| *** hamzy has quit IRC | 00:01 | |
| *** threestrands has quit IRC | 00:06 | |
| *** dingyichen has joined #openstack-infra | 00:10 | |
| *** hamzy has joined #openstack-infra | 00:14 | |
| *** longkb has joined #openstack-infra | 00:21 | |
| *** dingyichen has quit IRC | 00:38 | |
| *** dingyichen has joined #openstack-infra | 00:41 | |
| *** dingyichen has quit IRC | 00:44 | |
| *** dingyichen has joined #openstack-infra | 00:44 | |
| *** hongbin has joined #openstack-infra | 00:58 | |
| openstackgerrit | Ian Wienand proposed openstack-infra/zuul-sphinx master: Improve warning details https://review.openstack.org/599469 | 01:10 |
|---|---|---|
| openstackgerrit | Ian Wienand proposed openstack-infra/openstack-zuul-jobs master: [dnm] testing zuul-sphinx co-install https://review.openstack.org/599470 | 01:17 |
| *** SpamapS has joined #openstack-infra | 01:17 | |
| openstackgerrit | Ian Wienand proposed openstack-infra/openstack-zuul-jobs master: [dnm] testing zuul-sphinx co-install https://review.openstack.org/599470 | 01:21 |
| *** rcernin has quit IRC | 01:24 | |
| *** rcernin has joined #openstack-infra | 01:24 | |
| *** fried_rice is now known as efried | 01:27 | |
| openstackgerrit | Ian Wienand proposed openstack-infra/openstack-zuul-jobs master: [dnm] testing zuul-sphinx co-install https://review.openstack.org/599470 | 01:27 |
| *** apetrich has quit IRC | 01:28 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: web: add change status page https://review.openstack.org/599472 | 01:37 |
| openstackgerrit | Ian Wienand proposed openstack-infra/openstack-zuul-jobs master: [dnm] testing zuul-sphinx co-install https://review.openstack.org/599470 | 01:50 |
| *** yamamoto has quit IRC | 01:50 | |
| *** yamamoto has joined #openstack-infra | 02:11 | |
| *** larainema has joined #openstack-infra | 02:22 | |
| *** psachin has joined #openstack-infra | 02:41 | |
| *** jiapei has joined #openstack-infra | 02:55 | |
| *** jrist has joined #openstack-infra | 03:31 | |
| *** r-daneel has joined #openstack-infra | 03:34 | |
| *** ramishra has joined #openstack-infra | 03:48 | |
| *** psachin has quit IRC | 04:04 | |
| *** psachin has joined #openstack-infra | 04:06 | |
| *** udesale has joined #openstack-infra | 04:12 | |
| *** zhangfei has joined #openstack-infra | 04:14 | |
| *** Bhujay has joined #openstack-infra | 04:33 | |
| *** agopi has quit IRC | 04:35 | |
| *** janki has joined #openstack-infra | 04:42 | |
| *** zhangfei has quit IRC | 04:51 | |
| *** hongbin has quit IRC | 04:58 | |
| *** pcrews has quit IRC | 05:08 | |
| *** yboaron_ has joined #openstack-infra | 05:11 | |
| *** e0ne has joined #openstack-infra | 05:13 | |
| *** stakeda has joined #openstack-infra | 05:20 | |
| AJaeger | ianw, frickler, could you review one change to update template descriptions (https://review.openstack.org/#/c/598591/) and some of the open infra retirement changes, please (https://review.openstack.org/#/q/status:open+++topic:infra-retire)? | 05:22 |
| openstackgerrit | Tobias Henkel proposed openstack-infra/zuul master: Increase timeout on streaming tests https://review.openstack.org/599495 | 05:24 |
| *** tobiash has joined #openstack-infra | 05:25 | |
| openstackgerrit | Olivier Bourdon proposed openstack/diskimage-builder master: Fix CentOS image build failure when dib runs on debian based system and audit disabled https://review.openstack.org/559485 | 05:28 |
| openstackgerrit | Merged openstack-infra/activity-board master: Retire repo https://review.openstack.org/597389 | 05:29 |
| openstackgerrit | Merged openstack-infra/err2d2 master: Retire err2d2 https://review.openstack.org/597392 | 05:30 |
| openstackgerrit | Merged openstack-infra/vinz-webclient master: Retire vinz-webclient https://review.openstack.org/597398 | 05:31 |
| AJaeger | thanks, ianw ! | 05:31 |
| openstackgerrit | Merged openstack-infra/puppet-releasestatus master: Retire puppet-releasestatus https://review.openstack.org/597440 | 05:31 |
| openstackgerrit | Merged openstack-infra/puppet-docker_registry master: Retire puppet-docker_registry https://review.openstack.org/597631 | 05:32 |
| openstackgerrit | Merged openstack-infra/puppet-vinz master: Retire puppet-vinz https://review.openstack.org/597396 | 05:32 |
| openstackgerrit | Merged openstack-infra/vinz master: Retire vinz https://review.openstack.org/597397 | 05:32 |
| openstackgerrit | Merged openstack-infra/releasestatus master: Retire releasestatus https://review.openstack.org/597439 | 05:32 |
| openstackgerrit | Merged openstack-infra/puppet-featuretracker master: Retire puppet-featuretracker https://review.openstack.org/597650 | 05:32 |
| openstackgerrit | Merged openstack-infra/featuretracker master: retire featuretracker https://review.openstack.org/597649 | 05:33 |
| ianw | AJaeger: np, i just correlated those to the mail thread and all lgtm | 05:33 |
| openstackgerrit | Merged openstack-infra/openstack-zuul-jobs master: Update descriptions for templates https://review.openstack.org/598591 | 05:35 |
| AJaeger | do you have time to abandon remaining open jobs as well, ianw , please? | 05:36 |
| AJaeger | https://review.openstack.org/#/q/project:openstack-infra/vinz+is:open - 4 open | 05:36 |
| AJaeger | https://review.openstack.org/#/q/project:openstack-infra/puppet-docker_registry+is:open - 1 open | 05:36 |
| AJaeger | https://review.openstack.org/#/q/project:openstack-infra/trystack-site+is:open - 2 open | 05:37 |
| AJaeger | https://review.openstack.org/#/q/project:openstack-infra/zuul-packaging+is:open - 2 open | 05:37 |
| AJaeger | that should be all | 05:37 |
| ianw | sure | 05:40 |
| ianw | AJaeger: ok, should be done, i linked the abandon back to the retirement change (if i managed to copy-paste correctly :) | 05:47 |
| *** eumel8 has quit IRC | 05:48 | |
| *** pas-ha has quit IRC | 05:48 | |
| *** pas-ha has joined #openstack-infra | 05:48 | |
| *** dtroyer has quit IRC | 05:49 | |
| *** mugsie has quit IRC | 05:49 | |
| *** zigo has quit IRC | 05:49 | |
| *** dtroyer has joined #openstack-infra | 05:49 | |
| *** AJaeger has quit IRC | 05:50 | |
| *** AJaeger has joined #openstack-infra | 05:50 | |
| openstackgerrit | Tobias Henkel proposed openstack-infra/zuul master: Increase timeout on streaming tests https://review.openstack.org/599495 | 05:52 |
| *** annp has joined #openstack-infra | 05:55 | |
| openstackgerrit | Tobias Henkel proposed openstack-infra/zuul master: Increase hard test timeout https://review.openstack.org/599495 | 05:56 |
| *** elod has joined #openstack-infra | 06:06 | |
| *** hashar has joined #openstack-infra | 06:09 | |
| *** zhangfei has joined #openstack-infra | 06:10 | |
| *** agopi has joined #openstack-infra | 06:10 | |
| *** yboaron_ has quit IRC | 06:15 | |
| *** Bhujay has quit IRC | 06:16 | |
| openstackgerrit | Olivier Bourdon proposed openstack/diskimage-builder master: Fix CentOS image build failure when dib runs on system where audit disabled https://review.openstack.org/559485 | 06:23 |
| *** olivierb has quit IRC | 06:23 | |
| *** psachin has quit IRC | 06:25 | |
| *** psachin has joined #openstack-infra | 06:30 | |
| *** florianf has joined #openstack-infra | 06:31 | |
| *** rcernin has quit IRC | 06:33 | |
| openstackgerrit | Tobias Henkel proposed openstack-infra/zuul master: Cap cherrypy to below 18.0.0 https://review.openstack.org/599502 | 06:34 |
| *** Bhujay has joined #openstack-infra | 06:38 | |
| *** snapiri has joined #openstack-infra | 06:40 | |
| *** e0ne has quit IRC | 06:44 | |
| openstackgerrit | Markus Hosch proposed openstack-infra/zuul master: Add a sanity check for all refs returned by Gerrit https://review.openstack.org/599011 | 06:45 |
| *** apetrich has joined #openstack-infra | 06:59 | |
| *** dingyichen has quit IRC | 07:22 | |
| *** quiquell has joined #openstack-infra | 07:25 | |
| *** yboaron_ has joined #openstack-infra | 07:25 | |
| *** dtantsur|afk is now known as dtantsur | 07:26 | |
| *** ccamacho has joined #openstack-infra | 07:30 | |
| *** holser_ has joined #openstack-infra | 07:32 | |
| *** alexchadin has joined #openstack-infra | 07:33 | |
| *** ramishra has quit IRC | 07:38 | |
| *** apetrich has quit IRC | 07:41 | |
| *** tosky has joined #openstack-infra | 07:41 | |
| *** jpena|off is now known as jpena | 07:45 | |
| openstackgerrit | Tobias Henkel proposed openstack-infra/zuul master: Cap cherrypy to below 18.0.0 https://review.openstack.org/599502 | 07:51 |
| *** olivierb_ has joined #openstack-infra | 07:52 | |
| *** jpich has joined #openstack-infra | 07:57 | |
| *** ramishra has joined #openstack-infra | 08:00 | |
| *** Emine has joined #openstack-infra | 08:05 | |
| *** priteau has joined #openstack-infra | 08:12 | |
| *** apetrich has joined #openstack-infra | 08:13 | |
| *** alexchadin has quit IRC | 08:19 | |
| openstackgerrit | Chris Dent proposed openstack-infra/project-config master: Remove most job entries for microversion-parse https://review.openstack.org/599453 | 08:20 |
| *** derekh has joined #openstack-infra | 08:28 | |
| openstackgerrit | Thomas Morin proposed openstack-infra/project-config master: Add openstack/neutron-inter to neutron https://review.openstack.org/599429 | 08:30 |
| openstackgerrit | Thomas Morin proposed openstack-infra/project-config master: Add openstack/neutron-interconnection to neutron https://review.openstack.org/599429 | 08:32 |
| *** e0ne has joined #openstack-infra | 08:37 | |
| AJaeger | ianw: all looked fine - thanks | 08:38 |
| *** shabeeb has joined #openstack-infra | 08:41 | |
| *** yamamoto has quit IRC | 08:43 | |
| *** d0ugal has quit IRC | 08:43 | |
| *** psachin has quit IRC | 08:50 | |
| *** tonyb has quit IRC | 08:50 | |
| hwoarang | Good morning. Would it be possible to cut a new bindep release to fix opensuse Leap 15 detection (7e87972) ? | 08:51 |
| openstackgerrit | Thomas Morin proposed openstack-infra/project-config master: Add openstack/neutron-interconnection to neutron https://review.openstack.org/599429 | 08:57 |
| *** d0ugal has joined #openstack-infra | 09:03 | |
| *** psachin has joined #openstack-infra | 09:03 | |
| *** electrofelix has joined #openstack-infra | 09:14 | |
| *** gfidente has joined #openstack-infra | 09:16 | |
| *** yamamoto has joined #openstack-infra | 09:19 | |
| *** shabeeb has quit IRC | 09:20 | |
| *** stakeda has quit IRC | 09:21 | |
| quiquell | Hello is review.o.o down ? | 09:25 |
| bandini | same for me, I get timeouts as well | 09:25 |
| quiquell | bandini: ack | 09:26 |
| bandini | seems back up now | 09:40 |
| *** jaosorior has joined #openstack-infra | 09:47 | |
| *** holser_ has quit IRC | 09:48 | |
| *** holser_ has joined #openstack-infra | 09:52 | |
| *** tonyb has joined #openstack-infra | 09:53 | |
| openstackgerrit | Dmitry Tantsur proposed openstack-infra/project-config master: Remove job settings for metalsmith https://review.openstack.org/599547 | 09:54 |
| *** panda|rover has quit IRC | 09:54 | |
| *** panda has joined #openstack-infra | 09:59 | |
| *** panda has quit IRC | 10:04 | |
| *** psachin has quit IRC | 10:06 | |
| *** longkb has quit IRC | 10:06 | |
| openstackgerrit | Chris Dent proposed openstack-infra/openstack-zuul-jobs master: Add openstack-tox-functional-py36 job https://review.openstack.org/599553 | 10:10 |
| *** yamamoto has quit IRC | 10:11 | |
| *** yamamoto has joined #openstack-infra | 10:11 | |
| *** annp has quit IRC | 10:14 | |
| *** yamamoto has quit IRC | 10:20 | |
| openstackgerrit | Chris Dent proposed openstack-infra/openstack-zuul-jobs master: Add openstack-tox-functional-py36 job https://review.openstack.org/599553 | 10:28 |
| *** yamamoto has joined #openstack-infra | 10:44 | |
| sshnaidm | pabelanger, ping about CBS centos mirror on rdo cloud | 10:45 |
| *** mugsie has joined #openstack-infra | 10:46 | |
| *** bhavikdbavishi has joined #openstack-infra | 10:48 | |
| openstackgerrit | Chris Dent proposed openstack-infra/openstack-zuul-jobs master: Add openstack-tox-functional-py36 job https://review.openstack.org/599553 | 10:53 |
| *** pbourke has quit IRC | 10:55 | |
| *** pbourke has joined #openstack-infra | 10:57 | |
| *** takamatsu has joined #openstack-infra | 11:00 | |
| *** jpena is now known as jpena|lunch | 11:02 | |
| *** udesale has quit IRC | 11:24 | |
| *** zhangfei has quit IRC | 11:25 | |
| *** ansmith has quit IRC | 11:29 | |
| openstackgerrit | Trinh Nguyen proposed openstack-infra/openstack-zuul-jobs master: Add openstack-python36-jobs-horizon template https://review.openstack.org/599566 | 11:30 |
| *** beagles has joined #openstack-infra | 11:31 | |
| *** panda has joined #openstack-infra | 11:34 | |
| *** alexchadin has joined #openstack-infra | 11:34 | |
| openstackgerrit | Tobias Henkel proposed openstack-infra/gear master: Add support to keepalive to client https://review.openstack.org/599567 | 11:34 |
| openstackgerrit | Tobias Henkel proposed openstack-infra/gear master: Add support for keepalive to client https://review.openstack.org/599567 | 11:35 |
| *** jcoufal has joined #openstack-infra | 11:46 | |
| openstackgerrit | Tobias Henkel proposed openstack-infra/gear master: Add support for keepalive to client https://review.openstack.org/599567 | 11:50 |
| *** rh-jelabarre has joined #openstack-infra | 11:55 | |
| *** dims has joined #openstack-infra | 11:56 | |
| openstackgerrit | Tobias Henkel proposed openstack-infra/zuul master: Use gearman client keepalive https://review.openstack.org/599573 | 11:57 |
| *** boden has joined #openstack-infra | 11:59 | |
| *** jpena|lunch is now known as jpena | 12:03 | |
| *** ajo has joined #openstack-infra | 12:05 | |
| *** eharney has joined #openstack-infra | 12:06 | |
| *** trown|outtypewww is now known as trown | 12:17 | |
| *** tpsilva has joined #openstack-infra | 12:18 | |
| openstackgerrit | Javier Peña proposed openstack-infra/puppet-openstackci master: Add AFS mirror support https://review.openstack.org/529376 | 12:19 |
| openstackgerrit | Javier Peña proposed openstack-infra/puppet-openstackci master: Add AFS mirror support for RHEL/CentOS https://review.openstack.org/528739 | 12:19 |
| *** rlandy has joined #openstack-infra | 12:30 | |
| *** ccamacho has quit IRC | 12:34 | |
| ttx | any chance any project-config core could approve https://review.openstack.org/#/c/597037/ ? We'd like to have it on Gerrit before PTG | 12:36 |
| *** larainema has quit IRC | 12:37 | |
| *** jbryce has joined #openstack-infra | 12:37 | |
| *** jrist has quit IRC | 12:39 | |
| *** jrist has joined #openstack-infra | 12:40 | |
| mnaser | ttx: +w thank you | 12:41 |
| ttx | cool thx | 12:42 |
| AJaeger | config-core, could you review https://review.openstack.org/598232 and https://review.openstack.org/598131 and https://review.openstack.org/599269 , please? | 12:43 |
| AJaeger | config-core, there're some more unreviewed changes, I just selected some that seem to block more folks | 12:44 |
| AJaeger | thanks, mnaser | 12:44 |
| mnaser | AJaeger: didn't want to +A the last one as i'd like other config-core to check it out | 12:45 |
| mnaser | i should do a run on config stuff | 12:45 |
| AJaeger | thanks - ok, will ask another config-core to review https://review.openstack.org/599269 then... | 12:46 |
| mnaser | AJaeger: just for more awareness / eyes considering it's a pretty major wide change :p | 12:46 |
| *** jamesmcarthur has joined #openstack-infra | 12:46 | |
| openstackgerrit | Merged openstack-infra/project-config master: Add openstack-map repository https://review.openstack.org/597037 | 12:49 |
| *** mriedem has joined #openstack-infra | 12:53 | |
| openstackgerrit | Merged openstack-infra/openstack-zuul-jobs master: Add openstack-python36-jobs-horizon variant https://review.openstack.org/598131 | 12:54 |
| *** mdrabe has joined #openstack-infra | 12:56 | |
| openstackgerrit | Merged openstack-infra/project-config master: remove release job variants for networking-bgpvpn https://review.openstack.org/598232 | 12:56 |
| *** kgiusti has joined #openstack-infra | 12:57 | |
| *** r-mibu has joined #openstack-infra | 13:01 | |
| *** sthussey has joined #openstack-infra | 13:04 | |
| *** bobh has joined #openstack-infra | 13:04 | |
| *** jamesmcarthur has quit IRC | 13:05 | |
| *** ansmith has joined #openstack-infra | 13:11 | |
| *** lbragstad has joined #openstack-infra | 13:19 | |
| *** zxiiro-away is now known as zxiiro | 13:36 | |
| *** bhavikdbavishi has quit IRC | 13:37 | |
| ttx | fungi: could you seed me as openstack-map-core in gerrit: https://review.openstack.org/#/admin/groups/1940,members ? | 13:40 |
| fungi | ttx: done! have fun | 13:42 |
| ttx | thx! | 13:43 |
| *** jamesmcarthur has joined #openstack-infra | 13:45 | |
| *** ccamacho has joined #openstack-infra | 13:46 | |
| *** bobh_ has joined #openstack-infra | 13:54 | |
| *** mwhahaha has joined #openstack-infra | 13:55 | |
| *** aojea has joined #openstack-infra | 13:57 | |
| *** bobh has quit IRC | 13:58 | |
| openstackgerrit | Jeremy Stanley proposed openstack-infra/zuul-base-jobs master: Remove zuul-jobs namespace https://review.openstack.org/599607 | 13:58 |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/openstack-zuul-jobs master: Add warning job about readthedocs publishing https://review.openstack.org/599269 | 13:59 |
| *** udesale has joined #openstack-infra | 13:59 | |
| AJaeger | mnaser: it's not that major - but indeed was broken. Thanks frickler! | 13:59 |
| AJaeger | mnaser: the followup is more fun ;) | 13:59 |
| AJaeger | frickler, mnaser, want to check https://review.openstack.org/#/c/599269/ again, please? | 13:59 |
| AJaeger | infra-root, could you help unretiring cookbook-openstack-bare-metal? Please review https://review.openstack.org/596743 and once it merges, we need to make the repo writable in gerrit - since it's read-only, we cannot switch it via jeepyb | 14:03 |
| openstackgerrit | Merged openstack-infra/nodepool master: Fix node state checks after aquiring lock https://review.openstack.org/599343 | 14:04 |
| AJaeger | config-core, three repo retirements for review, please - https://review.openstack.org/#/c/597018/ https://review.openstack.org/599241 https://review.openstack.org/597636 | 14:04 |
| *** r-daneel has quit IRC | 14:07 | |
| *** ijw has joined #openstack-infra | 14:08 | |
| *** jcoufal has quit IRC | 14:10 | |
| *** jcoufal has joined #openstack-infra | 14:11 | |
| *** Bhujay has quit IRC | 14:17 | |
| openstackgerrit | Paul Bourke (pbourke) proposed openstack-infra/bindep master: Add OracleLinux support https://review.openstack.org/536355 | 14:19 |
| *** eharney has quit IRC | 14:20 | |
| *** quiquell has quit IRC | 14:21 | |
| *** alexchadin has quit IRC | 14:22 | |
| *** alexchadin has joined #openstack-infra | 14:26 | |
| *** jiapei has quit IRC | 14:34 | |
| openstackgerrit | Giulio Fidente proposed openstack-infra/irc-meetings master: Update the IRC room for TripleO https://review.openstack.org/599615 | 14:34 |
| corvus | AJaeger: why can't jeepyb make it read | 14:35 |
| corvus | AJaeger: why can't jeepyb make it read-write? | 14:35 |
| *** jamesmcarthur has quit IRC | 14:36 | |
| *** tmorin has joined #openstack-infra | 14:39 | |
| *** r-daneel has joined #openstack-infra | 14:39 | |
| tmorin | @infraroot: a failure of test-release-openstack-python3 is preventing a change to merge | 14:40 |
| *** gfidente has quit IRC | 14:40 | |
| tmorin | it seems like a syntax issue in an ansible task definition | 14:41 |
| tmorin | http://logs.openstack.org/22/599422/6/check/test-release-openstack-python3/5634ead/job-output.txt.gz#_2018-09-04_13_39_30_518528 | 14:41 |
| corvus | infra-root: ^ | 14:43 |
| tmorin | corvus: thanks... | 14:43 |
| tmorin | not a syntax issue in fact: "The task includes an option with an undefined variable. The error was: 'gpg_key' is undefined" (task is sign-artifacts/tasks/main.yaml) | 14:44 |
| openstackgerrit | Merged openstack-infra/project-config master: Unretire the openstack/cookbook-openstack-bare-metal repo https://review.openstack.org/596743 | 14:45 |
| corvus | tmorin: has that job been running in check for long? | 14:45 |
| pabelanger | yah, it seems like a check jobs is trying to use a secret from project-config | 14:47 |
| corvus | pabelanger: that job is the same as the release job, but it does not have the secret | 14:47 |
| tmorin | corvus: the job run is recent | 14:47 |
| corvus | pabelanger: it's defined in project-config, so i assume that's intentional | 14:47 |
| corvus | tmorin: did it work previously? | 14:48 |
| pabelanger | https://review.openstack.org/574853/ | 14:48 |
| pabelanger | seems to be when it was added | 14:48 |
| tmorin | (it finished 10 minutes ago) | 14:48 |
| tmorin | tmorin: it's the first time I see it, I suspect it's only triggered when setup.cfg changes and we haven't changes to this file very frequently | 14:48 |
| pabelanger | http://git.openstack.org/cgit/openstack-infra/project-config/tree/zuul.d/jobs.yaml#n262 will always fail, because the siging of a package, needs a gpg key, which we don't allow in check | 14:50 |
| corvus | dhellmann: ^ | 14:50 |
| pabelanger | so we likely can remove the post-run playbook | 14:50 |
| pabelanger | signing* | 14:50 |
| corvus | pabelanger: we can allow the use of the key in check | 14:50 |
| * dhellmann perks up his ears | 14:51 | |
| corvus | pabelanger: remember, secrets are bound to playbooks, and this playbook is in project-config. so it's possible. | 14:51 |
| corvus | dhellmann: we're trying to figure out what's up with test-release-openstack-python3 -- tmorin reported a failure above | 14:51 |
| dhellmann | I'm not sure we *need* to test the signing part. That's not usually what fails | 14:51 |
| dhellmann | oh, you know, I tested this job before adding the signing stuff | 14:51 |
| dhellmann | we should make the simplest change we can to get it working. Is that creating a separate playbook and job? or is that allowing the use of secrets in check? | 14:52 |
| corvus | ok, so that explains the change in behavior at least :) | 14:52 |
| dhellmann | yeah, we discovered the missing signing step as part of the release wrap-up | 14:52 |
| fungi | corvus: i think jeepyb could probably switch the project to active by using the api (assuming there's a method for that), but it can't push updated configuration to a read-only project to set the state to anything else (which is how it sets it to read-only) | 14:53 |
| fungi | corvus: for now the workaround has been to have a gerrit admin switch the project state to active before approving the acl change un-retiring it | 14:54 |
| corvus | fungi: ack. and i see that has already happened for that project | 14:54 |
| fungi | corvus: yes, i changed it and left a comment in the review when i approved | 14:55 |
| *** dklyle has joined #openstack-infra | 14:58 | |
| *** jamesmcarthur has joined #openstack-infra | 14:59 | |
| openstackgerrit | Doug Hellmann proposed openstack-infra/project-config master: move artifact signing to a separate post playbook https://review.openstack.org/599630 | 14:59 |
| dhellmann | corvus , pabelanger , tmorin : how about ^^ as a fix? | 15:00 |
| corvus | pabelanger, dhellmann: the sign-artifacts role writes the private key to the workspace on the executor. assuming all of our defenses are in place, a malicious change shouldn't be able to steal the key (as we prevent un-trusted code from running on the executor). so it should be okay to add the secret to that job. it's not that different from how the artifact upload secret is handled. | 15:00 |
| dhellmann | I'm not sure whether that really works because I don't know what the expectations are for things that appear in different playbooks in that list | 15:00 |
| dhellmann | ah, ok, so a simpler change would just add the secret to the job where it's missing? | 15:01 |
| corvus | dhellmann: that solution works too (i think that implements pabelanger's suggestion) | 15:01 |
| dhellmann | ok, I'm fine with either approach, so let know which you prefer. I can write the other patch to add the secret, too | 15:01 |
| corvus | dhellmann: right. i think either solution is valid and safe. but i'd like fungi and pabelanger to chime in on that too. :) the main difference is that adding the key makes it more like the job it's attempting to test (and therefore exercises our signing machinery more), at the risk of increasing the surface area where the key is exposed. | 15:03 |
| dhellmann | makes sense | 15:03 |
| dhellmann | this model treats signing a bit more like the pypi upload step, which seems appropriate, but I can see us wanting to test the signing code too | 15:04 |
| openstackgerrit | Doug Hellmann proposed openstack-infra/project-config master: add the gpg-key to test-release-openstack-python3 https://review.openstack.org/599632 | 15:05 |
| dhellmann | corvus , pabelanger , fungi : there's the alternative approach ^^ -- just let me know which to abandon :-) | 15:05 |
| *** yamamoto has quit IRC | 15:05 | |
| *** yamamoto has joined #openstack-infra | 15:05 | |
| *** gfidente has joined #openstack-infra | 15:06 | |
| *** beekneemech is now known as bnemec | 15:06 | |
| *** lamt has joined #openstack-infra | 15:07 | |
| tmorin | dhellmann, corvus: I haven't any opinion on which is the best fix, I can't say much more than: thanks for working on fixing that! ;) | 15:08 |
| corvus | i'll check back in after breakfast :) | 15:08 |
| fungi | dhellmann: corvus: i wonder if we need to worry about the possibility of proposed change coercing the executor to sign an arbitrary artifact with the official key | 15:09 |
| tmorin | thx pabelanger, fungi as well | 15:09 |
| fungi | if we want to test the signing role, we should probably generate a throw-away key on the fly instead for that? | 15:09 |
| dhellmann | fungi : the test job doesn't upload artifacts, but I can see your concern | 15:09 |
| dhellmann | oh, that's an idea, sure | 15:09 |
| dhellmann | how would we do that? :-) | 15:10 |
| dhellmann | that feels like it's likely to make the jobs even more different than just not testing the signing role | 15:10 |
| fungi | good question. we've done it in years past... i think gpg has an "unsafe" test option to use just its internal prng for key material | 15:11 |
| dhellmann | maybe we should take the first patch that removes the signing step for now, to unblock tmorin, and then work on the other change later? | 15:11 |
| fungi | wfm | 15:12 |
| dhellmann | ok, I'll abandon the one that adds the key | 15:12 |
| dhellmann | done | 15:13 |
| *** dklyle has quit IRC | 15:13 | |
| *** dklyle has joined #openstack-infra | 15:14 | |
| *** olivierb_ has quit IRC | 15:15 | |
| AJaeger | team, do we document the various bindep profiles anywhere? I've been discussing as part of https://review.openstack.org/#/c/599553 whether to add a py36 bindep profile for job tox-functions-py36 (which is consistent with the tox-py36 job but inconsinstent with tox-functional-py35 job). | 15:16 |
| fungi | dhellmann: aha, right, i always forget that the --quick-random and --debug-quick-random options are intentionally omitted from the gpg manpage ;) | 15:16 |
| AJaeger | I'm not aware of such doku - and can volunteer to do a first draft if there's interest and you have a good suggestion on where. | 15:16 |
| fungi | dhellmann: but those are the ones that do what i was remembering | 15:17 |
| dhellmann | fungi : ah, ok. is there already a role or playbook for generating and configuring a key using those? would the existing signing role use it, or does it assume that there is a gpg key installed? | 15:18 |
| dhellmann | I guess I'm not clear on which part of the job installs the key vs. which part uses it, and whether those are separable right now | 15:19 |
| pabelanger | corvus: dhellmann: tmorin: fungi: with adding the gpg key, won't we also need to update pipeline configuration to allow post-review: True, so zuul can decypt the secret? | 15:19 |
| pabelanger | for check | 15:19 |
| *** openstackgerrit has quit IRC | 15:20 | |
| fungi | dhellmann: i don't think we've got any job doing it currently, so no role written for that. the existing roles are linked from https://docs.openstack.org/infra/system-config/signing.html#at-a-glance | 15:20 |
| dhellmann | fungi : ok | 15:22 |
| dhellmann | I have to admit adding that isn't really high on my priority list. We see far more failures in the step trying to build the package, and that's the part I wanted us to be able to test | 15:23 |
| mriedem | clarkb: does gerrit index change-id? because we have a nova and placement repo change with the same change id but so far only the nova change shows up | 15:23 |
| mriedem | https://review.openstack.org/#/c/598988/ and https://review.openstack.org/#/c/596497/ | 15:23 |
| *** shachar has joined #openstack-infra | 15:23 | |
| corvus | pabelanger: no because the job is defined in project-config, a trusted project: https://zuul-ci.org/docs/zuul/user/config.html#attr-job.post-review | 15:24 |
| clarkb | mriedem: yes, it should, I wonder if that is cached and/or buggy | 15:24 |
| mriedem | clarkb: the placement change was just updated with that change id, | 15:25 |
| mriedem | so maybe needs to wait for the cache to refresh | 15:25 |
| pabelanger | dhellmann: also, can't we just delete the post-run playbook from test-release-openstack-python3 ? While we collect the tox logs, we just pull the artifacts back to the executor and don't publish them. Seems that last step could be skipped to save time / resources | 15:25 |
| clarkb | mriedem: oh if you look at the placement change it has a different change id | 15:25 |
| *** ramishra has quit IRC | 15:25 | |
| *** anteaya has joined #openstack-infra | 15:25 | |
| clarkb | mriedem: that is curious, but also explains why searchign by change id doesn't work the way you expect it | 15:25 |
| pabelanger | corvus: thanks, forgot about that. | 15:26 |
| pabelanger | need coffee | 15:26 |
| clarkb | mriedem: the ps1 change id seems to have persisted through to ps2, I'm not sure how that was managed | 15:26 |
| dhellmann | pabelanger : I'm not sure what you mean? | 15:26 |
| fungi | dhellmann: checking with codesearch, it looks like pbr, releases and reno have gpg key fixtures in their tests, but not being done at the job level: http://codesearch.openstack.org/?q=quick-random | 15:26 |
| clarkb | mriedem: maybe efried updating the change in gerrit itself? and that doesn't create a new chagne even if commit message change id is updated? | 15:26 |
| efried | clarkb: "the ps1 change id seems to have persisted through to ps2" <== how are you seeing this? | 15:27 |
| clarkb | mriedem: but that is why it isn't working the way you expect, the change id is not actually the change id in the commit message because reasons | 15:27 |
| *** snapiri has quit IRC | 15:27 | |
| fungi | dhellmann: but i agree, the work needed for safely testing that existing role isn't a high priority | 15:27 |
| dhellmann | pabelanger : oh, I see. for some reason I thought the post playbook was the one that actually built the tarballs | 15:27 |
| efried | eek, okay. | 15:27 |
| clarkb | efried: on the change screen there is a table with commit, parent, change-id in it, the change id there is the change id from ps1 | 15:27 |
| clarkb | efried: that is the actual changeid as far as gerrit sees it | 15:28 |
| pabelanger | dhellmann: http://git.openstack.org/cgit/openstack-infra/project-config/tree/playbooks/pti-python-tarball/post.yaml#n5 in http://git.openstack.org/cgit/openstack-infra/project-config/tree/zuul.d/jobs.yaml#n254 doesn't actully publish the tarballs, so not sure we actually need to do the post-run step | 15:28 |
| pabelanger | dhellmann: yah | 15:28 |
| mriedem | oh i see it yeah | 15:28 |
| dhellmann | pabelanger : ok, let me try this again | 15:28 |
| efried | oh, yeah, jeez | 15:28 |
| efried | okay, /me downloads... | 15:28 |
| AJaeger | mriedem, clarkb: efried updated the change via gerrit UI - and therefore it did not see that as a new one. Compared to an upload where it parses the commit message | 15:28 |
| mriedem | efried: so maybe need to push separately from 598988 but i'm not sure if it will let you | 15:29 |
| pabelanger | dhellmann: you could parent test-release-openstack-python3 to tox, for tox log collection too | 15:29 |
| mriedem | you'd probably have to go back to ps1 content, abandon and then do a new push | 15:29 |
| *** Miouge has left #openstack-infra | 15:29 | |
| *** Miouge has joined #openstack-infra | 15:29 | |
| dhellmann | pabelanger : that job doesn't use tox | 15:29 |
| fungi | efried: mriedem: worth noting, you can't alter the change-id in subsequent patchsets as the project,branch,change-id tuple is how gerrit figures out whether the push should go to an existing change or create a new one | 15:30 |
| *** openstackgerrit has joined #openstack-infra | 15:30 | |
| openstackgerrit | Doug Hellmann proposed openstack-infra/project-config master: skip collecting built artifacts in packaging test job https://review.openstack.org/599638 | 15:30 |
| dhellmann | pabelanger : is that better? ^ | 15:30 |
| *** armax has joined #openstack-infra | 15:31 | |
| *** Emine has quit IRC | 15:31 | |
| AJaeger | dhellmann: cloudkitty seems ready for python3-first, want to remove your WIP from https://review.openstack.org/598929 ? | 15:31 |
| efried | fungi: That makes... some sense. | 15:31 |
| openstackgerrit | Markus Hosch proposed openstack-infra/nodepool master: Add metric for image build result https://review.openstack.org/590412 | 15:31 |
| efried | fungi: But it's not going to be a problem to push the same change-id from a different repo? | 15:32 |
| openstackgerrit | sebastian marcet proposed openstack-infra/openstackid-resources master: added new endpoint ADD Organization https://review.openstack.org/599640 | 15:32 |
| pabelanger | dhellmann: yah, so if we think there is value in signing a test tarball, I would agree with corvus we likey can just add the gpg_key to the test job. I agree, we should have all the checks in place to prevent it from leaking | 15:32 |
| dhellmann | I think the value of figuring out how to do that is small compared to unblocking teams having an issue with the broken job right now. | 15:33 |
| dhellmann | so let's have a separate conversation about that | 15:33 |
| openstackgerrit | Merged openstack-infra/openstackid-resources master: added new endpoint ADD Organization https://review.openstack.org/599640 | 15:33 |
| dhellmann | thanks, AJaeger , done | 15:34 |
| fungi | efried: right, change-id is only unique per project+branch combination | 15:34 |
| AJaeger | config-core, please review https://review.openstack.org/598929 for python3-first of cloudkitty | 15:35 |
| *** alexchadin has quit IRC | 15:35 | |
| efried | Dig. Thanks fungi, AJaeger, clarkb - https://review.openstack.org/#/q/Ie21bddc186364db2b10af45546c35b9ad5a0208b now does what it ought. | 15:35 |
| fungi | efried: so you can have the same change-id be a separate change as long as the target branch or project is different | 15:35 |
| *** alexchadin has joined #openstack-infra | 15:36 | |
| AJaeger | pabelanger, fungi, do you want to approve dhellmann's change https://review.openstack.org/#/c/599630/ - or -1 it? | 15:36 |
| fungi | but to actually switch the change-id on an existing change, your only choice (afaik) is to abandon and re-propose | 15:36 |
| *** armax has quit IRC | 15:36 | |
| *** alexchadin has quit IRC | 15:36 | |
| dhellmann | AJaeger : I think we want https://review.openstack.org/599638 instead | 15:36 |
| *** alexchadin has joined #openstack-infra | 15:36 | |
| AJaeger | fungi: it works to download, change the change-id and git review again - that's what I think I did in the past. | 15:36 |
| *** alexchadin has quit IRC | 15:37 | |
| *** sshnaidm is now known as sshnaidm|afk | 15:37 | |
| AJaeger | dhellmann: ah, missed that - let me +2A that one... | 15:37 |
| fungi | AJaeger: right, but you get a new change and have to abandon the old change, right? | 15:37 |
| *** alexchadin has joined #openstack-infra | 15:37 | |
| AJaeger | fungi: yes, you have to abandon the old one | 15:37 |
| *** alexchadin has quit IRC | 15:37 | |
| fungi | that's what i meant by abandon and re-propose | 15:38 |
| *** alexchadin has joined #openstack-infra | 15:38 | |
| AJaeger | fungi: ah, so we're in agreement - I just didn't parse it initially ;( | 15:38 |
| *** alexchadin has quit IRC | 15:38 | |
| fungi | you can of course add comments in both changes linking to one another so people know where to find prior review history, if it's substantial enough to worry about | 15:38 |
| *** jamesmcarthur has quit IRC | 15:43 | |
| *** jamesmcarthur has joined #openstack-infra | 15:43 | |
| *** eharney has joined #openstack-infra | 15:45 | |
| openstackgerrit | Clark Boylan proposed openstack-infra/system-config master: Manage clouds.yaml files in ansible https://review.openstack.org/598329 | 15:49 |
| openstackgerrit | Clark Boylan proposed openstack-infra/system-config master: Add a run-nodepool job https://review.openstack.org/598403 | 15:49 |
| openstackgerrit | Clark Boylan proposed openstack-infra/system-config master: Add testinfra tests for bridge https://review.openstack.org/598404 | 15:49 |
| clarkb | corvus: ^ I squashed your include_role change down into that first cahnge so everything should work now I think | 15:49 |
| AJaeger | clarkb: I think we're ready for the final steps of the 18 repos to retire - could you put https://review.openstack.org/597636 on your review list. Any other config-core to review as well, please? | 15:51 |
| *** aojea has quit IRC | 15:52 | |
| *** ccamacho has quit IRC | 15:53 | |
| openstackgerrit | Doug Hellmann proposed openstack-infra/project-config master: remove job settings for barbican repositories https://review.openstack.org/599663 | 15:53 |
| corvus | clarkb: sounds good | 15:55 |
| hwoarang | fungi clarkb: hello, would it be possible to cut a new bindep release? We need 7e87972 which fixes opensuse leap 15 detection once again :/ | 15:56 |
| *** ginopc has quit IRC | 15:57 | |
| *** jlvilla-viva is now known as jlvillal | 15:58 | |
| *** yamamoto has quit IRC | 15:58 | |
| *** yamamoto has joined #openstack-infra | 15:58 | |
| clarkb | hwoarang: ya that should be possible, looks like the only commits since then are for job configs or to address suse behavior which is a small delta. Also openstack release happened so we won't impact that either | 15:59 |
| hwoarang | thank you clarkb | 15:59 |
| clarkb | fungi: ^ any objections to me pushing up a 2.8.1 tag | 15:59 |
| fungi | none whatsoever. .1 should be fine given what's landed in master | 16:00 |
| fungi | also we're probably free to finalize a new git-review release today | 16:00 |
| fungi | now that the release is past | 16:00 |
| clarkb | ok tagging bindep bacfe224388dbe8d2b2248bc606e82c0c11100cf as 2.8.1 now | 16:02 |
| clarkb | and done | 16:04 |
| *** tmorin has quit IRC | 16:06 | |
| *** alexchadin has joined #openstack-infra | 16:10 | |
| *** mdrabe has quit IRC | 16:10 | |
| openstackgerrit | Merged openstack-infra/project-config master: skip collecting built artifacts in packaging test job https://review.openstack.org/599638 | 16:12 |
| clarkb | any idea if we get mordred back before the PTG? curious if he will want to review those changes to write clouds.yaml with ansible | 16:14 |
| *** alexchadin has quit IRC | 16:14 | |
| *** holser_ has quit IRC | 16:15 | |
| Shrews | clarkb: i believe he said he was gone up through the ptg | 16:15 |
| AJaeger | fungi, pabelanger, could you review https://review.openstack.org/597636 to retire the infra repos, please? | 16:15 |
| clarkb | Shrews: thanks, we probably won't wait on mordred then :) | 16:15 |
| *** Bhujay has joined #openstack-infra | 16:17 | |
| fungi | AJaeger: oh, i had a note from sometime back that puppet-odsreg can be retired too, if we've got a new batch in the works | 16:18 |
| clarkb | Thinking about the meeting later today, if there are any PTG items you want to talk about please add them to the agenda. We'll have time during the meeting to make sure we are ready. | 16:19 |
| clarkb | Also if you can't attend the PTG but are interested in specific topics feel free to add those details to the etherpad and/or bring that up in the meeting so that we can be sure to take good notes | 16:19 |
| *** anteaya has quit IRC | 16:20 | |
| clarkb | https://etherpad.openstack.org/p/infra-ptg-denver-2018 is the etherpad | 16:20 |
| AJaeger | fungi: I can start a new batch ;) Just puppet-odsreg or anything else? | 16:21 |
| *** hashar is now known as hasharAway | 16:21 | |
| fungi | nothing i know of, just found my old note about that ojne | 16:22 |
| fungi | one | 16:22 |
| *** olivierb has joined #openstack-infra | 16:23 | |
| *** gyee has joined #openstack-infra | 16:24 | |
| Shrews | clarkb: do you happen to know the answer to the question i ask in https://review.openstack.org/593029 ? | 16:24 |
| Shrews | re: .pem file mode | 16:24 |
| clarkb | hwoarang: fungi https://pypi.org/project/bindep/ all done | 16:24 |
| hwoarang | great thank you! | 16:25 |
| AJaeger | fungi: will do it... | 16:25 |
| fungi | AJaeger: thanks! | 16:26 |
| clarkb | Shrews: responded it should be fine as is as the file content is public (it isn't the secret key, but the public cert) | 16:26 |
| *** janki has quit IRC | 16:26 | |
| *** janki has joined #openstack-infra | 16:27 | |
| Shrews | clarkb: cool. i figured it was ok, but wanted to point out the difference just in case | 16:27 |
| AJaeger | fungi: as you're one of the bindep experts, did you see my question earlier? Let me repeat it... | 16:27 |
| clarkb | Shrews: thank you for the review | 16:27 |
| AJaeger | team, do we document the various bindep profiles anywhere? I've been discussing as part of https://review.openstack.org/#/c/599553 whether to add a py36 bindep profile for job tox-functions-py36 (which is consistent with the tox-py36 job but inconsinstent with tox-functional-py35 job). | 16:27 |
| AJaeger | config-core, I'm not aware of such doku - and can volunteer to do a first draft if there's interest and you have a good suggestion on where. | 16:28 |
| clarkb | AJaeger: are the dependencies not the say for py35 and py36? The distros we run on use python3 package names which get you 3.5 or 3.6 depending on distro release iirc | 16:29 |
| clarkb | s/not the say/not the same/ | 16:29 |
| AJaeger | clarkb: http://git.openstack.org/cgit/openstack-infra/openstack-zuul-jobs/tree/zuul.d/jobs.yaml#n330 and http://git.openstack.org/cgit/openstack-infra/openstack-zuul-jobs/tree/zuul.d/jobs.yaml#n350 | 16:30 |
| AJaeger | and py27 has the same... | 16:30 |
| AJaeger | maybe that was wrong at that time already... | 16:30 |
| clarkb | huh, I'm happy to support that if we already have it. On some platforms where the package names are specific it may even be necessary | 16:31 |
| *** jpich has quit IRC | 16:31 | |
| clarkb | AJaeger: I don't know if such a document either, but maybe we do need one if we are adding more of these profiles | 16:31 |
| AJaeger | clarkb: I fear we need one - this is some kind of policy we should be clear on IMHO - and explain how to use them. | 16:32 |
| AJaeger | clarkb, fungi - but where to add it? Infra-manual? bindep? openstack-zuul-jobs? | 16:33 |
| AJaeger | checking infra-manual, I'll draft something... | 16:34 |
| clarkb | AJaeger: probably openstack-zuul-jobs since that is where we actually set and use them (bindep uses arbitrary profiles and Infra itself doesn't really care what various projects use) | 16:34 |
| Shrews | clarkb: what's the reasoning behind using include_role here? https://review.openstack.org/#/c/598329/10/playbooks/base.yaml | 16:35 |
| clarkb | Shrews: the previous patchset https://review.openstack.org/#/c/598329/9..10/playbooks/base.yaml used roles normally but the vars value persisted across the two invocations so we only ever wrote the one file and tests failed. corvus tested with include_role and found that allowed us to run the role more like a function multiple times | 16:36 |
| AJaeger | clarkb: was thinking about extending https://docs.openstack.org/infra/manual/drivers.html#package-requirements - but that can be a link to ozj as well... | 16:36 |
| Shrews | clarkb: ah. thx | 16:37 |
| openstackgerrit | Merged openstack-infra/project-config master: Infra repo retirement of 18 repos - step 3 https://review.openstack.org/597636 | 16:37 |
| *** janki has quit IRC | 16:38 | |
| *** trown is now known as trown|lunch | 16:39 | |
| clarkb | AJaeger: ya that looks like a good place for it currently given the context. I'm not sure that is the best place for it long term but we can figure that out as we work through winterscale | 16:39 |
| AJaeger | clarkb: yeah, let's see what winterscale will bring... | 16:40 |
| pabelanger | clarkb: ah, neat. ring.cx is from fine folks at SFL in montreal. TIL | 16:44 |
| *** ijw has quit IRC | 16:44 | |
| *** ijw has joined #openstack-infra | 16:44 | |
| pabelanger | they are big asterisk users | 16:44 |
| pabelanger | or were | 16:44 |
| fungi | AJaeger: clarkb: yeah, i hadn't noticed we were doing per-pyver bindep profiles... seems like maybe a good opportunity for an additional layer like py3 so that only package names which differ between python 3.5 and 3.6 environments need those specified? | 16:47 |
| *** bhavikdbavishi has joined #openstack-infra | 16:47 | |
| fungi | do you have any example bindep.txt files using these? | 16:48 |
| fungi | if codesearch is to be believed, nothing is relying on them anyway | 16:49 |
| AJaeger | fungi: only py37 in telemetry repos like http://git.openstack.org/cgit/openstack/ceilometer/tree/bindep.txt#n7 | 16:50 |
| fungi | http://codesearch.openstack.org/?q=py3&i=nope&files=bindep.txt | 16:50 |
| fungi | yep, just discovered the same | 16:50 |
| fungi | and looks like they're explicitly relying on those to get the python interpreter installed, nothing else | 16:50 |
| AJaeger | fungi: let me write a first draft - and we can discuss from there... | 16:50 |
| *** udesale has quit IRC | 16:51 | |
| AJaeger | so, do we want "py3 py37" as bindep profile? | 16:51 |
| fungi | well, it might be nice to start by thinking about what problem this is actually solving | 16:52 |
| *** bharath_t has joined #openstack-infra | 16:52 | |
| fungi | since we see it's not really been widely adopted yet (just a handful of telemetry repos using it to get python 3.7 installed onto rpm-based platforms) | 16:52 |
| AJaeger | fungi: yes, exactly | 16:53 |
| clarkb | pabelanger: I like that it is decentralized but the android client has never fully worked for me. Been about a year since I tested it though | 16:54 |
| fungi | "get the right python interpreter version installed for some particular tox testenv" seems like a bit of a corner case to me | 16:55 |
| fungi | and i wonder if there's not a better solution to that | 16:55 |
| bharath_t | Hi, do we need to use "https://test.pypi.org/legacy/" for uploading openstack python packages? | 16:55 |
| fungi | and also, will users who want to explicitly run tests in a python 3.7 tox testenv not know they need to install python 3.7? | 16:56 |
| bharath_t | or some other url? | 16:56 |
| fungi | bharath_t: test.pypi.org is a test copy of pypi.org | 16:56 |
| *** dkehn has quit IRC | 16:56 | |
| fungi | bharath_t: what openstack python packages are you uploading? we generally upload release artifacts through ci automation | 16:57 |
| bharath_t | I am trying to create new project based on https://docs.openstack.org/infra/manual/creators.html | 16:57 |
| *** hwoarang_ has joined #openstack-infra | 16:57 | |
| AJaeger | fungi, but also how to install the interpreter in our gates | 16:57 |
| bharath_t | here it suggested to upload initial version | 16:57 |
| pabelanger | clarkb: ah, it is a rename of SFLPhone: https://blog.savoirfairelinux.com/en-ca/e/ring/ which I've used in the past with asterisk and call centres | 16:58 |
| *** hwoarang__ has joined #openstack-infra | 16:58 | |
| *** yboaron_ has quit IRC | 16:58 | |
| clarkb | pabelanger: its more than that I think as its about ditching the server and doing point to point calls | 16:59 |
| *** dkehn has joined #openstack-infra | 16:59 | |
| Shrews | clarkb: corvus: the clouds.yaml stuff lgtm | 17:00 |
| *** anteaya has joined #openstack-infra | 17:00 | |
| Shrews | and tests seem happy | 17:00 |
| openstackgerrit | sebastian marcet proposed openstack-infra/openstackid master: Fixed logout without current user but with valid id token https://review.openstack.org/599692 | 17:00 |
| bharath_t | fungi, we shouldn't push manually for the first time according to https://docs.openstack.org/infra/manual/creators.html? | 17:00 |
| *** derekh has quit IRC | 17:01 | |
| *** hwoarang has quit IRC | 17:01 | |
| pabelanger | clarkb: yah | 17:01 |
| fungi | bharath_t: got it, so an initial upload to create the project on pypi? use `twine upload dist/* | 17:02 |
| *** r-mibu has quit IRC | 17:02 | |
| *** hwoarang_ has quit IRC | 17:02 | |
| fungi | bharath_t: you don't need the --repository-url https://test.pypi.org/legacy/ since that's for testing uploads to test.pypi.org and you want to use twine's default (which is pypi.org production, not test.pypi.org) | 17:02 |
| fungi | we should probably revisit this with the release team, but i feel like the entire initial upload step can be skipped. the release jobs will be running twine anyway | 17:03 |
| *** hwoarang has joined #openstack-infra | 17:03 | |
| bharath_t | fungi, got it, it worked! | 17:04 |
| bharath_t | fungi, oh is it, I thought it was recently updated | 17:04 |
| fungi | bharath_t: that's noted in the very last paragraph of the packaging tutorial: Keep in mind that this tutorial showed you how to upload your package to Test PyPI and Test PyPI is ephemeral. It’s not unusual for packages and accounts to be deleted occasionally. If you want to upload your package to the real Python Package Index you can do it by registering an account on https://pypi.org and following | 17:05 |
| fungi | the same instructions, however, use twine upload dist/* to upload your package and enter your credentials for the account you registered on the real PyPI. | 17:05 |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/infra-manual master: Add information about bindep profiles https://review.openstack.org/599694 | 17:06 |
| AJaeger | fungi, clarkb, first draft ^ - happy to rewrite based on feedback (and update jobs afterwards) | 17:06 |
| * AJaeger will WIP for now | 17:06 | |
| fungi | bharath_t: yeah, the creators guide section about setting up the project on pypi was revised recently when pypi changed its code base and disabled all the old methods for registering a new project, but in so doing they made separately registering a project somewhat unnecessary | 17:07 |
| *** hwoarang__ has quit IRC | 17:08 | |
| fungi | i like that the creators guide is now basically recommending people get familiar with python packaging basics, but it's also a bit bigger of a hurdle than previously | 17:08 |
| AJaeger | fungi, retire both odsreg and puppet-odsreg - or only puppet-odsreg? | 17:08 |
| bharath_t | fungi, I see, so I currently I uploaded using twine upload dist/* and added openstackci as colloborator | 17:08 |
| bharath_t | fungi, :) that's true | 17:08 |
| fungi | AJaeger: i think the odsreg codebase can be retired too. checking now but seems ttx stopped modifying it roughly a year ago | 17:10 |
| fungi | ttx: is the openstack-infra/odsreg repo still needed for anything or can it be retired? | 17:10 |
| *** ijw has quit IRC | 17:13 | |
| *** Adri2000 has quit IRC | 17:16 | |
| *** jpena is now known as jpena|off | 17:16 | |
| *** anteaya has quit IRC | 17:17 | |
| *** Adri2000 has joined #openstack-infra | 17:19 | |
| openstackgerrit | Jeremy Stanley proposed openstack-infra/zuul-website master: Add a get-started redirect https://review.openstack.org/599699 | 17:19 |
| AJaeger | ttx, fungi, I'll start with retiring both - and WIP first change waiting for ttx to +1. | 17:21 |
| *** openstackgerrit has quit IRC | 17:22 | |
| fungi | thanks AJaeger! | 17:22 |
| *** openstackgerrit has joined #openstack-infra | 17:23 | |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/project-config master: Retire odsreg and puppet-odsreg (step 2) https://review.openstack.org/599700 | 17:23 |
| openstackgerrit | Merged openstack-infra/zuul-website master: Add a promotional message banner and events list https://review.openstack.org/591870 | 17:25 |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/odsreg master: Retire odsreg repository https://review.openstack.org/599702 | 17:28 |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/puppet-odsreg master: Retire puppet-odsreg https://review.openstack.org/599706 | 17:30 |
| *** Bhujay has quit IRC | 17:31 | |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/openstack-zuul-jobs master: Remove puppet-odsreg from required-jobs https://review.openstack.org/599708 | 17:32 |
| openstackgerrit | bharaththiruveedula proposed openstack-infra/project-config master: Add new project Gyan https://review.openstack.org/599709 | 17:36 |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/project-config master: Retire odsreg and puppet-odsreg (step 4) https://review.openstack.org/599710 | 17:36 |
| AJaeger | fungi, pushed all changes with topic https://review.openstack.org/#/q/status:open+++topic:retire-odsreg | 17:37 |
| fungi | thanks again AJaeger! | 17:37 |
| AJaeger | clarkb: hope as PTL you agree with fungi - and my changes above ^ | 17:38 |
| AJaeger | config-core, small cleanup change to remove unused channel - please review https://review.openstack.org/598477 | 17:39 |
| AJaeger | config-core, and one change for python3-first: https://review.openstack.org/598929 cloudkitty is ready | 17:40 |
| *** e0ne has quit IRC | 17:44 | |
| openstackgerrit | Merged openstack-infra/openstackid master: Fixed logout without current user but with valid id token https://review.openstack.org/599692 | 17:45 |
| openstackgerrit | Jeremy Stanley proposed openstack-infra/zuul-website master: Add a get-started redirect https://review.openstack.org/599699 | 17:46 |
| *** diablo_rojo has joined #openstack-infra | 17:47 | |
| openstackgerrit | Merged openstack-dev/cookiecutter master: Clean up .gitignore references to personal tools https://review.openstack.org/592520 | 17:48 |
| clarkb | AJaeger: looking now | 17:50 |
| AJaeger | clarkb: we're waiting for ttx for confirmation that both are unused | 17:52 |
| AJaeger | fungi, clarkb, for bindep-profiles, I pushed a change to the manual - please put this on your review list: https://review.openstack.org/599694 | 17:53 |
| clarkb | AJaeger: ya I'm approving the changes but your WIP will prevent any of them from merging until you are ready | 17:54 |
| clarkb | AJaeger: you should be able to recheck when ttx confirms and get things moving on that | 17:54 |
| *** alexchadin has joined #openstack-infra | 17:56 | |
| AJaeger | clarkb: will do - thanks | 17:57 |
| *** electrofelix has quit IRC | 17:58 | |
| AJaeger | clarkb, fungi, governance change is https://review.openstack.org/599712 | 17:58 |
| clarkb | AJaeger: the doc draft lgtm | 17:59 |
| AJaeger | clarkb: want to +2A https://review.openstack.org/#/c/599708/ as well - that has no depends-on, we can move forward with it directly IMHO | 18:00 |
| clarkb | AJaeger: done | 18:00 |
| AJaeger | fungi, I like your idea with a proper "ensure" role on the doc draft. | 18:00 |
| *** alexchadin has quit IRC | 18:00 | |
| AJaeger | thanks, clarkb | 18:00 |
| clarkb | fungi: AJaeger: I could see a py3 and py2 env in bindep.txt that installed eg python and python3 on ubuntu. Then if you are on xenial you get 3.5 and on bionic 3.6 | 18:02 |
| fungi | yeah, if we tackle it from the job end, that seems like it should also be fine. i mean, the job already is intended to run some tests under a particular python version so having the job make sure that version of python is installed and available isn't a stretch | 18:02 |
| clarkb | then if tox -epy35 doesn't run because interpreter is missing it should error? let me test that | 18:02 |
| *** trown|lunch is now known as trown | 18:03 | |
| clarkb | heh zuul sets basepython to python3 so tox -epy35 does work here even though I only have 3.6 installed | 18:03 |
| clarkb | if I remove that basepython line in tox.ini then tox -epy35 does return an error becuase the interpreter is not found which should bubble up as a job error | 18:04 |
| pabelanger | infra-root: https://review.openstack.org/597297/ enables mitogen on bridge.o.o, if we are willing to give it a try. I'm mostly interested in if it will result in the performance improvements it claims. | 18:06 |
| pabelanger | possible we could also manually try to kick.sh a few servers and compare too | 18:06 |
| *** r-daneel has quit IRC | 18:06 | |
| clarkb | pabelanger: we do have the timing data logged now so we should be able to just compare that | 18:06 |
| dmsimard | interested in a benchmark if we have the data yes | 18:07 |
| *** r-daneel has joined #openstack-infra | 18:07 | |
| pabelanger | clarkb: ya, we are now profiling tasks in ansible logs | 18:07 |
| clarkb | pabelanger: does the mitogen_linear strategy conflict with anywhere we've set the strategy to free? | 18:07 |
| dmsimard | pabelanger: ara has task-level granular duration data fwiw | 18:07 |
| pabelanger | clarkb: I think it does, but so far we are not using free. well when I last checked | 18:07 |
| pabelanger | dmsimard: yah, we don't have ara on bridge.o.o yet, but think we could / should? ansible it and self host on bridge.o.o | 18:08 |
| clarkb | pabelanger: you or monty added it to https://review.openstack.org/#/c/593029/23/playbooks/base.yaml | 18:08 |
| pabelanger | clarkb: yah, we'd likely have to remove that or replace with mitogen_free | 18:09 |
| dmsimard | pabelanger: fwiw there is already a WIP ansible role for ara here: https://github.com/openstack/ansible-role-ara | 18:09 |
| clarkb | pabelanger: will it work as is or will it fail? (mostly I'm fine with it not using mitogen there as long as it won't break) | 18:09 |
| dmsimard | there's different topologies available (with nginx or mod_wsgi) | 18:09 |
| dmsimard | https://github.com/openstack/ansible-role-ara/tree/master/example-playbooks | 18:09 |
| pabelanger | clarkb: no, free should work. Just won't use mitogen there | 18:10 |
| clarkb | gotcha that should be fine then | 18:10 |
| openstackgerrit | Merged openstack-infra/openstack-zuul-jobs master: Remove puppet-odsreg from required-jobs https://review.openstack.org/599708 | 18:10 |
| pabelanger | I'm going to grab some food, but will be back shortly to help watch | 18:10 |
| *** jamesmcarthur has quit IRC | 18:12 | |
| AJaeger | clarkb: want to merge your change https://review.openstack.org/599065 - ianw and myself proposed to check that it works (babysit), so didn't approve ourselves | 18:12 |
| AJaeger | fungi, regarding the bindep profile - will you propose a change for discussion? | 18:14 |
| clarkb | pabelanger: https://mitogen.readthedocs.io/en/stable/shame.html#pbr that makes me wary (1. not heard of what the issue is and 2. setting pbr version like that may not be safe for our using of openstacky things) | 18:14 |
| *** ijw has joined #openstack-infra | 18:14 | |
| clarkb | I also don't understand why pbr reading things from disk presents a problem to independent softawre | 18:15 |
| clarkb | ah I see, pbr gracefully falls back if the metadata isn't there, but it does so by attempting to look at git which likely won't be there either which results in an error in pbr | 18:17 |
| clarkb | I think pbr does that intentionally to avoid pkg_resources slowness | 18:18 |
| clarkb | so we could fix it but then our use of mitogen may even be slower than it is without mitogen bceause we spend a second looking up package versions on every invocation | 18:19 |
| fungi | and note the second entry on their wall-o-shame is "anything which imports pkg_resources" | 18:19 |
| Shrews | hrm, i think reno 2.10.0 may be broken | 18:20 |
| Shrews | dhellmann: ^^ | 18:20 |
| AJaeger | Shrews: what's the problem? | 18:20 |
| Shrews | http://logs.openstack.org/33/590233/10/check/tox-docs/74bdac3/job-output.txt.gz#_2018-09-04_17_50_29_726314 | 18:20 |
| *** dtantsur is now known as dtantsur|afk | 18:21 | |
| Shrews | looks like it released 2 hours ago or so | 18:21 |
| AJaeger | mmh, problem with Sphinx 1.5.6? dhellmann ^ | 18:22 |
| clarkb | pabelanger: in any case I'm willing to try it, our use of shade/openstacksdk may have a sad though | 18:22 |
| *** alexchadin has joined #openstack-infra | 18:23 | |
| openstackgerrit | bharaththiruveedula proposed openstack-infra/project-config master: Add new project Gyan https://review.openstack.org/599709 | 18:24 |
| fungi | AJaeger: yeah, i'll see what i can come up with for a version-specific python interpreter installer role. i think there's some sort of package management in the ansible stdlib we can leverage, but not sure (i'm still not extremely familiar with ansible) | 18:24 |
| *** alexchadin has quit IRC | 18:27 | |
| fungi | pabelanger: clarkb: interesting. looks like mitogen lacks python 3.5 support (their python 3 support seems to start with 3.6) | 18:30 |
| fungi | i guess the assumption is that any systems which don't have a working python3.6 or later will have working python 2 | 18:30 |
| clarkb | fungi: that should be ok since bridge.o.o is bionic with python3.6 and it doesn't have requirements on the remote system | 18:30 |
| clarkb | I think | 18:31 |
| clarkb | (this is why pbr runs afoul of it, pbr is trying to do stuff ont he local system and mitogen is trying to avoid taht) | 18:31 |
| fungi | mitogen says you need an ssh connection and working remote python interpreter | 18:31 |
| clarkb | ah | 18:31 |
| clarkb | well in that case trusty and centos are python2 by default and I think my change to install python2 on xenial merged so we should still be good | 18:31 |
| fungi | yeah, xenial was the only place i was worried about it | 18:32 |
| *** jamesmcarthur has joined #openstack-infra | 18:32 | |
| clarkb | the next thing is whether or not it will use ansible_python on the remote though | 18:32 |
| clarkb | beacuse we do set that to python3 on some xenial nodes I think (or at least I proposed those changes) | 18:32 |
| clarkb | https://review.openstack.org/#/c/596911/ has not merged so we may be ok there as well | 18:33 |
| fungi | there was something in the latest mitogen release notes about ansible_python (i think) | 18:34 |
| fungi | ahh, no, it was the ansible_*_interpreter variables | 18:35 |
| fungi | e.g., ansible_python_interpreter | 18:35 |
| clarkb | that might be the var I'm thinking of | 18:36 |
| clarkb | ya it is https://review.openstack.org/#/c/596911/1/playbooks/group_vars/mirror.yaml | 18:36 |
| *** jamesmcarthur has quit IRC | 18:36 | |
| *** eharney has quit IRC | 18:36 | |
| dhellmann | Shrews , AJaeger : sigh | 18:38 |
| dhellmann | I'll take a look, thanks for the ping | 18:38 |
| openstackgerrit | bharaththiruveedula proposed openstack-infra/project-config master: Add new project Gyan https://review.openstack.org/599709 | 18:41 |
| dhellmann | Shrews : I suspect that's an issue with an old sphinx, as AJaeger points out. I think the solution is going to be to raise the minimum sphinx used in reno, which is going to break that job further because nodepool has sphinx capped. | 18:43 |
| clarkb | fungi: reading more Isuspect that we can't use mitogen + python3.5 on xenial | 18:45 |
| clarkb | fungi: which may be a worthwhile tradeoff if the performance improvement is good | 18:45 |
| *** gfidente is now known as gfidente|afk | 18:45 | |
| Shrews | dhellmann: i'm testing removing the sphinx cap now | 18:45 |
| openstackgerrit | Doug Hellmann proposed openstack-infra/nodepool master: uncap sphinx https://review.openstack.org/599742 | 18:45 |
| dhellmann | Shrews : ^^ works for me locally | 18:45 |
| Shrews | dhellmann: me too | 18:45 |
| dhellmann | I'll go update reno's minimum dependency, too, to be safe | 18:46 |
| AJaeger | dhellmann, Shrews let me update Zuul as well... | 18:47 |
| Shrews | dhellmann: we should probably either remove the minimum for sphinx, or set it to >1.5.6 | 18:47 |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/zuul master: Uncap sphinx https://review.openstack.org/599743 | 18:48 |
| AJaeger | Shrews, I set it to > 1.5.6 in Zuul ^ | 18:48 |
| dhellmann | is 1.5.6 the version that has the logger? | 18:50 |
| Shrews | AJaeger: you have >= | 18:50 |
| AJaeger | Shrews: ooh - will fix... | 18:51 |
| Shrews | dhellmann: good question | 18:51 |
| Shrews | we should probably find that version first | 18:51 |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/zuul master: Uncap sphinx https://review.openstack.org/599743 | 18:52 |
| dhellmann | I was just going to pick a relatively new version, but I could do a little bit of testing | 18:52 |
| Shrews | dhellmann: AJaeger: 1.6.1 is the first version that seems to work | 18:54 |
| Shrews | there was nothing between 1.5.6 and 1.6.1 | 18:54 |
| dhellmann | the reno unit tests don't fail with 1.5.1 so something isn't triggering the failure there | 18:54 |
| dhellmann | we have 1.6.1 excluded for some reason (probably cargo-culting) | 18:56 |
| AJaeger | dhellmann: do you need to update minimal requirement in requirements repo in this case? | 18:56 |
| dhellmann | AJaeger : reno doesn't claim to be co-installable so I don't think so | 18:57 |
| pabelanger | clarkb: ah, didn't see wall o shame | 18:58 |
| dhellmann | oh, test coverage for the sphinx extension is 0 | 18:58 |
| dhellmann | no wonder nothing tripped this | 18:58 |
| clarkb | Infra meeting in a couple minutes over in #openstack-infra. Feel free to join us, biggest topic of discussion likely to be the PTG (even if you aren't attending your input is still valuable) | 18:59 |
| Shrews | #openstack-meeting, even :) | 18:59 |
| clarkb | er ya that | 18:59 |
| *** bhavikdbavishi has quit IRC | 19:07 | |
| dhellmann | Shrews , AJaeger : the series ending with https://review.openstack.org/599750 adds the tests and fixes the minimum setting in reno | 19:09 |
| dhellmann | oh, zuul doesn't like that | 19:10 |
| *** olivierb has quit IRC | 19:10 | |
| *** jamesmcarthur has joined #openstack-infra | 19:11 | |
| dhellmann | names are hard | 19:11 |
| *** jamesmcarthur has quit IRC | 19:15 | |
| *** jamesmcarthur_ has joined #openstack-infra | 19:15 | |
| openstackgerrit | Jeremy Stanley proposed openstack-infra/zuul-website master: Add a get-started redirect https://review.openstack.org/599699 | 19:17 |
| openstackgerrit | Doug Hellmann proposed openstack-infra/nodepool master: uncap sphinx https://review.openstack.org/599742 | 19:19 |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/zuul master: Uncap sphinx https://review.openstack.org/599743 | 19:21 |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/bindep master: Update docs building https://review.openstack.org/599755 | 19:22 |
| openstackgerrit | Paul Belanger proposed openstack-infra/system-config master: Use package task over apt for base-server https://review.openstack.org/599756 | 19:23 |
| openstackgerrit | Mark Goddard proposed openstack-infra/project-config master: Add a group for kayobe projects https://review.openstack.org/599757 | 19:25 |
| *** e0ne has joined #openstack-infra | 19:27 | |
| openstackgerrit | Mark Goddard proposed openstack-infra/project-config master: Add new project config for kayobe-config-dev https://review.openstack.org/599760 | 19:28 |
| *** jamesmcarthur_ has quit IRC | 19:33 | |
| *** e0ne has quit IRC | 19:33 | |
| openstackgerrit | Mark Goddard proposed openstack-infra/project-config master: Add new project config for kayobe-config-dev https://review.openstack.org/599760 | 19:38 |
| *** eernst has joined #openstack-infra | 19:41 | |
| *** eernst has quit IRC | 19:47 | |
| *** e0ne has joined #openstack-infra | 19:52 | |
| *** bharath_t has quit IRC | 19:59 | |
| clarkb | to pick up the conversation about running ansible as root instead of making a new group with write perms to the various things on bridge.openstack.org I'm good with that we should just update our launch node docs and skim the system-config docs tree for any use of ansible that requires sudo now | 20:03 |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/openstack-zuul-jobs master: Remove puppet-docker_registry https://review.openstack.org/599824 | 20:04 |
| clarkb | ianw: I've approved the rtd failing changes | 20:04 |
| AJaeger | config-core, sorry, missed the following when retiring - please approve to unblock ozj ^ | 20:04 |
| AJaeger | clarkb: and it failed - 599824 will fix | 20:04 |
| clarkb | AJaeger: looks like 599824 also failed | 20:04 |
| clarkb | AJaeger: do you need to rebase? | 20:05 |
| AJaeger | clarkb: on it... | 20:05 |
| clarkb | corvus: ^ interesting zuul error reporting there, the base job failed to parse so it left comments on all the children jobs saying that the base job does not exist | 20:05 |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/openstack-zuul-jobs master: Remove retireted repos https://review.openstack.org/599824 | 20:06 |
| AJaeger | clarkb: hope that's all ^ | 20:07 |
| pabelanger | looks like limestone is having some issues launching nodes: http://grafana.openstack.org/dashboard/db/nodepool-limestone | 20:07 |
| pabelanger | checking nodepool | 20:07 |
| corvus | clarkb: that's more or less by design. zuul reports the first error it encounters in the cover message, but then reports all of the errors as inline comments. that way the cover message isn't too spammy, but if there's more than one error, we get to see it in the inline comments. | 20:08 |
| corvus | clarkb: we might be able to make it smart enough to know that an error was caused by another error, and omit those. | 20:08 |
| pabelanger | logan-: clarkb: I don't see anything specific in nl02.o.o logs about limestone, just error creating server | 20:09 |
| clarkb | AJaeger: looks like it just found the next missing repo in that list | 20:09 |
| ianw | clarkb / fungi : the whole thing with setgid on the cache/inventory directories was a little hand-wavy. i'm unsure how actively running ansible as root from cron interacts with sporadic unprivileged use | 20:09 |
| clarkb | AJaeger: probably needt o check the whole list against what we removed | 20:09 |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/openstack-zuul-jobs master: Remove retireted repos https://review.openstack.org/599824 | 20:09 |
| AJaeger | clarkb: yeah ;( | 20:09 |
| AJaeger | clarkb: I thought I did - pushed the one line and double check now... | 20:10 |
| pabelanger | we also seem to be hitting quota issue in packethost | 20:10 |
| pabelanger | openstack.exceptions.HttpException: HttpException: 403: Client Error for url: https://packet.platform9.net/nova/v2.1/ba163363643541d88a737f8af23cfa7f/servers, {"forbidden": {"message": "Quota exceeded for cores, instances: Requested 8, 1, but already used 800, 100 of 800, 100 cores, instances", "code": 403}} | 20:10 |
| fungi | ianw: you mean as far as one or the other holding an open write lock on the same file? | 20:11 |
| ianw | fungi: i mean as far as what files it decides to open/close/add/remove | 20:12 |
| clarkb | ianw: I think the only files we have to worry about are the log files as they may interleave? | 20:12 |
| clarkb | the caches are atomically written and read at the start of the run iirc | 20:13 |
| AJaeger | clarkb: https://review.openstack.org/599824 should be fine now... | 20:14 |
| clarkb | as a heads up to the channel I have approved https://review.openstack.org/#/c/599065/ which willremove the +rf mode on this channel. I plan to babysit it and make sure the update goes in properly as well as being ready to revert if the spam comes back | 20:14 |
| *** alexchadin has joined #openstack-infra | 20:16 | |
| *** ijw has quit IRC | 20:18 | |
| corvus | clarkb: i assume the timeout in 598329 is not change-related. i've +2d that series; i think we can +3 when that comes back (worth waiting just in case it happens again since we probably want all 4 changes to land ~together) | 20:20 |
| clarkb | corvus: ya, I looked at the subsequent changes and they ran that job in ~16 minutes | 20:20 |
| clarkb | I'm going to blame noisy neighbors for now | 20:20 |
| *** jamesmcarthur has joined #openstack-infra | 20:21 | |
| openstackgerrit | Merged openstack-infra/project-config master: Disable +r on #openstack-infra https://review.openstack.org/599065 | 20:23 |
| clarkb | I'm going to grab food while I wait for that to go in | 20:24 |
| *** holser_ has joined #openstack-infra | 20:26 | |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul-jobs master: add-build-sshkey: Remove only the master key https://review.openstack.org/599073 | 20:27 |
| *** kgiusti has quit IRC | 20:35 | |
| *** r-daneel_ has joined #openstack-infra | 20:38 | |
| *** r-daneel has quit IRC | 20:38 | |
| *** r-daneel_ is now known as r-daneel | 20:38 | |
| AJaeger | smcginnis, dhellmann , just commented on https://review.openstack.org/#/c/597289 - seems heat imported too many jobs . I don't see ricolin here. smcginnis if this cases release failures, can you reach out, please? | 20:39 |
| AJaeger | smcginnis: I'm signing off for today now... | 20:39 |
| ianw | corvus: has the upload-swift-logs work paused? | 20:40 |
| smcginnis | Thanks AJaeger | 20:40 |
| *** ijw has joined #openstack-infra | 20:41 | |
| openstackgerrit | Doug Hellmann proposed openstack-infra/project-config master: remove job settings for heat repositories https://review.openstack.org/597289 | 20:48 |
| *** ijw has quit IRC | 20:48 | |
| *** holser_ has quit IRC | 20:49 | |
| corvus | ianw: yes; i believe all pre-reqs are in place and we could start using it now, but i've shifted focus to the config-mgmt work. i didn't want to make major changes around release and wanted to help get cfg-mgmt ready for ptg. | 20:49 |
| *** ansmith has quit IRC | 20:49 | |
| *** ijw has joined #openstack-infra | 20:50 | |
| corvus | ianw: you may be interested in https://review.openstack.org/598406 | 20:50 |
| *** alexchadin has quit IRC | 20:51 | |
| *** eharney has joined #openstack-infra | 20:51 | |
| ianw | corvus: ok on the upping of the nodes; auristor pointed out that kafs will make it into the next fedora release. we may have a legitimate reason for control-plane fedora at that point, at least as a POC of that | 20:52 |
| corvus | ianw: that is awesome news :) | 20:53 |
| ianw | corvus: re the swift logs, if you could loop back on the stack from https://review.openstack.org/#/c/592853/ ; at the very least it leaves files everywhere as is. but i still think a download script is useful, and the other review is on hold waiting for swift logs, so that's why i attempted an implementation there too | 20:54 |
| openstackgerrit | Clark Boylan proposed openstack-infra/system-config master: Run cloud launcher on bridge.o.o https://review.openstack.org/597728 | 20:56 |
| clarkb | addressed comment on ^ and set up its depends on on aparent properly so that it can merge once those merge | 20:56 |
| *** ChanServ sets mode: -rf | 20:57 | |
| *** vaidy has joined #openstack-infra | 20:57 | |
| clarkb | infra-root does channel mode look correct to you now? It does to me | 20:57 |
| *** vipul has joined #openstack-infra | 20:57 | |
| corvus | yep | 20:57 |
| *** wolverineav has joined #openstack-infra | 20:58 | |
| *** caphrim007 has joined #openstack-infra | 20:58 | |
| *** wolverineav has quit IRC | 20:58 | |
| ianw | clarkb: LGTM ... fingers crossed. linaro-unregistered i noticed keeps getting a bunch of new crap ... one new feature seems to be lots of lookalike unicode characters obviously meant to trip up string matchers | 20:59 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: Add private key storage migration https://review.openstack.org/592213 | 20:59 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: Change project.private_key to private_secrets_key https://review.openstack.org/597302 | 20:59 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: Create per-project ssh keys https://review.openstack.org/597711 | 20:59 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: Provide per-project ssh key to executor https://review.openstack.org/597727 | 20:59 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: Serve project SSH keys and document https://review.openstack.org/599063 | 20:59 |
| *** wolverineav has joined #openstack-infra | 20:59 | |
| *** felipemonteiro_ has joined #openstack-infra | 20:59 | |
| *** felipemonteiro_ is now known as felipemonteiro | 20:59 | |
| *** felipemonteiro has quit IRC | 21:00 | |
| *** felipemonteiro has joined #openstack-infra | 21:00 | |
| *** bobh_ has quit IRC | 21:00 | |
| *** wolverineav has quit IRC | 21:01 | |
| *** wolverineav has joined #openstack-infra | 21:02 | |
| *** bobh has joined #openstack-infra | 21:02 | |
| *** imacdonn has quit IRC | 21:02 | |
| clarkb | corvus: https://review.openstack.org/#/c/597333/18 is a bridge.o.o quality of lfie change to rotate log files | 21:02 |
| clarkb | corvus: coudl you take a look? | 21:02 |
| clarkb | (our log files on that node are not small) | 21:02 |
| *** imacdonn has joined #openstack-infra | 21:02 | |
| *** caphrim007 has quit IRC | 21:03 | |
| *** trown is now known as trown|outtypewww | 21:03 | |
| corvus | ianw: ^ i thought we weren't using roles-test for playbooks/roles ? | 21:05 |
| *** bobh has quit IRC | 21:07 | |
| ianw | the logrotate role is in roles/* ? | 21:07 |
| ianw | although i can't see much use for non-long-lived servers, it didn't seem to be particularly specific | 21:07 |
| corvus | ianw: oh, sorry i saw the ansible-cron changes... | 21:07 |
| corvus | ianw: yeah, i'd prefer to err on the side of playbook/roles, and not double-test things | 21:08 |
| openstackgerrit | Merged openstack-infra/project-config master: Zuul: increase max-nodes-per-job https://review.openstack.org/598406 | 21:10 |
| pabelanger | corvus: clarkb: ianw: noticed another optimization for playbooks on bridge.o.o: https://review.openstack.org/599756 should result in 1 less task / network call to all our servers. hopefully saving some time | 21:10 |
| pabelanger | also uses generic package task for debian | 21:10 |
| corvus | pabelanger: set topic and +2 | 21:11 |
| clarkb | oh I +2'd as well | 21:11 |
| clarkb | I'll change that to a +3 | 21:11 |
| pabelanger | yay, I'll look more tonight at ansible logs, see where we can start to shave off some time | 21:11 |
| pabelanger | that was just something I quickly noticed | 21:12 |
| clarkb | ignoring potential speed ups that pattern is also easier to read imo | 21:12 |
| clarkb | so I like it | 21:12 |
| pabelanger | but agree, more forks will likely help too | 21:12 |
| corvus | i'm about to push up a forks change | 21:13 |
| pabelanger | going to grab some food with family | 21:13 |
| *** e0ne has quit IRC | 21:13 | |
| openstackgerrit | Merged openstack-infra/openstack-zuul-jobs master: Remove retireted repos https://review.openstack.org/599824 | 21:13 |
| clarkb | ianw: AJaeger I'm rechecking the rtd change now ^ | 21:13 |
| mgagne | We (inap-mtl01) would need to perform a maintenance tomorrow (time tbd) which would affect VM connectivity for (what we hope) ~10-15m. Would there be a way to disable inap-mtl01 in advance to drain the compute nodes beforehand? | 21:15 |
| *** wolverineav has quit IRC | 21:15 | |
| clarkb | mgagne: yup if you set max-servers to 0 we'll drain them down and stop booting new servers until you revert. Let me get a link to the file to update | 21:15 |
| clarkb | mgagne: http://git.openstack.org/cgit/openstack-infra/project-config/tree/nodepool/nl03.openstack.org.yaml#n65 do you want us to make that change or would you like to push it? | 21:16 |
| mgagne | I can do it =) | 21:16 |
| clarkb | thanks! | 21:16 |
| openstackgerrit | James E. Blair proposed openstack-infra/system-config master: Set Ansible forks to 50 https://review.openstack.org/599850 | 21:17 |
| fungi | that's an exciting number | 21:17 |
| mgagne | clarkb: how much time between merge and deployment? is it automated? | 21:18 |
| ianw | corvus: i mean obviously (since that's the way i wrote it) I've been thinking that if a role doesn't need anything special or do anything particularly infra-specific it is better homed in roles/*. given it seems unlikely to get a lot of development effort, it seems it won't trigger the zuul-based tests much? | 21:18 |
| clarkb | mgagne: it is automated, can be less than an hour or about an hour depending on when the chagne goes in relative to our cron job running things | 21:19 |
| clarkb | mgagne: you probably want to get the change merged today to be on the safe side | 21:19 |
| mgagne | ok, will look into the schedule on our side first | 21:19 |
| corvus | ianw: i think the opposite -- that things should be in playbook/roles unless we really intend to export them. and we should focus only on testinfra, and not individual role tests. i really really want to make this system easy for us to run. | 21:20 |
| clarkb | pabelanger: you have a vi process using 25% of the memory on bridge.o.o fwiw (double checked memory overhead with corvus' forks change above) | 21:20 |
| corvus | i don't want to have to write two kinds of tests for every role that we do | 21:21 |
| corvus | adminning our systems should be low-cost. | 21:21 |
| clarkb | looks like its opened our very large log file we are talking about rotating currently | 21:21 |
| openstackgerrit | Mathieu Gagné proposed openstack-infra/project-config master: Disable inap-mtl01 provider https://review.openstack.org/599852 | 21:23 |
| mgagne | WIP until we figure out the actual schedule and other technical details. | 21:24 |
| clarkb | I've +2'd it because the chagne is structurally sound. Just let us know what your schedule is when you know it | 21:25 |
| clarkb | ianw: re freenode spam, how often were you seeing it in the linaro channel? I'm wondering how long of a window we should have observing befor proposing wider removal of +r | 21:28 |
| pabelanger | clarkb: sorry, kille | 21:30 |
| pabelanger | killed* | 21:30 |
| *** bobh has joined #openstack-infra | 21:36 | |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard master: Support getting project groups by name https://review.openstack.org/590047 | 21:36 |
| mgagne | clarkb: ok, we will proceed with the maintenance tomorrow. I think it's fine to merge it now. | 21:36 |
| ianw | clarkb: http://paste.openstack.org/show/729453/ ... a rough grep of spammy messages for last few days | 21:37 |
| ianw | nothing like the recent levels | 21:37 |
| clarkb | ianw: based on that waiting a day should be platny | 21:38 |
| clarkb | er | 21:38 |
| clarkb | plenty | 21:38 |
| ianw | yep | 21:38 |
| *** ijw has quit IRC | 21:38 | |
| fungi | mgagne: thanks for confirming, approved now | 21:39 |
| mgagne | thanks! | 21:39 |
| openstackgerrit | Merged openstack-infra/system-config master: Use package task over apt for base-server https://review.openstack.org/599756 | 21:40 |
| *** ijw has joined #openstack-infra | 21:40 | |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard master: Support getting project groups by name https://review.openstack.org/590047 | 21:41 |
| fungi | pabelanger: i see the package task in 599756, do you happen to know if they maintain a cross-platform mapping of package names or anything? trying to come up with a sane mechanism for making sure a certain package is installed in a zuul-jobs role, but making it cross-platform (as there may be package naming differences) | 21:43 |
| clarkb | fungi: they don't | 21:43 |
| *** wolverineav has joined #openstack-infra | 21:43 | |
| fungi | :( | 21:44 |
| clarkb | fungi: I want to say bindep has actually been floated around as a possible tool to integrate to facilitate that | 21:44 |
| pabelanger | yah, user needs to define the package name | 21:44 |
| fungi | okay, so you add a platform conditional in the role i guess? | 21:44 |
| clarkb | fungi: or you do include vars with a hierarchy of release, distro, family vars files that specify the packages to install | 21:45 |
| * clarkb finds an example | 21:45 | |
| pabelanger | yup, check ansible_distribution then include_vars task | 21:45 |
| pabelanger | there is also with_first_found (I think) that also work | 21:46 |
| pabelanger | works* | 21:46 |
| clarkb | fungi: https://git.openstack.org/cgit/openstack-infra/system-config/tree/playbooks/roles/base-server/vars | 21:46 |
| clarkb | fungi: paired with https://git.openstack.org/cgit/openstack-infra/system-config/tree/playbooks/roles/base-server/tasks/main.yaml#n6 | 21:46 |
| fungi | ahh, neat. thanks | 21:46 |
| pabelanger | clarkb: fungi: also, https://github.com/ansible/ansible/pull/22159 adds a bindep task, into ansible. | 21:47 |
| pabelanger | but we need to rebsae I think | 21:47 |
| pabelanger | maybe something to prep for ansiblefest next month | 21:47 |
| pabelanger | but yah, I've longed for using something like bindep in ansible, just for multi-platform package names | 21:48 |
| ianw | corvus: just seems a shame that for very low cost (few lines of test code) we can provide interesting building-blocks but choose not to. while i don't exactly have a slam dunk use-case for the generic role, we're not inviting anyone to build interesting things with them. anyway, nobody else seems to have reviewed, so i i'll move it in the interests of getting things going | 21:51 |
| openstackgerrit | Ian Wienand proposed openstack-infra/system-config master: Add logrotate role and rotate ansible log files https://review.openstack.org/597333 | 21:55 |
| corvus | ianw: that sounds good, but my view is that every extra thing we do is a barrier to actually maintaining our systems, which have basically frozen in place due to the complexity they've accumulated over the years. i'm trying to stick to the spec here which just describes one way of testing our systems. | 21:56 |
| *** hasharAway has quit IRC | 21:56 | |
| corvus | the only way we can avoid ending up back in the same place is to agressively focus on simplicity | 21:57 |
| corvus | yes, i wrote the exim role completely generically and it should be usable by anyone. maybe some day we'll decide to publish it to galaxy or something. but i have no interest in managing anyone's systems but our own right now. | 21:58 |
| *** ijw has quit IRC | 22:02 | |
| *** ijw has joined #openstack-infra | 22:02 | |
| *** r-daneel has quit IRC | 22:08 | |
| clarkb | corvus: ^ latest patchset above did make your changes fwiw, would be good to get that in so pabelanger can read the log file without using 25% of system memory :) | 22:10 |
| *** bobh has quit IRC | 22:10 | |
| *** felipemonteiro_ has joined #openstack-infra | 22:10 | |
| corvus | +3 | 22:11 |
| *** felipemonteiro_ has quit IRC | 22:12 | |
| *** felipemonteiro has quit IRC | 22:15 | |
| *** boden has quit IRC | 22:15 | |
| *** priteau has quit IRC | 22:21 | |
| *** ianychoi has quit IRC | 22:22 | |
| clarkb | corvus: https://review.openstack.org/#/c/598329/ is +1 from zuul now. I've got to take kids to a first day of school thing in about 45 minutes so not in a great spot to watch that this evening | 22:29 |
| ianw | corvus: if you get a chance too, a look at https://review.openstack.org/599469 and a zuul-sphinx release to get the readme.rst detection would be good too and i can revert the special system-config checks | 22:29 |
| clarkb | probably the biggest thing is ensuring that the clouds.yaml is as we want because the next ansible run will use it to make an inventory | 22:29 |
| corvus | clarkb: yeah, i'll be leaving by the next ansible cycle too, so unless ianw wants to babysit the clouds.yaml thing, maybe we merge first thing tomorrow? | 22:34 |
| ianw | i can watch it, seeing as i probably want to add a cloud sometime today :) | 22:35 |
| corvus | ianw: cool, it's a stack of 4 changes; they should all be ready for +W | 22:36 |
| openstackgerrit | sebastian marcet proposed openstack-infra/openstackid-resources master: Fixed typo on PresentationCategorySerializer https://review.openstack.org/599863 | 22:37 |
| corvus | zuul-sphinx change looks cool +3 | 22:37 |
| openstackgerrit | Merged openstack-infra/openstackid-resources master: Fixed typo on PresentationCategorySerializer https://review.openstack.org/599863 | 22:38 |
| clarkb | I'll be around first thing tomorrow if ianw doesn't end up doing it | 22:38 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: Cap cherrypy and uncap sphinx https://review.openstack.org/599502 | 22:41 |
| ianw | i'll start with 593029 and make sure that rolls out ok to builder/launchers etc | 22:41 |
| ianw | i guess if it goes wrong and we drop authentication to every cloud we'll realise pretty quickly ;) | 22:41 |
| clarkb | ianw: ya ansible should stop running and nodepool will start erroring. I don't think anything will break beyond that we'll just stop doing new work | 22:42 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: Add private key storage migration https://review.openstack.org/592213 | 22:42 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: Change project.private_key to private_secrets_key https://review.openstack.org/597302 | 22:42 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: Create per-project ssh keys https://review.openstack.org/597711 | 22:42 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: Provide per-project ssh key to executor https://review.openstack.org/597727 | 22:42 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: Serve project SSH keys and document https://review.openstack.org/599063 | 22:42 |
| clarkb | https://review.openstack.org/#/c/598378/ is a dependency too fwiw | 22:43 |
| clarkb | I should fix the topic (though it is listed as a depends on | 22:43 |
| corvus | clarkb: yeah, i've been using a dashboard, so i'm generally not seeing changes out-of-topic | 22:44 |
| corvus | (though i did see that one and +2 it) | 22:44 |
| *** rlandy is now known as rlandy|bbl | 22:45 | |
| openstackgerrit | Merged openstack-infra/zuul-sphinx master: Improve warning details https://review.openstack.org/599469 | 22:46 |
| *** spartakos has joined #openstack-infra | 22:48 | |
| *** jamesmcarthur has quit IRC | 22:50 | |
| *** jamesmcarthur has joined #openstack-infra | 22:51 | |
| *** shardy has quit IRC | 22:52 | |
| *** jamesmcarthur has quit IRC | 22:52 | |
| *** shardy has joined #openstack-infra | 22:52 | |
| *** rcernin has joined #openstack-infra | 22:55 | |
| *** wolverineav has quit IRC | 22:59 | |
| openstackgerrit | Merged openstack-infra/system-config master: Install limestone CA on hosts using openstacksdk https://review.openstack.org/593029 | 23:05 |
| *** ijw has quit IRC | 23:06 | |
| openstackgerrit | Merged openstack-infra/project-config master: Disable inap-mtl01 provider https://review.openstack.org/599852 | 23:07 |
| openstackgerrit | Merged openstack-infra/openstack-zuul-jobs master: Add warning job about readthedocs publishing https://review.openstack.org/599269 | 23:08 |
| *** ijw has joined #openstack-infra | 23:12 | |
| *** gfidente|afk has quit IRC | 23:12 | |
| openstackgerrit | Ian Wienand proposed openstack-infra/system-config master: Add logrotate role and rotate ansible log files https://review.openstack.org/597333 | 23:16 |
| *** ianychoi has joined #openstack-infra | 23:16 | |
| openstackgerrit | Ian Wienand proposed openstack-infra/system-config master: Run cloud launcher on bridge.o.o https://review.openstack.org/597728 | 23:17 |
| *** tpsilva has quit IRC | 23:18 | |
| *** priteau has joined #openstack-infra | 23:22 | |
| *** sthussey has quit IRC | 23:34 | |
| *** eernst has joined #openstack-infra | 23:34 | |
| *** jcoufal has quit IRC | 23:36 | |
| *** eernst has quit IRC | 23:38 | |
| ianw | ok next run should integrate 593029 ... i'll watch that before approving more | 23:38 |
| *** jamesmcarthur has joined #openstack-infra | 23:44 | |
| *** bobh has joined #openstack-infra | 23:44 | |
| openstackgerrit | Merged openstack-infra/zuul master: Cap cherrypy and uncap sphinx https://review.openstack.org/599502 | 23:47 |
| ianw | fungi: whatever the heck is happening with the new gpg and throwing up a dialog when i try to open the passwords file on bridge it's not working :/ | 23:47 |
| *** tosky has quit IRC | 23:48 | |
| *** jamesmcarthur has quit IRC | 23:48 | |
| fungi | ianw: are you getting a local dialog, or an ncurses one? | 23:48 |
| ianw | not sure, it threw something up that appeared to take over emacs ... i've not seen it before | 23:49 |
| ianw | and it only randomly accepts keypresses, it seems | 23:49 |
| ianw | complicated by me running it in a screen, i'm sure | 23:49 |
| *** bobh has quit IRC | 23:49 | |
| fungi | gnupg 2.x (which ubuntu kindly links at /usr/bin/gpg now instead of gpg2) uses a configurable/modular pinentry frontend to take passphrase input now | 23:50 |
| ianw | great, now i'm running it out of a screen, and it just does .. nothing | 23:50 |
| ianw | root 32437 17766 0 23:44 ? 00:00:00 pinentry | 23:51 |
| ianw | i'm guessing i'm supposed to interact with that somehow | 23:51 |
| fungi | you can force loopback pinentry mode (which will just cause gnupg to provide a non-echoing password prompt) by adding --pinentry-mode=loopback | 23:51 |
| fungi | supposedly not providing the passphrase directly to the utility is safer, but i have a hard time with their justification | 23:52 |
| ianw | fungi: i used to be able to just use emacs to open hte password file and put in the password on the emacs status bar bit ... | 23:52 |
| ianw | i had to ~. quit the ssh session, and the pinentry process is still hanging around! | 23:53 |
| ianw | desire to debug multiple layers of tty problems is <0 | 23:54 |
| fungi | the argument is based on the theory that users should only be prompted for passwords by a consistent frontend, and so your local gpg will spawn the gnome pinentry frontend which your wireless password input and your root permission terminal and whatever also use | 23:54 |
| fungi | in practice users will just enter secrets into anything that asks them to | 23:55 |
| fungi | also if you're doing agent forwarding, you'll supposedly get a local window prompting you (but i've never tried that) | 23:56 |
| ianw | i'd be doubtful that would survive "sudo -s ; su -" anyway | 23:57 |
| ianw | something is totally borked. "gpg --decrypt passwords.gpg" just does nothing at all for me now | 23:59 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!