| ianw | mordred: maybe if there was some way to be incremental about it ... | 00:03 |
|---|---|---|
| fungi | renaming the module might allow us to switch vhosts over incrementally, one by one? | 00:08 |
| nibalizer | yea | 00:08 |
| nibalizer | could move it to openstack::apache::vhost | 00:09 |
| nibalizer | then install new apache module | 00:09 |
| nibalizer | and move stuff incrementally | 00:09 |
| clarkb | I dont want to move... | 00:10 |
| clarkb | I think old version is better... | 00:10 |
| nibalizer | thats the other issue | 00:10 |
| ianw | fungi / nibalizer : there is a part of "if it ain't broke" ... | 00:10 |
| clarkb | rename if we need to avoid collisions with other upstreams | 00:11 |
| ianw | clarkb: https://review.openstack.org/#/c/129096/ i think shows that it could use a little help, for example the ability to add an option without having to write out your own template | 00:11 |
| clarkb | but really 0.0.4 was best version | 00:11 |
| ianw | clarkb: but if it was forked, that's just a simple change to add | 00:11 |
| nibalizer | i think we have a responsibility to our downstreams to get off of 0.0.4 | 00:15 |
| nibalizer | if that means forking formally, fine | 00:16 |
| mordred | nibalizer: cause of namecollisino stufs? | 00:16 |
| nibalizer | yea | 00:16 |
| clarkb | I think shipping complete vhosts as we do now is better for downstreams | 00:17 |
| clarkb | its easy to reconsume regardless of config mgmt | 00:17 |
| nibalizer | so if we fork to openstack_apache that works good | 00:17 |
| mordred | ++ | 00:17 |
| clarkb | moving towards thicker puppet types is meh. renaming is fine | 00:17 |
| mordred | I tend to agree with clarkb - I don't want to configure apache with something that isn't an apache config file | 00:18 |
| nibalizer | yea the rename makes everyone happy i think | 00:18 |
| sunil_ | does anybody know where can I find the neutron-server's logs? I am getting an error in my setup q-vpn fails to start and the error is: "l3-agent cannot check service plugins enabled on the neutron server." I need to know what was going on inside neutron-server | 00:18 |
| mordred | although I keep meaning to make a flag or somethign that would plop down a "force-ssl redirect thing" | 00:18 |
| anteaya | sunil_: what are you running? | 00:19 |
| anteaya | now are you running neutron? | 00:20 |
| nibalizer | i honrstly dont care if we use a dumb vhost style or a smart puppet type, i just hate that our ancient apache mod keeps everyone on that | 00:20 |
| anteaya | as part of devstack? | 00:20 |
| sunil_ | yes, as part of the devstack | 00:20 |
| anteaya | *how are you running neutron | 00:20 |
| sunil_ | through the devstack-gate scripts | 00:20 |
| sunil_ | I am running it manually (very much like devstack-vm-gate-dev.sh) | 00:21 |
| anteaya | there should be a log directory | 00:21 |
| sunil_ | the goal is to reuse the devstack-gate scripts for runnign tests whenever we make changes in our plugin here | 00:22 |
| anteaya | /opt/log maybe? | 00:22 |
| sunil_ | there is /opt/stack/logs | 00:22 |
| sunil_ | but I can't find neutron logs there... | 00:22 |
| *** Longgeek has joined #openstack-infra | 00:23 | |
| anteaya | please paste the contents of ls -al /opt/stack/logs | 00:23 |
| anteaya | paste.openstack.org | 00:23 |
| anteaya | it has been so long since I looked I have forgotten | 00:23 |
| sunil_ | http://paste.openstack.org/show/121998/ | 00:25 |
| sunil_ | ignore the Oct 9 entries at the top | 00:25 |
| mordred | nibalizer: ++ | 00:25 |
| *** sdake_ has quit IRC | 00:25 | |
| clarkb | sunil_ screen-q* | 00:27 |
| *** sdake_ has joined #openstack-infra | 00:27 | |
| *** Longgeek has quit IRC | 00:27 | |
| sunil_ | oh...so, its scattered across those. I was looking for "*neutron*" | 00:27 |
| sunil_ | thanks. Let me troll those. But if someone has any clues to why q-vpn might fail to start with that message, pls let me know | 00:28 |
| fungi | q for "quantum" (the original name of neutron) | 00:29 |
| anteaya | and it keeps the q to avoid clashing with the n for nova | 00:30 |
| sunil_ | ok, good to know | 00:30 |
| mordred | fungi: we could call it "neutron-" to avoid clashing with "nova-" | 00:35 |
| *** unicell has joined #openstack-infra | 00:40 | |
| *** cody-somerville_ has quit IRC | 00:45 | |
| openstackgerrit | Ian Wienand proposed a change to openstack-infra/project-config: Add openstack-infra/puppet-apache https://review.openstack.org/129496 | 00:53 |
| ianw | ^ mordred / clarkb / nibalizer : first attempt; new area for me | 00:54 |
| ianw | (project configs, puppet classes, etc) | 00:54 |
| anteaya | lets talk about refs/meta/config | 00:58 |
| anteaya | ianw: why did you put it in your acl? | 00:58 |
| anteaya | is this a straight cargo cult? | 00:58 |
| anteaya | the puppet acls seem to do this and I don't know why | 00:59 |
| anteaya | as pleia2 mentioned in one of her reviews on another patch | 00:59 |
| *** cody-somerville_ has joined #openstack-infra | 01:02 | |
| ianw | anteaya: total cargo cult ... happy for pointers on how to do things right | 01:05 |
| pleia2 | ianw: I'll comment in the review, sec | 01:06 |
| *** stevemar has quit IRC | 01:06 | |
| *** stevemar has joined #openstack-infra | 01:06 | |
| *** amcrn has joined #openstack-infra | 01:08 | |
| *** rkukura has joined #openstack-infra | 01:12 | |
| nibalizer | ianw: so what do you see the name of the module being? | 01:13 |
| nibalizer | i need it to be something other than apache | 01:13 |
| nibalizer | so if we make it openstackci_apache(which i think we should) we should make the repo called puppet-openstackci_apache | 01:13 |
| ianw | nibalizer: you want the separate name to allow mixing of puppetlabs version and this one? | 01:16 |
| nibalizer | yup | 01:16 |
| ianw | nibalizer: the other thing i guess is put it in system-config/modules? then it would be openstack::apache? not sure about merging a fork into there, though | 01:19 |
| anteaya | hey I found this: https://blog.telekomcloud.com/openstack-es-gibt-keinen-anderen-weg-der-sinn-macht/ | 01:19 |
| anteaya | didn't see it before now | 01:19 |
| pleia2 | anteaya: nice photo! | 01:19 |
| * pleia2 google translates | 01:19 | |
| anteaya | thanks | 01:20 |
| nibalizer | ianw: yea we could hide it under openstack_project | 01:20 |
| anteaya | squinty | 01:20 |
| nibalizer | but putting it in a module openstackci_apache is better i think | 01:20 |
| anteaya | pleia2: I kind of gate-crashed the interview | 01:20 |
| anteaya | showed up to observe and just started talking | 01:21 |
| anteaya | sdad actually had some great things to say but unfortunately a lot of his stuff got cut | 01:21 |
| pleia2 | anteaya: haha, nice :) | 01:21 |
| pleia2 | aw | 01:21 |
| anteaya | and might got published | 01:21 |
| anteaya | yeah, aw | 01:21 |
| anteaya | would have liked to have seen more sdad in there | 01:21 |
| anteaya | mordred and sdad were really supportive of me being there | 01:22 |
| pleia2 | that photo of sdad and zaro is a bit funny | 01:22 |
| anteaya | so it would have been nice if the published article was a bit more sdad | 01:22 |
| anteaya | it is | 01:22 |
| anteaya | and kind of a good photo | 01:22 |
| anteaya | the photographer did a good job on the candids | 01:23 |
| anteaya | staring at the sun always makes a protrait difficult though | 01:23 |
| * pleia2 nods | 01:23 | |
| anteaya | my lapful of snoozy cat is hinting that hey sleep might be an idea | 01:27 |
| pleia2 | :D | 01:27 |
| anteaya | so I think I'll wander off now | 01:27 |
| anteaya | :D | 01:27 |
| anteaya | night | 01:27 |
| pleia2 | have a good night | 01:27 |
| anteaya | thanks you too | 01:28 |
| *** liu has quit IRC | 01:38 | |
| *** Longgeek has joined #openstack-infra | 01:38 | |
| *** ivar-laz_ has joined #openstack-infra | 01:39 | |
| *** jpeeler has quit IRC | 01:41 | |
| *** Longgeek has quit IRC | 01:43 | |
| *** zz_dimtruck is now known as dimtruck | 01:43 | |
| *** ivar-lazzaro has quit IRC | 01:43 | |
| *** dimtruck is now known as zz_dimtruck | 01:43 | |
| *** ivar-laz_ has quit IRC | 01:44 | |
| *** zz_dimtruck is now known as dimtruck | 01:45 | |
| openstackgerrit | David Pursehouse proposed a change to openstack-infra/jenkins-job-builder: Fix command to build docs in installation documentation https://review.openstack.org/129498 | 01:45 |
| *** nosnos has joined #openstack-infra | 01:54 | |
| *** pcrews has joined #openstack-infra | 01:55 | |
| *** yamahata has joined #openstack-infra | 01:57 | |
| *** jpeeler has joined #openstack-infra | 02:01 | |
| *** dimsum_ has quit IRC | 02:01 | |
| *** dimtruck is now known as zz_dimtruck | 02:01 | |
| *** dimsum_ has joined #openstack-infra | 02:02 | |
| *** jerryz has joined #openstack-infra | 02:03 | |
| *** dimsum_ has quit IRC | 02:06 | |
| *** fandi has quit IRC | 02:17 | |
| openstackgerrit | Ian Wienand proposed a change to openstack-infra/project-config: Add openstack-infra/puppet-openstackci_apache https://review.openstack.org/129496 | 02:19 |
| *** zul has quit IRC | 02:30 | |
| openstackgerrit | David Pursehouse proposed a change to openstack-infra/jenkins-job-builder: Add support for "Override Build Parameters" option in EnvInject https://review.openstack.org/129189 | 02:31 |
| openstackgerrit | David Pursehouse proposed a change to openstack-infra/jenkins-job-builder: Add tests for properties inject https://review.openstack.org/129502 | 02:31 |
| openstackgerrit | David Pursehouse proposed a change to openstack-infra/jenkins-job-builder: Ignore testr_subunit_log https://review.openstack.org/129503 | 02:31 |
| *** mmaglana has joined #openstack-infra | 02:31 | |
| *** harlowja_at_home has joined #openstack-infra | 02:35 | |
| *** bcourt has joined #openstack-infra | 02:43 | |
| openstackgerrit | Huang Rui proposed a change to openstack-infra/project-config: Create zvm-driver project on StackForge https://review.openstack.org/129506 | 02:46 |
| openstackgerrit | Ian Wienand proposed a change to openstack-infra/project-config: Add openstack-infra/puppet-openstackci_apache https://review.openstack.org/129496 | 02:47 |
| *** mmaglana has quit IRC | 02:59 | |
| *** mmaglana has joined #openstack-infra | 03:00 | |
| openstackgerrit | Ian Wienand proposed a change to openstack-infra/project-config: Add openstack-infra/puppet-openstackci_apache https://review.openstack.org/129496 | 03:01 |
| openstackgerrit | Ian Wienand proposed a change to openstack-infra/project-config: Adding puppet-pip as split out module https://review.openstack.org/126096 | 03:01 |
| *** jerryz has quit IRC | 03:01 | |
| *** julim has quit IRC | 03:05 | |
| *** jerryz has joined #openstack-infra | 03:07 | |
| *** pcrews has quit IRC | 03:12 | |
| *** marun has joined #openstack-infra | 03:15 | |
| *** hdd has joined #openstack-infra | 03:15 | |
| *** arnaud has joined #openstack-infra | 03:18 | |
| openstackgerrit | Steve Baker proposed a change to openstack-infra/project-config: Define experimental check-heat-dsvm-functional-postgres https://review.openstack.org/129509 | 03:21 |
| *** otter768 has quit IRC | 03:22 | |
| *** Longgeek has joined #openstack-infra | 03:22 | |
| *** Longgeek has quit IRC | 03:24 | |
| *** hdd has quit IRC | 03:24 | |
| *** Longgeek has joined #openstack-infra | 03:24 | |
| *** nosnos has quit IRC | 03:24 | |
| *** nosnos has joined #openstack-infra | 03:25 | |
| *** nosnos has quit IRC | 03:29 | |
| *** hdd has joined #openstack-infra | 03:32 | |
| *** marun has quit IRC | 03:38 | |
| *** ryanpetrello has quit IRC | 03:40 | |
| *** ryanpetrello has joined #openstack-infra | 03:41 | |
| *** changbl has quit IRC | 03:42 | |
| openstackgerrit | Ian Wienand proposed a change to openstack-infra/project-config: Add openstack-infra/puppet-openstackci_apache https://review.openstack.org/129496 | 03:45 |
| *** emagana has joined #openstack-infra | 03:46 | |
| *** nosnos has joined #openstack-infra | 03:48 | |
| *** HenryG has quit IRC | 03:49 | |
| openstackgerrit | Steve Baker proposed a change to openstack-infra/project-config: Define experimental check-heat-dsvm-functional-postgres https://review.openstack.org/129509 | 03:49 |
| *** marun has joined #openstack-infra | 03:57 | |
| *** ayoung is now known as ayoung-ZZZzzz | 03:57 | |
| *** emagana has quit IRC | 04:02 | |
| *** emagana has joined #openstack-infra | 04:03 | |
| *** harlowja_at_home has quit IRC | 04:05 | |
| *** marun has quit IRC | 04:06 | |
| *** emagana has quit IRC | 04:07 | |
| *** emagana has joined #openstack-infra | 04:08 | |
| *** emagana has quit IRC | 04:09 | |
| *** garyh has quit IRC | 04:09 | |
| *** yamamoto has quit IRC | 04:10 | |
| *** changbl has joined #openstack-infra | 04:16 | |
| *** andreykurilin_ has joined #openstack-infra | 04:17 | |
| *** yamamoto has joined #openstack-infra | 04:17 | |
| *** koolhead17 has quit IRC | 04:23 | |
| openstackgerrit | Ian Wienand proposed a change to openstack-infra/project-config: Add openstack-infra/puppet-openstackci_apache https://review.openstack.org/129496 | 04:27 |
| *** andreykurilin_ has quit IRC | 04:29 | |
| *** chandankumar has joined #openstack-infra | 04:29 | |
| *** emagana has joined #openstack-infra | 04:34 | |
| *** changbl has quit IRC | 04:39 | |
| *** garyh has joined #openstack-infra | 04:40 | |
| *** sdake has quit IRC | 04:41 | |
| *** fifieldt has joined #openstack-infra | 04:48 | |
| *** msabramo has joined #openstack-infra | 04:53 | |
| *** alexandra has joined #openstack-infra | 04:57 | |
| *** sressot has quit IRC | 04:57 | |
| *** sressot has joined #openstack-infra | 05:01 | |
| *** asettle has quit IRC | 05:01 | |
| *** emagana has quit IRC | 05:01 | |
| *** alexandra is now known as asettle | 05:01 | |
| *** emagana has joined #openstack-infra | 05:02 | |
| *** emagana has quit IRC | 05:06 | |
| *** BharatK has joined #openstack-infra | 05:08 | |
| *** Longgeek has quit IRC | 05:08 | |
| *** krtaylor has quit IRC | 05:09 | |
| *** koolhead17 has joined #openstack-infra | 05:09 | |
| *** krtaylor has joined #openstack-infra | 05:13 | |
| *** Longgeek has joined #openstack-infra | 05:23 | |
| *** alexpilotti_ has joined #openstack-infra | 05:28 | |
| *** alexpilotti has quit IRC | 05:29 | |
| *** alexpilotti_ is now known as alexpilotti | 05:29 | |
| *** amcrn has quit IRC | 05:34 | |
| *** yfried has joined #openstack-infra | 05:40 | |
| *** stevemar has quit IRC | 05:42 | |
| *** koolhead17 has quit IRC | 05:57 | |
| *** loki184 has joined #openstack-infra | 05:58 | |
| *** nosnos has quit IRC | 05:59 | |
| *** nosnos has joined #openstack-infra | 06:00 | |
| *** koolhead17 has joined #openstack-infra | 06:01 | |
| *** AJaeger has joined #openstack-infra | 06:04 | |
| *** punith has joined #openstack-infra | 06:04 | |
| punith | hello, | 06:05 |
| punith | i need to push a third party cinder storage plugin ! is it nessesary to setup our own CI ? | 06:06 |
| punith | fyi....i have passed the tempest test | 06:06 |
| *** doude has joined #openstack-infra | 06:06 | |
| *** chandankumar has quit IRC | 06:08 | |
| *** k4n0 has joined #openstack-infra | 06:11 | |
| AJaeger | punith, the requirement comes from the cinder team, see https://wiki.openstack.org/wiki/Cinder/how-to-contribute-a-driver | 06:11 |
| AJaeger | punith, there's a meeting for third party driver contributors every monday, see https://wiki.openstack.org/wiki/Meetings/ThirdParty | 06:12 |
| *** koolhead17 has quit IRC | 06:13 | |
| *** koolhead17 has joined #openstack-infra | 06:13 | |
| *** ildikov has joined #openstack-infra | 06:16 | |
| *** koolhead17 has quit IRC | 06:18 | |
| *** koolhead17 has joined #openstack-infra | 06:23 | |
| *** emagana has joined #openstack-infra | 06:23 | |
| *** emagana has quit IRC | 06:23 | |
| *** mmaglana has quit IRC | 06:32 | |
| *** r1chardj0n3s is now known as r1chardj0n3s_afk | 06:37 | |
| *** nosnos has quit IRC | 06:39 | |
| *** nosnos has joined #openstack-infra | 06:39 | |
| *** dimsum_ has joined #openstack-infra | 06:40 | |
| *** dimsum_ has quit IRC | 06:45 | |
| *** chandankumar has joined #openstack-infra | 06:46 | |
| *** koolhead17 has quit IRC | 06:48 | |
| *** koolhead17 has joined #openstack-infra | 06:48 | |
| *** alexpilotti has quit IRC | 06:49 | |
| *** doude has quit IRC | 06:49 | |
| *** sdake_ has quit IRC | 06:50 | |
| *** koolhead17 has quit IRC | 06:53 | |
| punith | hello, i'm setting up an third party CI setup for my cinder plugin, | 06:54 |
| punith | i'm currently following the given blog http://www.joinfu.com/2014/02/setting-up-an-external-openstack-testing-system/comment-page-1/ | 06:54 |
| punith | it seems we need an upstream account for this ! can anyone in the infra team set up an account for us ? | 06:55 |
| BharatK | punith, For that we need to send an email to OpenStack infra team with details like, email, account username, and ssh key | 06:56 |
| BharatK | Did you sent the mail to openstack-infra@lists.openstack.org | 06:57 |
| punith | yes, i have already mailed the infra team twice, but i didn't get any reply | 06:58 |
| BharatK | punith, Okay | 06:59 |
| BharatK | punith, I am also setting CI environment for my cinder driver | 06:59 |
| *** doude has joined #openstack-infra | 07:03 | |
| *** che-arne has joined #openstack-infra | 07:03 | |
| *** katyafervent_awa is now known as katyafervent | 07:06 | |
| openstackgerrit | James Polley proposed a change to openstack-infra/project-config: Add the prep-source-repos project https://review.openstack.org/129533 | 07:09 |
| AJaeger | punith, you mailed last Thursday AFAIU - and as documented at http://ci.openstack.org/third_party.html, it might take a week to get it setup... | 07:09 |
| AJaeger | punith, please join the 3rd party IRC meeting later today: https://wiki.openstack.org/wiki/Meetings/ThirdParty | 07:10 |
| *** rushiagr_away is now known as rushiagr | 07:10 | |
| *** koolhead17 has joined #openstack-infra | 07:10 | |
| *** Longgeek has quit IRC | 07:16 | |
| *** dizquierdo has joined #openstack-infra | 07:17 | |
| *** Longgeek has joined #openstack-infra | 07:19 | |
| openstackgerrit | Huang Rui proposed a change to openstack-infra/project-config: Create zvm-driver project on StackForge https://review.openstack.org/129506 | 07:23 |
| *** parimala has joined #openstack-infra | 07:24 | |
| *** skolekonov has joined #openstack-infra | 07:24 | |
| *** Longgeek has quit IRC | 07:24 | |
| parimala | Hello | 07:25 |
| parimala | I had submitted a patch for review, and this is my first patch | 07:26 |
| *** ihrachyshka has joined #openstack-infra | 07:28 | |
| *** jcoufal has joined #openstack-infra | 07:28 | |
| *** arnaud has quit IRC | 07:28 | |
| parimala | I have got the remarks as "You're fixing a bug, so add a separate line:" | 07:29 |
| openstackgerrit | Andreas Jaeger proposed a change to openstack-infra/project-config: Check requirements for manila https://review.openstack.org/129535 | 07:30 |
| *** koolhead17 has quit IRC | 07:30 | |
| *** koolhead17 has joined #openstack-infra | 07:31 | |
| openstackgerrit | Andreas Jaeger proposed a change to openstack/requirements: Add manila to projects.txt https://review.openstack.org/129536 | 07:32 |
| *** mmaglana has joined #openstack-infra | 07:32 | |
| *** dizquierdo has quit IRC | 07:35 | |
| *** koolhead17 has quit IRC | 07:35 | |
| *** Hal has joined #openstack-infra | 07:35 | |
| *** Hal is now known as Guest59036 | 07:36 | |
| *** yamamoto has quit IRC | 07:37 | |
| *** mmaglana has quit IRC | 07:37 | |
| *** vb has quit IRC | 07:37 | |
| *** parimala has quit IRC | 07:39 | |
| *** mpaolino has joined #openstack-infra | 07:40 | |
| *** arxcruz has joined #openstack-infra | 07:40 | |
| *** luqas has joined #openstack-infra | 07:42 | |
| *** jgallard has joined #openstack-infra | 07:42 | |
| *** dizquierdo has joined #openstack-infra | 07:47 | |
| *** koolhead17 has joined #openstack-infra | 07:49 | |
| *** koolhead17 has joined #openstack-infra | 07:49 | |
| *** jamielennox has joined #openstack-infra | 07:50 | |
| *** jpich has joined #openstack-infra | 07:53 | |
| *** parimala has joined #openstack-infra | 07:54 | |
| *** MaxV has joined #openstack-infra | 07:54 | |
| *** mpaolino has quit IRC | 07:55 | |
| *** hdd has quit IRC | 07:56 | |
| AJaeger | fungi, http://paste.openstack.org gives an internal error ;( | 07:58 |
| *** leopoldj has joined #openstack-infra | 08:01 | |
| nibalizer | AJaeger: works for me | 08:01 |
| nibalizer | as i understand it, paste is just flaky | 08:01 |
| AJaeger | nibalizer, ah, thanks | 08:01 |
| * AJaeger tried twice | 08:01 | |
| *** MaxV has quit IRC | 08:01 | |
| *** mpaolino has joined #openstack-infra | 08:04 | |
| *** rlandy has joined #openstack-infra | 08:04 | |
| *** MaxV has joined #openstack-infra | 08:04 | |
| *** _nadya_ has joined #openstack-infra | 08:07 | |
| *** koolhead17 has quit IRC | 08:10 | |
| *** jistr has joined #openstack-infra | 08:12 | |
| *** derekh has joined #openstack-infra | 08:12 | |
| openstackgerrit | A change was merged to stackforge/python-jenkins: Add get_plugins_info and get_plugin_info methods https://review.openstack.org/123932 | 08:14 |
| openstackgerrit | Jamie Lennox proposed a change to openstack-infra/project-config: Allow creating branches for requests-mock https://review.openstack.org/129541 | 08:15 |
| *** ihrachyshka has quit IRC | 08:19 | |
| *** afazekas has joined #openstack-infra | 08:19 | |
| *** plars has quit IRC | 08:20 | |
| *** yamahata has quit IRC | 08:20 | |
| *** Longgeek has joined #openstack-infra | 08:20 | |
| *** plars has joined #openstack-infra | 08:21 | |
| *** hashar has joined #openstack-infra | 08:24 | |
| *** Longgeek has quit IRC | 08:26 | |
| tchaypo | derekh: where does the overcloud-env.json file in the ci cloud come from? | 08:27 |
| tchaypo | Is that something we generated by hand, or is there some tooling I've forgotten or overlooked that generates it? | 08:27 |
| StevenK | tchaypo: devtest_overcloud.sh creates it, which is run by toci ... ? | 08:28 |
| * tchaypo re-reads devtest_ovcercloud.sh | 08:29 | |
| tchaypo | ah, it does to. Reads in at the top, writes out at the bottom. | 08:30 |
| derekh | tchaypo: ya, looks like its set to {"parameters":{}} and then generated from there | 08:30 |
| tchaypo | I saw that it was used as input and assumed that was all | 08:30 |
| tchaypo | So my problem currently in hp2 is that overcloud-env.json has: | 08:30 |
| tchaypo | NovaImage: '' | 08:30 |
| tchaypo | and that'd overriding the NovaImage in the template, and that fails as nova can't find an image called '' | 08:31 |
| tchaypo | i think if I just delete that line, the default should be inserted on the next pass.. | 08:31 |
| derekh | tchaypo: so nova image is comming from "NovaImage": "'"${OVERCLOUD_COMPUTE_ID}"'", | 08:31 |
| derekh | tchaypo: which is comming from load-image | 08:31 |
| derekh | OVERCLOUD_COMPUTE_ID=$(load-image -d $TRIPLEO_ROOT/overcloud-compute.qcow2) | 08:31 |
| derekh | shoud we be in #tripleo ? | 08:32 |
| tchaypo | mayhap | 08:32 |
| *** yfried is now known as yfried|afk | 08:32 | |
| *** jedimike has joined #openstack-infra | 08:33 | |
| *** isviridov_away is now known as isviridov | 08:36 | |
| openstackgerrit | Denis A. Kochetkov proposed a change to openstack-infra/jenkins-job-builder: Review issues fix. Duplicate key assigned-node removed. Key "node" should be used to restrict where job can be executed. https://review.openstack.org/129287 | 08:36 |
| *** mrmartin has joined #openstack-infra | 08:36 | |
| *** rushiagr is now known as rushiagr_away | 08:38 | |
| *** ihrachyshka has joined #openstack-infra | 08:38 | |
| *** dizquierdo has quit IRC | 08:42 | |
| *** e0ne has joined #openstack-infra | 08:43 | |
| *** Longgeek has joined #openstack-infra | 08:45 | |
| *** Longgeek has quit IRC | 08:47 | |
| *** Longgeek has joined #openstack-infra | 08:47 | |
| *** che-arne has quit IRC | 08:49 | |
| *** parimala has quit IRC | 08:54 | |
| *** _nadya_ has quit IRC | 08:55 | |
| *** che-arne has joined #openstack-infra | 08:56 | |
| *** rushiagr_away is now known as rushiagr | 08:57 | |
| *** hashar has quit IRC | 09:00 | |
| *** pelix has joined #openstack-infra | 09:05 | |
| *** yfried|afk is now known as yfried | 09:06 | |
| *** lttrl has joined #openstack-infra | 09:12 | |
| *** hashar has joined #openstack-infra | 09:13 | |
| *** pblaho has joined #openstack-infra | 09:14 | |
| *** yfried is now known as yfried|afk | 09:19 | |
| *** Guest59036 has quit IRC | 09:21 | |
| *** Guest59036 has joined #openstack-infra | 09:21 | |
| *** mmaglana has joined #openstack-infra | 09:22 | |
| *** mmaglana has quit IRC | 09:26 | |
| *** pbelanyi has joined #openstack-infra | 09:28 | |
| *** jp_at_hp has joined #openstack-infra | 09:29 | |
| *** maxbit has joined #openstack-infra | 09:30 | |
| *** k4n0 has quit IRC | 09:32 | |
| *** jlibosva has joined #openstack-infra | 09:32 | |
| *** monester_ has joined #openstack-infra | 09:33 | |
| *** k4n0 has joined #openstack-infra | 09:35 | |
| openstackgerrit | yolanda.robla proposed a change to openstack-infra/nodepool: Conditionally check for targets in reconfigureManagers https://review.openstack.org/129560 | 09:40 |
| *** mancdaz_away is now known as mancdaz | 09:52 | |
| *** achuprin_ has quit IRC | 09:53 | |
| *** achuprin_ has joined #openstack-infra | 09:54 | |
| *** cdent has joined #openstack-infra | 10:06 | |
| *** jgallard has quit IRC | 10:09 | |
| *** hashar has quit IRC | 10:10 | |
| openstackgerrit | yolanda.robla proposed a change to openstack-infra/nodepool: Conditionally check for targets in reconfigureManagers https://review.openstack.org/129560 | 10:10 |
| *** yamahata has joined #openstack-infra | 10:12 | |
| *** Longgeek has quit IRC | 10:14 | |
| *** Longgeek has joined #openstack-infra | 10:14 | |
| *** lttrl has quit IRC | 10:16 | |
| *** lttrl has joined #openstack-infra | 10:16 | |
| *** Guest59036 has quit IRC | 10:17 | |
| *** rushiagr is now known as rushiagr_away | 10:17 | |
| *** KanagarajM has joined #openstack-infra | 10:18 | |
| *** KurtMartin has quit IRC | 10:19 | |
| *** asselin has quit IRC | 10:20 | |
| *** sulochan has quit IRC | 10:20 | |
| *** eglynn-officeafk is now known as eglynn-office | 10:23 | |
| *** yfried|afk is now known as yfried | 10:27 | |
| *** punith has quit IRC | 10:29 | |
| *** chandankumar has quit IRC | 10:31 | |
| *** lttrl has quit IRC | 10:34 | |
| *** unicell has quit IRC | 10:38 | |
| *** chandankumar has joined #openstack-infra | 10:39 | |
| openstackgerrit | yolanda.robla proposed a change to openstack-infra/nodepool: Conditionally check for targets in reconfigureManagers https://review.openstack.org/129560 | 10:41 |
| *** unicell has joined #openstack-infra | 10:43 | |
| *** _nadya_ has joined #openstack-infra | 10:47 | |
| *** KanagarajM has quit IRC | 10:51 | |
| BharatK | Hi All, Is there a way to test CI environment without communicating Upstream? | 10:51 |
| jhesketh | BharatK: You can use the smtp reporter in zuul to check your results | 10:52 |
| BharatK | jhesketh, Oh for Upstream Service Account is required? | 10:53 |
| BharatK | *for that Upstream service account is required or not required? | 10:54 |
| jhesketh | BharatK: you shouldn't need one if you're not reporting, although you may need one to get the stream | 10:54 |
| jhesketh | you could use your own account but be careful to ensure you don't have any reporting turned on | 10:54 |
| *** simonmcc has joined #openstack-infra | 10:55 | |
| *** koolhead17 has joined #openstack-infra | 10:56 | |
| *** dimsum_ has joined #openstack-infra | 10:57 | |
| *** marcusvrn_ has joined #openstack-infra | 11:04 | |
| *** hdd has joined #openstack-infra | 11:05 | |
| *** slagle has joined #openstack-infra | 11:09 | |
| *** mmaglana has joined #openstack-infra | 11:10 | |
| *** denis_makogon has quit IRC | 11:11 | |
| *** dmakogon_ has joined #openstack-infra | 11:12 | |
| *** Longgeek has quit IRC | 11:14 | |
| *** aysyd has joined #openstack-infra | 11:14 | |
| *** mpaolino has quit IRC | 11:14 | |
| openstackgerrit | A change was merged to openstack-infra/jenkins-job-builder: Accept list of paths in 'path' arg. https://review.openstack.org/125854 | 11:15 |
| *** mmaglana has quit IRC | 11:15 | |
| *** Longgeek has joined #openstack-infra | 11:15 | |
| openstackgerrit | venkata anil proposed a change to openstack/requirements: Added apicapi package to global-requirements.txt https://review.openstack.org/129576 | 11:18 |
| *** dkranz has quit IRC | 11:21 | |
| *** rushiagr_away is now known as rushiagr | 11:21 | |
| openstackgerrit | Darragh Bailey proposed a change to openstack-infra/jenkins-job-builder: fix delete job command and add tests https://review.openstack.org/127370 | 11:21 |
| ZZelle | hi, jhesketh | 11:21 |
| *** Guest44335 has joined #openstack-infra | 11:22 | |
| *** Guest44335 is now known as freemanbrandon | 11:23 | |
| jhesketh | ZZelle: Hey | 11:23 |
| openstackgerrit | venkata anil proposed a change to openstack/requirements: Added apicapi package to global-requirements.txt https://review.openstack.org/129576 | 11:24 |
| *** loki184 has quit IRC | 11:24 | |
| *** lttrl has joined #openstack-infra | 11:27 | |
| *** katyafervent is now known as katyafervent_awa | 11:27 | |
| ZZelle | jhesketh, could you reapprove https://review.openstack.org/116035? I only resolve rebase conflict | 11:28 |
| *** enikanorov__ has quit IRC | 11:29 | |
| ZZelle | s/resolve/resolved/ | 11:29 |
| *** Longgeek has quit IRC | 11:30 | |
| *** Longgeek has joined #openstack-infra | 11:31 | |
| *** aysyd has quit IRC | 11:31 | |
| *** zul has joined #openstack-infra | 11:32 | |
| *** aysyd has joined #openstack-infra | 11:32 | |
| *** Longgeek_ has joined #openstack-infra | 11:32 | |
| *** pc_m has joined #openstack-infra | 11:32 | |
| *** rfolco has joined #openstack-infra | 11:35 | |
| *** Longgeek has quit IRC | 11:36 | |
| *** doug-fish has joined #openstack-infra | 11:38 | |
| BharatK | HI, Anybody tested https://github.com/rasselin/os-ext-testing scripts on fedora 20? | 11:39 |
| *** HenryG has joined #openstack-infra | 11:39 | |
| *** monester_ has quit IRC | 11:42 | |
| *** bookwar has quit IRC | 11:42 | |
| *** jgallard has joined #openstack-infra | 11:42 | |
| *** luqas has quit IRC | 11:42 | |
| *** bookwar has joined #openstack-infra | 11:47 | |
| *** eglynn-office is now known as eglynn-office-af | 11:50 | |
| *** eglynn-office-af is now known as eglynn-reg-offic | 11:50 | |
| *** amotoki has quit IRC | 11:50 | |
| openstackgerrit | Andreas Jaeger proposed a change to openstack-infra/jenkins-job-builder: Remove openstack-common.conf https://review.openstack.org/129585 | 11:50 |
| *** eglynn-reg-offic is now known as eglynn-office | 11:50 | |
| *** baoli has joined #openstack-infra | 11:51 | |
| openstackgerrit | Andreas Jaeger proposed a change to openstack-infra/subunit2sql: Remove openstack-common.conf https://review.openstack.org/129587 | 11:54 |
| *** koolhead17 has quit IRC | 11:57 | |
| openstackgerrit | A change was merged to openstack-infra/storyboard: Consolidated functional test data bootstrap https://review.openstack.org/127106 | 12:04 |
| openstackgerrit | A change was merged to openstack-infra/storyboard: Filter tasks by project, project_group, and various other fields https://review.openstack.org/127389 | 12:04 |
| *** lttrl has quit IRC | 12:05 | |
| *** mase_x200 has joined #openstack-infra | 12:05 | |
| jhesketh | ZZelle: lgtm, approved | 12:06 |
| *** lttrl has joined #openstack-infra | 12:06 | |
| *** lttrl has quit IRC | 12:08 | |
| *** amotoki_ has quit IRC | 12:08 | |
| *** lttrl has joined #openstack-infra | 12:08 | |
| *** lttrl has joined #openstack-infra | 12:09 | |
| *** davidhadas_ has joined #openstack-infra | 12:09 | |
| *** Guest59036 has joined #openstack-infra | 12:09 | |
| *** maxbit has quit IRC | 12:13 | |
| *** dprince has joined #openstack-infra | 12:16 | |
| *** dkliban_afk is now known as dkliban | 12:16 | |
| isviridov | SergeyLukjanov : could you please look at https://review.openstack.org/#/c/129269/ | 12:19 |
| *** bswartz has quit IRC | 12:20 | |
| *** mmedvede has joined #openstack-infra | 12:22 | |
| *** flaper87|afk is now known as flaper87 | 12:23 | |
| bcourt | hi, jjb question about multijob projects, I have a project that creates template jobs to build a component for multiple platforms - for example, the job group currently contains unittest-{component}-{arch}-provision, unittest-{component}-{arch}-runtest, unittest-{component}-{arch}-teardown. I would like to create a multijob project to start builds for one project across all arch's, is there a good way to use jjb to do that othe | 12:26 |
| bcourt | r than hardcoding the multijob phase with all the arch's? | 12:26 |
| *** dimsum_ has quit IRC | 12:28 | |
| *** dimsum_ has joined #openstack-infra | 12:28 | |
| bcourt | to rephrase slightly, one project for all the arch's for a particular component | 12:32 |
| *** kgiusti has joined #openstack-infra | 12:32 | |
| bcourt | I should know better than to try to type this early in my day. One job, that kicks off all the provision jobs for a given component | 12:33 |
| *** hashar has joined #openstack-infra | 12:35 | |
| *** mbacchi has joined #openstack-infra | 12:36 | |
| rcarrillocruz | fungi , clarkb: hi there, remember i asked you a week ago if the 'gerrig' user was created manually? I figured how to create users on the jenkins DB with jenkins-cli and groovy | 12:39 |
| rcarrillocruz | would it work for upstream a jenkins internal DB 'gerrig' instead of SSO ? if so, I can push a change for that | 12:40 |
| *** jaypipes has joined #openstack-infra | 12:40 | |
| rcarrillocruz | if it has to be SSO i can investigate how to create SSO users in a scripted manner | 12:40 |
| *** lttrl has quit IRC | 12:40 | |
| *** BharatK has quit IRC | 12:42 | |
| *** pc_m has quit IRC | 12:47 | |
| *** kgiusti has quit IRC | 12:48 | |
| *** pc_m has joined #openstack-infra | 12:48 | |
| *** pc_m has quit IRC | 12:48 | |
| *** hdd has quit IRC | 12:49 | |
| *** pc_m has joined #openstack-infra | 12:49 | |
| *** kgiusti has joined #openstack-infra | 12:49 | |
| *** flaper87 is now known as flaper87|afk | 12:49 | |
| *** bradjones has joined #openstack-infra | 12:52 | |
| *** sandywalsh has joined #openstack-infra | 12:52 | |
| openstackgerrit | Denis A. Kochetkov proposed a change to openstack-infra/jenkins-job-builder: Ability to restrict where job can be executed. Described how to check on the check-box "Restrict where this project can be run" and provide value for "Label Expression" field. https://review.openstack.org/129287 | 12:54 |
| *** radez_g0n3 is now known as radez | 12:55 | |
| openstackgerrit | A change was merged to openstack-infra/storyboard: Filter stories by project_id, status, project_group_id, more. https://review.openstack.org/127653 | 12:56 |
| *** bswartz has joined #openstack-infra | 12:58 | |
| *** mancdaz is now known as mancdaz_away | 12:58 | |
| *** mmaglana has joined #openstack-infra | 12:59 | |
| *** eharney has joined #openstack-infra | 12:59 | |
| *** changbl has joined #openstack-infra | 13:02 | |
| *** koolhead17 has joined #openstack-infra | 13:02 | |
| openstackgerrit | Dmitry Teselkin proposed a change to openstack-infra/system-config: Improve network support in launch-node.py https://review.openstack.org/129013 | 13:03 |
| *** mmaglana has quit IRC | 13:04 | |
| *** dkranz has joined #openstack-infra | 13:04 | |
| openstackgerrit | A change was merged to openstack-infra/storyboard-webclient: Subscription events now appear in UI https://review.openstack.org/114404 | 13:04 |
| *** kgiusti has quit IRC | 13:05 | |
| *** weshay has joined #openstack-infra | 13:05 | |
| *** miqui has joined #openstack-infra | 13:06 | |
| *** luqas has joined #openstack-infra | 13:06 | |
| ZZelle | jhesketh, thanks! | 13:07 |
| *** mjturek has joined #openstack-infra | 13:07 | |
| *** yolanda has joined #openstack-infra | 13:08 | |
| *** julim has joined #openstack-infra | 13:08 | |
| *** kgiusti has joined #openstack-infra | 13:09 | |
| *** bknudson has quit IRC | 13:12 | |
| *** dprince has quit IRC | 13:13 | |
| *** oramraz has joined #openstack-infra | 13:13 | |
| *** parimala has joined #openstack-infra | 13:14 | |
| parimala | Hello AJaeger: | 13:15 |
| *** dustins has joined #openstack-infra | 13:16 | |
| AJaeger | Hi parimala | 13:16 |
| *** _nadya_ has quit IRC | 13:16 | |
| parimala | this is for bug 1344295, I had submitted it several times, and I have got several review comments, which should I consider?? | 13:16 |
| uvirtbot | Launchpad bug 1344295 in openstack-manuals "Security Guide - Chapter 43. Trusted Images Intro" [Low,Confirmed] https://launchpad.net/bugs/1344295 | 13:16 |
| oramraz | hi Guys , can someone please contact me to a person experienced with Gearman daemon ? | 13:17 |
| AJaeger | parimala, you should submit a patch only once - and I suggest you continue with the oldest ;) | 13:17 |
| AJaeger | But whatever works for you :) | 13:18 |
| fungi | oramraz: which gearman daemon? there is more than one implementation | 13:18 |
| AJaeger | Just abandon - there'S a GUI button on http://review.openstack.org for those - the other two | 13:18 |
| AJaeger | And then update your patch as I suggested in the oldest one | 13:18 |
| *** bo_sh has joined #openstack-infra | 13:18 | |
| parimala | what was the issue? Should I give a blank line before and after the commit msg?? | 13:19 |
| *** dprince has joined #openstack-infra | 13:19 | |
| bo_sh | fungi: i am talking about log-gearman-client.py and log-gearman-worker.py | 13:19 |
| parimala | I do not see any abandon button in the GUI | 13:21 |
| fungi | bo_sh: oh, you're trying to index job longs into elasticsearch? | 13:21 |
| fungi | er, job logs | 13:21 |
| AJaeger | parimala, read my comments on https://review.openstack.org/#/c/129271/ | 13:21 |
| *** cpowell has joined #openstack-infra | 13:21 | |
| bo_sh | fungi: that's right | 13:21 |
| AJaeger | parimala, are you logged into gerrit? | 13:21 |
| *** katyafervent_awa is now known as katyafervent | 13:21 | |
| parimala | yes I am logged in | 13:22 |
| AJaeger | parimala, below "patch set 1" are three buttons - Review Cherry-Pick and Abandon | 13:22 |
| openstackgerrit | Michael Krotscheck proposed a change to openstack-infra/storyboard: Plugins may now register their own user preferences https://review.openstack.org/129061 | 13:23 |
| bo_sh | fungi: i have no idea how exactly to configure these two. the *.pp files doesn't help me to understand that too. | 13:23 |
| openstackgerrit | Michael Krotscheck proposed a change to openstack-infra/storyboard: Add User Preferences endpoint. https://review.openstack.org/128487 | 13:23 |
| openstackgerrit | Michael Krotscheck proposed a change to openstack-infra/storyboard: [WIP] Plugins may now register cron workers. https://review.openstack.org/129609 | 13:23 |
| openstackgerrit | Marton Kiss proposed a change to openstack-infra/system-config: Upgrade groups portal instances to use manifest based deployment https://review.openstack.org/129610 | 13:23 |
| fungi | bo_sh: have you seen the documentation we have on how we're using those? http://ci.openstack.org/logstash.html | 13:23 |
| *** dizquierdo has joined #openstack-infra | 13:24 | |
| bo_sh | fungi: yes i have read it. still doesn't give much details about the configuration file that i should use when running the client and the worker. moreover, do i have to run these two on the background? | 13:24 |
| *** yfried_ has joined #openstack-infra | 13:24 | |
| parimala | Yes, i have abandoned the recent submit, retained the oldest | 13:25 |
| bo_sh | fungi: what i am saying, more technical explanations are missing. | 13:25 |
| *** _nadya_ has joined #openstack-infra | 13:25 | |
| fungi | bo_sh: yes, you can run those in the foreground for debugging, but normally we launch them from initscripts | 13:25 |
| fungi | let me find you the configs we're using for them | 13:26 |
| *** yfried__ has joined #openstack-infra | 13:26 | |
| fungi | bo_sh: here is the manifest that instantiates that puppet class for us on our servers... http://git.openstack.org/cgit/openstack-infra/system-config/tree/modules/openstack_project/manifests/logstash_worker.pp | 13:27 |
| *** yfried__ has quit IRC | 13:27 | |
| *** mwagner_lap has quit IRC | 13:27 | |
| fungi | bo_sh: with these config files... http://git.openstack.org/cgit/openstack-infra/system-config/tree/modules/openstack_project/files/logstash | 13:27 |
| *** yfried__ has joined #openstack-infra | 13:27 | |
| *** yfried has quit IRC | 13:27 | |
| parimala | AJaeger: should I download the patch to my workarea and then make the changes? | 13:28 |
| fungi | bo_sh: also, the templated configs referenced in that manifest are at http://git.openstack.org/cgit/openstack-infra/system-config/tree/modules/openstack_project/templates/logstash | 13:28 |
| *** yfried__ is now known as yfried | 13:28 | |
| *** dprince has quit IRC | 13:29 | |
| *** yfried_ has quit IRC | 13:29 | |
| AJaeger | parimala, sure. Or reuse the existing patch if it's still in your workspace. | 13:29 |
| AJaeger | parimala, please read https://wiki.openstack.org/wiki/Documentation/HowTo/FirstTimers and https://wiki.openstack.org/wiki/Documentation/HowTo | 13:29 |
| bo_sh | fungi: gearman-host/port is the hostname/port of the machine that runs gearman-workers? | 13:30 |
| *** dprince has joined #openstack-infra | 13:30 | |
| *** otherwiseguy has joined #openstack-infra | 13:31 | |
| bo_sh | fungi: i can comment the crm114-script and crm114-data from the jenkins-log-worker.yaml, right? | 13:31 |
| fungi | bo_sh: yes, unless you want to use crm114 to classify success/failure indicator probabilities | 13:31 |
| bo_sh | fungi: what about the gearman-host and gearman-port in the jenkins-log-worker.yaml, are these related to the machine that runs the gearman-worker.py script? | 13:33 |
| bo_sh | or it is the hostname and port of the machine that runs the logstash shipper? | 13:34 |
| *** changbl has quit IRC | 13:34 | |
| fungi | bo_sh: i'm following the chain of puppet manifests now to double-check. clarkb is the one who wrote most of this so he has a lot more of it in his head than i do | 13:34 |
| *** mancdaz_away is now known as mancdaz | 13:35 | |
| bo_sh | fungi: thanks. you were very helpful. | 13:36 |
| *** mase_x200 has quit IRC | 13:36 | |
| fungi | bo_sh: ahh, okay so the gearman-host and gearman-port in jenkins-log-worker.yaml are the gearman service running on the machine where the jenkins client resides | 13:37 |
| openstackgerrit | Michael Krotscheck proposed a change to openstack-infra/storyboard: [WIP] Plugins may now register cron workers. https://review.openstack.org/129609 | 13:37 |
| openstackgerrit | Michael Krotscheck proposed a change to openstack-infra/storyboard: Plugins may now register their own user preferences https://review.openstack.org/129061 | 13:37 |
| openstackgerrit | Michael Krotscheck proposed a change to openstack-infra/storyboard: Add User Preferences endpoint. https://review.openstack.org/128487 | 13:37 |
| *** dane_leblanc has joined #openstack-infra | 13:38 | |
| *** bknudson has joined #openstack-infra | 13:39 | |
| fungi | it takes in zeromq messages from the event publisher plugin on the jenkins master(s) and then emits gearman work requests in the gearman queue | 13:39 |
| bo_sh | fungi: so it is the hostname, where the gearman-client is running. correct? | 13:40 |
| *** mriedem has joined #openstack-infra | 13:40 | |
| fungi | bo_sh: yes | 13:40 |
| bo_sh | fungi: great! that's nice to know. | 13:40 |
| fungi | bo_sh: also clarkb made this slide show deck i think is basically still accurate, if it helps... http://ci.openstack.org/logstash.html | 13:40 |
| fungi | er, sorry | 13:41 |
| *** pblaho has quit IRC | 13:41 | |
| fungi | bo_sh: http://docs.openstack.org/infra/publications/processing-ci-log-events/ | 13:41 |
| bo_sh | fungi: yeah i have seen that. | 13:42 |
| bo_sh | fungi: didn't help much, though. | 13:42 |
| *** mfer has joined #openstack-infra | 13:42 | |
| openstackgerrit | Dmitry Teselkin proposed a change to openstack-infra/system-config: Define default value for root_id_rsa https://review.openstack.org/129619 | 13:42 |
| *** superdan is now known as dansmith | 13:42 | |
| *** aysyd has quit IRC | 13:43 | |
| fungi | bo_sh: ahh, sorry. i'll readily admit this is a complex solution engineered for a very specific use case (chaining jenkins job logs through a webserver and then classifying them into an elasticsearch database with some other inline analysis along the way) | 13:44 |
| *** stevemar has joined #openstack-infra | 13:44 | |
| *** aysyd has joined #openstack-infra | 13:45 | |
| fungi | it's designed to handle the thousands of jobs an hour we run, which each generate hundreds of sizable logs (some many megabytes compressed) so the complexity comes from the various scaling points | 13:45 |
| bo_sh | fungi: thanks. you helped me a lot | 13:46 |
| *** bo_sh has quit IRC | 13:46 | |
| *** yfried is now known as yfried|afk | 13:47 | |
| *** yfried|afk is now known as yfried | 13:49 | |
| openstackgerrit | Mike Heald proposed a change to openstack-infra/infra-specs: Proposal for a 'fail-fast' option on jobs and project pipelines https://review.openstack.org/129255 | 13:50 |
| *** pc_m has quit IRC | 13:50 | |
| openstackgerrit | A change was merged to openstack-infra/project-config: Update stackforge/magnetodb acls https://review.openstack.org/129269 | 13:51 |
| *** pc_m has joined #openstack-infra | 13:52 | |
| *** pc_m has quit IRC | 13:53 | |
| *** pc_m has joined #openstack-infra | 13:53 | |
| *** dizquierdo has quit IRC | 13:54 | |
| *** k4n0 has quit IRC | 13:55 | |
| openstackgerrit | Dmitry Teselkin proposed a change to openstack-infra/system-config: Improve network support in launch-node.py https://review.openstack.org/129013 | 13:55 |
| *** Guest4574 is now known as mfisch | 13:56 | |
| *** mfisch has joined #openstack-infra | 13:56 | |
| *** pc_m has quit IRC | 13:58 | |
| *** pc_m has joined #openstack-infra | 13:58 | |
| *** sabeen1 has joined #openstack-infra | 13:58 | |
| *** rushiagr is now known as rushiagr_away | 13:58 | |
| openstackgerrit | A change was merged to openstack-infra/project-config: Add pylint check for neutron https://review.openstack.org/129493 | 13:59 |
| *** nelsnelson has joined #openstack-infra | 13:59 | |
| *** yfried is now known as yfried|afk | 13:59 | |
| openstackgerrit | A change was merged to openstack-infra/project-config: Create zvm-driver project on StackForge https://review.openstack.org/129506 | 14:00 |
| *** vladan has quit IRC | 14:00 | |
| *** thedodd has joined #openstack-infra | 14:01 | |
| *** mpaolino has joined #openstack-infra | 14:01 | |
| *** timrc-afk is now known as timrc | 14:02 | |
| *** vladan has joined #openstack-infra | 14:02 | |
| *** dizquierdo has joined #openstack-infra | 14:03 | |
| *** yfried|afk is now known as yfried | 14:03 | |
| AJaeger | fungi, do you have any idea why patches for manila do not set "In Progress" in launchpad. https://bugs.launchpad.net/manila/+bug/1383154 | 14:05 |
| uvirtbot | Launchpad bug 1383154 in manila "pep8 reports errors in manila/common twice" [Undecided,New] | 14:05 |
| *** signed8bit has joined #openstack-infra | 14:05 | |
| *** ZZelle has quit IRC | 14:06 | |
| fungi | AJaeger: usual reasons are either the infra bot account is not in the bug supervisor group for that project and it's trying to also reassign the bug when setting to in progress, or the gerrit/projects.yaml has the manila project in the wrong launchpad group | 14:06 |
| *** ZZelle has joined #openstack-infra | 14:07 | |
| AJaeger | fungi, strange thing is that abandon works, see https://bugs.launchpad.net/manila/+bug/1382187 | 14:07 |
| uvirtbot | Launchpad bug 1382187 in manila "manila still uses oslo-incubator gettextutils module" [Undecided,In progress] | 14:07 |
| *** vladan has quit IRC | 14:07 | |
| *** dmsimard_away is now known as dmsimard | 14:07 | |
| fungi | AJaeger: abandon doesn't need to assign the bug | 14:07 |
| fungi | AJaeger: bug assignment is a priviliged activity only allowed for members of the bug supervisor group on the project | 14:08 |
| *** dangers_away is now known as dangers | 14:08 | |
| AJaeger | Ah, so let me check the bug supervisor group | 14:08 |
| *** vladan has joined #openstack-infra | 14:08 | |
| *** nosnos has quit IRC | 14:09 | |
| *** chandankumar has quit IRC | 14:09 | |
| *** nosnos has joined #openstack-infra | 14:09 | |
| AJaeger | fungi, https://launchpad.net/~hudson-openstack needs to be in there, correct? | 14:10 |
| fungi | AJaeger: yeah | 14:10 |
| AJaeger | is missing from https://launchpad.net/~manila-core/+members#active - who can add hudson? | 14:10 |
| AJaeger | should I ask the manila team? | 14:10 |
| fungi | anyone who is marked as an "administrator" in that team can do it | 14:12 |
| AJaeger | ok, I ping the team... | 14:12 |
| *** yfried has quit IRC | 14:13 | |
| *** nosnos has quit IRC | 14:14 | |
| *** _nadya_ has quit IRC | 14:14 | |
| *** mmaglana has joined #openstack-infra | 14:16 | |
| *** beekneemech is now known as bnemec | 14:16 | |
| *** jheroux has joined #openstack-infra | 14:16 | |
| *** ayoung-ZZZzzz is now known as ayoung | 14:17 | |
| AJaeger | fungi, resolved - thanks for the help! | 14:17 |
| fungi | AJaeger: my pleasure--glad it was an easy problem for a change ;) | 14:18 |
| *** mmaglana has quit IRC | 14:21 | |
| AJaeger | ;) | 14:21 |
| *** msabramo has quit IRC | 14:22 | |
| *** dtroyer has joined #openstack-infra | 14:23 | |
| *** mmaglana has joined #openstack-infra | 14:24 | |
| *** pcrews has joined #openstack-infra | 14:24 | |
| *** lttrl has joined #openstack-infra | 14:25 | |
| *** _nadya_ has joined #openstack-infra | 14:26 | |
| *** jgrimm is now known as zz_jgrimm | 14:27 | |
| *** mmaglana has quit IRC | 14:28 | |
| *** waynr has quit IRC | 14:28 | |
| *** waynr has joined #openstack-infra | 14:30 | |
| parimala | Hi All, | 14:30 |
| *** rwsu has joined #openstack-infra | 14:30 | |
| *** yfried has joined #openstack-infra | 14:31 | |
| parimala | I do not have any info about who can be mentor for OPW? Is it necessary that I need to provide this info in my application? Any idea??? | 14:31 |
| fungi | parimala: you probably want to ask in the #openstack-opw channel | 14:34 |
| parimala | ok, thanks, I will ask there | 14:34 |
| fungi | parimala: #openstack-infra is the project server sysadmins | 14:34 |
| *** mmaglana has joined #openstack-infra | 14:34 | |
| parimala | ok | 14:35 |
| *** msabramo has joined #openstack-infra | 14:37 | |
| *** jerryz has quit IRC | 14:38 | |
| *** mmaglana has quit IRC | 14:39 | |
| *** cpallares has joined #openstack-infra | 14:40 | |
| *** mfer has quit IRC | 14:40 | |
| *** kgiusti has quit IRC | 14:42 | |
| *** mfer has joined #openstack-infra | 14:43 | |
| *** thedodd has quit IRC | 14:44 | |
| *** kgiusti has joined #openstack-infra | 14:44 | |
| *** BharatK has joined #openstack-infra | 14:47 | |
| DinaBelova | fungi, afaik parimala is interested in infra project itself :) | 14:47 |
| mordred | ooh! | 14:47 |
| mordred | well hello and welcome then parimala | 14:48 |
| fungi | DinaBelova: parimala: i think that's awesome! however the opw channel will definitely be a better resource for answering questions specific to opw itself | 14:48 |
| DinaBelova | fungi, but we don't know who'll be mentor in infra project | 14:48 |
| DinaBelova | and opw channel can't help here | 14:49 |
| DinaBelova | :) | 14:49 |
| DinaBelova | fungi, mordred, a-ha, annegentle came and is helping right now | 14:49 |
| mordred | woot | 14:49 |
| *** emagana has joined #openstack-infra | 14:50 | |
| fungi | DinaBelova: sounds like we need to put out a request for an infra opw mentor. i'll make sure to get it onto tomorrow's infra team meeting agenda if nobody volunteers before then | 14:50 |
| mordred | fungi: also, I think we should indicated that infra as a group is probably all interested in collectively helping - but there will probably need to be a specific more high-bandwidth contact point | 14:50 |
| DinaBelova | fungi, due to the https://wiki.openstack.org/wiki/OutreachProgramForWomen/Ideas#Community it looks like EmilienM is the one - as puppet lives in infra | 14:51 |
| DinaBelova | fungi, mordred - just found this thing in the ideas for opw | 14:51 |
| *** otherwiseguy has quit IRC | 14:52 | |
| *** zz_dimtruck is now known as dimtruck | 14:53 | |
| mordred | jeblair, fungi: I was thinking about what thing we could do with AFS now taht we have it that would be real enough to give us good data on how well it works, without being something that would take enough steps to implement that reverting in case of trouble would be hard | 14:53 |
| EmilienM | fungi: I wanted to be part of mentors list on puppet-dev, I've heard (by you maybe), there is some work to do in -infra things | 14:53 |
| fungi | DinaBelova: parimala: ahh, yes, i do recall EmilienM asking for ideas on potential puppet-related opw tasks or specs we need help implementing | 14:53 |
| *** otherwiseguy has joined #openstack-infra | 14:53 | |
| DinaBelova | fungi, :) | 14:54 |
| parimala | DinaBelova: I am interacting with EmilienM | 14:54 |
| AJaeger | Oh, the project-config approval bot is active - Hi SergeyLukjanov ! ;) | 14:55 |
| openstackgerrit | A change was merged to openstack-infra/gitdm: Update clarkb affiliation https://review.openstack.org/128484 | 14:55 |
| *** davidhadas_ has quit IRC | 14:55 | |
| DinaBelova | parimala, cool, thank everyone for quick response :) I was running from one channel to another only because time is running out before the last date of the application submission | 14:55 |
| mordred | jeblair, fungi: it made me think of pypi mirrors - in that we could continue running the bandersnatch crons we're running right now - but in addition, we could run one that writes to an AFS volume and vos releases it, plus add the AFS client to the current pypi mirror hosts | 14:55 |
| SergeyLukjanov | AJaeger, hey, yeah, I've find some time to go through the list of CRs :) | 14:55 |
| mordred | jeblair, fungi: so that switching between AFS-based and old is as simple as swapping a vhost definition | 14:55 |
| mordred | in case it goes south | 14:56 |
| AJaeger | Much appreciated, SergeyLukjanov - thanks! | 14:56 |
| mordred | morning SergeyLukjanov ! | 14:56 |
| fungi | mordred: actually that was the exact testbed i was going to suggest. definitely agree | 14:56 |
| mordred | fungi: woot! | 14:56 |
| mordred | it also gets our hands wet with vos release which is one of the things we thought would be nice | 14:56 |
| jeblair | yep also we can do some simple timing tests with log files | 14:56 |
| fungi | freemanbrandon has expressed an interest in learning to help out with our infrastructure management too... welcome freemanbrandon | 14:57 |
| mordred | jeblair, fungi: we could also include with that test having some of our australians consume the pypi mirror via AFS to see how the AFS caching works or doesn't work over their terrible overseas internet | 14:57 |
| *** kirscht has quit IRC | 14:57 | |
| mordred | morning freemanbrandon ! | 14:57 |
| *** cpallares has left #openstack-infra | 14:57 | |
| fungi | mordred: oh, that would be an excellent test. i didn't even think about that | 14:57 |
| mordred | jeblair: for timing tests with logfiles - you mean like a syntheic test of grabbing some of our log files, writing them and timing that activity? | 14:58 |
| jeblair | mordred: yeah, potentially running a local os-loganalyze on it | 14:59 |
| fungi | freemanbrandon: so, our primary entry point for documentation about how our systems are built and maintained is http://ci.openstack.org/ and we also have some slide presentations from past conference talks if those are much help to you... http://docs.openstack.org/infra/publications/ | 14:59 |
| *** baoli has quit IRC | 14:59 | |
| *** jistr has quit IRC | 14:59 | |
| *** mmaglana has joined #openstack-infra | 14:59 | |
| krotscheck | StoryBoard meeting in #openstack-meeting | 15:00 |
| jeblair | mordred: both upload/download times for logs | 15:00 |
| fungi | freemanbrandon: we maintain our systems through code review of configuration and automation, reconsuming the same development workflows we're supporting for the rest of openstack, so if you're not already an openstack contributor you probably need to start from https://wiki.openstack.org/wiki/How_To_Contribute#If_you.27re_a_developer | 15:01 |
| *** tonytan4ever has joined #openstack-infra | 15:01 | |
| mordred | jeblair: ++ | 15:01 |
| *** jistr has joined #openstack-infra | 15:01 | |
| openstackgerrit | A change was merged to openstack-infra/project-config: Setup translation jobs for openstackclient https://review.openstack.org/128245 | 15:01 |
| *** yfried has quit IRC | 15:01 | |
| *** kgiusti has quit IRC | 15:01 | |
| mordred | jedimike: you just broke my day | 15:03 |
| *** baoli has joined #openstack-infra | 15:03 | |
| jedimike | mordred, anything is possible at zombo.com! | 15:04 |
| *** mmaglana has quit IRC | 15:04 | |
| mordred | fungi: ^^ I believe jedimike is right | 15:04 |
| *** mfink has joined #openstack-infra | 15:04 | |
| fungi | jedimike: now you've got the zombocom audio track stuck in my head | 15:04 |
| * jedimike giggles | 15:04 | |
| *** dane_leblanc has quit IRC | 15:04 | |
| *** david-lyle has joined #openstack-infra | 15:05 | |
| *** kgiusti has joined #openstack-infra | 15:06 | |
| *** otherwiseguy has quit IRC | 15:06 | |
| *** parimala has quit IRC | 15:07 | |
| openstackgerrit | A change was merged to openstack-infra/system-config: Change wiki link https://review.openstack.org/128623 | 15:07 |
| openstackgerrit | A change was merged to openstack-infra/project-config: Build docs for openstack/governance https://review.openstack.org/125796 | 15:08 |
| openstackgerrit | A change was merged to openstack-infra/project-config: Add translation checks for training-guides https://review.openstack.org/128392 | 15:08 |
| *** eharney has quit IRC | 15:08 | |
| fungi | and yeah, zombo.com has been around for ages. a favorite of mine going way back | 15:09 |
| ashp | blkperl: You around at all? I have a dumb question and I'm hoping you just know the answer | 15:09 |
| *** emagana_ has joined #openstack-infra | 15:09 | |
| openstackgerrit | A change was merged to openstack-infra/project-config: Add link to translation proposals https://review.openstack.org/128569 | 15:12 |
| *** emagana has quit IRC | 15:12 | |
| openstackgerrit | Darragh Bailey proposed a change to openstack-infra/jenkins-job-builder: Support excludes when recursively processing paths https://review.openstack.org/68895 | 15:12 |
| *** emagana_ has quit IRC | 15:13 | |
| *** emagana has joined #openstack-infra | 15:13 | |
| *** oramraz has quit IRC | 15:14 | |
| *** BharatK has left #openstack-infra | 15:14 | |
| *** BharatK has quit IRC | 15:14 | |
| openstackgerrit | A change was merged to openstack-infra/project-config: python-congressclient: Enable Voting for python34 https://review.openstack.org/127408 | 15:15 |
| *** salv-orlando has joined #openstack-infra | 15:17 | |
| *** ildikov has quit IRC | 15:18 | |
| *** otherwiseguy has joined #openstack-infra | 15:18 | |
| *** emagana has quit IRC | 15:18 | |
| openstackgerrit | A change was merged to openstack-infra/project-config: Translation download: Update all existing translations https://review.openstack.org/128579 | 15:19 |
| *** zz_jgrimm is now known as jgrimm | 15:19 | |
| *** mancdaz is now known as mancdaz_away | 15:21 | |
| *** aysyd has quit IRC | 15:23 | |
| SergeyLukjanov | mordred, morning, /me returned to home, so, evening already :) | 15:23 |
| mordred | SergeyLukjanov: well, once again I did not accomplish my goal of "drink with SergeyLukjanov and DinaBelova" | 15:24 |
| *** david-lyle has quit IRC | 15:24 | |
| SergeyLukjanov | mordred, this summit should help you in this goal :) | 15:24 |
| mordred | SergeyLukjanov: YES | 15:25 |
| ashp | as blkperl isn't around can I ask a dumb question I don't get. I'm migrating a working gerrit/zuul/jenkins setup to a new updated set of infrastructure for our local openstack work but.. I cannot for the life of me understand how zuul is supposed to know where jenkins is to run the jobs. am I crazy? There's nothing in the setup docs about the jenkins location. | 15:25 |
| *** aysyd has joined #openstack-infra | 15:26 | |
| clarkb | ashp zuul doesnt know where jenkins is. it knows where the gearman server is | 15:26 |
| clarkb | you then point your jenkins at the gearman server tol | 15:27 |
| ashp | Ahhh ok, hmm, I had just used the internal gearman server because I was having trouble, so that probably explains part of what's confusing me here. :) | 15:27 |
| *** skolekonov has quit IRC | 15:27 | |
| clarkb | *too | 15:27 |
| openstackgerrit | yolanda.robla proposed a change to openstack-infra/storyboard-webclient: Dynamically refresh events on task changes https://review.openstack.org/126542 | 15:28 |
| ashp | also I'm blind because I just looked again and the old conf -clearly- has a jenkins section | 15:28 |
| ashp | monday mornings... | 15:28 |
| *** leopoldj has quit IRC | 15:31 | |
| *** kgiusti has quit IRC | 15:31 | |
| rainya | good regional time of day, openstack-infra | 15:31 |
| *** sdake has joined #openstack-infra | 15:31 | |
| *** sdake has quit IRC | 15:31 | |
| *** sdake has joined #openstack-infra | 15:31 | |
| *** teran has joined #openstack-infra | 15:32 | |
| fungi | ashp: we use the internal gearman server in zuul for production... but as clarkb says you configure your jenkins master's gearman plugin to connect to the gearman server for your zuul instance | 15:32 |
| dtroyer | clarkb: when you get a sec, could you cut Grenade's stable/juno branch at 800788d5d71f18bc6c62a1764ced4cb6ddc00288 "Move javelin execution…" ? | 15:33 |
| fungi | dtroyer: i can take care of it now | 15:33 |
| clarkb | woot I am barely awake and one of two things I wanted to get done today is happening :) | 15:34 |
| fungi | dtroyer: created | 15:34 |
| dtroyer | fungi: awesome, thanks | 15:34 |
| zaro | morning | 15:35 |
| *** kgiusti has joined #openstack-infra | 15:35 | |
| zaro | clarkb: gearman-plugin upgrade this week? | 15:36 |
| *** ttx has quit IRC | 15:37 | |
| clarkb | zaro ya thats the other thing. will stsrt on that today | 15:37 |
| *** ttx has joined #openstack-infra | 15:38 | |
| *** ttx has quit IRC | 15:38 | |
| *** ttx has joined #openstack-infra | 15:38 | |
| zaro | cool, it's got 2 +2 so how about approve it? | 15:38 |
| openstackgerrit | Matt Riedemann proposed a change to openstack-infra/elastic-recheck: Add query for neutron floating ip bug 1383347 https://review.openstack.org/129647 | 15:38 |
| uvirtbot | Launchpad bug 1383347 in neutron "test_network_basic_ops fails with: MismatchError: 'DOWN' != u'ACTIVE': FloatingIP is at status: ACTIVE. failed to reach status: DOWN" [Undecided,New] https://launchpad.net/bugs/1383347 | 15:38 |
| *** msabramo has quit IRC | 15:39 | |
| mordred | clarkb: when you wake up, https://review.openstack.org/#/q/status:open+project:openstack-infra/system-config+branch:master+topic:openafs,n,z are tested and ready to go | 15:41 |
| *** otherwiseguy has quit IRC | 15:41 | |
| mordred | clarkb: (mine, not jeblair's) | 15:41 |
| clarkb | zaro if there is no other reasons to push a new patch then yes we should approve | 15:41 |
| *** otherwiseguy has joined #openstack-infra | 15:41 | |
| jeblair | yeah, i just changed the topic of mine to keep marginally related work together | 15:41 |
| jeblair | mordred: i'm reviewing your changes now | 15:41 |
| *** msabramo has joined #openstack-infra | 15:42 | |
| mordred | jeblair: woot | 15:42 |
| jeblair | mordred: oh, did we set up a cname for the krb admin server? | 15:45 |
| *** todd_dsm has joined #openstack-infra | 15:45 | |
| *** jistr has quit IRC | 15:45 | |
| fungi | if you know what you want it named and to what name you want it resolving, i'm happy to add it | 15:45 |
| clarkb | zaro did you want to approve? I am going to breakfast and can do it after | 15:45 |
| *** kgiusti has quit IRC | 15:45 | |
| ashp | urgh, loose progress, I fixed the zuul connecting to gerrit bit, but of course now it freaks out with no json objects over and over | 15:46 |
| ashp | it's ... a start, I guess. | 15:46 |
| *** emagana has joined #openstack-infra | 15:46 | |
| *** gyee has joined #openstack-infra | 15:47 | |
| mordred | jeblair: NOPE | 15:47 |
| jeblair | fungi, mordred: there's a note in the etherpad that suggests we wanted to name it 'kdc.openstack.org' and point it to kdc01.openstack.org. but i'm not certain we bikeshedded fully on whether that should be kdc., kerberos., or kadmin. | 15:47 |
| ashp | I am guessing if I run 'gerrit stream-events' as the user I'm using for zuul that's -meant- to have the capability streamEvents? | 15:47 |
| fungi | i think whatever color the bikeshed is already is perfectly fine | 15:47 |
| clarkb | isnt there an 02 as well? | 15:47 |
| mordred | I'm pretty sure it shuld not be kadmin - but I think kdc. or kerberos. both make some sense | 15:47 |
| zaro | clarkb: ohh yeah, i will do. | 15:47 |
| *** mrmartin has quit IRC | 15:47 | |
| mordred | clarkb: there is - but only one of them is the admin server at a time | 15:47 |
| *** KurtMartin has joined #openstack-infra | 15:47 | |
| clarkb | ashp yes you need to be able to stream events | 15:48 |
| fungi | and the other(s) is/are slaved | 15:48 |
| blkperl | ashp: I know nothing about zuul/gerrit/jenkins | 15:48 |
| fungi | ashp: it's an acl permission in newer gerrit versions | 15:48 |
| *** zaphod42_ has joined #openstack-infra | 15:48 | |
| ashp | clarkb: Alright, once I convince the ops guys to make me an ssl cert against a non-expired CA (sigh) I'll be able to auth and actually add it :) | 15:48 |
| mordred | fungi: the bikeshed is not a color right now | 15:48 |
| ashp | I mean I can always force it in from the database, but, that's not pretty | 15:48 |
| ashp | blkperl: well what are you waiting for! now is your time! | 15:48 |
| *** mrmartin has joined #openstack-infra | 15:48 | |
| fungi | mordred: well, i meant kdc is perfectly fine by me | 15:49 |
| fungi | because that's what it is anyway | 15:49 |
| fungi | a kdc | 15:49 |
| *** kgiusti has joined #openstack-infra | 15:49 | |
| mordred | that logic works for me | 15:49 |
| fungi | that name makes sense to people who understand kerberos, and to everyone else... well... why we picked that name for the server is likely to be the least confusing part of all this | 15:50 |
| *** emagana has quit IRC | 15:50 | |
| mordred | fungi: to be fair, the number of people who will need to understand kerberos is fairly small | 15:50 |
| jeblair | yay, 3 votes for kdc then. :) | 15:51 |
| fungi | mordred: yep | 15:51 |
| *** yamahata has quit IRC | 15:51 | |
| fungi | and this name will also be mostly invisible to those people ;) | 15:51 |
| *** annegent_ has joined #openstack-infra | 15:51 | |
| *** emagana has joined #openstack-infra | 15:52 | |
| *** afazekas has quit IRC | 15:52 | |
| mordred | jeblair: so, shall we add the cname and update the puppet change? | 15:52 |
| freemanbrandon | Hello all, I am looking to assist with the infra teams efforts. Could someone give me a 1000 foot overview of the direction I should start in? (how to contribute/review software that the team codes on/ setup lauchpad for bug reports) | 15:53 |
| fungi | mordred: i've added the cname just now | 15:53 |
| jeblair | mordred: may want to give me a few more mins to finish reviewing the puppet change | 15:54 |
| mordred | jeblair: :) | 15:54 |
| *** mrmartin has quit IRC | 15:54 | |
| fungi | freemanbrandon: we maintain our systems through code review of configuration and automation, reconsuming the same development workflows we're supporting for the rest of openstack, so if you're not already an openstack contributor you probably need to start from https://wiki.openstack.org/wiki/How_To_Contribute#If_you.27re_a_developer | 15:54 |
| fungi | freemanbrandon: our primary entry point for documentation about how our systems are built and maintained is http://ci.openstack.org/ and we also have some slide presentations from past conference talks if those are much help to you... http://docs.openstack.org/infra/publications/ | 15:55 |
| fungi | freemanbrandon: also, welcome! | 15:56 |
| *** bcourt has quit IRC | 15:57 | |
| *** mpaolino has quit IRC | 15:58 | |
| *** koolhead17 has quit IRC | 15:59 | |
| *** arnaud__ has joined #openstack-infra | 16:00 | |
| openstackgerrit | Jay Faulkner proposed a change to openstack-infra/devstack-gate: Set SWIFT_TEMPURL_KEY for Ironic agent_ssh checks https://review.openstack.org/129653 | 16:00 |
| openstackgerrit | A change was merged to openstack-infra/gearman-plugin: update to work with Jenkins LTS ver 1.565.3 https://review.openstack.org/125755 | 16:01 |
| openstackgerrit | Khai Do proposed a change to openstack-infra/jenkins-job-builder: add documentation for the delete command https://review.openstack.org/128105 | 16:02 |
| fungi | jeblair: mordred: are kdc01 and kdc02 expected to be split between regions or service providers for resilience (or do we have other spofs which make that pointless effort)? | 16:02 |
| *** annegent_ has quit IRC | 16:02 | |
| *** baoli has quit IRC | 16:02 | |
| jeblair | mordred: um. you know how i kept forgetting to put things in ord? | 16:02 |
| jeblair | mordred: i bet i forgot to put kdc02 in ord. | 16:02 |
| * fungi is glad he asked ;) | 16:03 | |
| *** changbl has joined #openstack-infra | 16:03 | |
| JayF | If any of you kind sirs and madams have a moment, https://review.openstack.org/#/c/129653/1 is simply adding a single var to devstack-gate runs for ironic-python-agent; I even linked to the logs showing it's needed. Should be the easiest review of the day ;) | 16:03 |
| jeblair | the weird thing is we weren't even drinking that much. | 16:03 |
| clarkb | cnames seem like a poor way to front a master/slave | 16:03 |
| *** zaphod42_ has quit IRC | 16:03 | |
| mordred | jeblair: hehe | 16:03 |
| *** fifieldt has quit IRC | 16:03 | |
| clarkb | but will mostly work so :) | 16:03 |
| fungi | oh, you two were colocated while working on this? that 'splains the flurry of activity | 16:03 |
| *** arxcruz has quit IRC | 16:03 | |
| *** zaphod42 has joined #openstack-infra | 16:04 | |
| ashp | so, uh, earlier you guys mentioned configuring the jenkins plugin for gearman | 16:04 |
| ashp | it seems that somehow we're not using that, which is impressive | 16:04 |
| ashp | I guess we must be so far behind infrastructure wise that we predate gearman or something? :/ | 16:04 |
| jeblair | clarkb: essentially, in our environment, i don't ever expect us to promote the slave to master. however, we can physically do it if necessary. | 16:04 |
| openstackgerrit | Emilien Macchi proposed a change to openstack-infra/project-config: puppet-sahara: update Gerrit permissions https://review.openstack.org/129655 | 16:04 |
| clarkb | jeblair works for me | 16:05 |
| fungi | clarkb: there is also a round-robin srv for kerberos.openstack.org pointing to the service ports on both kdcs | 16:05 |
| *** koolhead17 has joined #openstack-infra | 16:05 | |
| *** koolhead17 has joined #openstack-infra | 16:05 | |
| jeblair | yeah. if the master goes offline, the realm is read-only (can't add principals or change passwords) | 16:05 |
| jeblair | JayF: ack, will do. also... i have a question you might be able to help with | 16:06 |
| *** gondoi is now known as zz_gondoi | 16:06 | |
| JayF | jeblair: I'll give it a shot; what's up? | 16:06 |
| jeblair | JayF: we're running up against a cinder quota in dfw.... do you know someone we can talk to about that? | 16:06 |
| JayF | I know someone on the Cloud Block Storage team, let me ping them on your behalf | 16:07 |
| *** ildikov has joined #openstack-infra | 16:07 | |
| fungi | er, _kerberos._udp _kerberos-master._udp _kerberos-adm._tcp _kpasswd._udp are the relevant srv records actually | 16:07 |
| *** baoli has joined #openstack-infra | 16:07 | |
| jeblair | (one of my comments on mordreds change is to mention those dns entries in the docs) | 16:07 |
| jroll | jeblair: should just be able to put in a ticket, too | 16:07 |
| *** _nadya_ has quit IRC | 16:08 | |
| *** prad has joined #openstack-infra | 16:09 | |
| *** e0ne has quit IRC | 16:10 | |
| anteaya | was out chatting with a neighbour, found out she has been sick all summer :( | 16:10 |
| anteaya | back now and getting oriented | 16:10 |
| openstackgerrit | Khai Do proposed a change to openstack-infra/jenkins-job-builder: add a JJB command reference https://review.openstack.org/128106 | 16:10 |
| *** hdd has joined #openstack-infra | 16:11 | |
| JayF | jeblair: I pinged some folks on IRC; but I echo what jroll says; often times explicitly using support is a better path :) | 16:11 |
| jeblair | JayF, jroll: cool, will do, thanks | 16:11 |
| jeblair | JayF: there's a tab in your change. it doesn't bother me, but it's bound to upset someone ;) | 16:11 |
| *** jgrimm is now known as zz_jgrimm | 16:12 | |
| *** Hefeweizen has joined #openstack-infra | 16:12 | |
| jeblair | JayF: one quick question though | 16:12 |
| jeblair | JayF: is there any way that change can go in devstack? | 16:12 |
| JayF | jeblair: aha, I'll fix it. Working on new vim configs and I must have missed the one that converts all tabs into less-evil spaces | 16:12 |
| JayF | jeblair: I don't know? | 16:12 |
| JayF | jroll: ^ | 16:12 |
| JayF | jeblair: I have someone queued to help you; pending your DDI and filing a ticket | 16:13 |
| openstackgerrit | Khai Do proposed a change to openstack-infra/jenkins-job-builder: add documentation for globbed parameter feature https://review.openstack.org/128476 | 16:13 |
| JayF | jeblair: if you can PM me the DDI and file the ticket real quick, I'll point that support rep at you | 16:13 |
| jeblair | JayF: the general principal we like to go by is to keep as much "how to configure openstack" knowledge in devstack, and then "how to configure devstack for the particular gate job" in devstack-gate | 16:13 |
| JayF | jeblair: nothing Ironic (or anything else in all of devstack afaik) requires swift temp urls | 16:14 |
| pleia2 | good morning | 16:14 |
| jedimike | morning pleia2 | 16:14 |
| anteaya | morning pleia2 | 16:14 |
| anteaya | jpich: are you around to chat about https://review.openstack.org/#/c/128325/ ? | 16:15 |
| *** hashar has quit IRC | 16:15 | |
| jpich | anteaya: Not for much longer but let's try :) | 16:15 |
| anteaya | jpich: thanks | 16:15 |
| anteaya | so we have established that the reason you need to install the lib is that the integration tests are on thin nodes | 16:16 |
| anteaya | so that makes sense to me | 16:16 |
| anteaya | just wondering if there is a reason to run the integration tests on thick nodes? | 16:16 |
| zaro | fungi: i want to reapply puppet to jenkins-dev.o.o, what's the command to re-apply puppet config? | 16:16 |
| jpich | It seems to be the default for devstack nodes | 16:16 |
| anteaya | would that be possible? | 16:16 |
| anteaya | thick nodes are the default for devstack nodes? | 16:17 |
| *** MaxV has quit IRC | 16:17 | |
| openstackgerrit | Jay Faulkner proposed a change to openstack-infra/devstack-gate: Set SWIFT_TEMPURL_KEY for Ironic agent_ssh checks https://review.openstack.org/129653 | 16:17 |
| clarkb | zaro I think its currently broken on that node | 16:17 |
| jpich | anteaya: I do not know the answer to this, and I'm not sure it's necessary when only one additional package is needed | 16:17 |
| JayF | jeblair: ^ took out that tab | 16:17 |
| anteaya | jpich: fair enough | 16:17 |
| zaro | clarkb: you mean the puppet or the config? | 16:17 |
| clarkb | zaro puppet fails not sure which bit but likely puppet itself | 16:17 |
| anteaya | fungi clarkb what is the decision making factor for use of thick or thin devstack nodes? | 16:17 |
| jeblair | JayF: cool. | 16:18 |
| jeblair | JayF: ticket 141020-ord-0000932 | 16:18 |
| anteaya | fungi clarkb context: https://review.openstack.org/#/c/128325/ | 16:18 |
| jeblair | i don't know why it says 'ord' in there... | 16:18 |
| *** ayoung is now known as ayoung-dogwalkin | 16:18 | |
| zaro | clarkb: will restarting the node fix it? | 16:18 |
| clarkb | zaro no I dont think so. | 16:18 |
| clarkb | anteaya devstack runs on thin. unittests docs et al run on thick | 16:19 |
| JayF | jeblair: Cory should be helping you out soon. | 16:19 |
| anteaya | jpich: I'm not saying the approach is wrong, I just don't know the criteria for deciding which node is used under what conditions | 16:19 |
| jeblair | JayF: thanks! | 16:19 |
| anteaya | clarkb: ah thanks | 16:19 |
| anteaya | jpich: and you are using devstack for your integration tests, so therefore must use thin and install the lib | 16:19 |
| anteaya | jpich: good enough | 16:19 |
| zaro | clarkb: ohh wait, i meant that i wanted to re-apply puppet on zuul-dev.o.o not jenkins-dev | 16:20 |
| jeblair | mordred: kerberos change reviewed | 16:20 |
| clarkb | zaro oh that should work I think | 16:20 |
| *** dimsum_ has quit IRC | 16:20 | |
| *** dimsum_ has joined #openstack-infra | 16:21 | |
| zaro | clarkb: i took a look at the connection problem between gearman-plugin (jenkins-dev) and gear server (zuul-dev). i can't see why it's not working. only thing i can guess is that something is wrong with zuul-dev. | 16:21 |
| zaro | clarkb: so what's the command to re-apply puppet? i'll give that a try on zuul-dev and see if i can make a connection. | 16:22 |
| jpich | anteaya: Glad that's clarified, I've been working off a number of assumptions too (e.g. that one based on https://git.openstack.org/cgit/openstack-infra/project-config/tree/nodepool/scripts/prepare_node_devstack.sh#n22 ) | 16:22 |
| clarkb | zaro `puppet agent --test` | 16:22 |
| jpich | anteaya: Thanks for following up! | 16:22 |
| hogepodge | Regarding this patch, https://review.openstack.org/#/c/129655/ , we've cleared changing owners with the openstack-sahara owners. | 16:22 |
| anteaya | jpich: sure, it all looked good I just realized that I didn't understand the logic behind the decision and used your patch to ask questions to learn the motivation | 16:23 |
| anteaya | now I know more | 16:23 |
| anteaya | jpich: thank you | 16:23 |
| openstackgerrit | A change was merged to openstack-infra/elastic-recheck: Add query for neutron floating ip bug 1383347 https://review.openstack.org/129647 | 16:24 |
| uvirtbot | Launchpad bug 1383347 in neutron "test_network_basic_ops fails with: MismatchError: 'DOWN' != u'ACTIVE': FloatingIP is at status: ACTIVE. failed to reach status: DOWN" [High,Confirmed] https://launchpad.net/bugs/1383347 | 16:24 |
| *** davidhadas has joined #openstack-infra | 16:24 | |
| jeblair | hogepodge: cool; maybe you could levae a review comment to that effect on the change? | 16:24 |
| jpich | :) | 16:24 |
| *** asselin has joined #openstack-infra | 16:24 | |
| anteaya | hogepodge EmilienM great, let's address the typo in the commit message as well | 16:24 |
| *** dimsum_ has quit IRC | 16:25 | |
| jesusaurus | clarkb: jeblair: i'd like to get the bashate change for system-config in. i just rebased against master and confirmed that the test still passes, can you review 118944? | 16:25 |
| jeblair | mordred, fungi: so i vote we land the kerberos changes to puppet, and exercise them by bringing up a replacement kdc02 in ord. | 16:25 |
| anteaya | clarkb: you are still -1 on this patch and while you have a reply to your concerns, I'm not sure if it addresses your concerns: https://review.openstack.org/#/c/126510/ | 16:26 |
| jeblair | jesusaurus: ack | 16:26 |
| anteaya | clarkb: so I am waiting until you are happy | 16:26 |
| fungi | jesusaurus: sounds like a great idea | 16:26 |
| fungi | er, jeblair | 16:26 |
| *** jpich has quit IRC | 16:26 | |
| fungi | sorry jesusaurus, your idea sounds fine too | 16:27 |
| jesusaurus | :) | 16:27 |
| *** lttrl has quit IRC | 16:29 | |
| *** lttrl has joined #openstack-infra | 16:30 | |
| clarkb | anteaya: mostly I dislike how these new projects with almost zero interaction with the toolchain show up and create a half dozen repos | 16:30 |
| clarkb | anteaya: then figure out that managing the cross repo dependencies is complicated | 16:30 |
| *** cody-somerville_ is now known as cody-somerville | 16:31 | |
| clarkb | anteaya: I have said my bit and I think thats all I can do (its not worht a -2 or not merging over) | 16:31 |
| *** oramraz has joined #openstack-infra | 16:31 | |
| *** enikanorov has quit IRC | 16:32 | |
| clarkb | zaro: looks like we are still waiting on the gearman plugin post job to run and create an hpi/jpi for us | 16:32 |
| jeblair | clarkb: well, it's getting easier with zuul-cloner | 16:32 |
| *** marun has joined #openstack-infra | 16:32 | |
| mmedvede | Hello, I am trying to figure out why Nodepool would not receive an event from Jenkins to mark a slave as USED. Am I correct at assuming ZMQ is used for this and Nodepool is subscribing to Jenkins? | 16:33 |
| clarkb | zaro: before we deploy on jenkinsXX I would like tht to complete so that we can deploy off a common hpi/jpi. But don't let me stop your jenkins-dev stuff | 16:33 |
| clarkb | jeblair: thats a good point | 16:33 |
| jeblair | mmedvede: that's correct | 16:33 |
| mmedvede | jeblair, would you recommend a good way to check if the message is being sent by Jenkins? | 16:34 |
| zaro | clarkb: i ran the puppet command, it seemed to do it's thing but still cannot connect gearman-plugin to gear server on zuul-dev | 16:34 |
| clarkb | zaro: ok I will take a look in a minute | 16:34 |
| jeblair | mmedvede: if nodepool is logging at debug level, it will log them... otherwise, let me see if i can find a quick zmq script | 16:35 |
| *** otherwiseguy has quit IRC | 16:35 | |
| *** terryw has joined #openstack-infra | 16:35 | |
| anteaya | clarkb: I agree | 16:35 |
| zaro | clarkb: i can't figure out what's going on. GP on jenkins-dev can connect to zuul.o.o and my own dev gear server. just can't connec to zuul-dev | 16:35 |
| clarkb | dtroyer: jogo https://review.openstack.org/#/c/128959/ is next in the grenadificaion of juno | 16:35 |
| anteaya | clarkb: so the response doesn't address what you need? (colour me unsurprised) | 16:35 |
| jeblair | mmedvede: hah, i even checked it in. :) http://git.openstack.org/cgit/openstack-infra/nodepool/tree/tools/zmq-stream.py | 16:35 |
| clarkb | dtroyer: jogo https://etherpad.openstack.org/p/juno-grenade-fun is the etherpad where I have listed out the things we need | 16:35 |
| mmedvede | jeblair, jenkins log shows the event, while nodepool does not show anything from NodeUpdateListener | 16:35 |
| mmedvede | jeblair, thanks! I would try it now | 16:36 |
| mmedvede | :) | 16:36 |
| jeblair | mmedvede: if you modify that with your host/port, you'll see all the zmq events | 16:36 |
| jeblair | mmedvede: you might run that both on your nodepool host and on your jenkins master | 16:36 |
| zaro | clarkb: i can make a connection from jenkins-dev command line (telnet) to zuul-dev though, just not from the GP jenkins configuration UI. | 16:36 |
| jeblair | mmedvede: to determine whether it's a network problem or a zmq problem | 16:36 |
| openstackgerrit | K Jonathan Harker proposed a change to openstack-infra/project-config: Make bashate voting on system-config https://review.openstack.org/129667 | 16:37 |
| *** jamielennox has quit IRC | 16:37 | |
| pleia2 | jesusaurus: ugh, the diffs on 118944 made me sad | 16:37 |
| pleia2 | (not well-aligned) | 16:37 |
| jeblair | mmedvede: i think zmq needs to be enabled in the global jenkins config too. | 16:37 |
| clarkb | zaro: looking at zuul-dev it is running gearmand not geard | 16:38 |
| clarkb | zaro: I do not know why yet. But that is he only odd thing I see there | 16:38 |
| clarkb | zaro: iptabes is happy | 16:38 |
| zaro | clarkb: http://paste.openstack.org/show/122261/ | 16:38 |
| *** otherwiseguy has joined #openstack-infra | 16:38 | |
| mmedvede | jeblair, all good advise, appreciated | 16:38 |
| clarkb | zaro: I am looking at jenkins-dev now to see what error it gives | 16:39 |
| zaro | clarkb: ohh, gearmand. puppet configs for zuul and zuul-dev use the same zuul.pp | 16:39 |
| *** terryw has quit IRC | 16:39 | |
| clarkb | zaro: jeblair: can I kill gearmand there to allow geard to start? | 16:39 |
| jeblair | clarkb: all yours afaic | 16:40 |
| zaro | clarkb: shouldn't reapplying puppet do that? | 16:40 |
| jesusaurus | pleia2: yeah, it's not as bad as the change for project-config was, but it's still sad | 16:40 |
| *** andreykurilin_ has joined #openstack-infra | 16:41 | |
| clarkb | zaro: not when things are sideways | 16:41 |
| *** arnaud__ has quit IRC | 16:42 | |
| *** todd_dsm has quit IRC | 16:42 | |
| zaro | really? i guess i underestimated puppet power | 16:42 |
| fungi | puppet won't stop arbitrary services it's not configured to manage | 16:43 |
| jogo | clarkb: thanks, about to approve https://review.openstack.org/#/c/128959 | 16:43 |
| clarkb | zaro: puppet only knows how to assert what it wants. If other things have started without puppet knowing it doesn't really know how to assert what it wants from that state | 16:43 |
| fungi | if someone starts gearmand on zuul-dev and it binds a listening socket to the same port number zuul's gearman service wants to use, puppet has no way of knowing that | 16:43 |
| clarkb | also zuul's geard is really unhappy it keeps doing [zuul-server] <defunct> | 16:43 |
| clarkb | and that is after I killed gearmand | 16:44 |
| zaro | isn't there a puppet command to reconfigure me from scratch? | 16:44 |
| clarkb | zaro: no | 16:44 |
| freemanbrandon | fungi: thank you! | 16:44 |
| clarkb | against because puppet has no way of knowing all that external state | 16:44 |
| fungi | zaro: puppet doesn't have a "from scratch." what you describe is basically deleting the current server and building another to replace it | 16:45 |
| *** chandankumar has joined #openstack-infra | 16:45 | |
| clarkb | and no logs for zuul so its really dying early | 16:45 |
| fungi | (at least that would be the closest thing we have to repuppeting a server from scratch) | 16:46 |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/system-config: Add Kerberos config https://review.openstack.org/129444 | 16:46 |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/system-config: Add AFS https://review.openstack.org/129450 | 16:46 |
| zaro | shall we go from scratch route on this one? | 16:46 |
| fungi | clarkb: i started to look at this last week... i wonder whether the layout/config have not been kept up to date with latest changes to the zuul server source code? | 16:46 |
| clarkb | TypeError: __init__() got an unexpected keyword argument 'statsd_port' | 16:46 |
| *** thedodd has joined #openstack-infra | 16:47 | |
| *** mrmartin has joined #openstack-infra | 16:48 | |
| *** mmaglana has joined #openstack-infra | 16:48 | |
| clarkb | OOOHHHHH what appears to have happened is we installed gear from source at one point | 16:48 |
| clarkb | and now isn't installing it from pypi | 16:48 |
| clarkb | because the sha1 version number causes problems | 16:48 |
| *** groknix has joined #openstack-infra | 16:48 | |
| clarkb | let me surgery | 16:48 |
| fungi | ahh, yep. i see that | 16:49 |
| jeblair | clarkb, fungi: also, that's true on zuul prod too. it has an increased timeout value, pending one of the non-blocking io changes landing to geard. | 16:49 |
| fungi | we're installing gear from source on zuul.o.o too, but new enough gear to make proper version numbers now that it's had a release | 16:49 |
| jeblair | (i mean, zuul.o.o has a manually installed version with a locally applied dirty fix for that) | 16:50 |
| clarkb | oh ok | 16:50 |
| *** bcourt has joined #openstack-infra | 16:50 | |
| fungi | the gear installed on zuul-dev is from before the initial release tag | 16:50 |
| jeblair | i guess it's not the same thing at all then. nm. :) | 16:50 |
| clarkb | so what I am going to do is uninstall gear, update gear repo, install from there with --upgrade | 16:50 |
| clarkb | gear==0.5.5.17.g026fe3b that looks much better now | 16:52 |
| mordred | jeblair: I manually renamed kprop-openstack.sh to run-kprop.sh on kdc01 and in the crontab | 16:52 |
| *** mmaglana has quit IRC | 16:53 | |
| clarkb | woot new errors. ssh key needs updating for review-dev and apparently the opnstack_functions.py file does not exist | 16:53 |
| EmilienM | anteaya: ok | 16:54 |
| *** patrickeast has joined #openstack-infra | 16:54 | |
| *** eharney has joined #openstack-infra | 16:54 | |
| *** yolanda_ has joined #openstack-infra | 16:54 | |
| EmilienM | anteaya: which typo? | 16:55 |
| clarkb | zaro: ok I think I have gearman plugin on jenkins-dev connected to my manually run in forground zuul + geard | 16:56 |
| *** reed has joined #openstack-infra | 16:56 | |
| clarkb | zaro: I am going to start it as a service now and reconnect | 16:56 |
| *** yolanda has quit IRC | 16:56 | |
| openstackgerrit | Emilien Macchi proposed a change to openstack-infra/project-config: puppet-sahara: update Gerrit permissions https://review.openstack.org/129655 | 16:57 |
| EmilienM | anteaya: ^ | 16:57 |
| *** sarob has joined #openstack-infra | 16:57 | |
| clarkb | jogo: dtroyer thanks a bunch | 16:57 |
| *** dizquierdo has quit IRC | 16:57 | |
| jeblair | mordred: so we ended up with the client configured and enabled but not running on the db servers. we ended up with the client not configured and not running on the fileservers (despite being installed).... | 16:58 |
| clarkb | jogo: dtroyer: once that merges I can recheck my d-g changes and they should be self testing | 16:58 |
| jeblair | mordred: and the /etc/openafs directory contents are different on the fileservers than the client (there's that symlink of cellservdb into the server directory) | 16:58 |
| zaro | clarkb: cool. thanks. does what you did manually fix this going forward? | 16:58 |
| anteaya | EmilienM: you found it, thanks | 16:59 |
| *** AJaeger_ has joined #openstack-infra | 16:59 | |
| *** luqas has quit IRC | 16:59 | |
| anteaya | EmilienM: yes, I wasn't clear, was I, sorry about that | 16:59 |
| zaro | clarkb: or is there a required puppet change that needs to happen? | 16:59 |
| jeblair | mordred: i'm not sure what the final state should be. part of me says we should not run the client on any servers. | 16:59 |
| *** emagana has quit IRC | 16:59 | |
| clarkb | zaro: I think so | 16:59 |
| clarkb | zaro: the gear versions are now able to upgrade properly | 17:00 |
| *** emagana has joined #openstack-infra | 17:00 | |
| bcourt | clarkb, using nodepool, if I have a test job that requires multiple nodes, what is the preferred way to handle that? For example, an integration test that uses one jenkins node to run the test harness but requires three servers for testing. | 17:00 |
| clarkb | bcourt: nodepool can spin up groups of servers that must all come up in order for the main node to be attached to jenkins | 17:01 |
| cpowell | is there going to be a third-party meeting today? | 17:01 |
| clarkb | bcourt: each of the nodes (main node and subnodes) has files written in /etc/nodepool iirc that tell them about the others | 17:02 |
| nibalizer | q/win 16 | 17:02 |
| clarkb | bcourt: if you look for subnodes in the nodepool docs you should see how this works | 17:02 |
| bcourt | clarkb, thanks! | 17:02 |
| clarkb | bcourt: http://ci.openstack.org/nodepool/configuration.html#labels | 17:02 |
| jogo | clarkb: excellent | 17:02 |
| *** Longgeek_ has quit IRC | 17:02 | |
| *** teran has quit IRC | 17:02 | |
| fungi | bcourt: an example change to devstack-gate to try to take advantage of that, currently still a work in progress, is https://review.openstack.org/106043 | 17:03 |
| *** gyee has quit IRC | 17:03 | |
| *** _buzz has joined #openstack-infra | 17:03 | |
| mordred | jeblair:the CellServDB doesn't need to be a symlink - it can also just be a file - and the difference seems to be that servers have the server dir and clients do not | 17:03 |
| *** e0ne has joined #openstack-infra | 17:03 | |
| *** Guest59036 has quit IRC | 17:03 | |
| *** katyafervent is now known as katyafervent_awa | 17:03 | |
| *** annegent_ has joined #openstack-infra | 17:03 | |
| mordred | jeblair: but the question of do we want to run the client on the servers is a good one - any reason _not_ to? | 17:04 |
| mordred | jeblair: (it's obviously an easy fix to the puppet) | 17:04 |
| *** radez is now known as radez_g0n3 | 17:04 | |
| jeblair | mordred: the thing in the back of my mind is whether it would cause any problems if the cell is down | 17:04 |
| jeblair | mordred: it should not, unless we actually try to use it. | 17:04 |
| *** emagana has quit IRC | 17:04 | |
| fungi | bcourt: it's worth pointing out that only the main node of that subnode group will be running the jenkins agent and registered as a jenkins slave, so job calls would need to be proxied through it to the others in some way (for example via ssh calls) | 17:04 |
| jeblair | so, i mean, we can't _depend_ on it on the fileservers. but perhaps it would not hurt. | 17:04 |
| clarkb | bcourt: http://ci.openstack.org/nodepool/scripts.html has info on what data is provided to all of the nodes | 17:05 |
| jeblair | mordred: i reviewed the afs change, mostly with the assumption that we don't want to run the clients. if we decide we want to try that, i'm fine with it. | 17:05 |
| *** Longgeek has joined #openstack-infra | 17:05 | |
| mordred | jeblair: k. I'm personally fine either way - and can make the change if we decide to go the other way | 17:05 |
| bcourt | fungi, that's fine, the only item that could be a little bit of a challenge for me is that I might need sub-nodes of different types | 17:05 |
| jeblair | fungi: thoughts on running the client on a fileserver? | 17:06 |
| clarkb | zaro: I think everything is ready for zuul-dev + jenkins-dev gearman plugin testing now | 17:06 |
| *** Longgeek has quit IRC | 17:06 | |
| clarkb | zaro: let me knwo if you need more infos | 17:06 |
| *** roaet_ has quit IRC | 17:07 | |
| fungi | jeblair: running the openafs client on an afs cell server? | 17:08 |
| jeblair | fungi: yeah | 17:08 |
| fungi | jeblair: doesn't seem necessary, no. though maybe i'm missing a use case | 17:08 |
| jeblair | fungi: the puppet is easier if it is :) | 17:08 |
| *** dkehn has quit IRC | 17:08 | |
| jeblair | looks like that's the default on freebsd | 17:09 |
| jeblair | though you can disable it | 17:09 |
| *** derekh has quit IRC | 17:09 | |
| fungi | jeblair: "we want all our puppeted machines consistent" sounds like a great use case to me | 17:09 |
| *** kgiusti has quit IRC | 17:09 | |
| jeblair | yeah, i'm starting to lean toward trying it. | 17:09 |
| fungi | for some definition of "use case" i have no objection at any rate | 17:10 |
| *** davidhadas has quit IRC | 17:11 | |
| *** thedodd has quit IRC | 17:11 | |
| jeblair | mordred: okay, published comments on afs change; feel free to ignore the ones about not running the client | 17:11 |
| mordred | woot | 17:11 |
| *** kgiusti has joined #openstack-infra | 17:12 | |
| *** roaet_ has joined #openstack-infra | 17:12 | |
| mordred | yah - in general, I have a hunch that once we're on board with this as a thing, that openstack_project::server should just install and configure the AFS client | 17:12 |
| mordred | but *hand wave* | 17:12 |
| clarkb | so does afs mount things using fuse? | 17:13 |
| *** jgallard has quit IRC | 17:13 | |
| clarkb | or is client there to negotiate kerberos then kernel does things? | 17:13 |
| *** julim has quit IRC | 17:14 | |
| *** bradjones has quit IRC | 17:14 | |
| *** arnaud__ has joined #openstack-infra | 17:14 | |
| mordred | clarkb: no fuse | 17:14 |
| *** harlowja_away is now known as harlowja | 17:14 | |
| *** zz_jgrimm is now known as jgrimm | 17:14 | |
| jesusaurus | mordred: wait, did you actually convince everyone to get on the afs bandwagon? | 17:14 |
| *** armax has joined #openstack-infra | 17:14 | |
| mordred | jesusaurus: jeblair and I have been convinced for the AFS bandwagon for ages | 17:14 |
| clarkb | and they spetn all weekend doing the owrk | 17:14 |
| jesusaurus | oh cool | 17:15 |
| *** davidhadas_ has joined #openstack-infra | 17:15 | |
| clarkb | so post queue being low priority | 17:15 |
| clarkb | I think this is a bug | 17:15 |
| clarkb | fungi: I know you had concerns with it going in, but I am ready to say we should normal queue it | 17:15 |
| jeblair | clarkb: why? | 17:15 |
| clarkb | jeblair: there is a docs change there that has been waiting for over 4 hours now to run its job | 17:15 |
| clarkb | jeblair: and docs are heavily dependent on post queue for publishing | 17:15 |
| fungi | clarkb: i didn't have concerns with post pipeline being normal priority | 17:16 |
| mordred | jeblair: re: subscribing to userlist and cellservdb - I put that in there specifically to allow for managing userlist with puppet and not with bos adduser - but we also could _totally_ just not do that and do the manual bos adduser on each of the hosts | 17:16 |
| clarkb | fungi: no you had them for low priority iirc | 17:16 |
| fungi | clarkb: i think sdad was worried it would further delay check results | 17:16 |
| *** chandankumar has quit IRC | 17:16 | |
| mordred | jeblair: (it was kindof a straw man "here's what this would look like") | 17:16 |
| fungi | clarkb: my main concerns were over making sure release, pre-release and tag stayed at high priority | 17:17 |
| *** bradjones has joined #openstack-infra | 17:17 | |
| clarkb | mordred: so if not fuse what does a client do? | 17:17 |
| mordred | clarkb: the other thing you said | 17:17 |
| jeblair | clarkb: i'm not sure the 4h thing is due to priority | 17:17 |
| mordred | clarkb: there is a kernel module and a daemon | 17:17 |
| *** pc_m has quit IRC | 17:17 | |
| fungi | clarkb: oh, i get you. yes, i advocated for keeping the post pipeline normal priority, not dropping it to low priority | 17:17 |
| jeblair | clarkb: we're not at nodepool copacity. | 17:18 |
| jeblair | clarkb: so nothing should be waiting more than 10 mins | 17:18 |
| *** pc_m has joined #openstack-infra | 17:18 | |
| *** arnaud has joined #openstack-infra | 17:18 | |
| clarkb | jeblair: I think errors have us effectively as capacity but that is only based on my looking at some graphite graphs | 17:19 |
| *** ihrachyshka has quit IRC | 17:19 | |
| fungi | clarkb: i assumed it was that we merged a large wad of changes, so the low-priority fifo is working its way through several hours worth of merge-check jobs and post jobs start to accumulate behind them | 17:20 |
| fungi | right now merge-check is up over 1000 | 17:21 |
| *** atiwari has joined #openstack-infra | 17:21 | |
| clarkb | fungi: ya but the oldest there is only 1.5 hours ish | 17:21 |
| clarkb | but it would certainly help make the problme worse | 17:21 |
| clarkb | (it is why I brought up the priority) | 17:22 |
| fungi | oh, also we' | 17:22 |
| jeblair | clarkb: we only have 600 nodes in use right now | 17:22 |
| fungi | grrr | 17:22 |
| jeblair | that's not near capacity | 17:22 |
| *** tonytan4ever has quit IRC | 17:22 | |
| fungi | we're not at node capacity but high and medium priority pipelines may be starving mergers | 17:22 |
| fungi | ? | 17:22 |
| fungi | looking at graphs right now | 17:23 |
| jeblair | the top jobs in post have already run merge jobs | 17:23 |
| clarkb | fungi: maybe? I note that rax-dfw has a very high error rate | 17:23 |
| clarkb | the other regions appear to be mostly happy | 17:23 |
| *** davidhadas_ is now known as davidhadas | 17:23 | |
| jeblair | so the reason that bec6ce7 is not running is not because of merge starvation, and it should not be because of node starvation. | 17:23 |
| fungi | i'm wondering if the top jobs in post hadn't gotten merges until a few minutes ago, though yes that training-guides change is an outlier and i agree it's likely waiting for a job restart | 17:24 |
| clarkb | maybe the job started and was cancelled so back to the end of the queue with it? | 17:25 |
| jeblair | fungi: that seems reasonable. merge starvation combined with a restart | 17:25 |
| clarkb | I can grep zuul logs | 17:25 |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/system-config: Add AFS https://review.openstack.org/129450 | 17:25 |
| jeblair | clarkb: it just started | 17:25 |
| mordred | jeblair: ^^ k. I removed the UserList management but left in the run-client stuff | 17:25 |
| jeblair | https://jenkins07.openstack.org/job/training-guides-tox-doc-publishlang/1/ | 17:25 |
| mordred | we can always add UserList management if we want later | 17:26 |
| *** ayoung-dogwalkin is now known as ayoung | 17:26 | |
| fungi | and network or similar errors in a provider would be a likely explanation for the node for that job going out for lunch | 17:26 |
| jeblair | mordred: could probably put in userlist management; i think the cellservdb is what i most objected to | 17:26 |
| *** bdpayne has joined #openstack-infra | 17:26 | |
| mordred | jeblair: oh - well, I expect that to change almost never, no? | 17:26 |
| AJaeger_ | this is the first time the publishlang job for the training-guide is run, SergeyLukjanov approved my patch a few hours ago... | 17:27 |
| jeblair | mordred: yeah. either way. :) | 17:27 |
| mordred | jeblair: I removed the subscribe to CellServDB too | 17:27 |
| fungi | mordred: did you see my comment about https://web.mit.edu/kerberos in the rst file on 129444? | 17:27 |
| clarkb | my grepping doesn't find anything obvious (though maybe 15 lines of after context isn't enough) | 17:27 |
| *** annegent_ has quit IRC | 17:28 | |
| mordred | fungi: I did not - lemme go look | 17:28 |
| jeblair | clarkb: uuid 1dab0f360c4e446a8498288b261d1027 | 17:28 |
| *** isviridov is now known as isviridov_away | 17:28 | |
| mordred | fungi: it's just http isn't it? | 17:28 |
| fungi | mordred: it should be http, yes, not https | 17:28 |
| clarkb | jeblair: ya I see that one | 17:28 |
| clarkb | jeblair: which just started. Was hoping to find something from ~4 hours ago for why it hadn't started until just now | 17:29 |
| *** pc_m has quit IRC | 17:29 | |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/system-config: Add Kerberos config https://review.openstack.org/129444 | 17:29 |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/system-config: Add AFS https://review.openstack.org/129450 | 17:29 |
| mordred | fungi: thanks | 17:29 |
| *** mmaglana has joined #openstack-infra | 17:29 | |
| jeblair | clarkb: you should see when it was enqueued | 17:29 |
| jeblair | (eg, if it was enqueued 4h ago, then no restart) | 17:30 |
| openstackgerrit | Matt Riedemann proposed a change to openstack-infra/elastic-recheck: Add query for neutron functional job bug 1383391 https://review.openstack.org/129678 | 17:30 |
| uvirtbot | Launchpad bug 1383391 in neutron "gate-neutron-dsvm-functional fails with "Length too long:"" [Undecided,New] https://launchpad.net/bugs/1383391 | 17:30 |
| clarkb | jeblair: oh I see. Ya it was enqueued around 15:22:45,660 | 17:31 |
| clarkb | so about 2 hours ago | 17:31 |
| jeblair | clarkb: i think that confirms the merge-starve + restart hypothesis | 17:31 |
| fungi | long after the change was enqueued. right | 17:31 |
| jeblair | it currently takes 100 seconds to merge a nova change | 17:31 |
| jeblair | on a merger | 17:31 |
| *** julim has joined #openstack-infra | 17:31 | |
| jeblair | i think a merger with a fresh clone takes 20-30 seconds | 17:31 |
| *** yjiang5 has quit IRC | 17:32 | |
| jeblair | so we could do some housekeeping and delete the nova clones on the mergers | 17:32 |
| clarkb | ya grepping for the job itself shows that the first time it was enqueued was 1522 | 17:32 |
| clarkb | so there wasn't anything before that point | 17:32 |
| *** yjiang5 has joined #openstack-infra | 17:32 | |
| marcusvrn_ | hi all! I'm trying to clone openstack-infra repo...but it returns the following error: "fatal: https://review.openstack.org/p/openstack-infra/config.git/info/refs not found: did you run git update-server-info on the server?" | 17:33 |
| clarkb | marcusvrn_: we renamed the repo to openstack-infra/system-config on friday | 17:33 |
| clarkb | marcusvrn_: you will need to update your git remotes | 17:33 |
| *** yjiang5 has left #openstack-infra | 17:34 | |
| marcusvrn_ | clarkb: hmmm nice! thx!! | 17:34 |
| marcusvrn_ | I'll do it | 17:34 |
| fungi | marcusvrn_: http://lists.openstack.org/pipermail/openstack-infra/2014-October/002012.html | 17:34 |
| jeblair | clarkb, fungi: if we want to do that, we can shut down the mergers one at a time, wait 10 minutes, then rm the nova repo and start them again... what do you think? | 17:35 |
| clarkb | jeblair: sounds good | 17:35 |
| jeblair | also, we really should shrink the size of the earlier servers and add some more | 17:35 |
| clarkb | ++ | 17:35 |
| *** annegent_ has joined #openstack-infra | 17:36 | |
| fungi | jeblair: clarkb: is that likely to adversely impact jobs which were waiting for nodes between when the merge was calculated on that merger and when it finally gets to the clone commands? | 17:36 |
| *** pc_m has joined #openstack-infra | 17:36 | |
| jeblair | clarkb, fungi, mordred: our dfw cinder quota is increased | 17:36 |
| marcusvrn_ | fungi: thx! :) | 17:36 |
| jeblair | JayF, jroll: thanks! | 17:36 |
| fungi | jeblair: yay! that was _fast_ | 17:36 |
| jeblair | fungi: yes; was thinking the 10 min delay would mitigate that | 17:36 |
| fungi | jeblair: oh, right. stop the merge worker but not apache | 17:37 |
| *** yjiang5 has joined #openstack-infra | 17:37 | |
| jeblair | yep | 17:37 |
| fungi | jeblair: sounds good then | 17:37 |
| jeblair | okay, starting with 04. | 17:37 |
| clarkb | ya as long as all jobs fetch the refs before we rm nova it will be fine | 17:37 |
| *** yjiang5 has quit IRC | 17:38 | |
| jeblair | groups.openstack.org | 17:38 |
| jeblair | 198.61.212.109 | 17:38 |
| jeblair | is shutoff according to the web ui | 17:38 |
| jeblair | assume that's an old server and it's been replaced | 17:38 |
| *** flaper87|afk is now known as flaper87 | 17:39 | |
| fungi | jeblair: yes, it should be fine to delete now | 17:39 |
| fungi | jeblair: i'll take care of it | 17:39 |
| fungi | just didn't want to do so until i have mrmartin a chance to confirm the new server wasn't broken | 17:40 |
| openstackgerrit | Matt Riedemann proposed a change to openstack-infra/elastic-recheck: Add query for python-openstackclient test bug 1383396 https://review.openstack.org/129679 | 17:40 |
| uvirtbot | Launchpad bug 1383396 in python-openstackclient "tests.test_identity.IdentityV3Tests.test_token_issue fails with "ERROR: openstack 'NoneType' object has no attribute 'service_catalog'"" [Undecided,New] https://launchpad.net/bugs/1383396 | 17:40 |
| fungi | er, s/have/gave/ | 17:40 |
| *** yjiang5 has joined #openstack-infra | 17:40 | |
| JayF | jeblair: no problem! | 17:40 |
| jeblair | np, just noting in the highly unlikely event it was forgotten. :) | 17:40 |
| mrmartin | fungi, feel free to delete | 17:40 |
| *** ihrachyshka has joined #openstack-infra | 17:40 | |
| fungi | jeblair: it was semi-forgotten, so thanks for the prompt ;) | 17:40 |
| fungi | mrmartin: thanks! | 17:41 |
| *** cdent has quit IRC | 17:41 | |
| jeblair | fungi: want me to delet it now? i'm right there | 17:41 |
| fungi | jeblair: i already did | 17:41 |
| jeblair | fungi: http://cacti.openstack.org/cacti/graph_view.php?action=tree&tree_id=1&leaf_id=135 | 17:41 |
| jeblair | fungi: zm04 is 4G; do you think we can get away with 2G on a merger? | 17:41 |
| openstackgerrit | A change was merged to openstack-infra/elastic-recheck: Add query for neutron functional job bug 1383391 https://review.openstack.org/129678 | 17:42 |
| clarkb | jeblair: what is the cpu difference? | 17:42 |
| uvirtbot | Launchpad bug 1383391 in neutron "gate-neutron-dsvm-functional fails with "Length too long:"" [High,Confirmed] https://launchpad.net/bugs/1383391 | 17:42 |
| jeblair | clarkb: 2=2 4=4 | 17:42 |
| *** eglynn-office is now known as eglynn-officeafk | 17:42 | |
| clarkb | looking at http://cacti.openstack.org/cacti/graph_view.php?action=tree&tree_id=1&leaf_id=135&page=2 we probably want at least 2 then | 17:43 |
| *** afazekas has joined #openstack-infra | 17:43 | |
| clarkb | and most of the meory is just cache use so 2GB ram wise looks fine | 17:43 |
| *** amcrn has joined #openstack-infra | 17:43 | |
| *** zaphod42_ has joined #openstack-infra | 17:43 | |
| *** tonytan4ever has joined #openstack-infra | 17:44 | |
| jeblair | i mean, we'll use some of that cache for sure, but 1g of cache is probably reasonable for a few git repos :) | 17:44 |
| fungi | jeblair: clarkb: yeah i think the kernel is caching most of those reads for far, far longer than they're relevant | 17:44 |
| *** ivar-lazzaro has joined #openstack-infra | 17:44 | |
| clarkb | jeblair: yeah the graph I am looking at makes it seem like 2G is plent | 17:44 |
| fungi | so in theory we could replace all our 8g mergers with 4x as many to much greater effect | 17:44 |
| mordred | ++ | 17:45 |
| jeblair | yeah. why don't i create zm05-zm08 as 2g machines, then instead of doing this restart cycle, i just replace zm01-zm04? | 17:46 |
| *** ivar-lazzaro has quit IRC | 17:46 | |
| clarkb | jeblair: that sounds like an excellent plan | 17:46 |
| *** ivar-lazzaro has joined #openstack-infra | 17:46 | |
| *** zaphod42 has quit IRC | 17:47 | |
| fungi | i think that's a much better use of everyone's time | 17:47 |
| *** pc_m has quit IRC | 17:47 | |
| clarkb | zaro: how goes the testing? | 17:48 |
| mriedem | i'm guessing this was a short term infra blip last week? http://goo.gl/vUUMm5 | 17:48 |
| clarkb | (still waiting on jenkins to build the final artifact for us) | 17:48 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/system-config: Add zm05-zm08 zuul mergers https://review.openstack.org/129684 | 17:48 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/system-config: Use zm05-zm08 https://review.openstack.org/129685 | 17:48 |
| *** zaphod42_ has quit IRC | 17:48 | |
| clarkb | mriedem: yes it turns out that apache documentation isnt really great around SSLProtocol | 17:48 |
| mriedem | clarkb: ok | 17:48 |
| clarkb | mriedem: so our POODLE mitigation caused a blip | 17:48 |
| mriedem | just trying to get our classification rates up | 17:48 |
| mriedem | i figured | 17:49 |
| mriedem | clarkb: was a bug used to fix that do you know? | 17:49 |
| fungi | mriedem: particularly i didn't think to test our proposed configuration change on centos's apache+openssl versions | 17:49 |
| clarkb | mriedem: no I think we just ninja'd it | 17:49 |
| mriedem | ok | 17:49 |
| clarkb | fungi: well because apache and redhat docs said it was fine :P | 17:49 |
| fungi | mriedem: so i had only manually verified it was working on an ubuntu machine | 17:49 |
| openstackgerrit | yolanda.robla proposed a change to openstack-infra/storyboard-webclient: Allow assigning tasks without expanding the edition box https://review.openstack.org/125643 | 17:49 |
| clarkb | apache 2.2 + openssl 1.0.1 == goodness | 17:50 |
| clarkb | except for when it doesnt | 17:50 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/system-config: Stop using zm01-zm04 https://review.openstack.org/129687 | 17:50 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/system-config: Use zm01-zm04 https://review.openstack.org/129688 | 17:50 |
| fungi | 2.2.23 and later == greatness, but 2.2 versions before 2.2.23 less great | 17:50 |
| fungi | if memory serves | 17:50 |
| clarkb | that sounds correct | 17:51 |
| clarkb | and centos is 2.2.15 | 17:51 |
| jeblair | clarkb, fungi, mordred: if you want to instapprove https://review.openstack.org/#/c/129684/ i can kick off the builds | 17:51 |
| clarkb | jeblair: any reason to not collapse those nodes into a regex? | 17:51 |
| jeblair | i've also restarted the zuul merger on 04 | 17:51 |
| jeblair | clarkb: swell idea. | 17:51 |
| *** pc_m has joined #openstack-infra | 17:51 | |
| fungi | ahh, yep we're at the point where they're turning into an unwieldy list of repetition | 17:53 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/system-config: Stop using zm01-zm04 https://review.openstack.org/129687 | 17:54 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/system-config: Make the zm.. host entries a regex https://review.openstack.org/129684 | 17:54 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/system-config: Use zm05-zm08 https://review.openstack.org/129685 | 17:54 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/system-config: Use zm01-zm04 https://review.openstack.org/129688 | 17:54 |
| clarkb | jeblair: approved the first one. +2 on the others which I figure you can approve as appropriate while building/deleting nodes | 17:57 |
| jeblair | cool, thx. | 17:57 |
| anteaya | devananda: so: https://review.openstack.org/#/c/126627/ https://review.openstack.org/#/c/129348/ what do we want to do? | 17:57 |
| anteaya | adam_g: | 17:57 |
| *** jedimike has quit IRC | 17:57 | |
| * devananda looks | 17:57 | |
| anteaya | thanks | 17:58 |
| *** miqui has quit IRC | 17:58 | |
| adam_g | devananda, anteaya i vote we just try to get 126627 merged with appropriate approval | 17:58 |
| openstackgerrit | Matt Riedemann proposed a change to openstack-infra/elastic-recheck: Add query for infra TLS packet bug 1383403 https://review.openstack.org/129691 | 17:59 |
| uvirtbot | Launchpad bug 1383403 in openstack-ci "fatal: unable to access 'https://git.openstack.org/openstack-infra/devstack-gate/': gnutls_handshake() failed: A TLS packet with unexpected length was received." [Undecided,Fix committed] https://launchpad.net/bugs/1383403 | 17:59 |
| anteaya | adam_g: it is your decision | 17:59 |
| anteaya | adam_g: whichever you choose, abandon or wip the other one | 17:59 |
| * devananda rebases 126627 | 17:59 | |
| adam_g | anteaya, sure | 17:59 |
| *** VijayTripathi has joined #openstack-infra | 18:00 | |
| fungi | jeblair: want to lower the ttl on zm01-04 or just planning to wait an hour for dns updates? | 18:00 |
| anteaya | adam_g: thanks | 18:00 |
| anteaya | I just need to know which one to review | 18:01 |
| anteaya | since it isn't my decision which one you want | 18:01 |
| openstackgerrit | A change was merged to openstack-infra/project-config: remove non-standard-reqs flag from check-tempest-dsvm-f20-docker https://review.openstack.org/129252 | 18:01 |
| fungi | jeblair: not that an hour is a particularly long time anyway | 18:01 |
| clarkb | the neutron full job is really slow... | 18:01 |
| clarkb | salv-orlando: mtreinish ^ is that a known thing? iirc it was expected to be slower but it seems to be averaging > 1 hour now | 18:02 |
| mtreinish | clarkb: yeah it should be slower, and it runs more tests. But I didn't think it was >1hr | 18:03 |
| jeblair | fungi: i'll just wait :) | 18:03 |
| *** sabeen2 has joined #openstack-infra | 18:03 | |
| *** Ryan_Lane has joined #openstack-infra | 18:03 | |
| *** dane_leblanc has joined #openstack-infra | 18:03 | |
| *** dkehn has joined #openstack-infra | 18:04 | |
| clarkb | mtreinish: jenkins is currently predicting 1 hour and 6ish minutes for the one I am waiting on | 18:04 |
| adam_g | that reminds me, the ironic forward grenade jobs are going to be *really* slow. the current sideways grenade runtime is 50min-1hr with only a single run of tempest (against target). forward is going to be 2x that | 18:04 |
| openstackgerrit | Devananda van der Veen proposed a change to openstack-infra/project-config: Update Ironic jobs post-graduation https://review.openstack.org/126627 | 18:04 |
| anteaya | adam_g: can you put a comment in the code with that information | 18:04 |
| mtreinish | anteaya: it should only run smoke on grenade which should be far fewer tests | 18:05 |
| mtreinish | adam_g: ^^ sry | 18:05 |
| *** sabeen1 has quit IRC | 18:05 | |
| adam_g | anteaya, in what code? | 18:05 |
| *** todd_dsm has joined #openstack-infra | 18:05 | |
| anteaya | adam_g: in the sideways job | 18:05 |
| devananda | adam_g: also, making the ironic parallel job voting is going to slow things down as well. the current dsvm-ironic-pxe_ssh job takes ~25 min, not an hour | 18:05 |
| adam_g | mtreinish, right but each server can take minutes to boot | 18:05 |
| jeblair | clarkb: want to review the kerberos change? (it has 2x+2 now) | 18:06 |
| anteaya | devananda adam_g so we like 126627 for review? | 18:06 |
| *** erlon_ has joined #openstack-infra | 18:06 | |
| clarkb | jeblair: sure | 18:06 |
| *** sarob has quit IRC | 18:06 | |
| adam_g | anteaya, yes plz | 18:06 |
| devananda | anteaya: I believe so. I would like to make sure mikal is OK with it before it is landed | 18:06 |
| mtreinish | clarkb: hmm, that is definitely slower than I'd like, 1hr should be the upper limit for a job | 18:06 |
| * devananda adds mikal to the review | 18:07 | |
| anteaya | devananda: thanks and can you add a comment ot that effect as well, so the approver waits for a comment from mikal? | 18:07 |
| zaro | clarkb: distracted by a code review. will test GP now. | 18:08 |
| *** afazekas has quit IRC | 18:08 | |
| *** jlibosva has quit IRC | 18:08 | |
| jeblair | and i'm +2 on the afs change too | 18:08 |
| anteaya | so dhellmann apparently this patch is waiting on a comment from you: https://review.openstack.org/#/c/129433/1 | 18:08 |
| clarkb | https://review.openstack.org/#/c/129487/1 for changes like that when do we make them their own modules vs put in system-config? | 18:09 |
| devananda | anteaya: done | 18:09 |
| clarkb | same question with kerberos | 18:09 |
| clarkb | I really think we shouldn't dig a deeper hole | 18:09 |
| anteaya | devananda: thank you | 18:09 |
| clarkb | especially since the other thing has been labeled a priority | 18:09 |
| jeblair | anyone else want to review the bashate change? | 18:10 |
| *** thedodd has joined #openstack-infra | 18:10 | |
| jeblair | clarkb: that's a really good question. otoh, this is an experiment, and we might delete the whole thing. | 18:11 |
| openstackgerrit | A change was merged to openstack-infra/system-config: Split haveged into its own class https://review.openstack.org/129487 | 18:11 |
| *** dimsum_ has joined #openstack-infra | 18:11 | |
| *** e0ne has quit IRC | 18:12 | |
| clarkb | jeblair: I think I wouldn't care too much if splitting out modules isn't one of our priorities | 18:12 |
| jeblair | clarkb: we could make sure it's mordred and i who do the split of this, if we decide to keep it, so that we aren't making too much extra work for others. | 18:13 |
| *** dbite has quit IRC | 18:13 | |
| fungi | i agree i didn't think to ask the question and should have. on the other hand not adding a separate repo just yet until we're satisfied with the results is a good argument | 18:13 |
| fungi | (two separate repos, in fact_ | 18:13 |
| jeblair | clarkb: but this in particular is such a big experiment, i think i want to avoid committing ourselves to it in any way. | 18:13 |
| *** dimtruck is now known as zz_dimtruck | 18:13 | |
| clarkb | ok | 18:13 |
| openstackgerrit | A change was merged to openstack-infra/system-config: Make the zm.. host entries a regex https://review.openstack.org/129684 | 18:14 |
| openstackgerrit | A change was merged to openstack-infra/project-config: Add docs jobs for subunit2sql https://review.openstack.org/129409 | 18:14 |
| jeblair | honestly, i wasn't even expecting puppet at all so soon, but mordred was like "oh, i have a thing that's almost what we need" and he ended up puppeting what we were doing as we went along :) | 18:15 |
| *** emagana has joined #openstack-infra | 18:15 | |
| *** afazekas has joined #openstack-infra | 18:16 | |
| clarkb | does ubuntu not ship an init script for kpropd? | 18:16 |
| jeblair | clarkb, fungi, SergeyLukjanov, mordred: bashate has 2x+2. if you want to review it, let me know. | 18:16 |
| jeblair | clarkb: nope. it is traditionally most often run from inetd. | 18:16 |
| clarkb | jeblair: I am ok with bashate going through without my review | 18:16 |
| clarkb | (its a big change and lots going on so may not get there today) | 18:16 |
| fungi | yeah, two reviews of whitespace is plenty already i think | 18:16 |
| jeblair | clarkb: i thought it'd be better to run it as a daemon than be the first thing we ran out of inetd. :) | 18:17 |
| fungi | clarkb: i actually asked the same question on patchset 11 | 18:17 |
| jeblair | (also, daemon mode is apparently somehow required for incremental propogation though i do not understand the details. yet. and we are not doing that. yet.) | 18:18 |
| *** mwagner_lap has joined #openstack-infra | 18:19 | |
| jeblair | jesusaurus: +3 | 18:19 |
| *** hdd has quit IRC | 18:19 | |
| *** hdd has joined #openstack-infra | 18:20 | |
| jesusaurus | jeblair: awesome, thanks | 18:20 |
| clarkb | mtreinish: https://jenkins02.openstack.org/job/gate-tempest-dsvm-neutron-full/3372/ well into the slow zone | 18:20 |
| jesusaurus | jeblair: there is also 129667 to follow it up and make the test voting | 18:20 |
| *** mpaolino has joined #openstack-infra | 18:21 | |
| jeblair | zm05-zm08 building | 18:21 |
| *** e0ne has joined #openstack-infra | 18:21 | |
| *** chandankumar has joined #openstack-infra | 18:22 | |
| mtreinish | clarkb: hmm it looks like the testr slowest output isn't being displayed on the console anymore... | 18:22 |
| jeblair | jesusaurus: thanks, will +3 when the other lands | 18:22 |
| *** chandankumar has quit IRC | 18:22 | |
| *** baoli has quit IRC | 18:22 | |
| jeblair | mordred: i'll write some afs docs unless you have already started? | 18:22 |
| *** pelix has quit IRC | 18:22 | |
| *** otherwiseguy has quit IRC | 18:23 | |
| *** baoli has joined #openstack-infra | 18:23 | |
| clarkb | mtreinish: we log the subunit log right? we can always grab that and post process | 18:23 |
| clarkb | mtreinish: unsure why it wouldn't show slowest for you though | 18:23 |
| mtreinish | clarkb: yeah I already did that, we just call setup.py test ith --slowest, and it used to be there | 18:24 |
| clarkb | maybe your filter thing is killing it? | 18:24 |
| mtreinish | yeah, that's what I'm thinking | 18:25 |
| *** wenlock has joined #openstack-infra | 18:25 | |
| openstackgerrit | Michael Krotscheck proposed a change to openstack-infra/storyboard: Add User Preferences endpoint. https://review.openstack.org/128487 | 18:26 |
| *** pelix has joined #openstack-infra | 18:26 | |
| *** Guest59036 has joined #openstack-infra | 18:26 | |
| *** thedodd has quit IRC | 18:27 | |
| mtreinish | clarkb: http://paste.openstack.org/show/122288/ | 18:27 |
| mtreinish | it's the lvm bug which is pushing things super long I think | 18:27 |
| clarkb | lvm bug? | 18:27 |
| clarkb | the only one I know if is fixed | 18:27 |
| mtreinish | clarkb: https://bugs.launchpad.net/cinder/+bug/1373513 | 18:28 |
| uvirtbot | Launchpad bug 1373513 in cinder "Lvm hang during tempest tests" [Critical,Confirmed] | 18:28 |
| clarkb | gah my grenade change decided to fail | 18:28 |
| *** spzala has joined #openstack-infra | 18:29 | |
| *** emagana has quit IRC | 18:31 | |
| mtreinish | clarkb: looks like rax mirror failure | 18:31 |
| mtreinish | http://logs.openstack.org/59/128959/1/gate/gate-grenade-dsvm-partial-ncpu/d245375/logs/grenade.sh.txt.gz#_2014-10-20_18_25_31_913 | 18:31 |
| *** emagana has joined #openstack-infra | 18:31 | |
| *** torandu has joined #openstack-infra | 18:32 | |
| *** torandu has quit IRC | 18:32 | |
| openstackgerrit | A change was merged to openstack-infra/elastic-recheck: Add query for python-openstackclient test bug 1383396 https://review.openstack.org/129679 | 18:33 |
| uvirtbot | Launchpad bug 1383396 in python-openstackclient "tests.test_identity.IdentityV3Tests.test_token_issue fails with "ERROR: openstack 'NoneType' object has no attribute 'service_catalog'"" [Undecided,Confirmed] https://launchpad.net/bugs/1383396 | 18:33 |
| openstackgerrit | A change was merged to openstack-infra/elastic-recheck: Add query for infra TLS packet bug 1383403 https://review.openstack.org/129691 | 18:33 |
| uvirtbot | Launchpad bug 1383403 in openstack-ci "fatal: unable to access 'https://git.openstack.org/openstack-infra/devstack-gate/': gnutls_handshake() failed: A TLS packet with unexpected length was received." [Undecided,Fix committed] https://launchpad.net/bugs/1383403 | 18:33 |
| openstackgerrit | Michael Krotscheck proposed a change to openstack-infra/storyboard: Add User Preferences endpoint. https://review.openstack.org/128487 | 18:33 |
| *** hemnass is now known as hemna | 18:33 | |
| openstackgerrit | Michael Krotscheck proposed a change to openstack-infra/storyboard: Add User Preferences endpoint. https://review.openstack.org/128487 | 18:34 |
| *** mrmartin has quit IRC | 18:35 | |
| openstackgerrit | A change was merged to openstack-infra/system-config: Clean up bashate failures https://review.openstack.org/118944 | 18:35 |
| *** emagana_ has joined #openstack-infra | 18:35 | |
| *** emagana has quit IRC | 18:36 | |
| clarkb | mordred: why a separate puppet run for afs? | 18:36 |
| jeblair | krotscheck: old patchset of 128487 reviewed. | 18:36 |
| clarkb | mordred: also what tells afsfs where the root of the afs is? | 18:36 |
| jeblair | clarkb: explained in the ansible file -- so that the db servers maintain quorum | 18:36 |
| krotscheck | jeblair: Thanks, will look | 18:36 |
| jeblair | clarkb: i may need more info on the second question, but i think i know where you're coming from and will take a stab | 18:37 |
| clarkb | jeblair: https://review.openstack.org/#/c/129450/12/modules/openstack_project/files/ansible/remote_puppet_afs.yaml I don't see that info there? | 18:38 |
| jeblair | clarkb: the puppet manifest currently does not do any managing of things related to the actual data served by a fileserver | 18:38 |
| *** afazekas has quit IRC | 18:38 | |
| clarkb | oh its in the sh script | 18:39 |
| jeblair | clarkb: is that the general direction you were heading with that question? | 18:39 |
| clarkb | jeblair: ya | 18:39 |
| jeblair | clarkb: sorry, yeah, i knew i saw that somewhere | 18:39 |
| clarkb | jeblair: I assume you say afs serve /afs/root somewhere | 18:39 |
| *** tonytan4ever has quit IRC | 18:39 | |
| jeblair | clarkb: so the missing info related to that is that fileservers have data stored on mounted partitions | 18:40 |
| jeblair | clarkb: our servers each have one lvm volume mounted at /vicepa | 18:40 |
| *** oramraz has quit IRC | 18:40 | |
| jeblair | clarkb: the mountpoints are standardized in afs and hardcoded, so it just knows to look there (vicepa, vicepb, vicepc..) | 18:40 |
| clarkb | gotcha | 18:40 |
| *** torandu has joined #openstack-infra | 18:41 | |
| *** torandu has quit IRC | 18:41 | |
| fungi | viceps are like a fancy kind of forceps ;) | 18:41 |
| jeblair | yuk yuk | 18:41 |
| *** Guest52830 is now known as mgagne | 18:42 | |
| *** mgagne has joined #openstack-infra | 18:42 | |
| *** pelix has quit IRC | 18:42 | |
| anteaya | or vicegrips | 18:42 |
| fungi | indeed | 18:43 |
| zaro | clarkb: tested GP on jenkins-dev. looks good. | 18:43 |
| *** zz_dimtruck is now known as dimtruck | 18:43 | |
| clarkb | zaro: awesome. As soon as jenkins builds an artifact for us I will start planning to roll it out | 18:43 |
| clarkb | (job is running now then board meeting then lunch so it may be a bit >_>) | 18:43 |
| clarkb | https://jenkins04.openstack.org/job/gearman-plugin-hpi-artifact/1/ | 18:43 |
| jeblair | clarkb: when we start talking about afs volumes, and how clients actually navigate around the filesystem, it's going to blow your mind. (it's so much better than anything like nfs) | 18:43 |
| clarkb | jeblair: I will prepare my skill | 18:44 |
| clarkb | er skull | 18:44 |
| *** jp_at_hp has quit IRC | 18:44 | |
| clarkb | zaro: looks like a test failed but the artifact was uploaded anyways | 18:45 |
| clarkb | zaro: is that expected? | 18:45 |
| *** andreykurilin_ has quit IRC | 18:45 | |
| openstackgerrit | A change was merged to openstack-infra/subunit2sql: Remove openstack-common.conf https://review.openstack.org/129587 | 18:45 |
| *** kgiusti has quit IRC | 18:45 | |
| zaro | np. i will need to be afk for a few hrs in about 2hrs. | 18:45 |
| zaro | clarkb: yes, | 18:46 |
| zaro | clarkb: i think it's flaky so jeblair set to ignore failure. | 18:46 |
| krotscheck | jeblair: So, about ‘deleting’ preferences. I… kindof feel that in some cases, a blank string could be a valid preference value. | 18:46 |
| krotscheck | jeblair: So I’m thinking of switching that to “value is None” to force the user to explicitly send null. | 18:47 |
| jeblair | krotscheck: sounds like a plan | 18:47 |
| *** baoli has quit IRC | 18:47 | |
| *** yfried has joined #openstack-infra | 18:48 | |
| devananda | question for ya'll - how does Jenkins get attributed as the author of code? | 18:48 |
| devananda | https://github.com/openstack/ironic/blame/master/ironic/tests/conductor/test_rpcapi.py#L131-L145 | 18:48 |
| *** yfried has quit IRC | 18:48 | |
| devananda | this method landed without being in either parent of that merge, as far as I can tell, and fails our pep8 checks | 18:48 |
| fungi | devananda: code that was changed by a merge commit in this case | 18:49 |
| devananda | so I must be missing something | 18:49 |
| *** yfried has joined #openstack-infra | 18:49 | |
| devananda | fungi: a whole new function? | 18:49 |
| jeblair | 0fc3ad85 ironic/tests/conductor/test_rpcapi.py (Devananda van der Veen 2014-01-04 14:57:05 -0800 59) def test_get_topic_for_known_driver(self): | 18:49 |
| *** kgiusti has joined #openstack-infra | 18:49 | |
| jeblair | devananda: file a bug with github? :) | 18:49 |
| devananda | jeblair: it's in our git tree | 18:49 |
| fungi | yeah, i think their ability to separate merge commits is broken-ish | 18:49 |
| devananda | github is just giving me a UI to show the blame, where as I'm not sure how to do that in our cgit | 18:49 |
| fungi | devananda: did you try with the git cli? | 18:50 |
| devananda | yes | 18:50 |
| devananda | git blame ironic/tests/conductor/test_rpcapi.py | grep "131)" | 18:50 |
| devananda | deb21ef7 ironic/tests/conductor/test_rpcapi.py (Jenkins 2014-10-20 15:35:46 +0000 131) def test_get_topic_for_driver_doesnt_cache(self): | 18:50 |
| jeblair | oh, sorry, i was looking at the wrong test | 18:50 |
| clarkb | I prefer git log -p | 18:51 |
| clarkb | (over blame as it gives you more history) | 18:51 |
| *** jgrimm is now known as zz_jgrimm | 18:52 | |
| clarkb | devananda: 3807e9447634c5ce6d9f65cac48ddc9bc0f9d295 | 18:52 |
| *** yolanda_ has quit IRC | 18:52 | |
| fungi | devananda: git show on the merge commit indicates it's bringing in 3807e94 | 18:52 |
| fungi | er, the one clarkb just linked | 18:53 |
| devananda | right - that's on proposed/juno | 18:53 |
| *** yfried has quit IRC | 18:53 | |
| clarkb | its master according to git log -p | 18:53 |
| devananda | maybe i'm just confused by seeing a block of code attributed to Jenkins as the author of it in our git history | 18:54 |
| devananda | rather than a human | 18:54 |
| clarkb | git log --oneline --graph shows where it ended up in that merge | 18:54 |
| clarkb | devananda: because "Jenkins" hit the submit button in gerrit so the merge commit is attributed to it | 18:55 |
| clarkb | devananda: but jenkins didn't write the code | 18:55 |
| *** yfried has joined #openstack-infra | 18:55 | |
| *** jcoufal has quit IRC | 18:55 | |
| fungi | and unlike the majority of merge commits you'll see, this one was non-null (the two parent commits modified the same file) | 18:56 |
| *** rushiagr_away is now known as rushiagr | 18:57 | |
| jeblair | let's back up a step. | 18:57 |
| jeblair | i believe we hit that rare gerrit bug fungi found | 18:57 |
| jeblair | i think that a juno commit was merged into master because the master commit dependended on the juno commit. | 18:58 |
| fungi | yes, this is quite possibly the same behavior | 18:58 |
| jeblair | https://review.openstack.org/#/c/129174/ | 18:58 |
| jeblair | https://review.openstack.org/#/c/129174/ (in master) depends on https://review.openstack.org/#/c/126984/1 (in juno) | 18:59 |
| fungi | yep | 18:59 |
| devananda | ah | 18:59 |
| fungi | that's definitely the same issue we spotted before | 18:59 |
| fungi | so 129174 was mis-proposed to master starting from a commit on a different branch | 18:59 |
| jeblair | and that's why there was a non-empty merge commit from jenkins | 19:00 |
| *** ZZelle_ has joined #openstack-infra | 19:00 | |
| devananda | oooh. I see. | 19:00 |
| fungi | likely .gitreview had master as the defaultbranch in proposed/juno | 19:00 |
| devananda | it wasn't apparent in https://review.openstack.org/#/c/129174/ that it was depending on not-master | 19:00 |
| devananda | thanks guys | 19:00 |
| * devananda jumps into the weekly meeting | 19:00 | |
| fungi | devananda: well, it was, just not something you'd be likely to keep an eye out for | 19:00 |
| devananda | s/apparent/obvious/ :) | 19:00 |
| fungi | devananda: you can see there is says the dependency is on branch proposed/juno | 19:00 |
| clarkb | is that something we should add a job check for? | 19:01 |
| fungi | but yes, easy to overlook | 19:01 |
| *** pc_m has quit IRC | 19:01 | |
| devananda | yep | 19:01 |
| fungi | it definitely doesn't scream at you | 19:01 |
| *** pc_m has joined #openstack-infra | 19:02 | |
| *** yfried is now known as yfried|afk | 19:02 | |
| *** _nadya_ has joined #openstack-infra | 19:02 | |
| fungi | ideally i think it's either gerrit not operating as intended, or it's sorely lacking in a configuration option we'd like to set to disable it | 19:02 |
| jeblair | it should at least be in red :) | 19:03 |
| fungi | unfortunately gerrit developer weigh-in on the bug report ranged from not understanding why it was a problem to suggesting there might be use cases where people want this to work the way it is now | 19:03 |
| fungi | devananda: https://code.google.com/p/gerrit/issues/detail?id=1107 | 19:04 |
| *** emagana_ has quit IRC | 19:05 | |
| *** emagana has joined #openstack-infra | 19:05 | |
| *** dane_leblanc has quit IRC | 19:06 | |
| fungi | though i think what happened in ironic's repository only tickled a little of that... the issue extends to even allowing people to propose existing commits from other branches which the acls should be expressly forbidding (for example, merge commits) | 19:07 |
| fungi | er, not propose, but propose changes which depend on | 19:07 |
| fungi | and which don't yet exist on the target branch and which they lack the permission to be able to propose otherwise | 19:08 |
| openstackgerrit | Matt Riedemann proposed a change to openstack-infra/elastic-recheck: Add query for javelin KeyError bug 138339 https://review.openstack.org/129704 | 19:08 |
| *** zz_jgrimm is now known as jgrimm | 19:08 | |
| *** oramraz has joined #openstack-infra | 19:08 | |
| fungi | and thinking on it, i can see where not allowing this would also break some relatively obvious workflows | 19:09 |
| fungi | including some of ours | 19:09 |
| *** emagana has quit IRC | 19:10 | |
| jeblair | zm05-08 are up; merging the firewall change for them | 19:10 |
| *** oramraz has quit IRC | 19:10 | |
| clarkb | jeblair: don't forget cacti updates | 19:10 |
| jeblair | clarkb: oh thanks! | 19:11 |
| clarkb | I just remembered them :) | 19:11 |
| jeblair | er, where did they go? | 19:12 |
| clarkb | zm01-04 should be fine | 19:12 |
| clarkb | but 05-08 need new entries right? | 19:12 |
| *** rushiagr is now known as rushiagr_away | 19:13 | |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/system-config: Use zm05-zm08 https://review.openstack.org/129685 | 19:13 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/system-config: Add zm05-zm08 to cacti https://review.openstack.org/129706 | 19:13 |
| jeblair | er, oops. | 19:13 |
| *** otherwiseguy has joined #openstack-infra | 19:13 | |
| *** emagana has joined #openstack-infra | 19:14 | |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/system-config: Add zm05-zm08 to cacti https://review.openstack.org/129706 | 19:14 |
| jeblair | clarkb: that should be able to be instamerged | 19:14 |
| clarkb | done | 19:15 |
| *** dane_leblanc has joined #openstack-infra | 19:16 | |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/system-config: Stop using zm01-zm04 https://review.openstack.org/129687 | 19:16 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/system-config: Use zm01-zm04 https://review.openstack.org/129688 | 19:16 |
| clarkb | ok I have new gearman-plugin hpi uploaded to all jenkins non dev servers can be found in /home/clarkb/plugins/gearman-plugin | 19:19 |
| clarkb | any reason for me to not take 07 and 06 offline now to begin the upgrade dance? | 19:19 |
| *** dimtruck is now known as zz_dimtruck | 19:19 | |
| clarkb | er put them in shutdown mode | 19:19 |
| *** stevemar has quit IRC | 19:19 | |
| fungi | seems fine--we're at a point where spreading the current node demand across fewer masters shouldn't be dangerous | 19:21 |
| fungi | you could probably safely do 3 or 4 if you want | 19:21 |
| mriedem | clarkb: how hard would it be to get a field in logstash that says whether a job is voting or not? i figure we already have build_queue, so a voting field shouldn't be crazy? | 19:22 |
| *** mpaolino has quit IRC | 19:22 | |
| clarkb | mriedem: its relatively hard. jenkins has none of that info so can't provide it to logstash today | 19:23 |
| mriedem | but jenkins has build_queue info? | 19:23 |
| clarkb | yes zuul provides that to jenknis | 19:23 |
| *** zz_dimtruck is now known as dimtruck | 19:23 | |
| *** sabeen2 has quit IRC | 19:23 | |
| *** andreykurilin_ has joined #openstack-infra | 19:23 | |
| *** sabeen has joined #openstack-infra | 19:23 | |
| mriedem | blarg, ok | 19:24 |
| mriedem | mtreinish: ^ | 19:24 |
| mriedem | mtreinish: so it's not trivial, unless we whitelist the voting gate jobs in e-r, which sucks | 19:24 |
| fungi | yeah, i think we talked about it previously, and it would need to be a parameter zuul puts into the gearman work request | 19:24 |
| clarkb | fungi: yup | 19:25 |
| fungi | so modifying zuul, the jenkins gearman plugin, and the jenkins logstash client i think? | 19:25 |
| *** yfried|afk is now known as yfried | 19:25 | |
| fungi | (to plumb that all the way in) | 19:25 |
| mtreinish | mriedem: well we could always revisit the idea of polling the layout.yaml from zuul and doing it that way. Although that adds it's own complexity | 19:25 |
| clarkb | fungi: I think just zuul and logstash client | 19:25 |
| clarkb | the gearman plugin already does the right hting for parameters | 19:25 |
| fungi | oh, so not quite as bad (no java need be written) | 19:26 |
| mriedem | heh, i know java :) | 19:26 |
| mriedem | are we running non-voting jobs in the gate on purpose? mtreinish thought we weren't. | 19:26 |
| *** markmcclain has joined #openstack-infra | 19:26 | |
| clarkb | mriedem: we shouldn't be | 19:26 |
| clarkb | but I think there are a few cases where we do | 19:27 |
| fungi | mriedem: i think when we do it's only a very short-term/temporary measure? | 19:27 |
| mriedem | http://goo.gl/ami8VW | 19:27 |
| clarkb | ya I think that is one exception | 19:27 |
| mriedem | i've opened at least 2 bugs today for gate failures in non-voting jobs | 19:27 |
| clarkb | but no one is working on making it work | 19:27 |
| clarkb | because :( | 19:27 |
| mriedem | this is the other gate-osc-dsvm-functional | 19:27 |
| mriedem | i think sdague created that pre-baby | 19:28 |
| *** andreaf has joined #openstack-infra | 19:28 | |
| fungi | right, i think both of these may be examples of a temporary state which has been extended slightly by unrelated real-world events | 19:28 |
| mtreinish | clarkb: so I know there is ongoing work on getting javelin2 and grenade to work with neutron, but that doesn't seem like a reason to run it non-voting in the gate pipeline | 19:29 |
| clarkb | mtreinish: I think the reason is it should e voting but doesn't work | 19:29 |
| clarkb | maybe we should just make it voting >_> | 19:30 |
| mtreinish | clarkb: it doesn't work | 19:30 |
| clarkb | I know | 19:30 |
| fungi | mtreinish: what better point in the cycle to block broken software! | 19:30 |
| *** radez_g0n3 is now known as radez | 19:30 | |
| fungi | "you must be this tall to ride the kilo" | 19:31 |
| openstackgerrit | A change was merged to openstack-infra/elastic-recheck: Add query for javelin KeyError bug 138339 https://review.openstack.org/129704 | 19:31 |
| fungi | but yeah, that would be pretty painful | 19:32 |
| clarkb | ok marking 07 and 06 for shutdown mode now | 19:32 |
| openstackgerrit | Doug Hellmann proposed a change to openstack-infra/project-config: Add jenkins-project to the default env list https://review.openstack.org/126579 | 19:32 |
| clarkb | mtreinish: basically we said at some point you must grenade | 19:33 |
| clarkb | mtreinish: turned on the test | 19:33 |
| clarkb | then quickl went gah it doesn't work! make it non voting | 19:33 |
| mtreinish | fungi: yeah it also wouldn't be productive because it would block grenade patches too not just neutron. Especially considering we're trying to land the kilo migration there | 19:33 |
| clarkb | thinking it would be fixed ASAP | 19:33 |
| mtreinish | clarkb: sigh, yeah that's how it normally goes. I'm just thinking we should drop it from the gate pipeline and just leave it non-voting on check | 19:34 |
| mtreinish | or move it to experimental | 19:34 |
| fungi | i'd be fine with either of those. it's just a waste of resources in the gate pipeline at the moment | 19:35 |
| *** wenlock has quit IRC | 19:35 | |
| jroll | hey y'all, we had a really weird merge happen, I outlined it a bit here: https://bugs.launchpad.net/ironic/+bug/1383436 | 19:35 |
| uvirtbot | Launchpad bug 1383436 in ironic "Duplicate method breaks pep8" [Critical,In progress] | 19:35 |
| jroll | tl;dr a merge commit added code not in the commit being merged | 19:35 |
| mtreinish | fungi: yeah, that was my only point that we were wasting resources | 19:35 |
| mtreinish | and confusing mriedem... | 19:35 |
| mtreinish | I'll push out a patch that drops it from gate | 19:36 |
| fungi | jroll: https://code.google.com/p/gerrit/issues/detail?id=1107 | 19:36 |
| anteaya | AJaeger_: might you happen to be about? | 19:36 |
| fungi | jroll: is it the same commit devananda pointed out where it was parented to the wrong branch? | 19:37 |
| anteaya | AJaeger_: I'm looking at: https://review.openstack.org/#/c/129541/1/gerrit/acls/stackforge/requests-mock.config | 19:37 |
| anteaya | AJaeger_: and you had a comment and then withdrew it, and I am wondering why you withdrew it | 19:37 |
| *** dimtruck is now known as zz_dimtruck | 19:37 | |
| anteaya | AJaeger_: and currently have no vote on the patch | 19:37 |
| *** yfried has quit IRC | 19:38 | |
| *** Guest59036 has quit IRC | 19:38 | |
| *** miqui has joined #openstack-infra | 19:38 | |
| jroll | fungi: yeah, I didn't see that conversation | 19:38 |
| *** wenlock has joined #openstack-infra | 19:39 | |
| openstackgerrit | Matthew Treinish proposed a change to openstack-infra/project-config: Remove grenade neutron non-voting from the gate https://review.openstack.org/129711 | 19:40 |
| fungi | jroll: so, the tl;dr is that the change was committed with a parent which was on proposed/juno instead of master, then the change was proposed to master. gerrit displays that relationship but it's not obvious if you're not looking for it. if you approve it, gerrit will allow it to merge (and will adjust the target branch with other required commits from the parent branch as necessary to do so) | 19:40 |
| mtreinish | fungi, clarkb, mriedem: ^^^ it was actually only in the gate for grenade... | 19:40 |
| jroll | fungi: cool, thanks | 19:40 |
| openstackgerrit | A change was merged to openstack-infra/system-config: Use zm05-zm08 https://review.openstack.org/129685 | 19:41 |
| mriedem | mtreinish: throw related-bug: 1383439 in there | 19:41 |
| openstackgerrit | Matthew Treinish proposed a change to openstack-infra/project-config: Remove grenade neutron non-voting from the gate https://review.openstack.org/129711 | 19:42 |
| mriedem | +1 | 19:42 |
| mtreinish | mriedem: ok done, although it's not really related to the bug :) | 19:42 |
| mriedem | mtreinish: well, the bug is a good reason why we can't make it non-voting | 19:42 |
| *** tonytan4ever has joined #openstack-infra | 19:44 | |
| *** baoli has joined #openstack-infra | 19:44 | |
| *** baoli has quit IRC | 19:45 | |
| *** baoli has joined #openstack-infra | 19:46 | |
| dhellmann | anteaya, fungi, AJaeger_: I'm not sure about https://review.openstack.org/#/c/129433/1 but I think we need to keep oslo libs testing under 2.6 for now to maintain support between master versions of the libs and stable versions of the apps? | 19:46 |
| fungi | dhellmann: but not incubator, correct? | 19:47 |
| *** hashar has joined #openstack-infra | 19:47 | |
| dhellmann | fungi: it should be safe to drop 2.6 testing in the incubator, yes | 19:47 |
| fungi | dhellmann: i was just thrown by the fact that the two libs listed had stable branches (one has a stable/icehouse and the other has a stable/juno) | 19:47 |
| dhellmann | fungi: well, on master, but not on the stable branches | 19:47 |
| fungi | dhellmann: what are those branches being used for? | 19:47 |
| dhellmann | fungi: yeah, we ended up needing patch releases | 19:47 |
| dhellmann | fungi: afaik, we don't continue to release from those branches, we just used them to prepare the patch releases | 19:48 |
| *** zz_dimtruck is now known as dimtruck | 19:48 | |
| fungi | dhellmann: aha. we could likely delete those branches if the tips of then ended up as release tags | 19:48 |
| dhellmann | we could, conceivably, use them to release more patches for juno as we make alpha releases during kilo, but I don't think we really have the review teams to support that | 19:48 |
| fungi | er, of them | 19:48 |
| *** sabeen has quit IRC | 19:49 | |
| fungi | dhellmann: but also, it may make more sense to name branches after versions you're branching to make the point releases | 19:49 |
| *** groknix has quit IRC | 19:49 | |
| openstackgerrit | A change was merged to openstack-infra/system-config: Add zm05-zm08 to cacti https://review.openstack.org/129706 | 19:49 |
| fungi | dhellmann: for the integrated release servers we name them after release codenames, granted, but the oslo libs aren't part of the integrated release, so having branches named for those is sort of misleading | 19:49 |
| dhellmann | yeah, ttx and I went back and forth a bit on that and decided to go with the usual standard, I think because the oslo.messaging repo had it as a precedent | 19:50 |
| dhellmann | fungi: I guess. We follow the cadence, and that was the cycle when the release was needed.That's easier to keep up with than the numbers. | 19:50 |
| openstackgerrit | Antoine Musso proposed a change to openstack-infra/zuul: Update merge status after merge:merge is submitted https://review.openstack.org/128921 | 19:50 |
| fungi | dhellmann: okay, makes some sense at least | 19:50 |
| *** whoops has joined #openstack-infra | 19:50 | |
| *** markmcclain has quit IRC | 19:52 | |
| openstackgerrit | Matthew Treinish proposed a change to openstack-infra/subunit2sql: Flush out the db api documenation https://review.openstack.org/129405 | 19:53 |
| openstackgerrit | Matthew Treinish proposed a change to openstack-infra/subunit2sql: Reorganize the python api documentation https://review.openstack.org/129406 | 19:53 |
| openstackgerrit | Matthew Treinish proposed a change to openstack-infra/subunit2sql: Add section on the basic data model https://review.openstack.org/129407 | 19:53 |
| mordred | jeblair: just for the record, I have just listed files in AFS and created one while in a plane | 19:53 |
| openstackgerrit | Matthew Treinish proposed a change to openstack-infra/system-config: Add subunit2sql gearman workers https://review.openstack.org/108003 | 19:53 |
| mordred | jeblair: and it worked well | 19:54 |
| openstackgerrit | Matthew Treinish proposed a change to openstack-infra/system-config: Add subunit2sql server configuraion https://review.openstack.org/127060 | 19:54 |
| *** koolhead17 has quit IRC | 19:55 | |
| mtreinish | clarkb, fungi: ^^^ I just had a bunch of merge conflicts reported from gerrit but there weren't any that I could see | 19:56 |
| mtreinish | I rebased to make it go away | 19:56 |
| clarkb | mtreinish: probably a jgit thing | 19:56 |
| *** emagana has quit IRC | 19:56 | |
| hashar | same deal here | 19:56 |
| openstackgerrit | Michael Krotscheck proposed a change to openstack-infra/storyboard: Add User Preferences endpoint. https://review.openstack.org/128487 | 19:56 |
| dhellmann | fungi: do you have a few minutes to help me figure out why the cliff and stevedore doc jobs aren't publishing to docs.openstack.org/developer? | 19:57 |
| dhellmann | fungi: http://logs.openstack.org/72/728aea721bf6b05402c0e06566df3ff6cc0b996c/ | 19:57 |
| dhellmann | fungi: actually this shows transfered 0 files: http://logs.openstack.org/72/728aea721bf6b05402c0e06566df3ff6cc0b996c/post/cliff-docs/fb7a791/console.html.gz#_2014-10-16_21_44_12_253 | 19:58 |
| * dhellmann wonders if he missed a configuration setting somewhere | 19:58 | |
| *** dprince has quit IRC | 19:58 | |
| clarkb | dhellmann: where do the docs build? | 19:58 |
| clarkb | dhellmann: should be doc/build/html iirc | 19:59 |
| dhellmann | clarkb: ah, maybe that's it. I think they're going to "docs/..." not "doc/..." | 19:59 |
| hashar | your zuul might well be in trouble Queue lengths: 306 events, 140 results. :/ | 20:00 |
| *** emagana_ has joined #openstack-infra | 20:01 | |
| clarkb | hashar: thats normal for us now | 20:01 |
| clarkb | hashar: we generate a lot of internal zuul events when code merges | 20:01 |
| *** sdake_ has joined #openstack-infra | 20:01 | |
| clarkb | hashar: this is how we drive the merge check pipeline | 20:01 |
| hashar | yeah but Jenkis report that no change can be merged | 20:01 |
| openstackgerrit | A change was merged to openstack-infra/reviewstats: Sync the infra subprojects to the governance list https://review.openstack.org/129440 | 20:02 |
| hashar | https://review.openstack.org/#/q/status:open,n,z is:open is full of v-1 | 20:02 |
| hashar | """This change was unable to be automatically merged with the current state of the repository. Please rebase your change and upload a new patchset.""" | 20:02 |
| clarkb | oh so that I don't think is related to the pipelines. more likely is related ot the work jeblair is doing | 20:02 |
| clarkb | jeblair: ^ | 20:02 |
| hashar | back to 0 events / 0 results :] | 20:03 |
| *** stevemar has joined #openstack-infra | 20:04 | |
| jeblair | well hrm. | 20:05 |
| openstackgerrit | Colleen Murphy proposed a change to openstack-infra/project-config: Only run stackforge/puppet-* unit tests on trusty https://review.openstack.org/129717 | 20:05 |
| clarkb | jeblair: hashar its not everything | 20:05 |
| openstackgerrit | Nicolas Simonds proposed a change to openstack-infra/system-config: Allow launch-node.py to use non-Keystone identity services https://review.openstack.org/112355 | 20:05 |
| clarkb | my grenade change is being tested so the merge stuff worked for it | 20:05 |
| crinkle | clarkb: https://review.openstack.org/129717 | 20:06 |
| openstackgerrit | Michael Krotscheck proposed a change to openstack-infra/storyboard: Plugins may now register their own user preferences https://review.openstack.org/129061 | 20:06 |
| jeblair | the most recent merge on 05 looks good | 20:06 |
| *** cpowell has left #openstack-infra | 20:06 | |
| hashar | clarkb: seems it was a transient issues with the mergers :D | 20:06 |
| pleia2 | it's a lot of things, my inbox is unusually full of merge conflicts | 20:06 |
| *** annegent_ has quit IRC | 20:06 | |
| pleia2 | 19:51 - 19:53 UTC seems to be the timing on the bulk of them | 20:07 |
| *** pc_m has quit IRC | 20:07 | |
| clarkb | crinkle: lgtm | 20:07 |
| jeblair | ah zm08 has a problem | 20:09 |
| hashar | if it can help for investigation https://review.openstack.org/#/c/128921/ PS2 has the tip of the branch of parent so should really be merge able :) | 20:09 |
| jeblair | GitCommandError: 'git clone -v ssh://jenkins@review.openstack.org:29418/openstack/nova /var/lib/zuul/git/openstack/nova' returned exit status 128: Host key verification failed. | 20:09 |
| hashar | :-/ | 20:09 |
| jeblair | so. um. yay for that. | 20:09 |
| krotscheck | Urm. .... | 20:10 |
| *** ildikov has quit IRC | 20:10 | |
| krotscheck | That... | 20:10 |
| krotscheck | Well then. | 20:10 |
| jeblair | so 05-07 are fine, but 08 did not end up with a known hosts file. | 20:10 |
| *** harlowja is now known as harlowja_away | 20:10 | |
| hashar | don't you provide the known host file via puppet? | 20:10 |
| jeblair | yep. looking back over the logs, puppet had an error. | 20:10 |
| hashar | cause it might be possible to have the service to require that known host file | 20:11 |
| jeblair | the error did not cause it to exit with an error. | 20:11 |
| clarkb | jeblair: ya we need detailed exit codes in that script probably | 20:11 |
| jeblair | root cause was a bad ubuntu mirror | 20:12 |
| jeblair | we might also want to make the return values from the merger more nuanced, so that zuul could reschedule a merge job | 20:13 |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/system-config: Add Kerberos config https://review.openstack.org/129444 | 20:13 |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/system-config: Add AFS https://review.openstack.org/129450 | 20:13 |
| mordred | clarkb: I thought we had ... oh, launch_node | 20:14 |
| *** groknix has joined #openstack-infra | 20:14 | |
| mordred | jeblair, clarkb: you guuys both +2'd the above chain, I rebasd to fix merge conflict - you have any issues with me +A and then sheperding in? | 20:14 |
| jeblair | mordred: ++ | 20:15 |
| clarkb | no opposition heer | 20:15 |
| *** _nadya_ has quit IRC | 20:15 | |
| jeblair | i'm making a list of changes that zuul -1d due to merge failure | 20:15 |
| pleia2 | thanks jeblair | 20:16 |
| jeblair | oh, also i fixed 08 | 20:16 |
| openstackgerrit | Ramy Asselin proposed a change to openstack-infra/system-config: Add jenkins job builder error logging conf option https://review.openstack.org/119934 | 20:16 |
| openstackgerrit | Matthew Treinish proposed a change to openstack-infra/subunit2sql: Add section on the basic data model https://review.openstack.org/129407 | 20:16 |
| pleia2 | running off for a longish lunch to do errands-before-travel (7AM flight tomorrow on my way to Raleigh, woo) | 20:17 |
| clarkb | pleia2: have fun | 20:17 |
| anteaya | pleia2: safe travels | 20:18 |
| fungi | dhellmann: sorry, stepped away for a moment | 20:18 |
| *** david-lyle has joined #openstack-infra | 20:18 | |
| *** markmcclain has joined #openstack-infra | 20:18 | |
| openstackgerrit | Andreas Jaeger proposed a change to openstack-infra/project-config: Check requirements for manila https://review.openstack.org/129535 | 20:18 |
| dhellmann | fungi: no problem, clarkb pointed out that the docs might be building to the wrong dir and they are | 20:19 |
| dhellmann | although now I'm hitting a pbr/sphinx integration issue | 20:19 |
| *** baoli has quit IRC | 20:20 | |
| *** hashar has quit IRC | 20:20 | |
| dhellmann | fungi, clarkb: oh, this is a markupsafe issue actually. Have you seen this one? https://jenkins04.openstack.org/job/gate-cliff-docs/36/console | 20:21 |
| *** AJaeger_ has quit IRC | 20:21 | |
| clarkb | I have not | 20:21 |
| fungi | mordred: did you approve the wrong change in your stack? | 20:21 |
| *** r1chardj0n3s_afk is now known as r1chardj0n3s | 20:22 | |
| clarkb | down to one job on 06 and two jobs on 07 | 20:22 |
| openstackgerrit | A change was merged to openstack-infra/project-config: Add jenkins-project to the default env list https://review.openstack.org/126579 | 20:22 |
| fungi | dhellmann: that's not familiar-looking. are you testing locally with MarkupSafe==0.23? | 20:22 |
| *** HenryG has quit IRC | 20:23 | |
| *** markmcclain has quit IRC | 20:23 | |
| *** markmcclain has joined #openstack-infra | 20:23 | |
| fungi | that's the most recent version, released in may | 20:23 |
| dhellmann | let me try again | 20:24 |
| clarkb | ok 06 is done running jobs. I am going to kill the offline nodes on it | 20:24 |
| clarkb | then shutdown and upgrade gearman-plugin | 20:24 |
| clarkb | let me know if I shoulnd't | 20:24 |
| *** markmcclain has quit IRC | 20:24 | |
| openstackgerrit | Matthew Treinish proposed a change to openstack-infra/system-config: Add subunit2sql server configuration https://review.openstack.org/127060 | 20:24 |
| dhellmann | fungi: yes, that's the version I have | 20:24 |
| openstackgerrit | Andreas Jaeger proposed a change to openstack-infra/system-config: Fix indention in gerrit.config.erb https://review.openstack.org/90031 | 20:25 |
| anteaya | well the -1's shortened my review queue quickly | 20:25 |
| anteaya | :D | 20:25 |
| * anteaya reviews | 20:25 | |
| clarkb | stopping 06 now | 20:26 |
| mordred | fungi: nope. I approved in reverse order to land it as a stack | 20:26 |
| mordred | because I'm strange | 20:26 |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/system-config: Run puppet even if git servers fail https://review.openstack.org/129722 | 20:27 |
| mordred | clarkb: I think this ^^ is what we talked about a few days ago | 20:27 |
| clarkb | ok its starting nowwith new plugin | 20:28 |
| clarkb | zaro: ^ | 20:28 |
| clarkb | at least one job has succeeded | 20:29 |
| fungi | mordred: ahh, okay. just thought maybe you meant to approve the krb addition | 20:29 |
| clarkb | and the one failure I have seen was appropriate | 20:30 |
| clarkb | please let me know if you see anything weird | 20:30 |
| clarkb | and gearman-plugin.hpi.bak can be moved back into place and its what we had before | 20:30 |
| mordred | fungi: both - none of the servers are getting auto-updates yet, so I can run puppet by hand on each of them | 20:31 |
| fungi | k | 20:31 |
| fungi | i saw it between when you approved them i guess | 20:32 |
| mordred | speaking of which ... jeblair, did you add the new zuul mergers to the inventory file on puppetmaster? I noticed we're still on a static file | 20:32 |
| *** julim has quit IRC | 20:32 | |
| jeblair | what? we're owha? | 20:32 |
| jeblair | why are we on a static file? | 20:32 |
| clarkb | ok 07 is done rnning jobs. I see no immediate isuses on 06 so I will do 07 now | 20:32 |
| mordred | we did it for the puppetmaster transition | 20:32 |
| jeblair | did we miss a puppet3 change to revert off of that? | 20:32 |
| mordred | I think so | 20:33 |
| mordred | I can send in a patch | 20:33 |
| jeblair | i thought there was a change to do that. and yeah, i think we very much want that. | 20:33 |
| *** markmcclain has joined #openstack-infra | 20:33 | |
| mordred | jeblair: mind if I do it after I apply the afs/krb puppet? it's actually handy that it's broken in this way at the moment :) | 20:33 |
| jeblair | mordred: sure | 20:34 |
| openstackgerrit | Steve Baker proposed a change to openstack-infra/project-config: Define experimental check-heat-dsvm-functional-postgres https://review.openstack.org/129509 | 20:34 |
| *** oramraz has joined #openstack-infra | 20:34 | |
| clarkb | jenkins is starting on 07 now | 20:36 |
| openstackgerrit | Antoine Musso proposed a change to openstack-infra/zuul: Update merge status after merge:merge is submitted https://review.openstack.org/128921 | 20:36 |
| *** dkehnm has quit IRC | 20:36 | |
| openstackgerrit | A change was merged to openstack-infra/system-config: Add Kerberos config https://review.openstack.org/129444 | 20:36 |
| clarkb | I will let 06 and 07 simmer | 20:37 |
| clarkb | jeblair: zaro: anthing you want to have checked post gearman-plugin upgrade? | 20:37 |
| *** dimsum_ has quit IRC | 20:37 | |
| *** unicell has quit IRC | 20:38 | |
| clarkb | this will let me context switch back to grenade stuff as soon as my change merges | 20:38 |
| *** dimsum_ has joined #openstack-infra | 20:38 | |
| *** esker has joined #openstack-infra | 20:38 | |
| *** dane_leblanc has quit IRC | 20:41 | |
| *** dimsum_ has quit IRC | 20:42 | |
| openstackgerrit | A change was merged to openstack-infra/system-config: Add AFS https://review.openstack.org/129450 | 20:43 |
| jeblair | okay, i have a script with commands to re-enqueue 537 changes into check. | 20:43 |
| clarkb | even more reason to not take out more jenkinses :) | 20:43 |
| clarkb | we are at full jenkins complement now and I will avoid taking more out until the check queue is happy again | 20:43 |
| openstackgerrit | Spencer Krum proposed a change to openstack-infra/system-config: Livegrep https://review.openstack.org/129726 | 20:44 |
| jeblair | here goes | 20:44 |
| *** dkehnm has joined #openstack-infra | 20:45 | |
| *** gyee has joined #openstack-infra | 20:45 | |
| *** dane_leblanc has joined #openstack-infra | 20:45 | |
| openstackgerrit | A change was merged to openstack-infra/project-config: Remove grenade neutron non-voting from the gate https://review.openstack.org/129711 | 20:45 |
| *** freemanbrandon has quit IRC | 20:45 | |
| jeblair | zm08 is functioning normally | 20:45 |
| mordred | jeblair: puppet applied to kdcs, I can still kinit | 20:48 |
| jeblair | neato. i'll start spinning up a replacement kdc02 | 20:48 |
| dansmith | hmm, | 20:49 |
| mordred | jeblair: I'll be interested to see how that goes | 20:49 |
| dansmith | is something rechecking all the things again? | 20:49 |
| clarkb | dansmith: yes, jeblair did that because a zuul merger eroneously reported that there were merge failures on all the things | 20:50 |
| dansmith | ah | 20:50 |
| clarkb | dansmith: so this is a one time shot for oops fixing | 20:50 |
| dansmith | gotcha, | 20:50 |
| dansmith | just wanted to raise it in case it was another ... you know. | 20:50 |
| clarkb | yup no problem | 20:50 |
| asselin | patrickeast krtaylor rfolco mmedvede jesusaurus ociuhandu nibalizer: ok so for subtree I need to first upgrade to latest git. I followed this: http://stackoverflow.com/questions/19109542/installing-latest-version-of-git-in-ubuntu | 20:51 |
| asselin | sudo add-apt-repository ppa:git-core/ppa | 20:51 |
| *** tonytan4ever has quit IRC | 20:52 | |
| jeblair | Warning: You can create 19 Cloud Servers before reaching your limit. View your limits or request an increase. | 20:52 |
| jeblair | clarkb, fungi, mordred: ^ | 20:52 |
| clarkb | jeblair: is that in ord? | 20:52 |
| mordred | uhm. neat? | 20:52 |
| *** teran has joined #openstack-infra | 20:52 | |
| anteaya | asselin: what version of git did you have? | 20:52 |
| *** radez is now known as radez_g0n3 | 20:52 | |
| jeblair | clarkb: it's in the web ui. dunno what region(s) it's referring to | 20:52 |
| jeblair | oh | 20:53 |
| jeblair | dfw | 20:53 |
| jeblair | it goes away when i select ord | 20:53 |
| asselin | anteaya, git version 1.7.9.5 | 20:53 |
| clarkb | fungi: mordred https://review.openstack.org/#/c/128973/1 that change is ready for review | 20:53 |
| anteaya | asselin: what os are you on? | 20:53 |
| clarkb | and is at the bottom of my grenadification with d-g stack | 20:53 |
| asselin | anteaya, that's the latest available on ubuntu 12.04 | 20:53 |
| mordred | asselin: what are you wanting to use subtrees on? | 20:53 |
| fungi | jeblair: good to be aware of though | 20:53 |
| clarkb | would be great if I could get another core review on it | 20:53 |
| clarkb | just to get it out of the way as I work on the proper grenadification | 20:53 |
| asselin | mordred, http://specs.openstack.org/openstack-infra/infra-specs/specs/puppet-modules.html | 20:53 |
| clarkb | mordred: its an easy way to filter branch out puppet modules | 20:54 |
| jeblair | fungi: soon to be 20 when i move kdc02 :) | 20:54 |
| mordred | clarkb: gotcha | 20:54 |
| * fungi nods | 20:54 | |
| anteaya | asselin: any possiblitity you can use a trusty image? | 20:54 |
| jeblair | mordred: initial catalog run of kdc02 completed with no errors | 20:54 |
| asselin | yes.... | 20:54 |
| mordred | jeblair: woot | 20:54 |
| anteaya | asselin: I'll suggest using a trusty image then | 20:55 |
| openstackgerrit | A change was merged to openstack-infra/project-config: Only run stackforge/puppet-* unit tests on trusty https://review.openstack.org/129717 | 20:55 |
| anteaya | asselin: makes a better starting point | 20:55 |
| asselin | ok | 20:55 |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/system-config: Fix one missed afs->openafs rename https://review.openstack.org/129729 | 20:55 |
| anteaya | asselin: I have git 1.9.1 on my trusty and it has the subtree help | 20:55 |
| mordred | stupid requires capitalization ^^ | 20:55 |
| anteaya | so am going with subtree is available on git 1.9.1 | 20:56 |
| mordred | clarkb:, jeblair: mind if I self-+3 that? | 20:56 |
| asselin | anteaya, after the add repository, but git version is 2.1.1 | 20:56 |
| *** david-lyle has quit IRC | 20:56 | |
| *** mfink has quit IRC | 20:56 | |
| *** radez_g0n3 is now known as radez | 20:57 | |
| openstackgerrit | Mikhail S Medvedev proposed a change to openstack-infra/devstack-gate: Ensure dir permissions are 755 for cleanup to work https://review.openstack.org/129730 | 20:57 |
| *** david-lyle has joined #openstack-infra | 20:57 | |
| clarkb | mordred I gave you +2 | 20:57 |
| asselin | anteaya, nibalizer jesusaurus also this command in the spec should be double-dash: git subtree split –prefix=modules/$MODULE –branch=$MODULE | 20:58 |
| clarkb | its simple should be ok for you to self approve | 20:58 |
| asselin | anteaya, nibalizer jesusaurus http://specs.openstack.org/openstack-infra/infra-specs/specs/puppet-modules.html | 20:58 |
| anteaya | asselin: if you are using trusty, why do you need to add the repository? | 20:58 |
| anteaya | asselin: you can make a patch to the spec to fix the command to what it should be | 20:59 |
| asselin | anteaya, b/c this was before you mentioned to use trusty....I guess it's an fyi... | 20:59 |
| jeblair | mordred: i'm going to change dns for kdc02 now. | 20:59 |
| nibalizer | asselin: i think it is https://github.com/openstack-infra/infra-specs/blob/master/specs/puppet-modules.rst | 21:00 |
| nibalizer | it is a double-dash, just rst is 'helping' | 21:00 |
| nibalizer | we should probably make that a shell command, or <pre> or something | 21:00 |
| asselin | nibalizer, yes | 21:01 |
| *** andreykurilin_ has quit IRC | 21:01 | |
| clarkb | use ``foo`` to literal quote | 21:02 |
| *** dkliban is now known as dkliban_biab | 21:02 | |
| mordred | jeblair: ossum | 21:02 |
| mordred | jeblair: while you're in there, you wanna do the cname for kdc.openstack.org ? | 21:02 |
| fungi | i added the cname for kdc earlier today when we agreed on it | 21:03 |
| mordred | oh. neat | 21:03 |
| mordred | nevermind | 21:03 |
| *** dhellman_ has joined #openstack-infra | 21:03 | |
| *** dhellman_ has quit IRC | 21:03 | |
| *** andreykurilin_ has joined #openstack-infra | 21:04 | |
| jeblair | mordred: okay, i've made the change in dns and manually propogated the db to the new server | 21:04 |
| jeblair | mordred: aiui, since the old server is still running, that means we should basically have seamless service while clients (and the kdc itself with its kprop script) switch over | 21:05 |
| *** dane_leblanc has quit IRC | 21:05 | |
| jeblair | mordred: (as long as we don't add principals) | 21:05 |
| jeblair | mordred: and in an hour, we should be able to just safely delete the old server | 21:05 |
| mmedvede | nibalizer, asselin: I assume 'git subtree split' would not follow the history if somewhere down the line the directory we are splitting out was renamed. Would this be an issue? | 21:05 |
| *** tonytan4ever has joined #openstack-infra | 21:05 | |
| *** rlandy has quit IRC | 21:06 | |
| mordred | jeblair: woot | 21:07 |
| lifeless | dstufft: greghaynes: you should talk | 21:07 |
| nibalizer | mmedvede: hrm, interesting point | 21:07 |
| lifeless | dstufft: get-pip.py doesn't honour multiple pip indices. | 21:08 |
| fungi | mmedvede: if it uses the same underlying implementation as git filter-branch on file patterns, it should keep any referring objects through git mv occurrences (also git tracks files, not directories, so it would technically be a git mv of all the files to change their paths) | 21:08 |
| jesusaurus | mmedvede: yes, under the hood it is filter-branching out any history that does not belong to the prefix, so if dir foobar was renamed to blort, then when we git-subtree blort we will lose the history of foobar in the new branch (though that history will always be in -infra/system-config) | 21:09 |
| mmedvede | nibalizer, fungi: I have used filter-branch before. And I remember I had to manually script it to maintain history for moves sometimes. This is why I am asking. But I need to research how the subtree split is implemented | 21:09 |
| *** cpowell has joined #openstack-infra | 21:09 | |
| clarkb | jesusaurus filter branch should retain that history | 21:09 |
| nibalizer | mmedvede: if you want to use filter branch you can | 21:09 |
| jesusaurus | clarkb: oh will it? | 21:09 |
| clarkb | if you git mv yes | 21:10 |
| greghaynes | dstufft: yes, this is causing much pain as we are trying to make a mirror of testing packages with pypi as the fallback mirror | 21:10 |
| *** mfer has quit IRC | 21:10 | |
| mordred | I love that the word blort was used in a conversation taht was meaningful | 21:10 |
| clarkb | lifeless greghaynes it does... | 21:10 |
| clarkb | just not in the way you want | 21:10 |
| fungi | (or if git file duplicate detection makes it a git mv for you instead of an rm and an add) | 21:10 |
| lifeless | clarkb: greghaynes just readthe code and said it didn't | 21:10 |
| clarkb | that isnt how pip works | 21:10 |
| mordred | get-pip.py is just pip under the covers really | 21:10 |
| clarkb | lifeless yes but only because your expectation is wrong | 21:10 |
| lifeless | clarkb: how so? details please. | 21:11 |
| clarkb | fallback indices only help in a small case | 21:11 |
| *** dmsimard is now known as dmsimard_away | 21:11 | |
| clarkb | that is where x==1.1.1 isnt in the first index but is in second | 21:11 |
| clarkb | if first index fails then pip fails | 21:11 |
| clarkb | no fallback there | 21:11 |
| anteaya | mmedvede: what do you mean manually script filter branch to maintain history | 21:12 |
| anteaya | mmedvede: have you an example? | 21:12 |
| *** oramraz has quit IRC | 21:12 | |
| lifeless | clarkb: I don't believe you. | 21:13 |
| *** andreykurilin_ has quit IRC | 21:13 | |
| lifeless | clarkb: I just tested, and my test disagrees with your assertion. | 21:13 |
| mmedvede | anteaya, yes. For example, before I do filter branch, I can use --follow for some of the files inside the directory | 21:13 |
| *** sdake_ has quit IRC | 21:13 | |
| *** cpowell has quit IRC | 21:13 | |
| mmedvede | anteaya, and then it would have to be manually specified in filter branch script | 21:13 |
| mmedvede | anteaya, as in 'if we have this directory -> move it there' | 21:14 |
| *** andreykurilin_ has joined #openstack-infra | 21:14 | |
| mordred | I agree with lifeless. the thing it doesnt' do is the inverse | 21:14 |
| clarkb | mordred ? | 21:14 |
| mordred | if it finds it in the first idex, it STILL checks the second index | 21:14 |
| mordred | as in, it always queries all indexes | 21:14 |
| mordred | which is NOT what we want | 21:14 |
| clarkb | mordred but say pypi.p.o is down and its your first index | 21:15 |
| *** emagana_ has quit IRC | 21:15 | |
| clarkb | pip will just fail | 21:15 |
| mordred | yes | 21:15 |
| *** bswartz has quit IRC | 21:15 | |
| lifeless | ok I don't like the new paste.o.o | 21:15 |
| mordred | because it always queries all indexes | 21:15 |
| clarkb | ya so you agree :) | 21:15 |
| lifeless | http://paste.openstack.org/show/122347/ | 21:15 |
| lifeless | is not spam. | 21:15 |
| anteaya | mmedvede: can you share an example in paste? | 21:15 |
| mordred | lifeless: that's not new | 21:15 |
| lifeless | mordred: ok well I don't like it :) | 21:15 |
| mmedvede | anteaya, I have an actual script if you're interested. But I was asking jesusaurus and nibalizer more on how important it is to maintain history. Because it could be more work if split-tree is not robust for renames/moves | 21:15 |
| clarkb | mordred the issue is their pypi first index does not exist | 21:15 |
| anteaya | mmedvede: yes I am interested in the actual script | 21:15 |
| clarkb | so pip fails | 21:15 |
| clarkb | has nothing to do with get pip | 21:16 |
| lifeless | clarkb: so I just tested that | 21:16 |
| lifeless | clarkb: and it works fine with real pip | 21:16 |
| lifeless | clarkb: see my paste | 21:16 |
| anteaya | mmedvede: and I am asking since I want to understand what you are saying since I used filter branch to create project-config from what is now system-config | 21:16 |
| nibalizer | mmedvede: my impression is that history is nice-to-have | 21:16 |
| lifeless | clarkb: note the typo in my test that means I tested what you just said | 21:16 |
| nibalizer | so if its easy, great | 21:16 |
| anteaya | mmedvede: and what to understand what you are talking about | 21:16 |
| nibalizer | but in the case of a dir rename, meh | 21:16 |
| nibalizer | i also don't really think its very likely that module names changed | 21:17 |
| jesusaurus | mmedvede: so part of the problem with tracking moves is that we are re-rooting the history into the module, so if we moved the directory, the new history would be outside of the git root, and I don't know if git can handle that | 21:17 |
| jesusaurus | but since we will still have all of our current history in the system-config project, i don't think losing some in the new module project is terrible, though clarkb jeblair and fungi should weigh in on that | 21:18 |
| jeblair | fungi: maybe i should status notice the merge conflict thing | 21:19 |
| fungi | jeblair: possibly | 21:19 |
| mmedvede | anteaya, http://paste.openstack.org/show/122348/ (keep in mind the script was used for a particular task, but it should be possible to understand) | 21:19 |
| jeblair | status notice Zuul erroneously marked some changes as having merge conflicts. Those changes have been added to the check queue to be rechecked and will be automatically updated when complete. | 21:20 |
| *** flaper87 is now known as flaper87|afk | 21:20 | |
| anteaya | mmedvede: thank you | 21:20 |
| *** emagana has joined #openstack-infra | 21:20 | |
| jeblair | fungi, clarkb: ^ look good? | 21:20 |
| fungi | jeblair: yep! | 21:20 |
| anteaya | jeblair: looks good to me | 21:20 |
| *** markmcclain has quit IRC | 21:20 | |
| jeblair | #status notice Zuul erroneously marked some changes as having merge conflicts. Those changes have been added to the check queue to be rechecked and will be automatically updated when complete. | 21:20 |
| openstackstatus | jeblair: sending notice | 21:20 |
| *** markmcclain has joined #openstack-infra | 21:20 | |
| krotscheck | What version of mysql is on our build machines? | 21:20 |
| -openstackstatus- NOTICE: Zuul erroneously marked some changes as having merge conflicts. Those changes have been added to the check queue to be rechecked and will be automatically updated when complete. | 21:21 | |
| fungi | krotscheck: whatever comes in ubuntu trusty+security updates | 21:21 |
| openstackstatus | jeblair: finished sending notice | 21:22 |
| JayF | This is still in need of +2A to fix ironic-python-agent tempest jobs (NV atm, but we want to change that) if anyone has a spare moment. https://review.openstack.org/#/c/129653/ | 21:22 |
| fungi | krotscheck: mysql-server 5.5.40-0ubuntu0.14.04.1 on a bare-trusty node i just checked | 21:23 |
| mmedvede | jesusaurus, anteaya, nibalizer, fungi: thanks for the input. What I got is this: I'll try to keep module history as good as I can, but not to the point of spending 8 hours hunting down a single file that has been moved :) | 21:25 |
| anteaya | mmedvede: yeah I'm not familiar with having filter branch mv stuff around, just rm stuff | 21:25 |
| mordred | fungi, krotscheck: however, our production MySQL is 5.6 I believe ... because it's whatever rackspace is running behind their trove | 21:25 |
| * mordred WANTS HIS PATCH TO MERGE NOW | 21:25 | |
| * mordred stops up and down | 21:25 | |
| jeblair | mordred: what patch? | 21:25 |
| mordred | sigh. s/stop/stomp/ | 21:26 |
| jesusaurus | mmedvede: +1 | 21:26 |
| mordred | jeblair: the Afs/Openafs patch | 21:26 |
| anteaya | you were stomping up and down | 21:26 |
| fungi | anteaya: well, the question is about what happens if you make and modify a file in a repository, then rename that file at some point, then later you do a filter-branch on the then-current name of the file, do you lose the history of changes to that file from before it was renamed | 21:26 |
| anteaya | on a plane I believe | 21:26 |
| *** markmcclain has quit IRC | 21:26 | |
| anteaya | fungi: yes | 21:26 |
| mordred | jesusaurus: I missed one rename in renaming the afs module to openafs - the part where puppet inexplicably likes references to be leading-caps'd | 21:26 |
| mordred | GAH | 21:26 |
| mordred | jeblair: ^ | 21:26 |
| krotscheck | mordred, fungi: Well, that explains why adding more test coverage made our builds break. | 21:26 |
| anteaya | fungi: and I was just checking to see if I had to deal with that issue myself | 21:26 |
| fungi | mordred: makes sense. i assumed krotscheck meant nodepool nodes when he said "build machines" | 21:27 |
| jeblair | mordred: how come a test did not catch that? | 21:27 |
| anteaya | fungi: and if I did, it didn't register with me consiously | 21:27 |
| * krotscheck just punished himself for adding more test coverage. | 21:27 | |
| anteaya | do we do a lot of renaming of files within modules in system-config? | 21:27 |
| anteaya | I don't recall that we do | 21:27 |
| mordred | jeblair: I don't know why puppet apply test did not - I did not re-test with noop after the rename, which was silly, clearly | 21:28 |
| anteaya | but we might and I may have missed it | 21:28 |
| fungi | anteaya: yeah, git's supposed to figure that out automatically. sounds like mmedvede's saying sometimes he's seen it fail to do that (i would wager it's when you have a mv and modify of a file in the same commit) | 21:28 |
| jeblair | jesusaurus: if you have some time -- i'm going increasingly suspicious that the apply test is not doing anything anymore. <-- nibalizer too. | 21:28 |
| anteaya | fungi: ah | 21:28 |
| jeblair | /going/growing/ | 21:28 |
| mordred | jeblair: ++ | 21:28 |
| jesusaurus | jeblair: noted, i'll see if i can find time to poke it with a stick | 21:29 |
| clarkb | lifeless: where is the typo? | 21:29 |
| lifeless | clarkb: I made a dir foo to use as the index and then passed file://$(pwd) in | 21:30 |
| lifeless | clarkb: rather than file://$(pwd)/foo | 21:30 |
| fungi | krotscheck: would that failure serve as an indicator that the test (or the software being tested) was too tightly coupled to a specific mysql version, or are we relying on particular features of mysql newer than ubuntu ships on trusty? | 21:31 |
| krotscheck | fungi: We are relying on particular features. In particular, FullText indexes on InnoDB tables. | 21:31 |
| *** dkranz has quit IRC | 21:31 | |
| krotscheck | fungi: Those are new as of 5.6.4 | 21:32 |
| lifeless | oh wow the p.o.o. capture stuff is broken - it has no replay protection | 21:32 |
| fungi | krotscheck: neat--i didn't realize they'd added support for tat | 21:32 |
| fungi | that | 21:32 |
| clarkb | lifeless: ok I have tested locally and new pip does seem to work | 21:32 |
| clarkb | lifeless: old pip definitely did not work this way | 21:32 |
| clarkb | so get-pip.py maybe needs an update | 21:32 |
| krotscheck | fungi: Yeah, well, it’s gimpy mysql fulltext. Full word match only, no good n-gram support or root word detection. | 21:32 |
| lifeless | clarkb: I have no opinion on old-pip | 21:33 |
| lifeless | clarkb: get-pip is the question | 21:33 |
| clarkb | lifeless: sure | 21:33 |
| clarkb | except get pip is just pip | 21:33 |
| clarkb | so it becomes a question of what version of pip | 21:33 |
| lifeless | clarkb: thus my wanting to talk to dstufft :) | 21:33 |
| clarkb | and now I need to run errands | 21:34 |
| *** dmellado has joined #openstack-infra | 21:34 | |
| clarkb | back later | 21:34 |
| *** dmellado has quit IRC | 21:35 | |
| *** mase_x200 has joined #openstack-infra | 21:35 | |
| openstackgerrit | Michael Krotscheck proposed a change to openstack-infra/storyboard: Add User Preferences endpoint. https://review.openstack.org/128487 | 21:36 |
| *** dmellado has joined #openstack-infra | 21:36 | |
| nibalizer | hi everybody, pleia2 and I would like to know why there is a 'meta' section in this acl file http://git.openstack.org/cgit/openstack-infra/project-config/tree/gerrit/acls/openstack-infra/puppet-vcsrepo.config#n6 | 21:36 |
| *** dmellado has quit IRC | 21:37 | |
| *** emagana has quit IRC | 21:37 | |
| *** mase_x200_ has joined #openstack-infra | 21:37 | |
| *** mase_x200 has quit IRC | 21:37 | |
| jesusaurus | nibalizer: that is an access section for the acl file itself | 21:37 |
| *** emagana has joined #openstack-infra | 21:37 | |
| anteaya | jesusaurus: what does that mean? | 21:38 |
| mordred | nibalizer: I think hysterical raisins | 21:38 |
| anteaya | can we remove it yet then, mordred | 21:38 |
| anteaya | since every puppet acl file cargo cults it | 21:38 |
| nibalizer | and more to the point, can we make new puppet modules without it | 21:38 |
| *** prad has quit IRC | 21:38 | |
| nibalizer | ya what anteaya said | 21:38 |
| jogo | is there a patch up that drops python2 jobs from nova? | 21:39 |
| fungi | i'm curious where that originated | 21:39 |
| jesusaurus | if you checkout that ref, you'll see the acl file as project.config in the root of the git dir | 21:39 |
| jogo | that I can +1 | 21:39 |
| mordred | I see no reason for its existence - but I'd like to have fungi/clarkb/jeblair double check me on that | 21:39 |
| jesusaurus | im not sure why its there | 21:39 |
| anteaya | jogo: if there is a patch that drops py27 jobs, I haven't seen it | 21:39 |
| fungi | mordred: yep, i'm blaming away in the history now to try and find its origin | 21:39 |
| anteaya | jogo: there is one of krotscheck that drops py26 jobs | 21:39 |
| *** mmedvede has quit IRC | 21:39 | |
| anteaya | jogo: or at least take is out of template | 21:39 |
| mordred | anteaya: I beleive that's what he's looking for | 21:39 |
| *** imcsk8 has quit IRC | 21:39 | |
| krotscheck | oh hai. | 21:39 |
| *** imcsk8 has joined #openstack-infra | 21:40 | |
| jogo | anteaya: lol, 26* | 21:40 |
| mordred | krotscheck: you're going to be popular with that patch | 21:40 |
| anteaya | jogo: whew | 21:40 |
| *** gokrokve has joined #openstack-infra | 21:40 | |
| *** mase_x200_ has quit IRC | 21:40 | |
| anteaya | jogo: like I know we are in a hurry | 21:40 |
| mordred | we're all excited about it | 21:40 |
| anteaya | jogo: but we aren't there yet | 21:40 |
| krotscheck | mordred: My patch doesn’t really do anything. | 21:40 |
| *** openstackgerrit has quit IRC | 21:40 | |
| krotscheck | Except, apparently, conflict with the merge. | 21:40 |
| krotscheck | argh | 21:40 |
| jogo | bye openstackgerrit | 21:40 |
| anteaya | krotscheck: it sets up the ability to do things though | 21:40 |
| anteaya | krotscheck: which we like | 21:40 |
| jogo | anteaya: thanks I'll look for the patch | 21:41 |
| mordred | krotscheck: we need more patches that don't do things | 21:41 |
| mordred | we do too manythings | 21:41 |
| anteaya | jogo: https://review.openstack.org/#/c/128736/ | 21:41 |
| krotscheck | Speaking of which, clarkb should look at this. https://review.openstack.org/#/c/128736/ | 21:41 |
| anteaya | set up to do things | 21:42 |
| krotscheck | AJaeger: Had to rebase. Want me to update yours while I’m at it? | 21:42 |
| *** emagana has quit IRC | 21:42 | |
| *** mfink has joined #openstack-infra | 21:42 | |
| jogo | anteaya: yeah just found that, and the follow up patch drops 26 from master | 21:43 |
| * jogo is looking forward to saying goodbye to 26 | 21:43 | |
| *** packet has joined #openstack-infra | 21:43 | |
| anteaya | jogo: cool, happy reviewing | 21:43 |
| krotscheck | AJaeger: n/m. Just rebased. | 21:43 |
| *** packet is now known as Guest29212 | 21:43 | |
| mordred | jogo: ++ | 21:44 |
| *** Guest29212 has quit IRC | 21:44 | |
| *** emagana has joined #openstack-infra | 21:45 | |
| *** dimsum_ has joined #openstack-infra | 21:45 | |
| *** todd_dsm has quit IRC | 21:46 | |
| fungi | nibalizer: anteaya: mordred: the trail lands squarely in a history lesson... https://review.openstack.org/#/c/15352/24/modules/openstack_project/files/gerrit/acls/openstack-ci-puppet.config | 21:46 |
| fungi | so long ago i barely remember reviewing that change | 21:46 |
| mordred | fungi: wow. that was a long time ago | 21:47 |
| *** mriedem has quit IRC | 21:47 | |
| fungi | mordred: so basically the tracks go cold at the point where we imported the acls from what was previously being hand-managed in gerrit | 21:48 |
| *** radez is now known as radez_g0n3 | 21:48 | |
| *** dims_ has joined #openstack-infra | 21:49 | |
| mordred | yeah -I think it was there beacuse we were hand managing it | 21:49 |
| mordred | we'd need that acl to hand manage the acl from git | 21:49 |
| mordred | as one of us | 21:49 |
| nibalizer | if you were hand managing it it makes sense to have that acl | 21:49 |
| jesusaurus | how do i tell gertty to refresh its list of projects? it doesnt seem to know about {project,system}-config yet | 21:50 |
| *** dims_ has quit IRC | 21:51 | |
| fungi | jesusaurus: it lacks a rename feature for projects... http://lists.openstack.org/pipermail/openstack-infra/2014-October/002012.html | 21:51 |
| *** VijayTripathi has quit IRC | 21:51 | |
| jesusaurus | fungi: ah, thanks! | 21:51 |
| *** dims_ has joined #openstack-infra | 21:51 | |
| jeblair | jesusaurus: and it should refresh the list of projects on startup (so you can subscribe to project-config) | 21:52 |
| *** dimsum_ has quit IRC | 21:52 | |
| *** todd_dsm has joined #openstack-infra | 21:52 | |
| *** dustins has quit IRC | 21:56 | |
| anteaya | mordred fungi so since we aren't hand managing anymore | 21:56 |
| anteaya | do we need that section? | 21:56 |
| *** otherwiseguy has quit IRC | 21:56 | |
| mordred | no | 21:56 |
| anteaya | so that was a temp fix for the transition from hand managing to not hand managing | 21:57 |
| jesusaurus | fungi: jeblair: awesome, i now have updated projects :) | 21:57 |
| anteaya | and we have been doing chalk drawings ever since | 21:57 |
| anteaya | fungi: nice detective work | 21:57 |
| *** changbl has quit IRC | 21:58 | |
| nibalizer | woot thanks all! | 21:59 |
| anteaya | nibalizer: thank you | 21:59 |
| *** emagana has quit IRC | 22:00 | |
| fungi | anteaya: i don't think it was necessary even then... i think it was probably left over from a manually-initiated experiment, never cleaned up, then imported into the git repo | 22:00 |
| *** jheroux has quit IRC | 22:01 | |
| anteaya | ah | 22:01 |
| anteaya | glad to kill some cruft then | 22:01 |
| anteaya | go nibalizer and pleia2 | 22:01 |
| fungi | anteaya: it was likely something one of us added to the openstack-ci-puppet acl when testing out what eventually became manage-projects, and we eventually just settled on using a super-user account to accomplish the acl pushing anyway | 22:02 |
| jhesketh | Morning | 22:02 |
| anteaya | ah | 22:04 |
| anteaya | morning jhesketh | 22:04 |
| anteaya | fungi: so meta/config is only for being able to make changes to acl files? | 22:05 |
| mordred | achuprin_: yes | 22:06 |
| mordred | gah | 22:06 |
| mordred | anteaya: yes | 22:06 |
| mordred | achuprin_: sorry | 22:06 |
| anteaya | mordred fungi so the only acl file that still needs a meta/config stanza would be this one, yes? http://git.openstack.org/cgit/openstack-infra/project-config/tree/dev/gerrit/acls/test-manage-project.config | 22:06 |
| *** ihrachyshka has quit IRC | 22:07 | |
| anteaya | the other three with a meta/config stanza are puppet-*.config files | 22:08 |
| fungi | anteaya: actually we don't use an acl with that in it specifically | 22:09 |
| fungi | or rather we don't rely on that ibeing in any acl for manage-projects to work | 22:10 |
| anteaya | fungi: awesome, thanks | 22:10 |
| anteaya | pleia2: so if you can remove all the meta/config stanzas I will review your patches | 22:11 |
| fungi | instead the All-Projects acl (which is not managed by puppet) grants basically unlimited rights to members of the "Project Bootstrappers" group which contains the "OpenStack Project Creator" account used by manage-projects | 22:12 |
| anteaya | ah ha | 22:12 |
| anteaya | all power derives from there | 22:12 |
| anteaya | all hail Project Bootstrappers | 22:12 |
| fungi | so that allows it to modify /refs/meta/* without ever having to specify it explicitly (it's managed at the refs/* level instead) | 22:12 |
| anteaya | cool | 22:13 |
| anteaya | in the individual acl files | 22:13 |
| anteaya | would there be any need to have a permision on refs/* | 22:13 |
| anteaya | this was something else I was looking at today | 22:13 |
| anteaya | since the ability to create a branch is on refs/heads/* | 22:13 |
| fungi | if individual projects want to grant their release managers access to make new branches, i believe they end up needing create permissions on refs/* instead of refs/heads/* | 22:14 |
| anteaya | oh | 22:14 |
| fungi | though i have not personally tested the latter as an alternative so am not positive | 22:14 |
| *** mmedvede has joined #openstack-infra | 22:14 | |
| anteaya | so on https://review.openstack.org/#/c/129541/1/gerrit/acls/stackforge/requests-mock.config | 22:15 |
| fungi | we give _our_ "Release Managers" group create on refs/* in the All-Projects acl at any rate | 22:15 |
| anteaya | AJaeger: had a comment to say have the permission on refs/* | 22:15 |
| *** weshay has quit IRC | 22:15 | |
| anteaya | and then recinded but I don't know why | 22:15 |
| *** eharney has quit IRC | 22:16 | |
| anteaya | and the proposed branch is refs/heads/proposed | 22:16 |
| anteaya | so I had thought that refs/heads/* would be sufficient for new branch creation | 22:16 |
| anteaya | now I don't know | 22:16 |
| *** HenryG has joined #openstack-infra | 22:16 | |
| anteaya | zaro: any thoughts on the correct gerrit permission for allow new branch creation? | 22:17 |
| fungi | i don't know either. testing would be needed unless someone happens to have the answer. i *do* know that create on refs/* works though | 22:17 |
| anteaya | okay | 22:17 |
| anteaya | not sure I want to review for that if we can get away with refs/heads/* | 22:17 |
| anteaya | I would need a jenkins to test correct? | 22:17 |
| anteaya | or could I use review-dev? | 22:17 |
| anteaya | sorry, I would need a gerrit to test | 22:18 |
| *** dimtruck is now known as zz_dimtruck | 22:18 | |
| fungi | you would basically need a gerrit where you have an account with the ability to modify an acl and then another (presumably) account without permission to create branches on every project | 22:18 |
| anteaya | hmmm okay thanks | 22:19 |
| fungi | so fairly easy with your own test gerrit deployment, if you have one | 22:20 |
| *** zz_dimtruck is now known as dimtruck | 22:20 | |
| *** esker has quit IRC | 22:21 | |
| *** dmsimard_away is now known as dmsimard | 22:22 | |
| anteaya | I don't have one right now | 22:22 |
| anteaya | but I should | 22:22 |
| anteaya | so I will create anotherone | 22:22 |
| *** dmsimard is now known as dmsimard_away | 22:22 | |
| zaro | anteaya: AJaeger said he took the comment back? | 22:22 |
| zaro | ref/heads/* should be all that's needed. | 22:23 |
| anteaya | k, thanks | 22:24 |
| anteaya | I'll set up a test gerrit tomorrow | 22:24 |
| anteaya | the patch needs rebasing anyway | 22:24 |
| *** david-lyle_ has joined #openstack-infra | 22:26 | |
| *** ZZelle_ has quit IRC | 22:26 | |
| *** david-lyle has quit IRC | 22:27 | |
| *** mfer has joined #openstack-infra | 22:29 | |
| jeblair | mordred: oh, we don't have NetInfo in the afs module | 22:30 |
| mordred | jeblair: correct. we need machine specific discoverable informatino for that | 22:34 |
| mordred | jeblair: I was going to do that as a follow on | 22:34 |
| jeblair | cool. i'm leaving the instruction to manually create the netinfo file for fileservers in the docs im writing for now | 22:34 |
| jeblair | mordred: ^ | 22:34 |
| *** david-lyle_ is now known as david-lyle | 22:35 | |
| mordred | jeblair: ++ | 22:35 |
| *** jgrimm is now known as zz_jgrimm | 22:37 | |
| *** mfer has quit IRC | 22:41 | |
| jeblair | mordred, fungi, clarkb: https://review.openstack.org/129748 AFS docs | 22:42 |
| *** openstackgerrit has joined #openstack-infra | 22:42 | |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/system-config: Fix kprop cron https://review.openstack.org/129749 | 22:46 |
| nibalizer | wow i wasn't paying attention and now AFS is super serious :D | 22:47 |
| *** e0ne has quit IRC | 22:47 | |
| fungi | nibalizer: it was always super serious | 22:47 |
| *** whoops has quit IRC | 22:47 | |
| fungi | nibalizer: it's just that, now, it's also actually running here | 22:47 |
| *** harlowja_away is now known as harlowja | 22:48 | |
| jeblair | fungi, mordred: i have deleted kdc02 from dfw. the new #2 is in ord. | 22:48 |
| fungi | excellent | 22:48 |
| jeblair | fungi, mordred: also, OMG, was it just easier to spin up a replacement KDC server than just about anything else we run? | 22:48 |
| *** andreykurilin_ has quit IRC | 22:49 | |
| fungi | jeblair: decades of experience has gone into figuring out how to install and run kerberos reliably... unlike just about anything else we run ;) | 22:50 |
| jeblair | fungi: indeed! | 22:50 |
| jeblair | also, i'm a fan of the afs db servers, which just quietly, without any fuss whatsoever, start talking to each other and figure out how to make themselves into a HA database :) | 22:51 |
| jesusaurus | nibalizer: right? | 22:53 |
| jogo | are the governance docs published yet? | 22:53 |
| *** marun has quit IRC | 22:54 | |
| jogo | I know there were plans to | 22:54 |
| anteaya | the patch I believe is still in review | 22:55 |
| jogo | anteaya: link? I can't seem to find it | 22:55 |
| anteaya | jogo: it merged: https://review.openstack.org/#/c/125796/ | 22:55 |
| jogo | anteaya: thats what I thought | 22:55 |
| jogo | anteaya: right we build the docs, but as per the commit message we need to publish them too | 22:56 |
| anteaya | right | 22:56 |
| anteaya | doesn't look like that has happened yet | 22:56 |
| jogo | :/ | 22:56 |
| jogo | anteaya: context https://review.openstack.org/#/c/127971/4/doc/source/index.rst | 22:57 |
| jogo | want to reference something in there | 22:57 |
| anteaya | jogo: nope, not published yet: http://git.openstack.org/cgit/openstack-infra/project-config/tree/zuul/layout.yaml#n2101 | 22:57 |
| *** msabramo has quit IRC | 22:57 | |
| anteaya | jogo: yep I feel ya | 22:58 |
| anteaya | we never figured out a way for git to render rst files pretty | 22:58 |
| anteaya | but I think that ensuring the tc like where the governance docs are published | 22:58 |
| anteaya | is worth waiting for the tc to say | 22:58 |
| anteaya | jogo: you might ping ttx and ask him if he can work an item on to the agenda to get them to say | 22:59 |
| anteaya | then you can make up a patch and I can review | 22:59 |
| anteaya | and since we can have a log of the conversation as reference we shouldn't have obstacles | 22:59 |
| jogo | anteaya: makes sense | 23:01 |
| jogo | ttx: ping ^ | 23:01 |
| openstackgerrit | Michael Tupitsyn proposed a change to openstack-infra/jenkins-job-builder: Fix mistyping in gerrit trigger parameters https://review.openstack.org/126081 | 23:01 |
| *** zaphod42 has joined #openstack-infra | 23:02 | |
| *** atiwari has quit IRC | 23:02 | |
| *** gokrokve_ has joined #openstack-infra | 23:03 | |
| *** gokrokve has quit IRC | 23:03 | |
| *** emagana has joined #openstack-infra | 23:03 | |
| jeblair | anteaya, jogo: are you certain that hasn't already happened? :) | 23:07 |
| *** plars has quit IRC | 23:07 | |
| anteaya | jeblair: the publishing of the governance docs? according to how I am reading the jobs it doesn't look like they are | 23:07 |
| jeblair | anteaya: no, the discussion about where to publish them. | 23:07 |
| *** bswartz has joined #openstack-infra | 23:07 | |
| anteaya | but based upon how you ask, it appears you know more | 23:07 |
| anteaya | ah, no I am not certain | 23:08 |
| jogo | jeblair: I don't see anything on https://wiki.openstack.org/wiki/Governance/TechnicalCommittee | 23:08 |
| jeblair | i seem to recall having it at least once | 23:08 |
| jogo | jeblair: and I couldn't find any open reviews | 23:09 |
| jeblair | jogo: i think it happened a long time ago | 23:09 |
| *** bknudson has quit IRC | 23:09 | |
| *** plars has joined #openstack-infra | 23:09 | |
| jogo | jeblair: any way of digging it up? | 23:09 |
| jeblair | jogo: google? | 23:10 |
| fungi | vague recollection is that we should set up yet another rackspace cloudsites site for it, and migrate it with the rest of the docs sites as part of the docs publishing revamp spec | 23:10 |
| jeblair | jogo: i'm sorry i can't be more help right now..... | 23:10 |
| jeblair | fungi: not another cloud site, just put it in the existing cloud site | 23:10 |
| pleia2 | anteaya: instead of creating a patch that fixes each of the puppet acls, I think we want to just update their acls to all point at puppet-system.config once 126096 merges | 23:10 |
| jeblair | fungi: docs/governance or something like that | 23:10 |
| fungi | jeblair: oh, docs.o.o/governance or something? | 23:10 |
| fungi | yeah | 23:10 |
| * pleia2 realizes half into a patch | 23:10 | |
| *** msabramo has joined #openstack-infra | 23:10 | |
| jeblair | jogo: i'd provide a citation if i had one handy, but i wasn't driving this, other people were and they sort of dropped it | 23:11 |
| fungi | i guess as the ptl for the docs team is on the tc, the vote will reflect the willingness to comingle that | 23:11 |
| anteaya | pleia2: do you want to create a patch that removes it from project-config/dev:something-manage-projects.config | 23:11 |
| jogo | fungi: about to push a WIP patch up to publish to docs.o.o/developer/goverance | 23:11 |
| anteaya | jogo: sounds like a jumping off point | 23:12 |
| anteaya | then they have something to discuss | 23:12 |
| fungi | jogo: i guess the governance repo, as an extension of the tc, is developer-oriented so that basically works | 23:12 |
| *** dims_ has quit IRC | 23:12 | |
| pleia2 | anteaya: you mean from from everything but the puppet modules? | 23:12 |
| jogo | fungi: and it requires the smallest patch to get started | 23:13 |
| fungi | jogo: though ideally we'd spell governance right when we do ;) | 23:13 |
| jeblair | jogo: http://eavesdrop.openstack.org/meetings/tc/2014/tc.2014-01-07-20.02.log.html | 23:13 |
| jogo | fungi: don't count on it :/ | 23:13 |
| *** dimsum_ has joined #openstack-infra | 23:13 | |
| fungi | hence "ideally" | 23:13 |
| jeblair | fungi: ^ that too | 23:13 |
| pleia2 | anteaya: looks like it's just in puppet-apparmor.config puppet-dashboard.config and puppet-vcsrepo.config | 23:14 |
| jogo | annegentle: ping | 23:14 |
| fungi | "20:20:07 <ttx> #topic Governance docs publication" | 23:14 |
| anteaya | pleia2: ./dev/gerrit/acls/test-manage-project.config:[access "refs/meta/config"] | 23:14 |
| anteaya | pleia2: jsut giving you the chance to create the patch | 23:14 |
| anteaya | you don't have to if you don't want to | 23:15 |
| pleia2 | oh, on dev! | 23:15 |
| anteaya | yes | 23:15 |
| pleia2 | eyeballs missed that part, can do | 23:15 |
| anteaya | thanks | 23:15 |
| jogo | fungi: well that meeting went in a very different direction | 23:16 |
| jogo | jeblair: thanks for the link | 23:16 |
| fungi | jogo: indeed it did | 23:16 |
| fungi | jeblair: in positive news, i *just* some hours ago today saw, with my own eyes, the source code for www.openstack.org | 23:16 |
| jeblair | jogo: i'm afraid at this point, anything that involves the foundation website is a dead end | 23:16 |
| *** dimtruck is now known as zz_dimtruck | 23:16 | |
| mordred | jeblair: yeah - the hardest part of dealing with krb/afs is figuring out how to do it from the tools which are designed to make running _other_ services that don't do those things easier | 23:16 |
| fungi | jeblair: so its publication is inching closer | 23:16 |
| *** dimsum_ has quit IRC | 23:17 | |
| fungi | (more like millimetering closer) | 23:17 |
| openstackgerrit | Joe Gordon proposed a change to openstack-infra/project-config: Publish governance docs to docs.o.o/developer/governance https://review.openstack.org/129753 | 23:18 |
| anteaya | fungi: wow | 23:18 |
| jeblair | jogo: why did you do that? you know annegentle is opposed. | 23:18 |
| jeblair | jogo, fungi: so we should probably just do governance.o.o since it's actually something that we can move forward with. | 23:18 |
| openstackgerrit | A change was merged to openstack-infra/pypi-mirror: Force --no-use-mirror installaions https://review.openstack.org/105650 | 23:19 |
| fungi | while it does require manually adding the site definition to each and every jenkins master, it's something i can knock out in half an hour tomorrow | 23:19 |
| jogo | jeblair: I am not sure if she is opposed to only docs.o.o/governance or docs.o.o/developer/governance as well | 23:19 |
| jeblair | fungi: we could use scp and static. | 23:20 |
| jeblair | static.o.o | 23:20 |
| anteaya | jeblair: yeah, governance.o.o is something that appears to meet criteria | 23:20 |
| jogo | jeblair: that and to get the ball rolling on this again | 23:20 |
| jogo | anything works for me, anything we can move forward with | 23:20 |
| fungi | jeblair: if it's not going to need path mangling at all, then yes scp uploading to and serving from static.o.o would work fine | 23:21 |
| fungi | given that specs.o.o was wrangled onto static.o.o, governance can probably go similarly | 23:22 |
| jeblair | jogo: left comment on change; if you can wear annegentle down and convince her docs.o.o will be okay, that works for me. otherwise, governance.o.o (either as a cloud site or static.o.o vhost) are the things we can move most quickly with. | 23:22 |
| jeblair | fungi: yeah, i think that's a good model | 23:22 |
| jeblair | jogo: we can continue to try to proceed with governance.o.o, but that involves some people who are normally very busy and currently even more busy. :( | 23:23 |
| openstackgerrit | Elizabeth K. Joseph proposed a change to openstack-infra/project-config: Remove confusing access "refs/meta/config" ACL def https://review.openstack.org/129754 | 23:23 |
| fungi | i'm in favor of a vhost on static.o.o as the primary option, cloudsites as a distant second | 23:23 |
| jogo | jeblair: thanks, anything that allows us to publish these docs in the near future works for me | 23:24 |
| *** todd_dsm has quit IRC | 23:24 | |
| mordred | vhost | 23:25 |
| * mordred wants to move EVERYTHING off of cloud sites - so certainly doesn't want to put more things there | 23:25 | |
| *** signed8bit is now known as signed8bit_ZZZzz | 23:25 | |
| jeblair | afs | 23:25 |
| mordred | jeblair: right. but that means actual work now | 23:25 |
| mordred | rather than just snark | 23:25 |
| mordred | jogo: you could write the change to make governance the first thing we publish to and serve from AFS ... | 23:26 |
| jeblair | mordred: :) i'm being whimsical and jumping ahead a bit. if we actually like afs and keep it and want to use it for docs, the next step is to update the docs publishing spec for it. | 23:26 |
| mordred | yes | 23:26 |
| mordred | jeblair: to be fair, it's going pretty darned well so far ... | 23:27 |
| mordred | I mean, ignoring the fact that we're not using it for things yet | 23:27 |
| *** armax has quit IRC | 23:27 | |
| mordred | when's the last time we installed a new complex service and you made a comment about how easy replacing one of the servers in it was? | 23:27 |
| jeblair | mordred: fungi said that about zuul mergers once :) | 23:28 |
| *** armax has joined #openstack-infra | 23:28 | |
| *** wenlock has quit IRC | 23:28 | |
| mordred | jeblair: heh | 23:28 |
| mordred | s/complex service/complex service we did not write/ | 23:28 |
| *** mnaser has left #openstack-infra | 23:29 | |
| *** hemna is now known as hemnass | 23:29 | |
| anteaya | mordred: my goodness you are ready to make a commitment before the first blush has worn off | 23:30 |
| *** emagana has quit IRC | 23:30 | |
| *** emagana has joined #openstack-infra | 23:31 | |
| mordred | anteaya: I was ready to commit about 10 years ago | 23:31 |
| anteaya | ah | 23:31 |
| anteaya | I'm late to the game | 23:31 |
| jeblair | i'm spinning up a shell server where i will install the afs client so we can do some shared testing and poking around | 23:32 |
| *** emagana_ has joined #openstack-infra | 23:32 | |
| *** emagana has quit IRC | 23:33 | |
| anteaya | so I was reading the docs | 23:33 |
| anteaya | who should be installing the afs client? | 23:33 |
| anteaya | devs? | 23:33 |
| anteaya | infra-core? | 23:33 |
| anteaya | who is the target market? | 23:33 |
| jeblair | anteaya: no one yet? | 23:34 |
| anteaya | okay | 23:34 |
| jeblair | anteaya: i think even if we ultimately use it we would not expect many people to install the client | 23:36 |
| anteaya | okay that was what I was feeling | 23:38 |
| fungi | jeblair: yes, you should take it as a compliment that adding a zuul merger is as similarly painless an experience as replacing a kdc | 23:38 |
| anteaya | since you need authentication to do anything with it | 23:38 |
| *** aysyd has quit IRC | 23:39 | |
| *** aysyd has joined #openstack-infra | 23:39 | |
| *** groknix has quit IRC | 23:40 | |
| *** pc_m has joined #openstack-infra | 23:41 | |
| *** pcm_ has joined #openstack-infra | 23:41 | |
| *** pc_m has quit IRC | 23:45 | |
| *** aysyd has quit IRC | 23:46 | |
| *** aysyd has joined #openstack-infra | 23:47 | |
| *** annegent_ has joined #openstack-infra | 23:48 | |
| *** todd_dsm has joined #openstack-infra | 23:48 | |
| *** Ryan_Lane has quit IRC | 23:49 | |
| *** Ryan_Lane has joined #openstack-infra | 23:49 | |
| *** gyee has quit IRC | 23:51 | |
| *** KurtMartin is now known as kmartin | 23:51 | |
| *** kgiusti has quit IRC | 23:51 | |
| *** hdd has quit IRC | 23:52 | |
| *** kgiusti has joined #openstack-infra | 23:53 | |
| *** r1chardj0n3s is now known as r1chardj0n3s_afk | 23:55 | |
| *** Kiall has quit IRC | 23:56 | |
| *** hdd has joined #openstack-infra | 23:56 | |
| *** Kiall has joined #openstack-infra | 23:58 | |
| *** yamamoto has joined #openstack-infra | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!