neither of them take the correct args | 18:37 |
| mordred | clarkb: https://docs.openstack.org/openstacksdk/latest/user/proxies/compute.html#openstack.compute.v2._proxy.Proxy.get_server | 18:38 |
| mordred | clarkb: there is some magic going on that is currently in play only for the compute proxy as a trial balloon - it may or may not stick around | 18:38 |
| clarkb | its more than some :P | 18:38 |
| clarkb | like I can see how _get and _get_resource work | 18:38 |
| clarkb | but nothing to link that back to a get_server | 18:39 |
| mordred | clarkb: http://git.openstack.org/cgit/openstack/openstacksdk/tree/openstack/_meta/_proxy_templates.py#n120 <-- is where the body of get_server is defined, and http://git.openstack.org/cgit/openstack/openstacksdk/tree/openstack/_meta/proxy.py is where it gets applied/added | 18:40 |
| clarkb | metadata.MetadataMixin on the resource type is doing the magic I aussume | 18:40 |
| clarkb | I have to say as a user of python-*client this was always my biggest issue with them... | 18:41 |
| mordred | my biggest issue is that they don't work properly | 18:41 |
| mordred | clarkb: but noted - there's a reason that's only live for one set of resources right now | 18:41 |
| clarkb | well that too but as a user you can never determine how anything works without jumping through 10 hoops | 18:41 |
| clarkb | and since these tools are user facing it is valuable to not do that imo | 18:42 |
| mordred | fair | 18:42 |
| clarkb | the first thing I do when trying to understand a set of code that calls get_server() is grep for get_server | 18:42 |
| clarkb | that doesn't actually seem to exist until you import the sdk into python | 18:42 |
| *** Swami has joined #openstack-infra | 18:42 | |
| mordred | in any case - I think we need to actually unwind the mess that is launch_node | 18:42 |
| *** sshnaidm has quit IRC | 18:43 | |
| *** jd_ has joined #openstack-infra | 18:44 | |
| mordred | or - maybe we don't - re-reading the README in the launch dir, I see that we're documenting that people should use a launch-env in their own dir | 18:44 |
| clarkb | ok so you set class level attributes which if a resource class is then checked for the list of allow_actions which are then compiled into python at run time | 18:44 |
| clarkb | its fancy but really not user friendly imo | 18:44 |
| mordred | clarkb: well, there are some reasons for and against | 18:45 |
| *** yamamoto has quit IRC | 18:45 | |
| mordred | that I'd prefer to dig in to in a different context that troubleshooting launch-node | 18:45 |
| clarkb | ya my launch node fail was due to old env too old to work at all against your new code. AttributeError: 'OpenStackCloud' object has no attribute 'compute' is what I got | 18:46 |
| clarkb | whcih was fine I didn't need the dns stuff as I had to manually configure dns anyways | 18:46 |
| clarkb | (moved cname) | 18:46 |
| mordred | cool | 18:46 |
| mordred | so - from running dns.py on puppetmaster with a good venv, I can confirm it works with latest shade | 18:47 |
| clarkb | cool | 18:47 |
| mordred | I can add a trap to dns that will print out "your shade is too old, please to update" if you'd like? | 18:47 |
| *** jd_ has joined #openstack-infra | 18:47 | |
| clarkb | mordred: ya maybe an if cloud does not have compute attr raise Exception('update shade/openstacksdk for working env') | 18:48 |
| mordred | ++ | 18:48 |
| mordred | patch coming | 18:48 |
| openstackgerrit | Monty Taylor proposed openstack-infra/system-config master: Add error messages to dns.py about old sdk https://review.openstack.org/558256 | 18:53 |
| openstackgerrit | Monty Taylor proposed openstack-infra/system-config master: Add error messages to dns.py about old sdk https://review.openstack.org/558256 | 18:54 |
| mordred | clarkb: ^^ how's that? | 18:54 |
| pabelanger | clarkb: sorry, I didn't have a backtrace for href, they were just empty | 18:55 |
| clarkb | pabelanger: when you ran launch-node.py it should've tracebacked? | 18:56 |
| pabelanger | https://review.openstack.org/558256 | 18:57 |
| pabelanger | that was the only traceback I got | 18:57 |
| pabelanger | using latest shade too | 18:57 |
| pabelanger | rr | 18:57 |
| pabelanger | https://review.openstack.org/558238/ | 18:57 |
| clarkb | pabelanger: cool that confirms mordreds earlier suspicion | 18:58 |
| clarkb | pabelanger: no compute object | 18:58 |
| clarkb | pabelanger: I think that means you need to update sdk/shade | 18:58 |
| clarkb | mordred: ^ I'm guessing based on pabelangers feedback it is no longer sufficient to install only shade and you need to install openstacksdk? | 18:59 |
| mordred | clarkb: latest shade depends on openstacksdk | 18:59 |
| *** VW_ has joined #openstack-infra | 18:59 | |
| clarkb | mordred: maybe it doesn't have a min so updating shade won't bring in newer sdk? <- pabelanger? | 19:00 |
| pabelanger | clarkb: to get review01.o.o online / dns info I reverted locally. But let me prepare the email for ML, then I can test more using latest info | 19:00 |
| mordred | clarkb: I literally just did "virtualenv --python=python2 launch-env-good ; source launch-env-good/bin/activate ; pip install shade" | 19:00 |
| *** VW_ has quit IRC | 19:00 | |
| mordred | clarkb: if you do pip install -U shade in a venv that has it already pip _should_ update all the depends | 19:00 |
| clarkb | mordred: ya because you made a new venv it would start from scratch and get latest sdk | 19:01 |
| *** VW_ has joined #openstack-infra | 19:01 | |
| clarkb | I'm not sure that necessarily would happen in pabelangers case | 19:01 |
| *** tesseract-RH has quit IRC | 19:01 | |
| mordred | pabelanger: mind if I poke in your home dir? | 19:01 |
| pabelanger | mordred: you can, but I deleted the latest launch-env folder | 19:02 |
| *** amoralej is now known as amoralej|off | 19:02 | |
| pabelanger | give me a few minutes and I can see if I can reproduce it | 19:02 |
| mordred | pabelanger: I appreciate that... I want to make sure there isn't another issue - currently I think we're at "pip is weird sometimes | 19:03 |
| *** VW has quit IRC | 19:03 | |
| *** VW_ has quit IRC | 19:05 | |
| pabelanger | clarkb: http://paste.openstack.org/show/718205/ is our new server for review01.o.o, how can I confirm the IPs are whitelisted for emails? | 19:05 |
| *** VW has joined #openstack-infra | 19:06 | |
| clarkb | pabelanger: check https://www.spamhaus.org/lookup/ | 19:06 |
| clarkb | corvus and fungi may have a more relaible method | 19:06 |
| corvus | that | 19:06 |
| pabelanger | thanks | 19:06 |
| *** e0ne has joined #openstack-infra | 19:09 | |
| *** jamesmcarthur has quit IRC | 19:10 | |
| *** david-lyle has joined #openstack-infra | 19:14 | |
| pabelanger | mordred: okay, reproduced issue, even with fresh virtualenv | 19:15 |
| pabelanger | AttributeError: 'OpenStackCloud' object has no attribute 'compute' | 19:15 |
| *** david-lyle is now known as dklyle | 19:16 | |
| smcginnis | So I noticed something a little concerning with bindep. | 19:16 |
| smcginnis | Apprently we had a package named wrong, but until we split things out between dpkg and rpm, it just silently ignored it. | 19:17 |
| smcginnis | https://review.openstack.org/#/c/557697/2/bindep.txt | 19:17 |
| clarkb | smcginnis: does it ignore it or did it print it out as missing? | 19:18 |
| smcginnis | clarkb: It may have printed something in the logs. I'll have to look. | 19:18 |
| mordred | pabelanger: awesome. which venv? | 19:18 |
| smcginnis | But as soon as we fixed the rpm package name, the ubuntu build actually failed because of it. | 19:19 |
| smcginnis | clarkb: So it's at least odd that there was a change in behavior. | 19:19 |
| clarkb | smcginnis: ya we should undersatnd what it is actually doing and if it is correct | 19:19 |
| smcginnis | Looking for a recent CI run to check now... | 19:19 |
| mordred | pabelanger: ~pabelanger/launch-env has a new enough sdk and shade | 19:19 |
| clarkb | smcginnis: my guess is there is no platform called 'dpkg rpm' so it skipped over it | 19:19 |
| clarkb | smcginnis: so there were two bugs in one that were affecting you | 19:20 |
| smcginnis | It should have been "platform:dpkg platform:rpm", right? | 19:20 |
| clarkb | smcginnis: yes | 19:20 |
| pabelanger | clarkb: fungi: do you think R-11 is okay for gerrit replacement? https://releases.openstack.org/rocky/schedule.html | 19:21 |
| fungi | beisner: pip shouldn't normally install prerelease versions unless you pass the --pre option... maybe you're setting that? | 19:21 |
| pabelanger | https://etherpad.openstack.org/p/HITPVWQ5Vr is what I am thinking for ML email | 19:21 |
| clarkb | one way to handle that may be to have bindep warn or error if it was asked to handle a platform it has no clue about | 19:21 |
| *** sshnaidm has joined #openstack-infra | 19:22 | |
| smcginnis | clarkb: Yeah, nothing in the logs. Here's a recent run: http://logs.openstack.org/41/556041/1/check/openstack-tox-py27/97f73b3/job-output.txt.gz#_2018-03-24_15_02_58_839324 | 19:22 |
| smcginnis | clarkb: Maybe that's good though to keep that flexibility. | 19:22 |
| smcginnis | clarkb: Maybe just add a check if there is a space in the name, warn that it probably isn't formatted properly? | 19:23 |
| fungi | pabelanger: R-11 is probably fine timing. this should be a really low-impact maintenance anyway since we're not changing the gerrit release just the distro/java version and maybe versions of some of the java libs | 19:23 |
| beisner | hi fungi - no --pre is in use; good news is the one pkg for which we used allow-unverified, appears to be fluff now. so, on with the day, without that cruft. :) | 19:24 |
| mordred | pabelanger: can you give me the command you ran to reproduce? I'm assuming you used ~pabelanger/launch-env yeah? | 19:25 |
| clarkb | smcginnis: if you run bindep --profiles against it that is close to what broke. It sees it as profiles: ['platform:dpkg', 'rpm'] but won't match on just 'rpm' | 19:25 |
| fungi | beisner: ahh, great. hopefully that somehow solves the gettnig-prerelease-pip problem (though also, you likely want to look into why pip beta releases aren't compatible with what you're running since there will be a 10.0.0 release in a couple weeks) | 19:25 |
| clarkb | smcginnis: but it is formated properly :) | 19:25 |
| pabelanger | mordred: yah, that env, and followed docs: ./launch-node.py $FQDN --flavor "$FLAVOR" \--cloud=$OS_CLOUD --region=$OS_REGION_NAME | 19:25 |
| *** eharney_ has joined #openstack-infra | 19:26 | |
| beisner | fungi: i think 10 is fine for us otherwise | 19:26 |
| fungi | oh, excellent | 19:26 |
| mordred | pabelanger: gotcha - so for you it broke when going all the way through launch-node | 19:26 |
| *** eharney has quit IRC | 19:26 | |
| *** eharney_ is now known as eharney | 19:26 | |
| smcginnis | clarkb: Is an entry without a "platform:" prefix valid there? | 19:26 |
| clarkb | smcginnis: yes I think so | 19:26 |
| clarkb | smcginnis: thats the profile name | 19:26 |
| mordred | yah. you can have one or more profiles | 19:27 |
| pabelanger | mordred: yah, server came online just bombed when we tried to print dns | 19:27 |
| *** jcoufal has quit IRC | 19:27 | |
| smcginnis | Guess I need to read up. Not aware of profiles. | 19:27 |
| smcginnis | Ah, "bindep test". | 19:27 |
| fungi | bindep supports 2 kinds of profiles right now... "platform profiles" which are autogenerated based on information determined by the platform and always get prefixed with platform:, and user profiles which are any arbitrarily-defined string (the grammar is a little more complex than that but essentially that) | 19:28 |
| mordred | pabelanger: WEIRD. that makes literlaly no sense to me | 19:28 |
| clarkb | I'd show examples but it appears that bindep is currently broken for my flavor of tumbleweed :/ | 19:31 |
| openstackgerrit | David Shrewsbury proposed openstack-infra/zuul master: Reorganize "Zuul From Scratch" document https://review.openstack.org/556988 | 19:31 |
| *** agopi has joined #openstack-infra | 19:35 | |
| *** andreas_s has joined #openstack-infra | 19:35 | |
| clarkb | and now I'm going down the rathole of fixing bindep for tumbleweed | 19:36 |
| *** harlowja has joined #openstack-infra | 19:38 | |
| *** spzala has joined #openstack-infra | 19:39 | |
| *** spzala has quit IRC | 19:39 | |
| clarkb | ok after hacking a fix for tumbleweed locally I've confirmed what I told smcginnis | 19:42 |
| clarkb | smcginnis: basically that is treated as a user defined profile which you can select at runtime | 19:42 |
| clarkb | smcginnis: so the use of bindep is buggy but not sure bindep did anythign wrong there | 19:42 |
| *** gouthamr has joined #openstack-infra | 19:42 | |
| *** andreas_s has quit IRC | 19:42 | |
| *** armaan has quit IRC | 19:43 | |
| *** armaan has joined #openstack-infra | 19:43 | |
| *** yamamoto has joined #openstack-infra | 19:45 | |
| *** VW has quit IRC | 19:49 | |
| *** yamamoto has quit IRC | 19:51 | |
| *** lpetrut has quit IRC | 19:51 | |
| *** VW has joined #openstack-infra | 19:53 | |
| *** armaan has quit IRC | 19:54 | |
| *** dsariel has joined #openstack-infra | 19:54 | |
| *** armaan has joined #openstack-infra | 19:54 | |
| *** armaan has quit IRC | 19:56 | |
| *** armaan_ has joined #openstack-infra | 19:56 | |
| smcginnis | clarkb: Yeah, I guess this was just a usage error. | 19:57 |
| smcginnis | It's too bad that there isn't some way to make it more visible if a mistake like that is made, but I can't think of a good way with the current usage. | 19:58 |
| smcginnis | clarkb: One other mistake like that, and it's in another Cinder repo. http://codesearch.openstack.org/?q=dpkg%20rpm&i=nope&files=bindep.txt&repos= | 19:58 |
| smcginnis | clarkb: At least it's an easy fix. | 19:58 |
| *** dtruong has quit IRC | 19:59 | |
| openstackgerrit | Clark Boylan proposed openstack-infra/bindep master: Better openSUSE Tumbleweed detection https://review.openstack.org/558260 | 20:00 |
| clarkb | fungi: smcginnis ^ bugfix that came out of this when trying to understand things locally | 20:00 |
| smcginnis | clarkb: Cool, at least something useful came out of it. :) | 20:01 |
| smcginnis | Other than my education. | 20:01 |
| clarkb | fungi: interestingly new lsb_release seems to output the cirs content on a single line rather than all on different lines | 20:03 |
| *** jamesmca_ has quit IRC | 20:03 | |
| clarkb | fungi: or maybe that is an ubuntu vs suse behavior. Thankfully the split in bindep already seems to handle it properly | 20:03 |
| *** chkumar246 has joined #openstack-infra | 20:03 | |
| *** chandankumar has quit IRC | 20:03 | |
| smcginnis | clarkb: Am I correct in understanding "platform:suse" is not valid? It would need to at least need to be "platform:opensuse". | 20:04 |
| clarkb | smcginnis: both are valid | 20:04 |
| clarkb | smcginnis: opensuse will match both but suse enterprise linux will only match suse aiui | 20:05 |
| smcginnis | OK. Looking through other examples for any other issues. I noticed some entries were "platform: type" with a space after the platform: tag. | 20:05 |
| smcginnis | clarkb: Would that be an unmatched platform with a profile of the tag? | 20:05 |
| clarkb | smcginnis: yes I think so | 20:06 |
| clarkb | worth noting that if those aren't working projects may want to reevaluate if they need the system deps in the first place | 20:06 |
| smcginnis | clarkb: Right, apparently it's been skipping those then, and no one has noticed. | 20:06 |
| *** rockyg has joined #openstack-infra | 20:08 | |
| *** chkumar246 has quit IRC | 20:08 | |
| *** dtruong has joined #openstack-infra | 20:11 | |
| clarkb | hogepodge: any idea if the OHS fedora multinode job is expected to pass? | 20:14 |
| *** iyamahat has quit IRC | 20:16 | |
| *** dprince has quit IRC | 20:16 | |
| *** jamesmcarthur has joined #openstack-infra | 20:16 | |
| *** yamahata has quit IRC | 20:17 | |
| clarkb | http://logs.openstack.org/14/558214/2/check/openstack-helm-multinode-fedora/090b77f/primary/pod-logs/openstack/nova-db-sync-d7lwm/nova-db-sync.txt looks like that is the error | 20:17 |
| clarkb | whcih I don't think is a limestone issue | 20:17 |
| clarkb | pabelanger: logan- do we think we are ready to bump up to more ndoes? I've tracked down a few of the failures in limestone since turning it back on and one is a legit job error and the other is ^ which I think is also the job failing properly | 20:18 |
| *** chkumar246 has joined #openstack-infra | 20:19 | |
| *** eharney_ has joined #openstack-infra | 20:21 | |
| *** jamesmcarthur has quit IRC | 20:21 | |
| *** eharney has quit IRC | 20:23 | |
| *** e0ne_ has joined #openstack-infra | 20:23 | |
| clarkb | I've pinged openstack-helm in their channel asking if it makes sense to move the multinode fedora job to experimental so we don't chew up 5 nodes on every patch if it is expected to fail currently | 20:26 |
| logan- | All looks good on my end, add more whenever you’re ready | 20:26 |
| clarkb | (but that also helps people like me debugging cloud health to see if it is or isn't expected to be reliable) | 20:26 |
| clarkb | logan-: 50 is the number ? | 20:26 |
| logan- | Yup | 20:26 |
| *** e0ne has quit IRC | 20:26 | |
| openstackgerrit | Clark Boylan proposed openstack-infra/project-config master: Ramp limestone up to 50 instances https://review.openstack.org/558266 | 20:27 |
| clarkb | there we go | 20:27 |
| pabelanger | logan-: wfm | 20:28 |
| *** eharney_ is now known as eharney | 20:28 | |
| *** e0ne_ has quit IRC | 20:30 | |
| *** e0ne has joined #openstack-infra | 20:30 | |
| *** iyamahat has joined #openstack-infra | 20:30 | |
| *** rcernin has joined #openstack-infra | 20:31 | |
| *** kgiusti has left #openstack-infra | 20:37 | |
| *** mattt has joined #openstack-infra | 20:40 | |
| mattt | is it possible to cancel a recheck? :P | 20:41 |
| corvus | mattt: nope | 20:41 |
| mattt | corvus: *cries* | 20:43 |
| corvus | mattt: is that a problem? | 20:43 |
| corvus | i ask because our assumption is that it shouldn't be a problem aside from "wasting" resources, but we don't bill projects for use of the ci system :) | 20:44 |
| mattt | corvus: the change is in the gate queue atm (almost done testing), if i inadvertently rechecked the review will it still restart the check even if the gate tests are successful? :) | 20:45 |
| openstackgerrit | Merged openstack-infra/project-config master: Ramp limestone up to 50 instances https://review.openstack.org/558266 | 20:46 |
| corvus | mattt: ah, as long as the gate tests finish first, the recheck shouldn't affect it | 20:46 |
| mattt | corvus: no real issue other than i may have just delayed this change going in by several hours :) | 20:46 |
| mattt | corvus: oh yay! | 20:46 |
| *** yamahata has joined #openstack-infra | 20:47 | |
| *** yamamoto has joined #openstack-infra | 20:47 | |
| corvus | mattt: it should be in both pipelines now; the only thing that could cause it to be removed from gate is if the check result comes back negative. at which point, well, it might be worth a look to see why it's intermittently failing. | 20:47 |
| *** Goneri has joined #openstack-infra | 20:47 | |
| mattt | corvus: i'll feel less bad if it actually fails the gate test :) | 20:48 |
| corvus | definitely worth a look if it fails both :) | 20:49 |
| mattt | thanks corvus :) | 20:51 |
| dmsimard | fungi, clarkb: we got another meetbot abuse in #openstack-operators.. I'll put something on the agenda for tomorrow's meeting | 20:52 |
| pabelanger | clarkb: did you see https://etherpad.openstack.org/p/HITPVWQ5Vr rough draft on revew.o.o replacement | 20:52 |
| *** yamamoto has quit IRC | 20:52 | |
| clarkb | pabelanger: april 11? | 20:53 |
| clarkb | pabelanger: guessing you need to update to the r-11 date if release team is happy with that week | 20:53 |
| pabelanger | clarkb: provisional, wanted to get thoughts | 20:53 |
| clarkb | pabelanger: oh earlier you had asked about r-11 I thought | 20:53 |
| clarkb | which is in june | 20:53 |
| pabelanger | oh, ha | 20:54 |
| *** bramwelt has joined #openstack-infra | 20:54 | |
| pabelanger | sorry, I meant R-21 | 20:54 |
| pabelanger | or maybe r-18? | 20:54 |
| pabelanger | sigh | 20:54 |
| pabelanger | R20, next week | 20:54 |
| clarkb | I think in the past we've tried to give more time for people to update firewalls | 20:55 |
| clarkb | closer to 4 weeks | 20:55 |
| pabelanger | yah, so that is into R-17 then | 20:55 |
| pabelanger | 3 weeks before summit | 20:56 |
| pabelanger | I'll add a topic for the meeting tomorrow | 20:57 |
| dmsimard | pabelanger: is there a reason why we couldn't recover the old floating IPs ? | 20:59 |
| dmsimard | I mean, that's kind of the point of floating IPs | 20:59 |
| dmsimard | Also, I guess 18.04 is due any time soon. Do we want to skip Xenial ? | 20:59 |
| corvus | dmsimard: are you talking about gerrit? rax doesn't have floating ips. | 21:00 |
| pabelanger | dmsimard: given that bionic isn't released yet, don't we can do review.o.o | 21:00 |
| pabelanger | dmsimard: plan is to get up to java 8 first, which is just a rebuild, then plan upgrade of gerrit | 21:01 |
| *** trown is now known as trown|outtypewww | 21:01 | |
| dmsimard | corvus: oh really ? I didn't realize that. At $oldjob (Internap) we ran provider networks too but you can still detach/attach ports manually though. I don't know to what extent we want to do this, just saying there's probably a way if we really want to keep the same IPs. | 21:01 |
| scas | according to ubuntu wiki, 18.04 is slated for april 26 | 21:01 |
| dmsimard | pabelanger: yeah I suppose 18.04 is still some time away | 21:02 |
| pabelanger | scas: I think we've discussed upgrading zuul-executors first to bionic, to pick up kernel fixes. I'd be a little shy to start with review.o.o as first server | 21:02 |
| scas | https://wiki.ubuntu.com/BionicBeaver/ReleaseSchedule is to what i refer | 21:02 |
| scas | pabelanger: a prudent measure | 21:02 |
| scas | doing the extremely visible one first is almost asking for sadness | 21:03 |
| *** esberglu has quit IRC | 21:03 | |
| pabelanger | April 2021 is EOL for xenial, so lots of time | 21:03 |
| *** chandankumar has joined #openstack-infra | 21:04 | |
| *** sree has joined #openstack-infra | 21:04 | |
| *** chkumar246 has quit IRC | 21:04 | |
| clarkb | dmsimard: rax doesn't expose the neutron api at all iirc | 21:06 |
| corvus | dmsimard: rax does not expose the necessary neutron api. we've looked extensively into this and not produced a suitable result. please do look into it yourself with fresh eyes, i hope you find a way. we would be extremely interested in utilizing it if we could. | 21:06 |
| *** rockyg has quit IRC | 21:07 | |
| *** rockyg has joined #openstack-infra | 21:07 | |
| *** rockyg has quit IRC | 21:07 | |
| *** sree has quit IRC | 21:09 | |
| *** chandankumar has quit IRC | 21:09 | |
| dmsimard | I'm not familiar with the rax specifics, I'll look | 21:11 |
| *** jamesmcarthur has joined #openstack-infra | 21:13 | |
| *** ldnunes has quit IRC | 21:14 | |
| *** e0ne_ has joined #openstack-infra | 21:17 | |
| *** jbryce has quit IRC | 21:18 | |
| clarkb | pabelanger: probably the thing to do is get in touch with release team on that date nowish, then if they are happy with it we can confirm it in infra meeting tomorrow | 21:18 |
| ianw | pabelanger: hey, back from easter, catching up. thanks for looking on the arm64 builds, i forgot that the old branch would try to get the old etcd. is that all fixed now? | 21:18 |
| *** e0ne has quit IRC | 21:18 | |
| *** jbryce has joined #openstack-infra | 21:18 | |
| clarkb | ianw: https://review.openstack.org/#/c/558136/2 it is gating I think | 21:19 |
| clarkb | ianw: then backports will need to be done if not already | 21:19 |
| pabelanger | ianw: clarkb: yah, I think that is the only fix. I guess we started caching etcd for queens | 21:20 |
| pabelanger | clarkb: wfm | 21:20 |
| ianw | ok, we could have reverted the override in the project-config too | 21:20 |
| *** chandankumar has joined #openstack-infra | 21:21 | |
| pabelanger | ianw: yah, I had a pause image build patch up, but likely abandon that now | 21:21 |
| clarkb | out of curiousity does anything actually use etcd yet? | 21:23 |
| clarkb | maybe it can be completely removed on the old branch? | 21:23 |
| *** eharney has quit IRC | 21:24 | |
| ianw | clarkb: on arm, or in general? | 21:24 |
| clarkb | in general | 21:25 |
| pabelanger | I want to say maybe tooz? | 21:25 |
| clarkb | projects were told they could use it but unsure if any did in that timeframe | 21:25 |
| pabelanger | I know we discussed at TC meeting on Friday at PTG, but cannot remember the project | 21:26 |
| *** gouthamr has quit IRC | 21:30 | |
| *** VW has quit IRC | 21:30 | |
| *** VW has joined #openstack-infra | 21:31 | |
| *** eernst has quit IRC | 21:35 | |
| *** VW has quit IRC | 21:35 | |
| *** eernst has joined #openstack-infra | 21:36 | |
| smcginnis | Yeah, tooz. | 21:37 |
| smcginnis | AFAIK, that's the only consumer of that service at the moment. | 21:37 |
| *** eernst has quit IRC | 21:37 | |
| clarkb | and it did so in queens? | 21:37 |
| *** eernst has joined #openstack-infra | 21:37 | |
| smcginnis | Some discussion of using it for some Cinder service communication/state sharing right now, but nothing implemented. | 21:37 |
| smcginnis | clarkb: Yep. | 21:37 |
| smcginnis | I think that was added in Pike. | 21:37 |
| clarkb | ok in that case I guess we keep caching it on x86 | 21:38 |
| *** eernst has quit IRC | 21:38 | |
| *** eernst has joined #openstack-infra | 21:38 | |
| smcginnis | Well, I should qualify that statement I guess. In pike tooz, used by Cinder, was enabled to use etcd for global distributed locking. | 21:39 |
| clarkb | ianw: https://review.openstack.org/#/c/557842/ if you can review that when you get a chance it would be much appreciated | 21:39 |
| smcginnis | That's a config option that we default to using local file locks. | 21:39 |
| clarkb | smcginnis: right but is there any way to configure devstack to run it that way? | 21:39 |
| smcginnis | clarkb: I am not aware of any devstack or tempest runs that enable cinder service clustering that need DLM. | 21:39 |
| openstackgerrit | megan guiney proposed openstack-infra/subunit2sql master: Add delete by uuid functions https://review.openstack.org/537775 | 21:41 |
| ianw | clarkb / pabelanger: so that symlink is setup during the dib build on bionic? | 21:43 |
| clarkb | ianw: yup | 21:43 |
| clarkb | ianw: and appears to come from the resolvconf package | 21:43 |
| *** iyamahat_ has joined #openstack-infra | 21:44 | |
| ianw | i feel like i've delt with this before for some reason ... | 21:44 |
| clarkb | ianw: and it points to /var/run/something/or/other that systemd mounts after booting so its really awkward to try and write to that location later | 21:44 |
| *** iyamahat has quit IRC | 21:44 | |
| ianw | but in general name resolution on bionic is broken? i'm wondering if this is more a dib thing | 21:44 |
| ianw | or at least early name resolution? | 21:45 |
| clarkb | ianw: I don't think it is broken in general. The default setup is to use systemd-resolved and it configures your etc/resolv.conf via the resolvconf method aiui | 21:45 |
| clarkb | ianw: it is possible that systemd-resolved isn't starting up early enough to work in some cases | 21:46 |
| *** rfolco|ruck is now known as rfolco|off | 21:47 | |
| pabelanger | yah, dns seems to work | 21:48 |
| *** yamamoto has joined #openstack-infra | 21:48 | |
| ianw | then i guess the other question is, with these workarounds we've got going on now, is it telling us we've got service dependencies not quite right | 21:49 |
| clarkb | our workaround is basically to just write /etc/resolv.conf to nameserver: 127.0.0.1 which happens every early in the boot process due to rc.local aiui | 21:51 |
| clarkb | is there antoher workaround? | 21:51 |
| *** Goneri has quit IRC | 21:51 | |
| *** HeOS has joined #openstack-infra | 21:52 | |
| ianw | probably i bet that systemd people would say this is why rc.local isn't a thing and get a .service file? | 21:53 |
| pabelanger | I'd be curious to see what still needs rc.local | 21:53 |
| pabelanger | and which providers are still injecting dns | 21:54 |
| pabelanger | our glean is setup to no longer do dns | 21:54 |
| *** yamamoto has quit IRC | 21:54 | |
| *** germs has joined #openstack-infra | 21:55 | |
| *** germs has quit IRC | 21:55 | |
| *** germs has joined #openstack-infra | 21:55 | |
| clarkb | ianw: I'm not suer I understand the conflict? | 21:56 |
| clarkb | ianw: we just need a method to put 127.0.0.1 in /etc/resolv.conf and avoid any cloud (like rax) overwriting that for us. | 21:56 |
| clarkb | the rax situation is why rc.local is/was used | 21:56 |
| clarkb | becuse you have to do it on boot to make sure the cloud's tendrils aren't breaking you | 21:57 |
| jhesketh | Morning | 21:57 |
| * mordred waves to jhesketh | 21:59 | |
| jhesketh | o/ | 21:59 |
| *** germs has quit IRC | 22:00 | |
| corvus | zuul meeting time in #openstack-meeting-alt | 22:00 |
| ianw | mordred: when you have a minute and feel like looking at some sql, and who doesn't, could you look in on https://review.openstack.org/#/c/552288/ | 22:01 |
| *** eernst has quit IRC | 22:05 | |
| *** eernst has joined #openstack-infra | 22:06 | |
| *** eernst has quit IRC | 22:06 | |
| *** eernst_ has joined #openstack-infra | 22:07 | |
| *** eernst_ has quit IRC | 22:07 | |
| *** e0ne_ has quit IRC | 22:08 | |
| mordred | ianw: the patch re-orders the columns in the primary key ... oh, I see from your statement | 22:08 |
| dmsimard | pabelanger: hey I see that in your windmill things you set up an SSH key.. but do you set it up on the remote nodes anywhere ? | 22:09 |
| mordred | ianw: so - how much doyou want me to provide feedback on this sql? | 22:09 |
| dmsimard | pabelanger: https://github.com/openstack/windmill/blob/master/playbooks/bastion.yaml | 22:09 |
| *** e0ne has joined #openstack-infra | 22:09 | |
| pabelanger | dmsimard: remote nodes? | 22:10 |
| dmsimard | pabelanger: so the bastion can connect to the other nodes | 22:10 |
| pabelanger | dmsimard: no, that is assumed the remote node has keys setup | 22:11 |
| ianw | mordred: tbh i'm not sure i want to get involved :) it's a yak shaving exercise. git-review broke because we're not testing against a modern git. it uses h2 in testing, so when i tried to update it, i got that | 22:11 |
| dmsimard | pabelanger: hmk, I'll have to hack something together then.. I'm knee deep in inception/nested ansible/ara setup | 22:12 |
| pabelanger | dmsimard: I did before, but recently removed it. If causing an issue, might be able to add it back | 22:12 |
| dmsimard | pabelanger: oh, I don't know if it's an issue right now but I'm looking for inspiration about some things :p | 22:12 |
| *** jamesmcarthur has quit IRC | 22:14 | |
| dmsimard | I'm sort of halfway there with http://git.openstack.org/cgit/openstack-infra/zuul-jobs/tree/roles/copy-build-sshkey | 22:14 |
| ianw | pabelanger: hmm, not so sure ... https://stgraber.org/2012/02/24/dns-in-ubuntu-12-04/ and yeah, the 12.04 in there is telling :) also i've found other bugs that say that resolv.conf now doesn't consider a non-symlinked /etc/resolv.conf as an indication to turn off | 22:15 |
| *** eernst has joined #openstack-infra | 22:15 | |
| *** bobh_ has quit IRC | 22:15 | |
| ianw | pabelanger: maybe it's suggesting to us we work with resolvconf ... put and entry in the "head" file? | 22:16 |
| *** bobh_ has joined #openstack-infra | 22:16 | |
| pabelanger | ianw: yah, I'd be okay with trying that to see if that is the correct way to handle it | 22:16 |
| pabelanger | vs rc.local | 22:17 |
| *** rpioso is now known as rpioso|afk | 22:18 | |
| clarkb | the reason we use rc.local is rax writes to your networking config on boot to force you to use their dns servers | 22:19 |
| clarkb | I don't know if removing novaagent from our images is sufficient for avoiding that or not | 22:20 |
| clarkb | but the rc.local script cleans up the stuff that rax writes to avoid it uses rax dns servers | 22:20 |
| pabelanger | clarkb: yah, we'd need to see if resolvconf happens after write by rax | 22:20 |
| clarkb | pabelanger: I believe it does, the write by rax is there very early | 22:20 |
| clarkb | I expect its there prior to boot even | 22:20 |
| mordred | ianw: wow fun | 22:21 |
| *** bobh_ has quit IRC | 22:21 | |
| clarkb | I think if we can show that rax doesn't interfere anymore because no more nova agent then likely unbound's munging of resolvconfg is sufficient for our needs | 22:22 |
| ianw | clarkb: wouldn't a resolvconf "head" entry override all that anyway? | 22:23 |
| clarkb | ianw: I'm not sure, the way rax sets it up your /etc/network/interfaces scripts (or something similar depending on distro) get written to write specific things into /etc/resolv.conf | 22:24 |
| clarkb | ianw: I want to say resolv.conf is additive so we'd likely have the rax resolvers and local unbound in that case | 22:25 |
| clarkb | rather than one or the other | 22:25 |
| ianw | clarkb: that would be the sed's on the config files removing those bits? | 22:26 |
| clarkb | ianw: ya | 22:27 |
| ianw | https://git.openstack.org/cgit/openstack-infra/project-config/tree/nodepool/elements/nodepool-base/finalise.d/89-unbound#n115 for reference | 22:27 |
| ianw | so i feel like a resolconf override would override even that. yes we'd have 127.0.0.1 then rax1 rax2 in resolv.conf | 22:28 |
| ianw | which is probably ok? | 22:28 |
| clarkb | it isn't ok to use the rax resolvers, they blacklist IPs that hit ratelimits then don't whitelist them when the instance's IP is reused by another user | 22:29 |
| *** rosmaita has quit IRC | 22:29 | |
| clarkb | so we have errors when some other users had hit the ratelimit and gets blacklisted they release the ip and we get it later and dns doesn't work | 22:29 |
| clarkb | (this is what precipitated the use of local unbound in the first place... I do wish cloud local resolvers were reliable though) | 22:30 |
| ianw | but you'd have 127.0.0.1 first in the resolv.conf ; it's in order right, not round-robbin? | 22:31 |
| clarkb | unbound's config is round robin. I don't know what libc does | 22:31 |
| ianw | libc i'm talking about | 22:32 |
| clarkb | ya reading manpage it is in order with libc | 22:32 |
| clarkb | so in that case using head is probably fine | 22:32 |
| ianw | If there are multiple servers, the resolver library queries them in the order listed | 22:33 |
| pabelanger | Does fedora / centos use NetworkManager? | 22:33 |
| clarkb | pabelanger: I don't think it does in the minimal install | 22:33 |
| ianw | pabelanger: no ... that was my next question on all this :) | 22:33 |
| pabelanger | yah, was trying to find all the ways that /etc/resolv.conf gets updated | 22:34 |
| pabelanger | and see some stuff about NetworkManager | 22:34 |
| *** e0ne has quit IRC | 22:36 | |
| ianw | i wonder if networkmanager is setup to use resolvconf underneath | 22:37 |
| *** gouthamr has joined #openstack-infra | 22:37 | |
| ianw | it may be that way on ubuntu? | 22:37 |
| clarkb | similar to how we don't have selinux enabled by default on centos and fedora images I want to say things like networkmanager require more work to pull in | 22:38 |
| clarkb | you'll get them both on a desktop image but not the minimal like cloud stuff | 22:38 |
| *** felipemonteiro_ has quit IRC | 22:39 | |
| fungi | i had to install a recentish xubuntu to do a kernel comparison on a device i'm hacking on, and /etc/systemd/resolved.conf is where i had to adjust local dns resolution | 22:43 |
| fungi | not sure if the cloud images would end up similarly | 22:43 |
| *** yamamoto has joined #openstack-infra | 22:50 | |
| *** yamamoto has quit IRC | 22:56 | |
| openstackgerrit | Merged openstack-infra/nodepool master: Test growroot in boot tests https://review.openstack.org/555103 | 23:03 |
| *** neiloy has quit IRC | 23:04 | |
| ianw | mordred: is your and jhesketh's question on 552288 "do we even need this at all"? | 23:06 |
| *** tosky has quit IRC | 23:08 | |
| *** eernst has quit IRC | 23:11 | |
| *** eernst has joined #openstack-infra | 23:14 | |
| *** caphrim007_ has joined #openstack-infra | 23:16 | |
| *** caphrim007_ has quit IRC | 23:17 | |
| *** hongbin has quit IRC | 23:17 | |
| *** eernst has quit IRC | 23:18 | |
| *** iyamahat_ has quit IRC | 23:18 | |
| *** eernst has joined #openstack-infra | 23:19 | |
| *** yamahata has quit IRC | 23:19 | |
| *** caphrim007_ has joined #openstack-infra | 23:19 | |
| *** caphrim007 has quit IRC | 23:20 | |
| *** eernst has quit IRC | 23:20 | |
| *** eernst has joined #openstack-infra | 23:20 | |
| *** caphrim007_ has quit IRC | 23:23 | |
| *** eernst has quit IRC | 23:24 | |
| *** lbragstad has quit IRC | 23:31 | |
| *** pahuang has joined #openstack-infra | 23:33 | |
| *** iyamahat has joined #openstack-infra | 23:36 | |
| *** olaph has joined #openstack-infra | 23:37 | |
| *** olaph1 has quit IRC | 23:38 | |
| *** gouthamr has quit IRC | 23:41 | |
| *** apetrich has quit IRC | 23:43 | |
| ianw | pabelanger / clarkb : https://bugs.launchpad.net/ubuntu/+source/systemd/+bug/1713803 ... tl;dr what a mess | 23:46 |
| openstack | Launchpad bug 1713803 in open-iscsi (Ubuntu) "replacement of resolvconf with systemd needs integration" [High,Triaged] | 23:46 |
| clarkb | ianw: ugh | 23:48 |
| clarkb | resolved just isn't great for when you know what you want your dns to do | 23:48 |
| clarkb | I like how recent the conversation around unbound + resolved is there though | 23:49 |
| clarkb | clearly an unsolved problem | 23:49 |
| openstackgerrit | Jesse Keating proposed openstack-infra/project-config master: Add github3.py for third party testing https://review.openstack.org/558321 | 23:51 |
| *** Swami has quit IRC | 23:51 | |
| *** yamamoto has joined #openstack-infra | 23:52 | |
| openstackgerrit | Jesse Keating proposed openstack-infra/project-config master: Add github3.py for third party testing https://review.openstack.org/558321 | 23:55 |
| *** yamahata has joined #openstack-infra | 23:55 | |
| *** germs has joined #openstack-infra | 23:56 | |
| *** germs has quit IRC | 23:56 | |
| *** germs has joined #openstack-infra | 23:56 | |
| *** yamamoto has quit IRC | 23:56 | |
| openstackgerrit | Jesse Keating proposed openstack-infra/project-config master: Add a zuul test to run on github3.py changes https://review.openstack.org/558323 | 23:57 |
| ianw | http://manpages.ubuntu.com/manpages/bionic/man8/systemd-resolved.service.8.html ... huh, "The native, fully-featured API systemd-resolved exposes on the bus." | 23:58 |
| ianw | i did not really realise this replaces getaddrinfo() | 23:58 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!