| *** EricGonczer_ has quit IRC | 00:02 | |
| *** yamamoto has joined #openstack-infra | 00:08 | |
| *** jascott1 has joined #openstack-infra | 00:09 | |
| *** jascott1_ has joined #openstack-infra | 00:10 | |
| *** jascott1 has quit IRC | 00:10 | |
| openstackgerrit | Ian Wienand proposed openstack/diskimage-builder master: Move setfiles to outside chroot with runcon https://review.openstack.org/485899 | 00:14 |
|---|---|---|
| *** gmann has quit IRC | 00:14 | |
| *** gmann has joined #openstack-infra | 00:14 | |
| *** jascott1_ has quit IRC | 00:19 | |
| *** jascott1 has joined #openstack-infra | 00:19 | |
| *** jascott1 has quit IRC | 00:20 | |
| *** jascott1_ has joined #openstack-infra | 00:20 | |
| *** dingyichen has joined #openstack-infra | 00:23 | |
| *** lbragsta_ has quit IRC | 00:28 | |
| *** thorst has joined #openstack-infra | 00:31 | |
| *** aeng has quit IRC | 00:35 | |
| *** hichihara has joined #openstack-infra | 00:35 | |
| *** gildub has joined #openstack-infra | 00:35 | |
| *** thorst has quit IRC | 00:36 | |
| *** aeng has joined #openstack-infra | 00:52 | |
| *** EricGonczer_ has joined #openstack-infra | 00:58 | |
| *** gouthamr has joined #openstack-infra | 01:01 | |
| *** LindaWang has joined #openstack-infra | 01:02 | |
| *** mriedem has joined #openstack-infra | 01:03 | |
| *** thorst has joined #openstack-infra | 01:03 | |
| *** thorst has quit IRC | 01:03 | |
| *** cuongnv has joined #openstack-infra | 01:08 | |
| *** r-daneel has joined #openstack-infra | 01:18 | |
| *** Daisy_ has joined #openstack-infra | 01:23 | |
| *** EricGonczer_ has quit IRC | 01:23 | |
| *** Daisy_ has quit IRC | 01:30 | |
| *** Daisy_ has joined #openstack-infra | 01:33 | |
| *** Daisy_ has quit IRC | 01:40 | |
| *** aeng has quit IRC | 01:43 | |
| *** aeng has joined #openstack-infra | 01:44 | |
| *** MatthewLi has joined #openstack-infra | 01:44 | |
| *** cody-somerville has joined #openstack-infra | 01:45 | |
| *** cody-somerville has joined #openstack-infra | 01:45 | |
| *** Daisy_ has joined #openstack-infra | 01:45 | |
| *** stakeda has joined #openstack-infra | 01:49 | |
| *** EricGonczer_ has joined #openstack-infra | 02:01 | |
| openstackgerrit | Merged openstack/diskimage-builder master: Move setfiles to outside chroot with runcon https://review.openstack.org/485899 | 02:04 |
| *** r-daneel has quit IRC | 02:06 | |
| *** armax has joined #openstack-infra | 02:07 | |
| *** funzo has joined #openstack-infra | 02:10 | |
| *** funzo has quit IRC | 02:15 | |
| *** r-daneel has joined #openstack-infra | 02:23 | |
| *** EricGonczer_ has quit IRC | 02:39 | |
| *** mriedem has quit IRC | 02:43 | |
| *** sbezverk has quit IRC | 02:44 | |
| *** lbragsta_ has joined #openstack-infra | 02:48 | |
| *** gouthamr has quit IRC | 02:49 | |
| *** ykarel has joined #openstack-infra | 02:58 | |
| *** pahuang has quit IRC | 03:00 | |
| *** Daisy_ has quit IRC | 03:03 | |
| *** thorst has joined #openstack-infra | 03:04 | |
| *** cody-somerville has quit IRC | 03:07 | |
| *** thorst has quit IRC | 03:10 | |
| *** funzo has joined #openstack-infra | 03:11 | |
| *** pahuang has joined #openstack-infra | 03:13 | |
| *** mtreinish has quit IRC | 03:14 | |
| *** funzo has quit IRC | 03:15 | |
| *** mtreinish has joined #openstack-infra | 03:20 | |
| *** ramishra has joined #openstack-infra | 03:39 | |
| *** aeng has quit IRC | 03:40 | |
| *** jesusaurum has quit IRC | 03:41 | |
| *** aeng has joined #openstack-infra | 03:41 | |
| *** links has joined #openstack-infra | 03:46 | |
| *** rodrigopaiva has joined #openstack-infra | 03:50 | |
| *** lbragsta_ has quit IRC | 04:02 | |
| *** lbragsta_ has joined #openstack-infra | 04:03 | |
| *** funzo has joined #openstack-infra | 04:12 | |
| *** funzo has quit IRC | 04:16 | |
| *** adisky__ has joined #openstack-infra | 04:28 | |
| *** lbragsta_ has quit IRC | 04:35 | |
| *** bhavik1 has joined #openstack-infra | 04:37 | |
| *** stakeda has quit IRC | 04:39 | |
| *** shu-mutou-AWAY is now known as shu-mutou | 04:40 | |
| openstackgerrit | Merged openstack-infra/project-config master: networking-odl: make fullstack with carbon voting https://review.openstack.org/486347 | 04:48 |
| *** prateek has joined #openstack-infra | 04:48 | |
| *** sbezverk has joined #openstack-infra | 04:51 | |
| openstackgerrit | Merged openstack-infra/project-config master: setup correct release jobs for networking-bigswitch https://review.openstack.org/427291 | 04:51 |
| *** dassistant is now known as chandankumar | 04:52 | |
| openstackgerrit | Ian Wienand proposed openstack/diskimage-builder master: Bump fedora-minimal DIB_RELEASE 26 https://review.openstack.org/482570 | 04:54 |
| *** Dinesh_Bhor has joined #openstack-infra | 04:56 | |
| *** thorst has joined #openstack-infra | 05:05 | |
| *** rodrigopaiva has quit IRC | 05:06 | |
| *** rodrigopaiva has joined #openstack-infra | 05:06 | |
| *** thorst has quit IRC | 05:10 | |
| *** jascott1_ has quit IRC | 05:12 | |
| *** funzo has joined #openstack-infra | 05:13 | |
| *** jascott1 has joined #openstack-infra | 05:13 | |
| *** pgadiya has joined #openstack-infra | 05:14 | |
| *** rodrigopaiva has quit IRC | 05:14 | |
| *** rodrigopaiva has joined #openstack-infra | 05:15 | |
| *** Julien-zte has quit IRC | 05:16 | |
| *** rodrigopaiva has quit IRC | 05:16 | |
| *** funzo has quit IRC | 05:17 | |
| *** Julien-zte has joined #openstack-infra | 05:17 | |
| *** jascott1 has quit IRC | 05:20 | |
| *** stakeda has joined #openstack-infra | 05:23 | |
| *** arturb has joined #openstack-infra | 05:26 | |
| *** psachin has joined #openstack-infra | 05:30 | |
| *** boris-42____ has joined #openstack-infra | 05:30 | |
| *** aeng has quit IRC | 05:34 | |
| *** sree_ has joined #openstack-infra | 05:36 | |
| *** sree_ is now known as Guest55915 | 05:36 | |
| *** dhajare has joined #openstack-infra | 05:45 | |
| *** wolverineav has joined #openstack-infra | 05:45 | |
| *** aeng has joined #openstack-infra | 05:47 | |
| *** wolverineav has quit IRC | 05:50 | |
| *** jascott1 has joined #openstack-infra | 05:57 | |
| *** bhavik1 has quit IRC | 05:57 | |
| *** aeng has quit IRC | 05:58 | |
| *** aeng has joined #openstack-infra | 05:59 | |
| *** cshastri has joined #openstack-infra | 06:00 | |
| *** jascott1 has quit IRC | 06:02 | |
| *** jbadiapa has joined #openstack-infra | 06:08 | |
| AJaeger | fungi, interesting - the number of defined jobs as output by JJB has decreased from 17170 to 12503. | 06:09 |
| *** funzo has joined #openstack-infra | 06:13 | |
| *** kong has quit IRC | 06:17 | |
| *** isaacb has joined #openstack-infra | 06:17 | |
| *** funzo has quit IRC | 06:18 | |
| *** coolsvap has joined #openstack-infra | 06:19 | |
| *** rcernin has joined #openstack-infra | 06:23 | |
| *** makowals has joined #openstack-infra | 06:23 | |
| *** thorst has joined #openstack-infra | 06:27 | |
| *** andreas_s has joined #openstack-infra | 06:27 | |
| *** hashar has joined #openstack-infra | 06:29 | |
| *** thorst has quit IRC | 06:31 | |
| *** tuanluong has joined #openstack-infra | 06:32 | |
| *** yamamoto has quit IRC | 06:34 | |
| *** jascott1 has joined #openstack-infra | 06:39 | |
| *** jascott1 has quit IRC | 06:43 | |
| *** hichihara has quit IRC | 06:59 | |
| openstackgerrit | Simon Kuhnle proposed openstack-infra/git-review master: Fix listing changes over SSH for 2.14 https://review.openstack.org/479192 | 07:00 |
| *** AJaeger has quit IRC | 07:00 | |
| *** AJaeger has joined #openstack-infra | 07:01 | |
| *** ccamacho has joined #openstack-infra | 07:02 | |
| *** nicolasbock has joined #openstack-infra | 07:03 | |
| *** yolanda_ has joined #openstack-infra | 07:05 | |
| *** yolanda_ has quit IRC | 07:06 | |
| *** 7ITABD5MB has joined #openstack-infra | 07:06 | |
| *** 07IAALFJ9 has joined #openstack-infra | 07:06 | |
| openstackgerrit | Thomas Herve proposed openstack-infra/project-config master: Filter out services in designate jobs https://review.openstack.org/486221 | 07:07 |
| *** 07IAALFJ9 has quit IRC | 07:07 | |
| *** 7ITABD5MB has quit IRC | 07:08 | |
| *** yolanda_ has joined #openstack-infra | 07:08 | |
| *** yolanda_ is now known as yolanda | 07:11 | |
| *** dizquierdo has joined #openstack-infra | 07:11 | |
| *** migi_ is now known as migi | 07:12 | |
| *** ralonsoh has joined #openstack-infra | 07:13 | |
| *** vtapia has quit IRC | 07:13 | |
| *** xarses has quit IRC | 07:14 | |
| *** vtapia has joined #openstack-infra | 07:14 | |
| *** xarses has joined #openstack-infra | 07:15 | |
| *** xarses has quit IRC | 07:15 | |
| *** xarses has joined #openstack-infra | 07:15 | |
| *** isaacb has quit IRC | 07:15 | |
| *** lennyb has quit IRC | 07:19 | |
| *** jascott1 has joined #openstack-infra | 07:20 | |
| *** ralonsoh_ has joined #openstack-infra | 07:22 | |
| *** ralonsoh has quit IRC | 07:22 | |
| *** isaacb has joined #openstack-infra | 07:23 | |
| *** Hal has joined #openstack-infra | 07:25 | |
| *** Hal is now known as Guest36716 | 07:25 | |
| *** jascott1 has quit IRC | 07:27 | |
| *** flepied has quit IRC | 07:28 | |
| *** lennyb has joined #openstack-infra | 07:32 | |
| *** iyamahat has joined #openstack-infra | 07:33 | |
| *** gildub has quit IRC | 07:35 | |
| *** iyamahat has quit IRC | 07:38 | |
| openstackgerrit | Merged openstack-infra/project-config master: Filter out services in designate jobs https://review.openstack.org/486221 | 07:41 |
| *** yamamoto has joined #openstack-infra | 07:45 | |
| *** egonzalez has joined #openstack-infra | 07:47 | |
| hwoarang | dirk: good morning do you have a second? | 07:49 |
| dirk | hwoarang: sure, good morning | 07:50 |
| hwoarang | dirk: hi in regards to your keystone comment on this https://review.openstack.org/#/c/485985/ | 07:50 |
| hwoarang | is it this https://bugs.launchpad.net/openstack-ansible/+bug/1705521 btw ? | 07:50 |
| openstack | Launchpad bug 1705521 in openstack-ansible "[os_keystone] uwsgi process segfaults on openSUSE" [Undecided,New] - Assigned to Markos Chandras (hwoarang) | 07:50 |
| *** david-lyle has quit IRC | 07:50 | |
| *** dklyle has joined #openstack-infra | 07:50 | |
| hwoarang | because i am trying to debug this failure which happens only on opensuse with cryptography-2.0 | 07:51 |
| *** alexchadin has joined #openstack-infra | 07:54 | |
| dirk | hwoarang: yes | 07:54 |
| dirk | hwoarang: yes, because cryptography 2.0 added a universal wheel with an embedded openssl 1.1 | 07:55 |
| dirk | which has symbol clashes with the opensuse provided openssl 1.0 | 07:55 |
| dirk | it doesn't happen on any other distro in infra because infra is rebuilding all wheels and provides a wheel mirror | 07:55 |
| dirk | https://review.openstack.org/#/c/486305/ is going to fix that | 07:56 |
| dirk | but its actually a terrible workaround | 07:56 |
| dirk | hwoarang: the --no-binary is imho more correct | 07:56 |
| hwoarang | aha | 07:57 |
| hwoarang | i also have an issue opened with upstream | 07:57 |
| hwoarang | maybe worth telling them all that and fix their wheel... | 07:58 |
| hwoarang | thank you dirk | 07:59 |
| *** rossella__ has joined #openstack-infra | 08:01 | |
| dirk | hwoarang: its like the 3rd or 4th time the cryptography upstream adds a manylinux1 wheel | 08:01 |
| dirk | even though it is clearly very much outside the defined abi of manylinux1. so I'm afraid they just don't care | 08:01 |
| *** flepied has joined #openstack-infra | 08:02 | |
| *** jascott1 has joined #openstack-infra | 08:04 | |
| hwoarang | hmm i see. i will them know and then see how to fix it for openstack-ansible in the meantime :/ | 08:06 |
| hwoarang | thank you for all the infor! | 08:06 |
| *** markus_z has joined #openstack-infra | 08:06 | |
| dirk | hwoarang: does openstack-ansible use devstack scripts? | 08:08 |
| hwoarang | no | 08:08 |
| AJaeger | dirk: for 485305, best work with pabelanger later today... | 08:08 |
| dirk | hwoarang: is there actually any openstack-ansible script that passes on opensuse? I was doing some experimental checks a few days ago and couldn't ifnd one that passed (before the cryptography 2.0 messup) | 08:08 |
| *** jascott1 has quit IRC | 08:09 | |
| hwoarang | the openstack-ansible-repo_server should pass | 08:09 |
| dirk | hwoarang: once the wheel mirror is there it should work again then | 08:09 |
| hwoarang | dirk: do all projects consume the stuff from the wheel mirrors? | 08:10 |
| hwoarang | or do they need to do something special to get them instead of the pypi ones | 08:10 |
| dirk | hwoarang: iirc all nodepool images have a WHEEL_MIRROR set | 08:10 |
| dirk | but I'm not exactly sure how that all works | 08:11 |
| *** dizquierdo has quit IRC | 08:11 | |
| hwoarang | hmm | 08:11 |
| dirk | see nodepool/scripts/configure_mirror.sh in project-config which is setting the wheel_mirror based on the AFS_SLUG | 08:11 |
| dirk | I believe that script is run on image instantiation inside a particular cloud | 08:12 |
| hwoarang | hmm i see | 08:12 |
| *** funzo has joined #openstack-infra | 08:15 | |
| *** yamahata has quit IRC | 08:15 | |
| *** markus_z has quit IRC | 08:16 | |
| *** danpawlik has quit IRC | 08:17 | |
| *** danpawlik has joined #openstack-infra | 08:17 | |
| *** funzo has quit IRC | 08:19 | |
| *** derekh has joined #openstack-infra | 08:23 | |
| *** lucas-afk is now known as lucasagomes | 08:26 | |
| *** e0ne has joined #openstack-infra | 08:27 | |
| *** thorst has joined #openstack-infra | 08:27 | |
| mordred | dirk, hwoarang it gets run once a node is booted but before jobs are run on the node | 08:31 |
| *** s-shiono_ has quit IRC | 08:32 | |
| mordred | and it makes sure pip and friends are configured to use the right mirror based on the region | 08:32 |
| *** thorst has quit IRC | 08:32 | |
| openstackgerrit | Daniel Mellado proposed openstack-infra/project-config master: Add nova service to kuryr gate https://review.openstack.org/486525 | 08:32 |
| *** dklyle has quit IRC | 08:32 | |
| mordred | there's also some docs about the metadata that is put on images: https://docs.openstack.org/infra/nodepool/scripts.html | 08:33 |
| *** dklyle has joined #openstack-infra | 08:33 | |
| mordred | which you can use to configure mirror if for some reason you can't directly rely on the pre-configured mirror information | 08:33 |
| *** dtantsur|afk is now known as dtantsur | 08:34 | |
| mordred | for instance, loci does this: http://git.openstack.org/cgit/openstack/loci/tree/scripts/run-tests.sh#n115 | 08:34 |
| mordred | since they need mirrors set inside of the containers they're building | 08:35 |
| mordred | (which obviously the base images don't know anything about | 08:35 |
| *** dklyle has quit IRC | 08:36 | |
| *** vponomaryov has joined #openstack-infra | 08:37 | |
| openstackgerrit | Daniel Mellado proposed openstack-infra/project-config master: Add nova service to kuryr gate https://review.openstack.org/486525 | 08:38 |
| *** jangutter has joined #openstack-infra | 08:42 | |
| *** jascott1 has joined #openstack-infra | 08:42 | |
| *** markus_z has joined #openstack-infra | 08:42 | |
| *** jascott1 has quit IRC | 08:47 | |
| *** david-lyle has joined #openstack-infra | 08:47 | |
| *** david-lyle has quit IRC | 08:49 | |
| *** david-lyle has joined #openstack-infra | 08:50 | |
| *** dingyichen has quit IRC | 08:52 | |
| *** dklyle has joined #openstack-infra | 08:53 | |
| *** david-lyle has quit IRC | 08:53 | |
| openstackgerrit | Gael Chamoulaud proposed openstack-infra/tripleo-ci master: WIP: Enable tripleo-validations tests https://review.openstack.org/481080 | 08:53 |
| openstackgerrit | Deepak Mourya proposed openstack-infra/os-loganalyze master: Stop ignoring H904 hacking rule in tox https://review.openstack.org/468761 | 08:57 |
| *** yamamoto has quit IRC | 09:02 | |
| *** jaosorior has joined #openstack-infra | 09:02 | |
| *** rlandy has joined #openstack-infra | 09:04 | |
| *** florianf has joined #openstack-infra | 09:05 | |
| hwoarang | mordred: thank you for the information | 09:14 |
| *** isaacb has quit IRC | 09:15 | |
| *** yamamoto has joined #openstack-infra | 09:15 | |
| *** funzo has joined #openstack-infra | 09:16 | |
| *** sambetts|afk is now known as sambetts | 09:16 | |
| *** isaacb has joined #openstack-infra | 09:16 | |
| *** samP has joined #openstack-infra | 09:16 | |
| samP | Hi o/.. Need help to create 3rd party CI group for project Masakari | 09:19 |
| *** boris-42____ has quit IRC | 09:19 | |
| *** funzo has quit IRC | 09:20 | |
| *** pblaho has joined #openstack-infra | 09:21 | |
| *** dizquierdo has joined #openstack-infra | 09:23 | |
| openstackgerrit | Kien Nguyen proposed openstack-infra/project-config master: Add topology flag to Heat job template https://review.openstack.org/482048 | 09:23 |
| *** electrofelix has joined #openstack-infra | 09:24 | |
| *** shu-mutou is now known as shu-mutou-AWAY | 09:35 | |
| *** claudiub has joined #openstack-infra | 09:38 | |
| *** jascott1 has joined #openstack-infra | 09:39 | |
| *** katkapilatova has joined #openstack-infra | 09:39 | |
| *** katkapilatova has left #openstack-infra | 09:40 | |
| *** jascott1 has quit IRC | 09:44 | |
| *** yamamoto has quit IRC | 09:47 | |
| *** wolverineav has joined #openstack-infra | 09:51 | |
| *** cuongnv has quit IRC | 09:58 | |
| *** tuanluong has quit IRC | 10:02 | |
| *** sdague has joined #openstack-infra | 10:02 | |
| *** yamamoto has joined #openstack-infra | 10:02 | |
| *** yamamoto has quit IRC | 10:02 | |
| openstackgerrit | Ronelle Landy proposed openstack-infra/tripleo-ci master: Add settings for OVB in RDO Cloud https://review.openstack.org/480950 | 10:06 |
| *** stakeda has quit IRC | 10:08 | |
| *** LindaWang has quit IRC | 10:13 | |
| *** kong has joined #openstack-infra | 10:16 | |
| *** funzo has joined #openstack-infra | 10:16 | |
| *** jascott1 has joined #openstack-infra | 10:21 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack-infra/project-config master: Make tripleo-ci-centos-7-scenario007 non-voting for networking-ovn https://review.openstack.org/486557 | 10:21 |
| *** Guest55915 has quit IRC | 10:21 | |
| *** funzo has quit IRC | 10:21 | |
| *** sree_ has joined #openstack-infra | 10:22 | |
| *** sree_ is now known as Guest4234 | 10:22 | |
| *** jascott1 has quit IRC | 10:25 | |
| *** Guest4234 has quit IRC | 10:26 | |
| openstackgerrit | yolanda.robla proposed openstack/diskimage-builder master: [WIP] LVM support for dib-block-device https://review.openstack.org/472065 | 10:29 |
| *** thorst has joined #openstack-infra | 10:29 | |
| *** rlandy has quit IRC | 10:31 | |
| *** LindaWang has joined #openstack-infra | 10:32 | |
| *** rlandy has joined #openstack-infra | 10:32 | |
| *** yamamoto has joined #openstack-infra | 10:33 | |
| *** thorst has quit IRC | 10:33 | |
| *** electrofelix has quit IRC | 10:38 | |
| *** electrofelix has joined #openstack-infra | 10:43 | |
| *** jkilpatr has quit IRC | 10:45 | |
| *** lucasagomes is now known as lucas-hungry | 10:52 | |
| *** r-daneel has quit IRC | 10:55 | |
| *** jkilpatr has joined #openstack-infra | 11:02 | |
| *** Dinesh_Bhor has quit IRC | 11:07 | |
| *** hashar is now known as hasharLunch | 11:13 | |
| *** jascott1 has joined #openstack-infra | 11:23 | |
| *** jascott1 has quit IRC | 11:27 | |
| *** flepied has quit IRC | 11:28 | |
| *** Guest36716 has quit IRC | 11:29 | |
| *** Hal has joined #openstack-infra | 11:36 | |
| *** Hal is now known as Guest23415 | 11:36 | |
| *** ldnunes has joined #openstack-infra | 11:41 | |
| *** HeOS has quit IRC | 11:48 | |
| *** sree has joined #openstack-infra | 11:52 | |
| *** thorst has joined #openstack-infra | 11:54 | |
| *** thorst_ has joined #openstack-infra | 11:56 | |
| *** thorst_ has quit IRC | 11:56 | |
| *** thorst_ has joined #openstack-infra | 11:57 | |
| *** thorst has quit IRC | 11:58 | |
| *** EricGonczer_ has joined #openstack-infra | 12:03 | |
| *** lucas-hungry is now known as lucasagomes | 12:03 | |
| *** flepied has joined #openstack-infra | 12:03 | |
| *** jascott1 has joined #openstack-infra | 12:05 | |
| *** Julien-zte has quit IRC | 12:09 | |
| *** jascott1 has quit IRC | 12:09 | |
| *** Julien-zte has joined #openstack-infra | 12:10 | |
| openstackgerrit | Ifat Afek proposed openstack-infra/project-config master: Run tempest tests on python35 devstack (non-voting for now) https://review.openstack.org/486591 | 12:15 |
| *** stevebaker has quit IRC | 12:16 | |
| *** stevebaker has joined #openstack-infra | 12:16 | |
| *** pgadiya has quit IRC | 12:17 | |
| *** funzo has joined #openstack-infra | 12:18 | |
| *** khappone has quit IRC | 12:19 | |
| *** yamamoto has quit IRC | 12:19 | |
| *** khappone has joined #openstack-infra | 12:20 | |
| *** funzo has quit IRC | 12:22 | |
| *** baoli has joined #openstack-infra | 12:24 | |
| *** askb has quit IRC | 12:24 | |
| *** jascott1 has joined #openstack-infra | 12:27 | |
| openstackgerrit | Ifat Afek proposed openstack-infra/project-config master: Run tempest tests on python35 devstack (non-voting for now) https://review.openstack.org/486591 | 12:29 |
| *** rhallisey has joined #openstack-infra | 12:30 | |
| *** jcoufal has joined #openstack-infra | 12:31 | |
| *** jascott1 has quit IRC | 12:32 | |
| *** baoli has quit IRC | 12:34 | |
| *** baoli has joined #openstack-infra | 12:34 | |
| *** yamamoto has joined #openstack-infra | 12:36 | |
| *** yamamoto has quit IRC | 12:36 | |
| *** trown|outtypewww is now known as trown | 12:38 | |
| *** gordc has joined #openstack-infra | 12:39 | |
| electrofelix | jamielennox: have you had a chance to look over https://review.openstack.org/475474 since I updated it? | 12:40 |
| *** ldnunes has quit IRC | 12:44 | |
| electrofelix | anyone around to review an elastic-recheck change: https://review.openstack.org/320988 | 12:49 |
| *** chlong has joined #openstack-infra | 12:53 | |
| *** kgiusti has joined #openstack-infra | 12:53 | |
| *** ralonsoh_ is now known as ralonsoh | 12:54 | |
| *** bobh has joined #openstack-infra | 12:54 | |
| *** ldnunes has joined #openstack-infra | 12:56 | |
| openstackgerrit | Markos Chandras (hwoarang) proposed openstack-infra/project-config master: jenkins: openstack-ansible: Add openSUSE 42.3 jobs https://review.openstack.org/486602 | 12:56 |
| *** mriedem has joined #openstack-infra | 12:59 | |
| *** hasharLunch is now known as hashar | 13:01 | |
| *** jaypipes has joined #openstack-infra | 13:04 | |
| *** xarses has quit IRC | 13:05 | |
| *** esberglu has joined #openstack-infra | 13:07 | |
| *** r-daneel has joined #openstack-infra | 13:09 | |
| *** jascott1 has joined #openstack-infra | 13:09 | |
| *** ihrachys has joined #openstack-infra | 13:09 | |
| *** bobh has quit IRC | 13:10 | |
| *** EricGonczer_ has quit IRC | 13:10 | |
| *** eharney has joined #openstack-infra | 13:11 | |
| *** r-daneel has quit IRC | 13:13 | |
| *** jascott1 has quit IRC | 13:13 | |
| *** gagehugo has quit IRC | 13:15 | |
| *** ykarel has quit IRC | 13:17 | |
| *** jamesdenton has joined #openstack-infra | 13:18 | |
| *** links has quit IRC | 13:20 | |
| *** gagehugo has joined #openstack-infra | 13:21 | |
| *** prateek has quit IRC | 13:24 | |
| *** jamesdenton has quit IRC | 13:25 | |
| *** jamesdenton has joined #openstack-infra | 13:25 | |
| *** xarses has joined #openstack-infra | 13:31 | |
| *** yamamoto has joined #openstack-infra | 13:36 | |
| *** bobh has joined #openstack-infra | 13:39 | |
| *** Dinesh_Bhor has joined #openstack-infra | 13:40 | |
| *** gouthamr has joined #openstack-infra | 13:41 | |
| *** gordc has quit IRC | 13:42 | |
| *** yamamoto has quit IRC | 13:44 | |
| *** xyang1 has joined #openstack-infra | 13:44 | |
| *** ykarel has joined #openstack-infra | 13:50 | |
| *** Goneri has joined #openstack-infra | 13:50 | |
| *** eharney has quit IRC | 13:50 | |
| *** camunoz has joined #openstack-infra | 13:52 | |
| *** trown is now known as trown|brb | 13:52 | |
| *** eharney has joined #openstack-infra | 13:55 | |
| *** trown|brb is now known as trown | 13:57 | |
| *** isaacb has quit IRC | 13:58 | |
| *** efoley has joined #openstack-infra | 14:04 | |
| *** alexchadin has quit IRC | 14:06 | |
| *** iyamahat has joined #openstack-infra | 14:10 | |
| *** isaacb has joined #openstack-infra | 14:12 | |
| *** rbrndt has joined #openstack-infra | 14:16 | |
| *** HeOS has joined #openstack-infra | 14:18 | |
| pabelanger | AJaeger: dirk: for 485305 we should consider waiting until after PTG. It might be a good thing to do in zuulv3 too so we don't need to stand up a opensuse static node | 14:19 |
| *** funzo has joined #openstack-infra | 14:19 | |
| *** dizquierdo has quit IRC | 14:20 | |
| *** annegentle has joined #openstack-infra | 14:20 | |
| *** gordc has joined #openstack-infra | 14:23 | |
| *** funzo has quit IRC | 14:24 | |
| *** kgiusti has quit IRC | 14:25 | |
| *** sree has quit IRC | 14:26 | |
| *** jcoufal has quit IRC | 14:27 | |
| *** jcoufal has joined #openstack-infra | 14:27 | |
| *** yamahata has joined #openstack-infra | 14:29 | |
| *** felipemonteiro has joined #openstack-infra | 14:30 | |
| *** baoli has quit IRC | 14:31 | |
| *** felipemonteiro_ has joined #openstack-infra | 14:31 | |
| *** felipemonteiro has quit IRC | 14:34 | |
| openstackgerrit | Sagi Shnaidman proposed openstack-infra/project-config master: Transit Heat convergence job to use quickstart https://review.openstack.org/483292 | 14:35 |
| *** cshastri has quit IRC | 14:39 | |
| *** marst has quit IRC | 14:39 | |
| openstackgerrit | Gael Chamoulaud proposed openstack-infra/tripleo-ci master: Enable tripleo-validations tests https://review.openstack.org/481080 | 14:40 |
| AJaeger | pabelanger: there's a problem with one of the wheels that hwoarang and dirk discussed this morning (cryptography-2.0), see backscroll. | 14:41 |
| AJaeger | pabelanger: is there a short time solution for these? | 14:42 |
| hwoarang | AJaeger: for openstack-ansible i am proposing --no-binary in https://review.openstack.org/#/c/486580/ | 14:42 |
| hwoarang | the upstream bug report is rather informative btw https://github.com/pyca/cryptography/issues/3804 | 14:42 |
| pabelanger | AJaeger: hwoarang: Ya, 486580 is best approach right now, adding opensuse wheel-mirror will require some work / planning | 14:43 |
| hwoarang | yeah | 14:43 |
| dirk | AJaeger: pabelanger : hwoarang : any review on https://review.openstack.org/#/c/485985/ appreciated | 14:45 |
| hwoarang | dirk: i didn't vote because of Ian's comment so I was expecting a new patchset to be honest | 14:45 |
| dirk | I just don't want to have things bitrot for months, as it takes time to recover the job into a green state when it keeps failing for extended period of time (and this sneaked in while the previous patch to unbreak it was under review) | 14:45 |
| *** EricGonczer_ has joined #openstack-infra | 14:45 | |
| dirk | hwoarang: right, but the code doesn't hurt once it is in dib | 14:46 |
| dirk | hwoarang: it will just become a noop then | 14:46 |
| dirk | pabelanger: how does zuulv3 help with that? I'm confused. | 14:47 |
| dirk | pabelanger: would there not be a need for wheels then anymore? or does zuulv3 spin up dynamic nodes? | 14:47 |
| *** marst has joined #openstack-infra | 14:47 | |
| pabelanger | dirk: right, we'd move wheel builds into nodepool and not on static nodes | 14:47 |
| pabelanger | I mean, we should be able to do that today too. Maybe we should talk about it at our weekly meeting | 14:48 |
| pabelanger | just need to deal with vos release (AFS) | 14:48 |
| pabelanger | ya, wheel-release can still be trusted | 14:49 |
| pabelanger | but wheel-build job, could use an untrusted node | 14:50 |
| *** jamesdenton has quit IRC | 14:51 | |
| *** makowals has quit IRC | 14:51 | |
| *** jamesdenton has joined #openstack-infra | 14:52 | |
| *** baoli has joined #openstack-infra | 14:52 | |
| *** sree has joined #openstack-infra | 14:53 | |
| clarkb | regardless of zuulv3 or zuul v2.5 you need working afs on suse right? which last I checked wasn't a thing? | 14:53 |
| *** makowals has joined #openstack-infra | 14:53 | |
| pabelanger | Ya, I haven't looked into that either | 14:54 |
| clarkb | pabelanger: fungi: will you be able to respond to citycloud? I am less familiar with the current situation (though they are looking into background info which I guess any infraer should be able to provide) | 14:54 |
| pabelanger | preformance aside, opensuse could upload to xenial and have xenial publish to afs | 14:55 |
| fungi | AJaeger: yeah, i believe it's because zuul-launcher registers both build:<jobname> and build:<jobname>:<nodetype> into gearman, so the number of build registrations will always be >2x the number of job definitions as long as all jobs get registered | 14:55 |
| pabelanger | clarkb: Ya, was holding off for fungi this morning | 14:55 |
| pabelanger | but I can reply now | 14:55 |
| clarkb | fungi: AJaeger yup that sounds correct | 14:57 |
| *** jamesdenton has quit IRC | 14:59 | |
| *** baoli has quit IRC | 14:59 | |
| fungi | dirk: hwoarang: AJaeger: if it matters, the wheel mirror we build isn't so much to work around problems with individual wheels as it is to prevent jobs from wasting a lot of time building platform-specific wheels. it provides a pretty massive speedup for devstack, for example | 14:59 |
| dirk | clarkb: there should be working afs I think (see the link I pasted) | 15:00 |
| *** dizquierdo has joined #openstack-infra | 15:00 | |
| *** annegentle has quit IRC | 15:00 | |
| fungi | one thing we're still missing is a step to exclude wheels which are already present on pypi. the goal with the wheel mirrors is to fill in the gaps for wheels pypi doesn't already carry, not rebuild and replace them with our own, which is unfortunately the end result | 15:00 |
| dirk | fungi: sure, but its one thing that is different for the opensuse job than for othrs, and others don't notice the problems with upstream wheels because they get rebuilt anyway | 15:01 |
| fungi | yep, i completely understand the issue | 15:01 |
| *** sree has quit IRC | 15:01 | |
| clarkb | dirk: are you sure that pip will work that way? we add our wheel mirror as an extra index whcih I think means it is not preferred | 15:01 |
| pabelanger | fungi: clarkb: replied to citycloud | 15:02 |
| fungi | thanks pabelanger! | 15:02 |
| *** EricGonczer_ has quit IRC | 15:02 | |
| *** sree has joined #openstack-infra | 15:02 | |
| *** jamesdenton has joined #openstack-infra | 15:03 | |
| clarkb | dirk: but that would certainly explain why the other distros don't have a problem in testing with cryptography | 15:03 |
| fungi | the issue is that these are non-pure-python wheels | 15:03 |
| clarkb | fungi: right but the pypi hosted wheel built against the simple linux from land before time should be preferred by pip because pypi is a regular index url, but maybe I misunderstand the behavipr there | 15:04 |
| fungi | so i believe what happens is that while pip is willing to use the manylinux1 wheels it prefers, the platform-specific ones even if they're found in the extra-index | 15:04 |
| fungi | er, what happens is that while pip is willing to use the manylinux1 wheels, it prefers the platform-specific ones even if they're found in the extra-index | 15:05 |
| fungi | bad comma placement ;) | 15:05 |
| clarkb | dirk: https://software.opensuse.org/package/openafs looks like all available packages are "unstable"? | 15:05 |
| *** sree has quit IRC | 15:06 | |
| clarkb | but I guess they are there in filesystems so maybe we don't care | 15:06 |
| *** markus_z has quit IRC | 15:07 | |
| *** isaacb has quit IRC | 15:08 | |
| *** baoli has joined #openstack-infra | 15:10 | |
| *** dhajare has quit IRC | 15:13 | |
| *** dklyle is now known as david-lyle | 15:14 | |
| fungi | interestingly, the ones built by our wheel builders are still named like cryptography-2.0-cp27-cp27mu-manylinux1_x86_64.whl so pip isn't finding a more specific wheel architecture after all? | 15:15 |
| *** jascott1 has joined #openstack-infra | 15:16 | |
| inc0 | good morning, qq, RAX jobs didnt seem to get volume on /opt | 15:18 |
| inc0 | and we ran out of space for ceph | 15:18 |
| inc0 | (with 30Gig disk) | 15:18 |
| inc0 | is this planned (and should figure out how to use less disk) or something went wrong? | 15:19 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Remove ZUUL_PROJECT https://review.openstack.org/486251 | 15:19 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Remove ZUUL_UUID https://review.openstack.org/486252 | 15:19 |
| *** annegentle has joined #openstack-infra | 15:20 | |
| jeblair | inc0: yes, rax doesn't have ephemeral disks, it just has large root disks | 15:20 |
| *** jascott1 has quit IRC | 15:20 | |
| jeblair | inc0: i thought there was more free space than that though | 15:20 |
| *** Apoorva has joined #openstack-infra | 15:21 | |
| pabelanger | 40GB is what I am seeing in rax-ord ATM | 15:21 |
| *** chlong_ has joined #openstack-infra | 15:21 | |
| inc0 | right, just partitioned | 15:22 |
| inc0 | http://logs.openstack.org/40/456140/15/check/gate-kolla-ansible-dsvm-deploy-ceph-ubuntu-source-ubuntu-xenial-2-node-nv/9516ad4/console.html#_2017-07-24_14_13_10_169301 | 15:23 |
| pabelanger | inc0: have you mounted /dev/xvde2? | 15:23 |
| inc0 | no | 15:23 |
| jeblair | on the node i looked at, we apparently have 38g total, 28g free | 15:23 |
| inc0 | it's mointed on / | 15:23 |
| *** chlong has quit IRC | 15:24 | |
| inc0 | anyway, I'll remove one OSD since it's not unplanned | 15:24 |
| pabelanger | IIRC, devstack-gate handles this right? And sets up /mnt on rackspace? | 15:25 |
| dtantsur | yolanda: hi! if you're still here, mind checking https://review.openstack.org/#/c/480332/ please? should be a quick one | 15:25 |
| fungi | pabelanger: /opt | 15:25 |
| jeblair | fungi, pabelanger: does rax have an ephemeral volume now? | 15:26 |
| fungi | that's what i'm checking, though i thought it always did unless you told it to boot without | 15:26 |
| dmsimard | Maybe I'm just impatient but I subscribed to a ml on lists.o.o and haven't gotten anything back to confirm the subscription in like 15 minutes ? :/ is legal-discuss moderated ? | 15:26 |
| pabelanger | fungi: jeblair: preformance1-8 is reporting 80GB ephemeral | 15:27 |
| pabelanger | | performance1-8 | 8 GB Performance | 8192 | 40 | 80 | 8 | N/A | | 15:27 |
| fungi | http://paste.openstack.org/show/616317/ is in the middle of running a devstack-gate based job in rax-ord | 15:28 |
| fungi | 72gb /opt mounted | 15:28 |
| fungi | 38gb rootfs | 15:28 |
| pabelanger | right, I think kolla just needs to do that | 15:29 |
| jeblair | oh, yeah, that's handled by devstack-gate | 15:29 |
| pabelanger | Isn't fix_disk_layout in d-g where we do that now? | 15:29 |
| fungi | d-g partitions /dev/xvde into two partitions there, one for swap and one for /opt | 15:29 |
| fungi | which is why it's not 80gb | 15:29 |
| fungi | dmsimard: i can check the mailman logs, just a sec | 15:29 |
| jeblair | inc0: ^ devstack-gate will set it up for you | 15:30 |
| *** Julien-zte has quit IRC | 15:30 | |
| inc0 | that's my point, df didn't seem to show that it did. But on second glance that also doesn't seem to be issue for me | 15:31 |
| *** Julien-zte has joined #openstack-infra | 15:31 | |
| dmsimard | fungi: got it just now, not sure if you did anything | 15:32 |
| fungi | dmsimard: nope, maybe your mta was greylisting us | 15:32 |
| inc0 | how about osic? | 15:33 |
| fungi | dmsimard: bingo... 2017-07-24 15:15:44 1dZf5D-0002m5-1L SMTP error from remote mail server after RCPT TO:<REDACTED>: host mx1.redhat.com [209.132.183.28]: 451 4.7.1 Greylisting in action, please come back in 00:05:00 | 15:34 |
| jeblair | inc0: in http://logs.openstack.org/40/456140/15/check/gate-kolla-ansible-dsvm-deploy-ceph-ubuntu-source-ubuntu-xenial-2-node-nv/9516ad4/logs/ where are the devstack-gate logs? | 15:34 |
| *** rcernin has quit IRC | 15:34 | |
| dmsimard | fungi: ah, there you go. At least I wasn't crazy :) | 15:34 |
| jeblair | inc0: should have names like workspace-setup or something | 15:35 |
| inc0 | hold on, scrach that, sorry guys I haven't get my caffeine flowing. Looked at wrong job again | 15:35 |
| clarkb | fungi: re unbound and opendns and google, I don't think order matters in unbound config, eg ther is no way to say use foo then fall back on bar. It will use foo and bar evenly | 15:35 |
| inc0 | http://logs.openstack.org/40/456140/15/check/gate-kolla-ansible-dsvm-deploy-ceph-ubuntu-binary-ubuntu-xenial-2-node-nv/89dd4e6/console.html <- that node had disk issues | 15:36 |
| *** LindaWang has quit IRC | 15:36 | |
| inc0 | http://logs.openstack.org/40/456140/15/check/gate-kolla-ansible-dsvm-deploy-ceph-centos-source-centos-7-2-node-nv/1930a51/console.html and that | 15:37 |
| inc0 | so rax-ord | 15:37 |
| openstackgerrit | Dima Kuznetsov proposed openstack-infra/project-config master: Dragonflow: generate HTML results in tempest job https://review.openstack.org/486664 | 15:38 |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul-jobs master: WIP: testing https://review.openstack.org/486665 | 15:38 |
| jeblair | inc0: i still can't find devstack-gate logs. do they get copied in those jobs? | 15:38 |
| fungi | inc0: jeblair: i think the "dsvm" names on those jobs are misleading. they don't seem to use devstack-gate at all | 15:38 |
| pabelanger | right | 15:39 |
| *** coolsvap has quit IRC | 15:39 | |
| inc0 | hmm...possible | 15:39 |
| jeblair | can someone explain like i'm 5 what we're debugging? | 15:39 |
| pabelanger | kolla needs to add d-g to there jobs, or setup ephemeral mounts themself | 15:40 |
| openstackgerrit | Merged openstack-infra/zuul-jobs master: Remove nodepool DIB specific logic https://review.openstack.org/485824 | 15:40 |
| fungi | well, before jumping to solutions, a clear outline of what the expectations for the jobs are would help me as well | 15:40 |
| inc0 | so if we're not devstack gates, which would make sense I guess | 15:40 |
| inc0 | we need to change name of jobs;) | 15:41 |
| jeblair | inc0: no, please use devstack-gate | 15:41 |
| *** ccamacho has left #openstack-infra | 15:41 | |
| inc0 | ok... | 15:41 |
| jeblair | it does *lots* of stuff to set up a node for testing | 15:41 |
| jeblair | like it took us 6 years to make that script | 15:41 |
| jeblair | save yourself 6 years and use it :) | 15:41 |
| jeblair | it doesn't have to run devstack | 15:41 |
| jeblair | i'm really sorry it has devstack in the name | 15:41 |
| inc0 | ok | 15:41 |
| inc0 | where do I define that? | 15:42 |
| jeblair | and we'll fix that in about a month and a half, but we can't change it right now :) | 15:42 |
| inc0 | since it's not just job name:) | 15:42 |
| jeblair | inc0: basically, we should set up a job like most of the other devstack-based jobs, and then override the 'gate hook' to do kolla things instead of devstack things | 15:42 |
| jeblair | lemme dig around for an example | 15:42 |
| inc0 | thank you jeblair | 15:43 |
| *** zara_the_lemur__ is now known as Zara | 15:43 | |
| *** kong has quit IRC | 15:45 | |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul-jobs master: Remove .txt suffix from tox logs https://review.openstack.org/486665 | 15:46 |
| *** Apoorva has quit IRC | 15:46 | |
| *** armax has quit IRC | 15:47 | |
| jeblair | inc0, clarkb, pabelanger: do we need the devstack-gate multinode networking setup for these jobs? | 15:48 |
| openstackgerrit | Jeremy Stanley proposed openstack-infra/infra-manual master: Extra disk space is not necessarily mounted https://review.openstack.org/486672 | 15:49 |
| fungi | jeblair: inc0: ^ clarification. our document was unfortunately misleading and made unmentioned assumptions | 15:49 |
| *** prateek has joined #openstack-infra | 15:50 | |
| openstackgerrit | Clark Boylan proposed openstack-infra/elastic-recheck master: Add query for bug 1706103 https://review.openstack.org/486673 | 15:51 |
| openstack | bug 1706103 in OpenStack-Gate "mkswap fails on cannot open /dev/xvde1" [Undecided,New] https://launchpad.net/bugs/1706103 | 15:51 |
| *** armax has joined #openstack-infra | 15:52 | |
| *** hashar is now known as hasharMeeting | 15:52 | |
| clarkb | jeblair: it depends on what the job is doing. Its possible they don't need it because their container stuff ships its own overlays (k8s will do this for example) | 15:52 |
| jeblair | inc0: this job looks like a good skeleton to work from: http://git.openstack.org/cgit/openstack-infra/project-config/tree/jenkins/jobs/devstack-gate.yaml#n1 | 15:53 |
| jeblair | inc0: gate-dg-hooks-dsvm -- it's a job that just exercises the hooks without doing anything, so it's easy to see the structure | 15:53 |
| jeblair | inc0: i'd start with that, delete the pre_ and post_ test hooks, and then replace gate_hook with some shell script to execute your actual job | 15:54 |
| inc0 | jeblair: so in our gate I should run everything as hooks instead of big shell like we do today? | 15:54 |
| jeblair | inc0: that will do all of the host setup (filesystems, mirrors, git repos, etc) and none of the devstack stuff. (the devstack/tempest stuff is what's normally in the gate_hook, so you'll be overriding that stuff) | 15:54 |
| jeblair | inc0: i would write just a tiny bit of shell to put in the gate hook which then invokes what you're doing now | 15:55 |
| jeblair | inc0: so like: function gate_hook {{ cd /somewhere/kolla; ./run-kolla-test.sh }} | 15:55 |
| inc0 | I'll look at it thanks | 15:55 |
| jeblair | keep that part simple, just get from devstack-gate to kolla asap :) | 15:56 |
| *** jamesden_ has joined #openstack-infra | 15:56 | |
| openstackgerrit | Clark Boylan proposed openstack-infra/devstack-gate master: Sync and partprobe after mkpart on ephemeral disk https://review.openstack.org/486676 | 15:56 |
| *** vhosakot has joined #openstack-infra | 15:56 | |
| inc0 | probably gonna do mounts before that to fix immediate issue, but I'll look at devstack gate as well | 15:56 |
| clarkb | this d-g change is semi related (noticed due to failres in e-r uncategorized queue) | 15:56 |
| *** nicolasbock has quit IRC | 15:57 | |
| openstackgerrit | Jeremy Stanley proposed openstack-infra/infra-manual master: Extra disk space is not necessarily mounted https://review.openstack.org/486672 | 15:57 |
| *** funzo has joined #openstack-infra | 15:57 | |
| *** funzo has quit IRC | 15:57 | |
| *** funzo has joined #openstack-infra | 15:57 | |
| *** jamesdenton has quit IRC | 15:57 | |
| jeblair | inc0: the only other sort of general-purpose thing that devstack-gate does but this *won't* do is setup a multinode network overlay. if you find you need that, we can look deeper into that. but that's tied up with the more devstacky parts of this, so it's not as straightforward. | 15:57 |
| fungi | another thing d-g does is try and set up a reasonable amount of swap memory (not sure if that got mentioned already) | 15:58 |
| inc0 | multinode network overlay as in tunnels | 15:58 |
| inc0 | ? | 15:58 |
| *** rcernin has joined #openstack-infra | 15:58 | |
| clarkb | inc0: yes so that multinode test nodes can have a shared layer 2 regardless of how they were actually booted in $cloud | 15:58 |
| fungi | inc0: yes, a layer 2 "tunnel" between nodes in the multi-node group so that they can be guaranteed layer 2 connectivity and the ability to use whatever ip addresses they want there | 15:58 |
| clarkb | inc0: this makes it easy to be able to ssh from a VM on one instance to a VM on another instance for example | 15:59 |
| inc0 | alternatively we can wait for zuulv3 and rework our gates to ansible bottoms-up right? | 15:59 |
| clarkb | since we can give openstack control of this virtual networking | 15:59 |
| inc0 | yeah we could use that to for vm network, but that's lesser priority | 16:00 |
| inc0 | question, will zuulv3 jobs look differently than jobs today? | 16:00 |
| *** Douhet has quit IRC | 16:01 | |
| inc0 | I mean, I know zuulv3 will be backwards compatible, but will zuulv3 native jobs be different that what's there today? | 16:01 |
| *** Douhet has joined #openstack-infra | 16:01 | |
| clarkb | inc0: there will be differences, distribution of encrypted secrets, nodepool node reservation api, ability to use ansible directly in job, etc | 16:02 |
| inc0 | what I'm asking is, should we work on d-g today or wait for zuulv3 and do d-g+zuulv3 changes | 16:02 |
| *** baoli has quit IRC | 16:02 | |
| fungi | inc0: https://git.openstack.org/cgit/openstack-infra/openstack-zuul-jobs/tree/ | 16:02 |
| pabelanger | we still need to re-write devstack-gate into ansible however. Still working on refactoring tox jobs | 16:02 |
| *** baoli has joined #openstack-infra | 16:02 | |
| *** lucasagomes is now known as lucas-afk | 16:04 | |
| clarkb | inc0: I think you'll want to use d-g either way. Only the manner of consumption will change. So probably makes sense to just go ahead and use d-g now if it will make your jobs more reliabel | 16:04 |
| *** armax has quit IRC | 16:04 | |
| fungi | inc0: our goal with the migration and maintaining some backward-compatibility is that nobody should wait to write jobs just so they can make then zuulv3-native (though they may be waiting to write jobs which depend on v3-specific features) | 16:05 |
| *** ijw has joined #openstack-infra | 16:05 | |
| fungi | anything that's done by executing literal shell instructions, which is nearly everything in our jjb config, can still be done by putting the literal shell into playbooks/roles | 16:06 |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul-jobs master: WIP: do not merge https://review.openstack.org/486679 | 16:06 |
| openstackgerrit | Merged openstack-infra/elastic-recheck master: Add query for bug 1706103 https://review.openstack.org/486673 | 16:06 |
| openstack | bug 1706103 in OpenStack-Gate "mkswap fails on cannot open /dev/xvde1" [Undecided,New] https://launchpad.net/bugs/1706103 | 16:06 |
| fungi | but for a lot of the common job patterns and macros we'll likely have them rewritten in equivalent ansible instead of literal shell script | 16:07 |
| *** baoli has quit IRC | 16:07 | |
| *** andreas_s has quit IRC | 16:08 | |
| *** egonzalez has quit IRC | 16:08 | |
| inc0 | clarkb fungi thing is, I don't want to make small revolution to our gates this close to release | 16:09 |
| *** ijw has quit IRC | 16:10 | |
| *** bhavik1 has joined #openstack-infra | 16:10 | |
| *** kgiusti has joined #openstack-infra | 16:10 | |
| inc0 | so I'll wait with d-g to after release anyway, and that, assuming you'll manage to land zuulv3 before PTG, will be pretty close | 16:10 |
| *** jascott1 has joined #openstack-infra | 16:10 | |
| inc0 | so might as well do it all since I'll need zuulv3 features anyway | 16:10 |
| fungi | inc0: makes sense. i'm saying don't wait to do it because of zuul v3. if you have other reasons to wait, then those are other reasons | 16:11 |
| inc0 | yeah, I agree | 16:11 |
| fungi | and at least now you know where the issue lies, and what the "right" solution is, and can plan accordingly | 16:11 |
| inc0 | I just wanted to make sure if there is anything in zuulv3 that would justify changes to gate, and I understand that is the case | 16:12 |
| inc0 | yeah | 16:12 |
| *** chlong__ has joined #openstack-infra | 16:13 | |
| *** chlong_ has quit IRC | 16:13 | |
| *** EricGonczer_ has joined #openstack-infra | 16:13 | |
| fungi | but i share jeblair's opinion as far as devstack-gate goes there. if you try to make an openstack integration testing framework from scratch, you get to make many of the same mistakes and reexperience most of the problems we solved over the years with devstack-gate | 16:14 |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul-jobs master: WIP: do not merge https://review.openstack.org/486679 | 16:14 |
| fungi | and you're unlikely to get a lot of sympathy or help in re-solving problems we already solved once | 16:15 |
| inc0 | oh I don't want to make this from scrach | 16:15 |
| inc0 | I'll just fix immediate issue | 16:15 |
| inc0 | and work on d-g as long term | 16:15 |
| pabelanger | fungi: ++ | 16:15 |
| inc0 | or rather d-g + zuulv3 | 16:15 |
| inc0 | probably worth a goal for Q | 16:15 |
| *** claudiub has quit IRC | 16:15 | |
| inc0 | I was sure we're using dvsm today;) | 16:16 |
| fungi | the mini-project pabelanger mentioned around ansiblification of devstack-gate will hopefully mean that we can provide a more decomposed set of modular solutions broken out of d-g in the future, so you can mix and match in your own jobs without needing the entire framework | 16:17 |
| fungi | but as demonstrated, you can hollow out most of the logic in it right now by substituting your own hooks | 16:18 |
| jeblair | pabelanger: re 486665 -- does our logserver handle serving ".log" files as text now? | 16:19 |
| pabelanger | jeblair: it worked with chrome browser, let me find the apache rewrite code | 16:19 |
| *** bhavik1 has quit IRC | 16:20 | |
| *** jascott1 has quit IRC | 16:20 | |
| pabelanger | jeblair: Oh, looks like you added it: https://review.openstack.org/#/c/330119/ | 16:21 |
| *** jascott1 has joined #openstack-infra | 16:21 | |
| jeblair | neat. glad someone's on the ball. | 16:21 |
| pabelanger | :) | 16:21 |
| fungi | send a thank-you beer back to your past self! | 16:21 |
| *** jamesden_ has quit IRC | 16:21 | |
| openstackgerrit | Merged openstack-infra/zuul-jobs master: Remove .txt suffix from tox logs https://review.openstack.org/486665 | 16:22 |
| jeblair | "future me is going to thank me for this someday" | 16:22 |
| *** ykarel has quit IRC | 16:23 | |
| clarkb | fwiw most of the setupy type stuff in d-g is in ansible playbooks now. So it is also theoretically possible to consume it that way today. The big missing piece is the network overlay stuff | 16:23 |
| *** jamesdenton has joined #openstack-infra | 16:23 | |
| openstackgerrit | Paul Belanger proposed openstack-infra/openstack-zuul-jobs master: Create openstack-py35 job with upper-constraints https://review.openstack.org/483987 | 16:23 |
| clarkb | (we don't have any examples of how something consuming d-g setup via ansible would look today that I know of though) | 16:23 |
| jeblair | clarkb: true. i'd caution against that because i don't think we want to expose that as a stable api for jobs yet. | 16:23 |
| jeblair | clarkb: ie, i'd like us to be able to forklift that into the openstack-zuul-jobs repo without having to have a deprecation period. :) | 16:24 |
| fungi | right, will get ugly if/when we reshuffle those out of the d-g repo | 16:24 |
| *** vhosakot has quit IRC | 16:24 | |
| pabelanger | adding the disk fixup stuff as a role could be a log hanging fruit for somebody | 16:24 |
| *** jascott1_ has joined #openstack-infra | 16:24 | |
| *** jascott1 has quit IRC | 16:24 | |
| *** jascott1_ has quit IRC | 16:25 | |
| *** jascott1 has joined #openstack-infra | 16:26 | |
| openstackgerrit | Sam Betts proposed openstack-infra/os-loganalyze master: Repair htmlify-log.py command https://review.openstack.org/383812 | 16:27 |
| openstackgerrit | Merged openstack-infra/infra-manual master: Extra disk space is not necessarily mounted https://review.openstack.org/486672 | 16:27 |
| clarkb | ah but fix_disk_layout is in setup_workspace not part of setup_host so it isn't ansible either | 16:28 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Allow loading additional variables file for site config https://review.openstack.org/447734 | 16:30 |
| *** ykarel has joined #openstack-infra | 16:31 | |
| *** e0ne has quit IRC | 16:31 | |
| *** jascott1 has quit IRC | 16:31 | |
| jeblair | clarkb: setup_<meaningless word> followed by setup_<other meaningless word> :) | 16:31 |
| *** jascott1 has joined #openstack-infra | 16:32 | |
| jeblair | i'm sure we could probably come up with a "reason" mounting disks is setting up a workspace rather than a host. but i'm guessing those words are just not sufficient to describe what those functions are actually doing after so much evolution. :) | 16:32 |
| *** jascott1 has quit IRC | 16:33 | |
| *** baoli has joined #openstack-infra | 16:33 | |
| *** jascott1 has joined #openstack-infra | 16:35 | |
| *** EricGonczer_ has quit IRC | 16:39 | |
| *** chlong_ has joined #openstack-infra | 16:40 | |
| openstackgerrit | Clark Boylan proposed openstack-infra/system-config master: Update the keystone files in logstash worker config https://review.openstack.org/486691 | 16:41 |
| *** chlong__ has quit IRC | 16:41 | |
| clarkb | sdague: mtreinish lbragstad ^ review on that would be great to make sure I didn't miss something re keystone log file indexing | 16:41 |
| *** Apoorva has joined #openstack-infra | 16:42 | |
| lbragstad | clarkb: looks ok to me | 16:42 |
| *** ykarel has quit IRC | 16:42 | |
| *** EricGonczer_ has joined #openstack-infra | 16:44 | |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul-jobs master: WIP: do not merge https://review.openstack.org/486679 | 16:44 |
| openstackgerrit | David Shrewsbury proposed openstack-infra/zuul feature/zuulv3: WIP: Implement autohold https://review.openstack.org/486692 | 16:45 |
| clarkb | mtreinish: good point I'll push a new patchset | 16:45 |
| clarkb | mtreinish: actually | 16:46 |
| clarkb | mtreinish: I'm not sure how systemd/journald + grenade will work out, eg will we have split logs at all? | 16:46 |
| clarkb | mtreinish: maybe its best to wait on that just so that we don't add in things that will never pan out | 16:46 |
| mtreinish | clarkb: ok, that's fair. I don't have an answer for what the grenade log story with systemd either | 16:47 |
| mtreinish | clarkb: then I'm +1 | 16:47 |
| mtreinish | sdague: ^^^ will we need to change something for grenade when we branch pike? | 16:47 |
| clarkb | mtreinish: sdague if we don't change anything I think we get all the logs in a single file | 16:48 |
| clarkb | which is probably fine? | 16:48 |
| *** jascott1 has quit IRC | 16:49 | |
| *** ykarel has joined #openstack-infra | 16:49 | |
| mtreinish | clarkb: well it might make it kinda confusing to see the upgrade boundary | 16:49 |
| sdague | honestly, a single set of logs would probably be easier to handle, the important thing is going to be making the boundary clear | 16:50 |
| openstackgerrit | Paul Belanger proposed openstack-infra/openstack-zuul-jobs master: Update tox.ini to use linters envlist https://review.openstack.org/486695 | 16:52 |
| openstackgerrit | Paul Belanger proposed openstack-infra/openstack-zuul-jobs master: Create openstack-py35 job with upper-constraints https://review.openstack.org/483987 | 16:52 |
| *** armax has joined #openstack-infra | 16:52 | |
| *** ykarel has quit IRC | 16:53 | |
| *** wolverineav has quit IRC | 16:53 | |
| *** ykarel has joined #openstack-infra | 16:54 | |
| clarkb | sdague: mtreinish should be simple to ensure each service logs a "I just started" message (or similar) to make the boundary clear | 16:54 |
| clarkb | I think most services already do that in the form of loading their config? | 16:54 |
| sdague | clarkb: they do | 16:55 |
| sdague | the oslo service startup is pretty distinctive | 16:55 |
| *** vhosakot has joined #openstack-infra | 16:56 | |
| *** derekh has quit IRC | 16:56 | |
| dirk | clarkb: fungi : sorry, lost contact on the pip /wheels discussion.. I'm certainly fine to postpone the wheels topic if we could get a workaround in place to have the opensuse job no longer bitrotting .. e.g. https://review.openstack.org/#/c/485985 should be doing that | 16:57 |
| *** ykarel has quit IRC | 16:58 | |
| *** felipemonteiro_ has quit IRC | 16:59 | |
| *** EricGonczer_ has quit IRC | 17:02 | |
| *** ralonsoh has quit IRC | 17:02 | |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul feature/zuulv3: Log an extra blank line to get space after each skip https://review.openstack.org/486698 | 17:04 |
| *** chlong_ has quit IRC | 17:08 | |
| *** ricardoas has joined #openstack-infra | 17:08 | |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul-jobs master: WIP: do not merge https://review.openstack.org/486679 | 17:09 |
| *** harlowja has joined #openstack-infra | 17:09 | |
| *** chlong_ has joined #openstack-infra | 17:10 | |
| *** ricardoas has quit IRC | 17:14 | |
| jeblair | clarkb: do you need any gerrit 2.13 testing help from me? | 17:14 |
| *** ricardoas has joined #openstack-infra | 17:14 | |
| *** efoley has quit IRC | 17:15 | |
| *** ramishra has quit IRC | 17:16 | |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul-jobs master: Create tox_environment_defaults variable for tox based jobs https://review.openstack.org/486679 | 17:17 |
| *** flepied has quit IRC | 17:17 | |
| *** trown is now known as trown|lunch | 17:18 | |
| *** jascott1 has joined #openstack-infra | 17:20 | |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul-jobs master: Create tox_environment_defaults variable for tox based jobs https://review.openstack.org/486679 | 17:20 |
| *** baoli has quit IRC | 17:20 | |
| *** baoli has joined #openstack-infra | 17:23 | |
| openstackgerrit | Ashwin Agate proposed openstack-infra/project-config master: Fix devstack py35 functional test for monasca-transform https://review.openstack.org/485756 | 17:25 |
| *** electrofelix has quit IRC | 17:27 | |
| clarkb | jeblair: I don't think I do right now other than individuals just checking that new gerrit doesn't break personal workflows. I'm back home now so will be working on getting noopy zuuls deployed (v2.5 and v3) | 17:27 |
| *** rbrndt has quit IRC | 17:28 | |
| clarkb | my plan for that is to just run them out of virtualenvs in my homedir on review-dev to start | 17:29 |
| clarkb | infra-root reviews on https://review.openstack.org/#/c/486691/ will help us be able to better track keystone related bugs with e-r | 17:29 |
| fungi | jeblair: actually, if you want to lend some mysql expertise, we seem to have experienced a regression in support for extended codepoints in review comments on review-dev | 17:31 |
| jeblair | clarkb: ok. let me know if you want help with any of that. you might consider using the 'git' zuulv3 driver for that, so you can put some review-dev zuulv3 config in a local git repo rather than bootstrapping it in gerrit itself. | 17:31 |
| jeblair | fungi: oh dear. my understanding of that is minimal, but it's not zero. maybe i can help push it slightly forward while we stall for time until mordred's back. :) | 17:32 |
| clarkb | I'll likely dig into it in force tomorrow. Thanks for tip on git driver. Today is catch up on email and all that | 17:32 |
| *** sambetts is now known as sambetts|afk | 17:33 | |
| openstackgerrit | Ashwin Agate proposed openstack-infra/project-config master: Fix devstack py35 functional test for monasca-transform https://review.openstack.org/485756 | 17:33 |
| fungi | not sure if something happened with the upgrade-related db migrations or if it's a behavior change in gerrit but it now seems to store a lot of characters (including some i tested which are 3-byte in utf-8) as a literal "?" in the database backend (i double-checked with mysqlclient) | 17:33 |
| clarkb | fungi: you said you observed that behavior on 2.11 and were checking if 2.13 was different? or is that new to 2.13? | 17:34 |
| *** eharney has quit IRC | 17:35 | |
| *** baoli has quit IRC | 17:37 | |
| *** thorst_ has quit IRC | 17:38 | |
| *** felipemonteiro has joined #openstack-infra | 17:43 | |
| *** felipemonteiro_ has joined #openstack-infra | 17:44 | |
| *** thorst has joined #openstack-infra | 17:44 | |
| *** baoli has joined #openstack-infra | 17:45 | |
| *** felipemonteiro has quit IRC | 17:47 | |
| *** HeOS has quit IRC | 17:49 | |
| *** thorst has quit IRC | 17:49 | |
| *** baoli has quit IRC | 17:50 | |
| *** eharney has joined #openstack-infra | 17:50 | |
| fungi | clarkb: the behavior waynr reported was a 500 error trying to insert a "see no evil monkey" glyph in a review comment | 17:50 |
| *** yamahata has quit IRC | 17:50 | |
| fungi | on production review.o.o | 17:50 |
| *** iyamahat has quit IRC | 17:51 | |
| clarkb | fungi: and now it isn't an error but replaces with "?" ? | 17:51 |
| fungi | so we were trying to see if it could be replicated on review-dev but it's doing substitutions there | 17:51 |
| clarkb | see no evil monkey is 4 byte so that makes sense on production | 17:51 |
| fungi | when it clearly wasn't at one point because we have other reviews from the 2.11 timeframe with utf-8 (snowman et cetera) | 17:51 |
| *** flepied has joined #openstack-infra | 17:52 | |
| fungi | 500 error on 4-byte utf-8 codepoints is a known behavior for our production deployment? for some reason i thought we had it working there | 17:52 |
| fungi | but i could be conflating that with etherpad | 17:52 |
| clarkb | fungi: neither of our mysql deployments support 4 byte utf8 because the mysql version behind those servers is too old for the utf8mb4 encoding | 17:53 |
| fungi | aha, okay so the only surprising bit is the (recent?) behavior change on review-dev then | 17:54 |
| fungi | maybe we should plan to incorporate a db upgrade into the gerrit upgrade plan while we're at it? (or as a separate maintenance?) | 17:54 |
| clarkb | if/when we upgrade to newer mysql we can update the tablse to utf8mb4 fairly safely since its a strict superset of current data | 17:54 |
| *** felipemonteiro_ has quit IRC | 17:54 | |
| *** felipemonteiro has joined #openstack-infra | 17:54 | |
| clarkb | I think a separate maintenance would be best | 17:54 |
| clarkb | fungi: what characters did you try on review-dev? | 17:55 |
| fungi | seems like the maintenance for that would be pretty quick. mostly just the mysqldump/source time | 17:55 |
| fungi | clarkb: snowman, some kanji and some (confirmed 3-byte) kana | 17:55 |
| *** wolverineav has joined #openstack-infra | 17:56 | |
| clarkb | ya just confirmed snowman becomes ? | 17:56 |
| clarkb | https://bugs.chromium.org/p/gerrit/issues/detail?id=1499 perhaps misconfigured? | 17:56 |
| fungi | also some non-ascii which fits in the latin-1 codepage and that came through fine | 17:56 |
| clarkb | but I wouldn't expect my.cnf to affect us that way | 17:57 |
| clarkb | hrm maybe review-dev didn't get its db tables updated ? | 17:57 |
| clarkb | or the new account review db thing is not utf8? | 17:57 |
| clarkb | that would be annoying | 17:58 |
| *** claudiub has joined #openstack-infra | 17:58 | |
| AJaeger | project-config cores, here's a one line cleanup review for https://review.openstack.org/485110 - could I get a second +2, please? | 17:59 |
| *** wolverineav has quit IRC | 17:59 | |
| clarkb | AJaeger: will that update the subnodes config too? | 18:01 |
| clarkb | yup confirmed it should affect all nodes local.conf | 18:04 |
| clarkb | since it is applied before we do the split between sub and primary in the localrc generation logic | 18:04 |
| *** e0ne has joined #openstack-infra | 18:04 | |
| fungi | clarkb: apparently i was able to snowman a review-dev change comment as recently as december, so something definitely changed on upgrade: https://review-dev.openstack.org/5335 | 18:05 |
| *** dtantsur is now known as dtantsur|afk | 18:05 | |
| AJaeger | thanks, clarkb | 18:05 |
| clarkb | fungi: that likely points at the new database on 2.13, we should inspect the tables and see if gerrit is overriding defaults there or something | 18:05 |
| clarkb | fungi: etherpad indicates I definitely altered the db to be utf8 by default, but gerrit could be creating the tables with an override maybe? | 18:06 |
| *** jbernard has quit IRC | 18:06 | |
| clarkb | fungi: or maybe we restored a database at some point and lost a latin1 to utf8 transition? zaro do you know? | 18:06 |
| fungi | clarkb: more interesting is that it's a write-related behavior change, because review-dev is still serving that snowman intact | 18:07 |
| clarkb | huh | 18:07 |
| fungi | but it translates newer ones to "?" when writing to the same table | 18:08 |
| *** gtrxcb has quit IRC | 18:08 | |
| clarkb | maybe we lack some new gerrit side config that says "it is safe to utf8" | 18:08 |
| bswartz | Is lib/neutron-legacy required by devstack or is there a way to avoid using it? | 18:08 |
| clarkb | bswartz: you'll likely get better answer to that question in #openstack-qa (devstack is qa project) or in #openstack-neutron (I think neutron devs did a bunch of the work to get off of the legacy scripts) | 18:09 |
| fungi | bswartz: i think that's related to the lib/neutron rewrite, but probably better to ask in #openstack-qa. it could be cruft at this stage, i don't know | 18:09 |
| bswartz | k ty | 18:09 |
| *** ricardoas has left #openstack-infra | 18:11 | |
| *** iyamahat has joined #openstack-infra | 18:11 | |
| *** jbernard has joined #openstack-infra | 18:11 | |
| *** thorst has joined #openstack-infra | 18:12 | |
| openstackgerrit | Merged openstack-infra/project-config master: networking-ovn: Remove DEVSTACK_SUBNODE_CONFIG https://review.openstack.org/485110 | 18:14 |
| *** psachin has quit IRC | 18:16 | |
| *** rcernin has quit IRC | 18:16 | |
| *** prateek has quit IRC | 18:17 | |
| *** trown|lunch is now known as trown | 18:17 | |
| *** dizquierdo has quit IRC | 18:18 | |
| *** rbrndt has joined #openstack-infra | 18:20 | |
| *** florianf has quit IRC | 18:22 | |
| *** felipemonteiro_ has joined #openstack-infra | 18:24 | |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul-jobs master: WIP: Move subunit processing https://review.openstack.org/485840 | 18:25 |
| *** felipemonteiro has quit IRC | 18:27 | |
| *** EricGonczer_ has joined #openstack-infra | 18:29 | |
| *** dimak has joined #openstack-infra | 18:29 | |
| openstackgerrit | John Fulton proposed openstack-infra/tripleo-ci master: Create a loop device in multinode subnodes https://review.openstack.org/484963 | 18:30 |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul-jobs master: WIP: Move subunit processing https://review.openstack.org/485840 | 18:30 |
| *** yamahata has joined #openstack-infra | 18:30 | |
| *** gtrxcb has joined #openstack-infra | 18:31 | |
| *** baoli has joined #openstack-infra | 18:32 | |
| *** cody-somerville has joined #openstack-infra | 18:33 | |
| *** watersoul has quit IRC | 18:34 | |
| *** baoli has quit IRC | 18:37 | |
| *** EricGonczer_ has quit IRC | 18:41 | |
| openstackgerrit | Merged openstack-infra/tripleo-ci master: Nic config for containerized undercloud https://review.openstack.org/442098 | 18:41 |
| *** EricGonczer_ has joined #openstack-infra | 18:43 | |
| *** dougwig has joined #openstack-infra | 18:44 | |
| *** watersoul has joined #openstack-infra | 18:45 | |
| *** jamesden_ has joined #openstack-infra | 18:49 | |
| *** jamesdenton has quit IRC | 18:49 | |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul-jobs master: WIP: Move subunit processing https://review.openstack.org/485840 | 18:51 |
| *** camunoz has quit IRC | 18:54 | |
| openstackgerrit | Emilien Macchi proposed openstack-infra/project-config master: Introduce experimental TripleO scenario008 https://review.openstack.org/486726 | 18:54 |
| *** claudiub has quit IRC | 18:56 | |
| *** hasharMeeting is now known as hasharDinner | 18:56 | |
| *** EricGonczer_ has quit IRC | 18:57 | |
| *** dhajare has joined #openstack-infra | 18:59 | |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul-jobs master: WIP: Move subunit processing https://review.openstack.org/485840 | 19:00 |
| *** annegentle has quit IRC | 19:00 | |
| *** EricGonczer_ has joined #openstack-infra | 19:00 | |
| *** dhajare has quit IRC | 19:03 | |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul-jobs master: WIP: Move subunit processing https://review.openstack.org/485840 | 19:09 |
| openstackgerrit | John Fulton proposed openstack-infra/tripleo-ci master: Create a loop device in multinode subnodes https://review.openstack.org/484963 | 19:14 |
| *** thorst is now known as thorst_afk | 19:15 | |
| *** rossella__ has quit IRC | 19:16 | |
| *** rossella__ has joined #openstack-infra | 19:17 | |
| *** EricGonczer_ has quit IRC | 19:19 | |
| *** e0ne has quit IRC | 19:21 | |
| openstackgerrit | David Shrewsbury proposed openstack-infra/zuul feature/zuulv3: WIP: Implement autohold https://review.openstack.org/486692 | 19:21 |
| *** baoli has joined #openstack-infra | 19:22 | |
| *** hasharDinner has quit IRC | 19:23 | |
| *** wolverineav has joined #openstack-infra | 19:28 | |
| *** reed has quit IRC | 19:31 | |
| *** reed has joined #openstack-infra | 19:32 | |
| *** camunoz has joined #openstack-infra | 19:33 | |
| *** csomerville has joined #openstack-infra | 19:37 | |
| *** cody-somerville has quit IRC | 19:40 | |
| *** gtrxcb has quit IRC | 19:41 | |
| openstackgerrit | Dirk Mueller proposed openstack-infra/project-config master: Adding more openSUSE 42.3 variants to existing 42.2 jobs https://review.openstack.org/484499 | 19:44 |
| openstackgerrit | Dirk Mueller proposed openstack/os-client-config master: Manually sync with g-r https://review.openstack.org/484166 | 19:46 |
| *** danpawlik has quit IRC | 19:48 | |
| *** danpawlik has joined #openstack-infra | 19:51 | |
| *** jcoufal has quit IRC | 19:54 | |
| *** HeOS has joined #openstack-infra | 19:56 | |
| *** ldnunes has quit IRC | 19:58 | |
| openstackgerrit | Julia Kreger proposed openstack-infra/project-config master: Remove non-ironic governed gerrit annoucements https://review.openstack.org/486741 | 19:59 |
| pabelanger | mordred: mtreinish: clarkb: jeblair: fungi: Any reason that subunit2html (os-testr) is not a test-requirements.txt dependency in jobs that want to create testr_results.html ? | 20:00 |
| pabelanger | today we install os-testr into a virutualenv at DIB creation | 20:01 |
| *** EricGonczer_ has joined #openstack-infra | 20:01 | |
| *** e0ne has joined #openstack-infra | 20:01 | |
| *** ekhugen has quit IRC | 20:03 | |
| mtreinish | pabelanger: because it's not used by every project that runs tests | 20:03 |
| *** zz_ja has quit IRC | 20:03 | |
| mtreinish | it's a postprocessing step step like stackviz | 20:03 |
| mtreinish | it just happens to share a python package with a test runner and an output filter used by most projects | 20:05 |
| *** patriciadomin has quit IRC | 20:05 | |
| *** mdrabe has quit IRC | 20:05 | |
| *** ihrachys_ has joined #openstack-infra | 20:07 | |
| *** rossella__ has quit IRC | 20:07 | |
| *** Apoorva_ has joined #openstack-infra | 20:08 | |
| *** Apoorva_ has quit IRC | 20:09 | |
| *** ihrachys has quit IRC | 20:09 | |
| *** Apoorva_ has joined #openstack-infra | 20:09 | |
| *** mamitchl has joined #openstack-infra | 20:09 | |
| *** mdrabe has joined #openstack-infra | 20:09 | |
| *** patriciadomin has joined #openstack-infra | 20:09 | |
| *** ekhugen has joined #openstack-infra | 20:09 | |
| *** Apoorva_ has quit IRC | 20:10 | |
| *** zz_ja has joined #openstack-infra | 20:10 | |
| *** rossella__ has joined #openstack-infra | 20:10 | |
| *** Apoorva_ has joined #openstack-infra | 20:10 | |
| *** ldnunes has joined #openstack-infra | 20:10 | |
| *** Apoorva has quit IRC | 20:11 | |
| clarkb | pabelanger: fungi if https://review.openstack.org/#/c/486691/ could get reviews that would be helpful. For indexing keystone logs | 20:12 |
| *** EricGonczer_ has quit IRC | 20:14 | |
| *** rossella__ has quit IRC | 20:16 | |
| clarkb | fungi: I'm going to start poking at the new db gerrit made now | 20:17 |
| fungi | thanks clarkb! | 20:17 |
| *** rossella__ has joined #openstack-infra | 20:18 | |
| pabelanger | mtreinish: well, any job that has a .testrepository file today we run subunit2html. It seems like we could add os-testr as a dependency for jobs using .testrepository | 20:20 |
| clarkb | fungi: the new db appears to be utf8 as expected based on show table status output | 20:21 |
| fungi | clarkb: so presumably a gerrit behavior change then | 20:21 |
| pabelanger | mtreinish: I'm in the process of converting run-tox.sh into ansible playbooks, and trying to break out subunit2html into its own role | 20:21 |
| clarkb | fungi: ya that is what I am thinking | 20:21 |
| fungi | clarkb: the latest keystone service log path is logs/screen-keystone.txt? what does "screen" mean in this context? i thought latest devstack wasn't using screen by default any longer? | 20:21 |
| clarkb | fungi: the reviewdb tables are all utf as well | 20:21 |
| clarkb | fungi: we kept screen in the filenames for backward compat so that we wouldn't have to rewrite all the log parsing tools and configs | 20:22 |
| fungi | ahh, okay | 20:22 |
| jeblair | pabelanger: can you elaborate on what you mean by ""add os-testr as a dependency for jobs using .testrepository | 20:22 |
| jeblair | 20:21 < clarkb> fungi: the new db appears to be utf8 as expected based on show table status output | 20:22 |
| jeblair | paste fail | 20:22 |
| jeblair | but the first line was still correct :) | 20:22 |
| jeblair | pabelanger: i don't know what "dependency for jobs" means in this context | 20:23 |
| mtreinish | jeblair: I was assuming he meant add os-testr to the test-requirements for every project. But I could be mistaken in how I was reading it | 20:23 |
| pabelanger | jeblair: sure, adding os-testr to test-requirements.txt | 20:23 |
| clarkb | fungi: eventually when we stop having screen on old branches I think we will want to tackle addressing that. But right now with mixed content it seems easiest to pretend we still have screen for log processing | 20:23 |
| jeblair | pabelanger: okay, that's adding as a dependency to a project -- are you suggesting adding as a dependency to every project in our system? | 20:24 |
| pabelanger | jeblair: yes? What I am trying to determine is, when we should be running subunit2html for our tox role: http://git.openstack.org/cgit/openstack-infra/zuul-jobs/tree/roles/tox/tasks/main.yaml#n28 because, it is possible we could have tox jobs that don't want to gerenate it | 20:25 |
| fungi | to put it in context, that would be like adding zuul to test-requirements.txt of most projects because their jobs use zuul-cloner to checkout repos | 20:26 |
| jeblair | fungi: apt analogy | 20:26 |
| pabelanger | fungi: sure, this would be non-openstack projects using zuul-jobs tox role | 20:26 |
| jeblair | pabelanger: we made subunit2html so that we could stop using the jenkins xunit plugin because it had global locks that were preventing us from scaling. in other words, we did it to achieve feature parity with jenkins. | 20:27 |
| clarkb | digging in gerrit git history I'm not really seeing anything (at least immediately) that would impact unicode handling. Perhaps it is in the db driver itself? | 20:27 |
| fungi | might make sense to move the subunit2html filter calls to a separate playbook? or whatever the new equivalent of a jjb macro is (i'm still a little fuzzy on the difference between ansible roles, playbooks and plays) | 20:27 |
| pabelanger | jeblair: okay. Should we be using it for zuulv3? | 20:28 |
| jeblair | pabelanger: it's a feature of the ci system that we chose to provide to openstack projects. i don't know that all the projects would agree that generating html output from their tests is something that should happen in all cases (even if they do agree that it's nice to have in the ci system). | 20:28 |
| jeblair | pabelanger: absolutely! the experience is terrible without it. :) | 20:28 |
| jeblair | (to be fair, i don't think anyone has ever asked projects if they would like to opt-in to automatic html generation when they run their tests locally) | 20:29 |
| jeblair | (but i'm not optimistic the answer would be 100% yes) | 20:29 |
| fungi | pabelanger: it's what creates pages like http://logs.openstack.org/43/484443/1/gate/gate-cinder-python35/259237d/testr_results.html.gz which seems to me like a useful feature | 20:30 |
| jeblair | pabelanger: to directly answer the question at hand -- i think we should run subunit2html for all tox jobs in the world | 20:30 |
| jeblair | pabelanger: it is a gift from the openstack community to the world of zuul users running tox jobs. :) | 20:30 |
| jeblair | pabelanger: er, tox+testr i should say | 20:31 |
| jeblair | pabelanger: so basically, i think our opinion should be: if you are using zuul to run tox+testr, the default python unit test job in zuul-jobs will give you nice html output | 20:31 |
| *** baoli has quit IRC | 20:31 | |
| jeblair | i don't think there is a good reason to consider that an openstack-only thing | 20:32 |
| pabelanger | jeblair: right, it would make sense for test-requirements.txt to have os-testr. And if we find it, generate html | 20:33 |
| jeblair | pabelanger: no i disagree. we should not ask folks to add os-testr to test-requirements.txt | 20:33 |
| pabelanger | k | 20:34 |
| jeblair | i realize that's inconvenient for us because we now have to figure out how to install this thing. but it's not that much different than bindep in that regard. | 20:35 |
| clarkb | fungi: I am beginning to wonder if the old mysql install on review-dev for the on host database server is crossing some stream between jdbc and trove hosted mysql when determining connection details | 20:35 |
| *** wolverineav has quit IRC | 20:35 | |
| jeblair | clarkb, fungi: do they have the same trove profile thingy? | 20:36 |
| *** wolverineav has joined #openstack-infra | 20:36 | |
| jeblair | clarkb, fungi: dev / prod | 20:36 |
| clarkb | jeblair: yes I think they do, in order to address the connection time outs, but I will double check | 20:36 |
| jeblair | oh, right, 'worked in december' probably takes care of that. but still worth a sanity check. | 20:36 |
| clarkb | the jdbc connection apparently tries to automatically determine this stuff if not explicity set, and system my.cnf values can be used to do that (we have an /etc/mysql/my.cnf from when we ran a server there) | 20:37 |
| openstackgerrit | Merged openstack-infra/system-config master: Update the keystone files in logstash worker config https://review.openstack.org/486691 | 20:37 |
| fungi | argh, that's a terrible side effect behavior | 20:37 |
| *** wolverineav has quit IRC | 20:40 | |
| clarkb | jeblair: confirmed they both have the "sanity" configuration applied to them | 20:41 |
| *** rockyg has joined #openstack-infra | 20:41 | |
| jeblair | clarkb: review-dev and review have the same my.cnf | 20:42 |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul-jobs master: WIP: Move subunit processing into fetch-testr-output https://review.openstack.org/485840 | 20:43 |
| clarkb | jeblair: that makes gerrit 2.13 seem more a likely culprit then | 20:44 |
| *** baoli has joined #openstack-infra | 20:44 | |
| clarkb | also database appears to be storing a literal '?' so it isn't some presentation layer encoding problem I don't think | 20:44 |
| *** baoli has quit IRC | 20:45 | |
| openstackgerrit | Boris Pavlovic proposed openstack-infra/project-config master: Actually Disable Swift Service in Rally DSVM jobs (when needed) https://review.openstack.org/486749 | 20:45 |
| *** baoli has joined #openstack-infra | 20:45 | |
| clarkb | `select message from reviewdb.change_messages where change_id = 107957 and uuid = 'df33271e_c13ae2cb';` will show a literal ? that was intended to be a snowman | 20:46 |
| *** eharney has quit IRC | 20:47 | |
| *** e0ne has quit IRC | 20:50 | |
| clarkb | one thought is we provide an explicit jdbc url now for the accountpatchreviewdb perhaps that is affecting some global jdbc state? (accountpatchreviewdb doesn't appear to have anything to do with the actual comment content) | 20:51 |
| *** Goneri has quit IRC | 20:52 | |
| *** sileht has quit IRC | 20:52 | |
| *** rlandy has quit IRC | 20:53 | |
| clarkb | jdbc:mysql://localhost:3306/dbname?useUnicode=yes&characterEncoding=UTF-8 is apparently how you force it to utf8 via the connection url | 20:53 |
| *** annegentle has joined #openstack-infra | 20:53 | |
| clarkb | I could try updating that url really quickly and see if it fixes the problem. Any objections to trying that? | 20:54 |
| *** trown is now known as trown|outtypewww | 20:54 | |
| jeblair | clarkb: ++ | 20:59 |
| clarkb | ok stopping gerrit on review-dev now | 21:00 |
| jeblair | clarkb: note: both servers have "character_set_server = utf8" in /etc/mysql/conf.d/client.conf | 21:01 |
| jeblair | clarkb: so if this works, maybe that file is now being ignored | 21:02 |
| clarkb | jeblair: ya I'm wondering if the explicit url overrides that behavior | 21:02 |
| jeblair | clarkb: (otoh, i don't know if jdbc used it before) | 21:02 |
| *** rhallisey has quit IRC | 21:02 | |
| clarkb | I don't think that affected anything, still getting ? on snowman | 21:03 |
| clarkb | which is sane considering the db for which we have the url explicitly set doesn't contain comment data | 21:03 |
| *** jkilpatr has quit IRC | 21:04 | |
| *** armax has quit IRC | 21:05 | |
| *** EricGonczer_ has joined #openstack-infra | 21:06 | |
| clarkb | tobiash: you are running gerrit 2.13 as well. Any chance you use mysql as the database server? do you observe this behavior as well (gerrit storing unicode characters as literal ? in the database) | 21:07 |
| *** yamamoto_ has joined #openstack-infra | 21:09 | |
| *** hichihara has joined #openstack-infra | 21:10 | |
| *** baoli has quit IRC | 21:15 | |
| fungi | clarkb: have you tried tweaking (or removing) the /etc/mysql/my.cnf on the server yet? | 21:15 |
| *** ldnunes has quit IRC | 21:15 | |
| clarkb | fungi: I have not | 21:15 |
| fungi | since it sounds like gerrit may (surprisingly) inspect that | 21:15 |
| *** baoli has joined #openstack-infra | 21:16 | |
| clarkb | the other possibility is locale? I'm going to double check that really quick | 21:16 |
| jeblair | fungi: i think that's a standard mysql client lib behavior. the file is intended to be applicable to both | 21:16 |
| fungi | okay | 21:16 |
| clarkb | LANG=en_US.UTF-8 is in the gerrit process environment according to procfs | 21:17 |
| fungi | so it's not so much that gerrit assumes a local /etc/mysql/my.cnf means any database gerrit is using is the locally-configured one, but rather that a local /etc/mysql/my.cnf is intended as a way of configuring client applications to know what to expect from remote databases they might access? | 21:18 |
| clarkb | fungi: ya it has global client config in /etc/mysql as well as server config | 21:18 |
| fungi | and there's no ~gerrit2/.my.cnf either at least | 21:20 |
| jeblair | fungi: right (and the ~ would override that i believe) | 21:20 |
| *** yamamoto_ has quit IRC | 21:20 | |
| *** baoli has quit IRC | 21:20 | |
| fungi | clarkb: so apparently you can set database.url in the gerrit config too, not just accountPatchReviewDb.url | 21:24 |
| fungi | this looks like what wmf did with their gerrit last december | 21:25 |
| clarkb | fungi: rather than feeding it individual pecies of info for hostname and username and password and db name? | 21:25 |
| clarkb | also I double checked that error_log doesn't record an error when converting snowman to ? | 21:26 |
| clarkb | error_log looks really clean on review-dev nothing there | 21:26 |
| fungi | well, they seemed to already be using database.url in their config, but yes | 21:27 |
| jeblair | fungi: if you do that, does password still come from secret file? | 21:27 |
| jeblair | secure.conf | 21:28 |
| *** baoli has joined #openstack-infra | 21:28 | |
| clarkb | jeblair: I think you drop the password key value entirely and put the url in secure.conf like we did with accountpatchreviewdb | 21:29 |
| fungi | we set all those independent values in secure.conf right now, and the way i understand that works whatever you put in secure.conf is merged with what's in gerrit.conf | 21:29 |
| fungi | mainly i was pointing out that database.url is an option on modern gerrits rather than relying on all those separate values as individual settings | 21:29 |
| fungi | easy enough to test on review-dev regardless | 21:30 |
| clarkb | right so we could attempt to force utf8 via the jdbc url if we wanted | 21:30 |
| clarkb | like I tried with accountpatchreviewdb | 21:30 |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul-jobs master: WIP: Move subunit processing into fetch-testr-output https://review.openstack.org/485840 | 21:32 |
| fungi | yeah, same format. can even try setting it in secure.conf | 21:32 |
| *** Apoorva_ has quit IRC | 21:34 | |
| *** Apoorva has joined #openstack-infra | 21:34 | |
| fungi | looks like they actually set both? https://phabricator.wikimedia.org/source/operations-puppet/browse/production/modules/gerrit/templates/gerrit.config.erb;53545623f46161413f10dcb1256c457d2604c64f$74 | 21:36 |
| fungi | aha, and yeah they leave out the password so probably still setting that separately in secure.conf | 21:37 |
| fungi | that seems like a safe enough way to go | 21:37 |
| clarkb | fungi: do we want to try stopping puppet on review-dev and manually editing config to make that change? | 21:41 |
| openstackgerrit | Merged openstack-infra/tripleo-ci master: Fix stackviz collect data https://review.openstack.org/474559 | 21:41 |
| clarkb | also I thought I read java wants UTF-8 ? | 21:41 |
| clarkb | I suppose we can try both | 21:41 |
| *** baoli has quit IRC | 21:41 | |
| *** sileht has joined #openstack-infra | 21:43 | |
| clarkb | I've got a sandwich with my name on it but I can work on ^ after | 21:45 |
| *** esberglu has quit IRC | 21:46 | |
| *** Apoorva has quit IRC | 21:47 | |
| *** Apoorva has joined #openstack-infra | 21:48 | |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Allow loading additional variables file for site config https://review.openstack.org/447734 | 21:50 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Remove state_dir from setMountsMap https://review.openstack.org/486766 | 21:50 |
| *** marst has quit IRC | 21:52 | |
| *** jascott1 has quit IRC | 21:52 | |
| openstackgerrit | Merged openstack-infra/openstackid-resources master: Updated Eventbrite permission model https://review.openstack.org/482730 | 21:52 |
| *** jascott1 has joined #openstack-infra | 21:53 | |
| *** Goneri has joined #openstack-infra | 21:53 | |
| fungi | clarkb: or just ignore puppet? i think gerrit only rereads that file on service start anyway | 21:54 |
| *** xyang1 has quit IRC | 21:54 | |
| fungi | but yeah, i'm in the midst of chores/food as well (plus, zuul meeting starts in 5 minutes) | 21:55 |
| *** jascott1 has quit IRC | 21:55 | |
| *** kong has joined #openstack-infra | 21:55 | |
| *** armax has joined #openstack-infra | 21:56 | |
| clarkb | fungi: I think puppet may take that as a trigger to reinit gerrit (when the config changes) | 21:57 |
| openstackgerrit | Matt Riedemann proposed openstack-infra/elastic-recheck master: Add query for nova functional placement test bug 1706207 https://review.openstack.org/486773 | 21:57 |
| openstack | bug 1706207 in OpenStack Compute (nova) "resource-class-in-use_create_a_resource_class fails with "MismatchError: '204' not in ['201']" since 7/22" [High,Confirmed] https://launchpad.net/bugs/1706207 | 21:57 |
| clarkb | fungi: which is probably something we want to avoid | 21:57 |
| *** jascott1 has joined #openstack-infra | 21:58 | |
| clarkb | I'll go ahead and put it in the emergency list just to avoid any unexpected problems | 21:59 |
| *** felipemonteiro_ has quit IRC | 22:00 | |
| fungi | oh, i thought we adjusted gerrit config semi-regularly in production and it didn't reinit, but i have no objection to extra caution | 22:00 |
| jeblair | it's zuul meeting time in #openstack-meeting-alt | 22:00 |
| *** jascott1 has quit IRC | 22:01 | |
| *** yamamoto has joined #openstack-infra | 22:01 | |
| *** esberglu has joined #openstack-infra | 22:05 | |
| openstackgerrit | Merged openstack-infra/elastic-recheck master: Add query for nova functional placement test bug 1706207 https://review.openstack.org/486773 | 22:06 |
| openstack | bug 1706207 in OpenStack Compute (nova) "resource-class-in-use_create_a_resource_class fails with "MismatchError: '204' not in ['201']" since 7/22" [High,Confirmed] https://launchpad.net/bugs/1706207 | 22:06 |
| *** EricGonczer_ has quit IRC | 22:06 | |
| openstackgerrit | Clint 'SpamapS' Byrum proposed openstack-infra/zuul feature/zuulv3: Monitor job root and kill over limit jobs https://review.openstack.org/485902 | 22:07 |
| *** bobh has quit IRC | 22:08 | |
| *** kgiusti has left #openstack-infra | 22:09 | |
| *** esberglu has quit IRC | 22:10 | |
| *** jgriffith has quit IRC | 22:10 | |
| *** jgriffith has joined #openstack-infra | 22:10 | |
| *** jgriffith has quit IRC | 22:12 | |
| *** jgriffith has joined #openstack-infra | 22:15 | |
| *** jgriffith has quit IRC | 22:16 | |
| openstackgerrit | dane-fichter proposed openstack-infra/project-config master: Add barbican-tempest-plugin tests to Glance https://review.openstack.org/486781 | 22:16 |
| *** jgriffith has joined #openstack-infra | 22:19 | |
| *** jgriffith has quit IRC | 22:19 | |
| *** jkilpatr has joined #openstack-infra | 22:21 | |
| *** jgriffith has joined #openstack-infra | 22:23 | |
| *** rockyg has quit IRC | 22:26 | |
| *** gouthamr has quit IRC | 22:27 | |
| *** jaypipes has quit IRC | 22:32 | |
| *** EricGonczer_ has joined #openstack-infra | 22:32 | |
| *** ihrachys_ has quit IRC | 22:40 | |
| *** ihrachys_ has joined #openstack-infra | 22:40 | |
| *** gouthamr has joined #openstack-infra | 22:41 | |
| *** vhosakot has quit IRC | 22:50 | |
| *** Goneri has quit IRC | 22:50 | |
| *** rbrndt has quit IRC | 22:53 | |
| *** gtrxcb has joined #openstack-infra | 22:57 | |
| *** gordc has quit IRC | 22:58 | |
| openstackgerrit | Steve Baker proposed openstack-infra/tripleo-ci master: Use prepare command to build container images env file https://review.openstack.org/483178 | 23:00 |
| *** aeng has quit IRC | 23:03 | |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul-jobs master: Create tox_environment_defaults variable for tox based jobs https://review.openstack.org/486679 | 23:05 |
| *** danpawlik has quit IRC | 23:06 | |
| *** rossella__ has quit IRC | 23:14 | |
| *** rossella__ has joined #openstack-infra | 23:16 | |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul-jobs master: Create tox_environment_defaults variable for tox based jobs https://review.openstack.org/486679 | 23:17 |
| *** aeng has joined #openstack-infra | 23:20 | |
| clarkb | fungi: jeblair I'm going to hack in a url for the gerrit db now. Will paste what I end up with before restarting gerrit on review-dev | 23:22 |
| openstackgerrit | Paul Belanger proposed openstack-infra/openstack-zuul-jobs master: DNM https://review.openstack.org/486792 | 23:23 |
| fungi | thanks clarkb. and as i said, looks like you can leave out username and password and continue to define them in separate options (if wmf's production config is to be believed anyway) | 23:23 |
| clarkb | fungi: yup going to try that first | 23:24 |
| *** EricGonczer_ has quit IRC | 23:24 | |
| clarkb | actually why don't you just look at it on disk? | 23:24 |
| fungi | looking | 23:24 |
| clarkb | that way I don't ahve to copy the db host hostname here whcih is semi secret | 23:24 |
| fungi | lgtm, though do we want useunicode? | 23:25 |
| fungi | does that make gerrit store unicode instead of utf-8 in the fields? | 23:25 |
| fungi | i guess i could look up config docs for those options | 23:26 |
| clarkb | fungi: that comes from my earlier investigation of how to force a connection to use utf8 | 23:26 |
| clarkb | fungi: it may not be necessary though my research seemed to imply you may need it in some cases | 23:26 |
| fungi | ahh, certainly confusing | 23:27 |
| fungi | i mean, there is a defined mapping between unicode and utf-8 but they're entirely different binary representations after all | 23:27 |
| clarkb | ya I think in this case it means we are going to do unicode using a utf8 representation | 23:28 |
| *** aeng has quit IRC | 23:28 | |
| *** HeOS has quit IRC | 23:28 | |
| clarkb | fungi: should I go ahead and restart gerrit with that in place as is? | 23:28 |
| fungi | makes sense, and sure go for it | 23:29 |
| clarkb | ok here goes | 23:29 |
| fungi | ftr, for those following along the options being tested in the dburl are: ?characterSetResults=utf8&characterEncoding=utf8&connectionCollation=utf8_bin&useUnicode=yes | 23:29 |
| openstackgerrit | Paul Belanger proposed openstack-infra/shade master: WIP: Initial commit of zuulv3 jobs https://review.openstack.org/484027 | 23:29 |
| clarkb | it works \o/ | 23:30 |
| *** gildub has joined #openstack-infra | 23:30 | |
| clarkb | fungi: I changed the collation method compared to wmf too since _bin is what we have set on our tables | 23:30 |
| fungi | yep, i figured that was why | 23:31 |
| fungi | glad i noticed they were doing it that way | 23:31 |
| fungi | also it looks like the old extended characters are still read correctly: https://review-dev.openstack.org/5335 | 23:31 |
| fungi | obviously those written to the db a ? though are still ?: https://review-dev.openstack.org/107958 | 23:32 |
| clarkb | yup the previous ? are still ? because they went into the db that way | 23:32 |
| fungi | excellent | 23:32 |
| fungi | my concern was that we would end up telling it to read a different encoding than we'd written earlier, but the old comment in https://review-dev.openstack.org/5335 demonstrates that it didn't | 23:33 |
| clarkb | ok, I'll work on a patch to get that into our gerrit configs via puppet | 23:33 |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul-jobs master: Create tox_environment_defaults variable for tox based jobs https://review.openstack.org/486679 | 23:34 |
| fungi | clarkb: presumably we want to go ahead and roll that to production and just carefully monitor it after a restart | 23:34 |
| openstackgerrit | Paul Belanger proposed openstack-infra/shade master: Initial commit of zuulv3 jobs https://review.openstack.org/484027 | 23:34 |
| fungi | independent of any upgrade plans to 2.13 | 23:34 |
| clarkb | fungi: ya | 23:34 |
| clarkb | and we'll want to update the accountpatchreviewdb to use the same set of args for consistency | 23:35 |
| *** chlong__ has joined #openstack-infra | 23:36 | |
| *** chlong_ has quit IRC | 23:37 | |
| openstackgerrit | Merged openstack-infra/system-config master: Remove pholio.openstack.org https://review.openstack.org/477386 | 23:39 |
| clarkb | probably worth pushing up a patch to gerrit docs too | 23:41 |
| clarkb | mordred: speaking of gerrit patches did you ever manage to push your accountpatchreviewdb change up? | 23:41 |
| *** aeng has joined #openstack-infra | 23:41 | |
| openstackgerrit | Ian Wienand proposed openstack-infra/infra-specs master: Abandon pholio spec https://review.openstack.org/486794 | 23:45 |
| clarkb | ianw: yay for black magic disk partitioning I guess | 23:45 |
| fungi | clarkb: oh, right, we keep a copy of the config in our system-config docs. we should definitely update it accordingly | 23:45 |
| clarkb | ianw: except I think I need to use sudo in my change | 23:46 |
| clarkb | ianw: ya just double checked locally so I'll push up a new patchset | 23:46 |
| openstackgerrit | Clark Boylan proposed openstack-infra/devstack-gate master: Sync and partprobe after mkpart on ephemeral disk https://review.openstack.org/486676 | 23:47 |
| clarkb | ianw: ^ there we go | 23:47 |
| ianw | clarkb: ahh, good catch. yeah ... it's still not clear to me if a sleep() effectively does the same thing, but it doesn't hurt | 23:47 |
| ianw | although, the partprobe was getting us confused in dib, as we hard partprobe makeing /dev/loopXpY devices, then kpartx calls making the same /dev/mapper/ ... devices | 23:48 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!