| mordred | fungi: that's my theory completely | 00:00 |
|---|---|---|
| *** sarob has joined #openstack-infra | 00:02 | |
| *** salv-orl_ has quit IRC | 00:02 | |
| *** hdd has joined #openstack-infra | 00:03 | |
| *** achanda_ has quit IRC | 00:04 | |
| *** asselin_ has joined #openstack-infra | 00:04 | |
| *** achanda has joined #openstack-infra | 00:05 | |
| *** sarob has quit IRC | 00:07 | |
| *** doug-fish has left #openstack-infra | 00:07 | |
| mordred | btw - my laptop sucks | 00:15 |
| mordred | qemu-img destroys is | 00:15 |
| mordred | it | 00:15 |
| openstackgerrit | Spencer Krum proposed openstack-infra/system-config: Print the node definition before apply test https://review.openstack.org/167346 | 00:15 |
| fungi | 4 of 5 hpcloud bare-trusty updates failed and have been restarted. all the ones in rax worked though | 00:16 |
| *** ddieterly has joined #openstack-infra | 00:16 | |
| *** dboik_ has quit IRC | 00:17 | |
| *** ddieterly has quit IRC | 00:17 | |
| *** davide has quit IRC | 00:20 | |
| openstackgerrit | Stefano Maffulli 'reed' proposed openstack-infra/system-config: Create a list to coordinate internships https://review.openstack.org/167458 | 00:20 |
| *** achanda has quit IRC | 00:20 | |
| *** markvoelker has joined #openstack-infra | 00:25 | |
| *** bknudson has joined #openstack-infra | 00:30 | |
| *** yamamoto has quit IRC | 00:32 | |
| *** trey has quit IRC | 00:33 | |
| *** mtanino has quit IRC | 00:33 | |
| openstackgerrit | Spencer Krum proposed openstack-infra/system-config: Print the node definition before apply test https://review.openstack.org/167346 | 00:35 |
| *** trey has joined #openstack-infra | 00:35 | |
| *** melwitt has quit IRC | 00:35 | |
| *** mattfarina has quit IRC | 00:36 | |
| *** mfink__ has joined #openstack-infra | 00:44 | |
| *** jcoufal has quit IRC | 00:46 | |
| mordred | greghaynes: I'm going to stab something | 00:46 |
| *** mfink_ has quit IRC | 00:47 | |
| *** gyee has quit IRC | 00:48 | |
| *** sputnik13 has quit IRC | 00:49 | |
| * Shrews hides from mordred | 00:50 | |
| *** asselin_ has quit IRC | 00:50 | |
| Shrews | and what is your laptop that qemu-img breaks it???? | 00:50 |
| mordred | Shrews: HP | 00:51 |
| Shrews | OF COURSE | 00:51 |
| mordred | SpamapS: AHA! | 00:51 |
| *** emagana has joined #openstack-infra | 00:51 | |
| mordred | Shrews, greghaynes, SpamapS: I HAVE FOUND IT (except I probably haven't) | 00:51 |
| *** sdake has joined #openstack-infra | 00:53 | |
| *** emagana_ has joined #openstack-infra | 00:53 | |
| *** emagana has quit IRC | 00:55 | |
| *** zz_dimtruck is now known as dimtruck | 00:56 | |
| *** amitgandhinz has joined #openstack-infra | 00:58 | |
| mordred | Shrews: btw - I'm using an ansible playbook with our modules in anger tonight | 01:00 |
| *** amitgandhinz has quit IRC | 01:00 | |
| mordred | Shrews: and it's working well | 01:00 |
| *** amitgandhinz has joined #openstack-infra | 01:01 | |
| *** mbacchi-pi has quit IRC | 01:01 | |
| Shrews | mordred: that's a good sign | 01:02 |
| mordred | Shrews: I mean, except for the image re-upload not working | 01:02 |
| Shrews | "software that works, even in anger, is good software" -- Shrews | 01:02 |
| mordred | Shrews: https://github.com/emonty/ansible-build-image <-- using this role | 01:02 |
| *** salv-orlando has joined #openstack-infra | 01:03 | |
| mordred | Shrews: with this: https://review.openstack.org/#/c/140840/8/tools/test-dib-upload.yaml | 01:03 |
| mordred | BTW - for anyone in channel | 01:03 |
| Shrews | oh, this is a new thing | 01:03 |
| mordred | ansible "role" is a very confusing word | 01:03 |
| mordred | it does not describe a role like "this server is a webserver" | 01:03 |
| *** ddieterly has joined #openstack-infra | 01:03 | |
| mordred | it describes a reusable unit of execution | 01:03 |
| fungi | okay, all remaining bare-trusty image-updates i had to restart (some multiple times) have finally succeeded | 01:04 |
| *** jerryz has joined #openstack-infra | 01:04 | |
| mordred | fungi: woot! | 01:04 |
| mordred | Shrews: yes - a form of that test playbook I believe can be used for us to have an in-tree defition of our base images | 01:04 |
| mordred | Shrews: and it can build them one time, upload them to all of out clouds, and then forget about them | 01:04 |
| *** sarob has joined #openstack-infra | 01:05 | |
| fungi | notmyname: it's probably safe to recheck that change now, though there may be a few lingering nodes in the process of booting from the previous images. in a few more minutes it's pretty certain none will remain | 01:05 |
| mordred | te playbook works on rax and hp - I have shell scripts that use the python*client untilities that are brittle and break | 01:05 |
| notmyname | fungi: thanks | 01:05 |
| mordred | Shrews: so somewhere we're doing a good job | 01:05 |
| mordred | Shrews: also - Ipushed up a shade patch (I think) that re-adds adding a swift container when needed | 01:05 |
| mordred | Shrews: because it doesn't do what we expect on all of our clouds :( | 01:06 |
| *** armax has joined #openstack-infra | 01:07 | |
| *** tiswanso has joined #openstack-infra | 01:08 | |
| *** Sukhdev has joined #openstack-infra | 01:08 | |
| *** tiswanso has quit IRC | 01:08 | |
| *** tiswanso has joined #openstack-infra | 01:08 | |
| pleia2 | puppetcamp sf today was fun :) | 01:08 |
| pleia2 | but so tired, so finding some rest spot | 01:09 |
| Shrews | part of me wonders if we are bastardizing ansible to coerce it into doing unnatural things | 01:09 |
| *** tiswanso has quit IRC | 01:09 | |
| *** Longgeek has joined #openstack-infra | 01:09 | |
| *** tiswanso has joined #openstack-infra | 01:10 | |
| Shrews | mordred: that playbook defines clouds within the actual role? interesting | 01:11 |
| *** gokrokve has quit IRC | 01:11 | |
| mordred | Shrews: well - not the way you're thinking | 01:11 |
| mordred | Shrews: it defines which of teh clouds that are defined you want to upload things to | 01:11 |
| mordred | Shrews: because the clouds in clouds.yaml dont' really have an ansible-level iterator | 01:11 |
| mordred | and you may not want to upload your images to all of them | 01:11 |
| *** sarob has quit IRC | 01:12 | |
| Shrews | hrm... that seems... awkward | 01:12 |
| mordred | but ultimately, I think we're using it for exactly what it's meant for | 01:12 |
| mordred | Shrews: it's ... less awkward than the version of code that came befor eit | 01:12 |
| mordred | Shrews: it did make me think that we might want to have the ansible inventory define a global list called "clouds" or something, so that if you did need to iterate over them, awesome | 01:13 |
| mordred | Shrews: oh - but also - there is this other thing | 01:13 |
| mordred | Shrews: which is "format" and "properties" | 01:13 |
| Shrews | i would think that the iteration should naturally come from running the playbook with a different cloud | 01:13 |
| mordred | Shrews: playbooks don't know about clouds | 01:13 |
| mordred | they know about hosts | 01:13 |
| mordred | all of these are plays to run on the image building host | 01:14 |
| Shrews | they do when you define the 'cloud' variable between iterations | 01:14 |
| mordred | pleia2: woot! | 01:14 |
| mordred | well, yeah | 01:14 |
| *** baoli has joined #openstack-infra | 01:14 | |
| fungi | ansibo knows clouds | 01:14 |
| Shrews | fungi: bo don't know clouds | 01:15 |
| mordred | BWAHAHAHA | 01:15 |
| mordred | Shrews: in any case- we may want to think about extending os-client-config ... | 01:15 |
| mordred | Shrews: because there are two pieces of per-cloud info that do not come from the global cloud config | 01:16 |
| mordred | Shrews: that is "what image format does this cloud expect" | 01:16 |
| mordred | Shrews: and "if you're going to upload some images to this cloud, what properties do you pretty much ahve to set" | 01:16 |
| *** ivar-lazzaro has quit IRC | 01:17 | |
| *** tsg_ has quit IRC | 01:18 | |
| mordred | Shrews: those are the reasons I needed an extra data structure in the role | 01:18 |
| mordred | Shrews: because I needed diffrent answers for rax and hp and I couldn't extract them from the existing inventory | 01:19 |
| mordred | or global vars or whatnot | 01:19 |
| * Shrews pines for true vendor independence | 01:19 | |
| mordred | Shrews: that's what we're writing :) | 01:19 |
| Shrews | yeah | 01:19 |
| Shrews | you're welcome, openstack :) | 01:20 |
| *** sarob has joined #openstack-infra | 01:20 | |
| mordred | :) | 01:22 |
| *** claudiub has quit IRC | 01:23 | |
| *** ociuhandu has quit IRC | 01:23 | |
| mordred | Shrews, greghaynes, SpamapS, clarkb: BOOM! works on HP | 01:24 |
| fungi | notmyname: success! https://jenkins03.openstack.org/job/gate-swift-requirements/46/console | 01:27 |
| notmyname | fungi: yay! thanks :-) | 01:27 |
| fungi | notmyname: apologies for all the delays and thanks for sticking with us while we filed down the rough edges there | 01:27 |
| mordred | notmyname: woot! | 01:28 |
| notmyname | glad we got it :-) | 01:28 |
| notmyname | and by we, I mean you and clarkb ;-) | 01:28 |
| mordred | notmyname: by we, you mean fungi and clarkb | 01:29 |
| mordred | I was useless | 01:29 |
| notmyname | lol | 01:30 |
| mordred | Shrews, greghaynes, SpamapS, clarkb: I lied. ignore me | 01:30 |
| *** ayoung has joined #openstack-infra | 01:30 | |
| Shrews | /ignore mordred | 01:31 |
| * mordred shoots Shrews | 01:31 | |
| *** ujuc has joined #openstack-infra | 01:31 | |
| *** sarob has quit IRC | 01:32 | |
| *** otter768 has joined #openstack-infra | 01:33 | |
| * fungi throws a wet mordred at mordred | 01:33 | |
| *** mattfarina has joined #openstack-infra | 01:36 | |
| *** patrickeast has quit IRC | 01:37 | |
| *** otter768 has quit IRC | 01:38 | |
| *** baoli has quit IRC | 01:40 | |
| *** salv-orlando has quit IRC | 01:45 | |
| *** stevemar has quit IRC | 01:46 | |
| *** dims has quit IRC | 01:47 | |
| * mordred accepts fungi's wet cat | 01:51 | |
| *** hdd has quit IRC | 01:53 | |
| *** xyang1 has quit IRC | 01:53 | |
| nibalizer | fungi: metaaaa | 01:54 |
| openstackgerrit | Stephanie Miller proposed openstack-infra/puppet-zanata: Add apache configuration for zanata https://review.openstack.org/164011 | 01:56 |
| *** dimtruck is now known as zz_dimtruck | 02:00 | |
| *** emagana_ has quit IRC | 02:01 | |
| *** ghostpl_ has joined #openstack-infra | 02:01 | |
| *** asselin_ has joined #openstack-infra | 02:01 | |
| *** baoli has joined #openstack-infra | 02:01 | |
| *** dboik has joined #openstack-infra | 02:04 | |
| *** dboik_ has joined #openstack-infra | 02:04 | |
| *** ghostpl_ has quit IRC | 02:06 | |
| *** baoli has quit IRC | 02:06 | |
| *** dboik has quit IRC | 02:08 | |
| openstackgerrit | Spencer Krum proposed openstack-infra/system-config: Print the node definition before apply test https://review.openstack.org/167346 | 02:09 |
| *** ajmiller has joined #openstack-infra | 02:10 | |
| *** annegentle has joined #openstack-infra | 02:13 | |
| *** baoli has joined #openstack-infra | 02:13 | |
| *** agentle has joined #openstack-infra | 02:13 | |
| *** hdd has joined #openstack-infra | 02:13 | |
| *** annegentle is now known as Guest31292 | 02:13 | |
| *** bradjones has quit IRC | 02:16 | |
| *** harlowja is now known as harlowja_away | 02:17 | |
| *** Somay has quit IRC | 02:17 | |
| *** jyuso1 has quit IRC | 02:19 | |
| openstackgerrit | Stephanie Miller proposed openstack-infra/puppet-zanata: Add apache configuration for zanata https://review.openstack.org/164011 | 02:19 |
| zxiiro | any nodepool folks still around? I was wondering what permissions the jenkins user needs to have configured to successfully manage the nodes? | 02:20 |
| zxiiro | doesn't seem to be documented | 02:20 |
| *** bradjones has joined #openstack-infra | 02:21 | |
| *** bradjones has quit IRC | 02:21 | |
| *** bradjones has joined #openstack-infra | 02:21 | |
| *** kaisers has joined #openstack-infra | 02:25 | |
| *** sdake has quit IRC | 02:25 | |
| *** kaisers1 has quit IRC | 02:27 | |
| *** ajmiller_ has joined #openstack-infra | 02:27 | |
| *** thedodd has joined #openstack-infra | 02:28 | |
| *** asselin__ has joined #openstack-infra | 02:30 | |
| *** ajmiller has quit IRC | 02:30 | |
| *** sdake has joined #openstack-infra | 02:31 | |
| *** Sukhdev has quit IRC | 02:31 | |
| *** asselin_ has quit IRC | 02:33 | |
| *** asselin__ has quit IRC | 02:39 | |
| *** dhritishikhar has joined #openstack-infra | 02:40 | |
| *** dhritishikhar_ has joined #openstack-infra | 02:40 | |
| *** r-daneel has quit IRC | 02:41 | |
| *** salv-orlando has joined #openstack-infra | 02:44 | |
| *** achanda has joined #openstack-infra | 02:45 | |
| *** thedodd has quit IRC | 02:47 | |
| *** agentle has quit IRC | 02:48 | |
| *** Guest31292 has quit IRC | 02:48 | |
| *** changbl has joined #openstack-infra | 02:52 | |
| *** annegentle has joined #openstack-infra | 02:53 | |
| *** agentle has joined #openstack-infra | 02:53 | |
| *** annegentle is now known as Guest25284 | 02:53 | |
| *** Longgeek has quit IRC | 02:53 | |
| *** kevinbenton has joined #openstack-infra | 02:54 | |
| *** unicell has quit IRC | 02:55 | |
| *** yamamoto has joined #openstack-infra | 02:56 | |
| *** agentle has quit IRC | 02:57 | |
| *** Guest25284 has quit IRC | 02:57 | |
| *** Longgeek has joined #openstack-infra | 02:59 | |
| *** ChuckC has quit IRC | 03:00 | |
| *** samueldmq has joined #openstack-infra | 03:01 | |
| *** thingee has quit IRC | 03:01 | |
| *** spzala has quit IRC | 03:03 | |
| *** Longgeek has quit IRC | 03:04 | |
| *** tqtran has quit IRC | 03:04 | |
| *** armax has quit IRC | 03:06 | |
| *** tiswanso has quit IRC | 03:07 | |
| *** amitgandhinz has quit IRC | 03:11 | |
| openstackgerrit | Brian Hunter proposed openstack-infra/nodepool: Pretty Age in nodepool list https://review.openstack.org/161994 | 03:12 |
| *** samueldmq has quit IRC | 03:14 | |
| openstackgerrit | Brian Hunter proposed openstack-infra/nodepool: Pretty Age in nodepool list https://review.openstack.org/161994 | 03:15 |
| *** otter768 has joined #openstack-infra | 03:20 | |
| openstackgerrit | Spencer Krum proposed openstack-infra/system-config: Print the node definition before apply test https://review.openstack.org/167346 | 03:21 |
| mordred | SpamapS, greghaynes: I just spent 2 hours tracking down "directory didn't exist" | 03:22 |
| mordred | it's possible the cachaça has somethign to do with it | 03:22 |
| *** hrubi_ has joined #openstack-infra | 03:26 | |
| *** hrubi has quit IRC | 03:28 | |
| *** dhritishikhar has quit IRC | 03:28 | |
| *** cburgess has quit IRC | 03:28 | |
| *** morgabra has quit IRC | 03:28 | |
| *** cburgess has joined #openstack-infra | 03:28 | |
| *** dhritishikhar has joined #openstack-infra | 03:29 | |
| *** esker has joined #openstack-infra | 03:29 | |
| *** morgabra has joined #openstack-infra | 03:29 | |
| *** salv-orlando has quit IRC | 03:30 | |
| *** sdake has quit IRC | 03:34 | |
| *** yamamoto has quit IRC | 03:35 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 03:36 | |
| *** purp_2 has joined #openstack-infra | 03:36 | |
| *** purp has quit IRC | 03:37 | |
| *** achanda has quit IRC | 03:38 | |
| lifeless | mordred: with the directory not existing? | 03:40 |
| lifeless | mordred: or you taking 2 hours to not find it? | 03:40 |
| mordred | lifeless: yes | 03:40 |
| mordred | lifeless: it was one of those "too obvious, was looking at more complex issues" problems :) | 03:41 |
| *** purp_2 is now known as purp | 03:41 | |
| *** asselin_ has joined #openstack-infra | 03:43 | |
| *** fabo_ has joined #openstack-infra | 03:45 | |
| *** james_li has joined #openstack-infra | 03:46 | |
| *** dhritishikhar has quit IRC | 03:47 | |
| *** dhritishikhar_ has quit IRC | 03:47 | |
| *** fabo has quit IRC | 03:49 | |
| openstackgerrit | Brian Hunter proposed openstack-infra/nodepool: Pretty Age in nodepool list https://review.openstack.org/161994 | 03:52 |
| *** tkelsey has joined #openstack-infra | 03:53 | |
| *** bhunter71 has quit IRC | 03:54 | |
| *** Sukhdev has joined #openstack-infra | 03:54 | |
| *** tkelsey has quit IRC | 03:57 | |
| *** Longgeek has joined #openstack-infra | 04:00 | |
| *** yamamoto has joined #openstack-infra | 04:01 | |
| *** yamamoto has quit IRC | 04:02 | |
| *** ddieterly has quit IRC | 04:02 | |
| *** sputnik13 has joined #openstack-infra | 04:02 | |
| *** juggler_ has joined #openstack-infra | 04:03 | |
| *** baoli has quit IRC | 04:04 | |
| *** sabeen3 has joined #openstack-infra | 04:04 | |
| *** sabeen has quit IRC | 04:05 | |
| *** krtaylor has quit IRC | 04:05 | |
| *** juggler has quit IRC | 04:05 | |
| *** eharney has quit IRC | 04:06 | |
| *** eharney has joined #openstack-infra | 04:07 | |
| *** krtaylor has joined #openstack-infra | 04:07 | |
| *** tsg has joined #openstack-infra | 04:07 | |
| *** sputnik13 has quit IRC | 04:08 | |
| *** woodster_ has quit IRC | 04:10 | |
| *** hdd has quit IRC | 04:15 | |
| *** stevemar has joined #openstack-infra | 04:17 | |
| *** sputnik13 has joined #openstack-infra | 04:18 | |
| *** Longgeek has quit IRC | 04:19 | |
| mordred | SpamapS, greghaynes, Shrews: for real this time ... BOOM ... current patches pushed up work with and without config-drive for ubuntu on hp | 04:20 |
| mordred | SpamapS, greghaynes, Shrews: I don't recommend any of you including the infra element, as it will install our sshd_config which will prevent you from logging in from any server but puppetmaster.openstack.org | 04:21 |
| jhesketh | greghaynes: ping | 04:21 |
| mordred | I'll go back and figure out how to get that to suck less | 04:21 |
| mordred | for now - just don't use the infra element | 04:21 |
| *** asettle has joined #openstack-infra | 04:21 | |
| *** yamamoto has joined #openstack-infra | 04:23 | |
| *** bradjones has quit IRC | 04:24 | |
| mordred | greghaynes, SpamapS: I have now kicked off a centos build, and have also re-enabled rax upload | 04:24 |
| mordred | if centos works on hp, I'll test that both work on rax | 04:25 |
| *** unicell has joined #openstack-infra | 04:25 | |
| *** achanda has joined #openstack-infra | 04:27 | |
| openstackgerrit | Spencer Krum proposed openstack-infra/system-config: Print the node definition before apply test https://review.openstack.org/167346 | 04:28 |
| *** bradjones has joined #openstack-infra | 04:28 | |
| *** bradjones has quit IRC | 04:29 | |
| *** bradjones has joined #openstack-infra | 04:29 | |
| *** unicell has quit IRC | 04:30 | |
| pleia2 | cinerama: oh! I did learn that the open source version of puppet dashboard is alive again, with a version 2, we switched to puppetboard because puppet dashboard seemed abandoned in favor of the enterprise one | 04:37 |
| pleia2 | cinerama: turns out the guy who'd been doing most of the work on the open source side came down with a case of having 2 kids | 04:37 |
| StevenK | Haha | 04:37 |
| pleia2 | did a talk today on features of v2, and what's coming in the future, so that was cool | 04:38 |
| pleia2 | I still think puppetboard is good for us, but it's good to know | 04:38 |
| *** dboik_ has quit IRC | 04:38 | |
| pleia2 | I am reminded of this because packaging, he said he's not going to make distro packages anymore because it's not his forte and takes too long, wants to focus on the software dev itself | 04:39 |
| pleia2 | but the debian/ and whatever redhat uses exist, so someone could pick it back up | 04:40 |
| *** sarob has joined #openstack-infra | 04:44 | |
| *** sarob has quit IRC | 04:49 | |
| *** sandywalsh_ has quit IRC | 04:52 | |
| *** juggler_ is now known as juggler | 04:53 | |
| *** asselin_ has quit IRC | 04:55 | |
| *** sputnik13 has quit IRC | 04:57 | |
| *** reed has quit IRC | 05:00 | |
| *** ddieterly has joined #openstack-infra | 05:03 | |
| *** ajmiller_ has quit IRC | 05:03 | |
| *** sandywalsh has joined #openstack-infra | 05:04 | |
| *** sabeen3 has quit IRC | 05:06 | |
| mordred | greghaynes, SpamapS: centos works on hp | 05:07 |
| mordred | I'll test rax in the morning | 05:07 |
| greghaynes | mordred: ooo fun | 05:07 |
| greghaynes | jhesketh: pong | 05:07 |
| *** ddieterly has quit IRC | 05:08 | |
| *** Longgeek has joined #openstack-infra | 05:16 | |
| *** salv-orlando has joined #openstack-infra | 05:16 | |
| *** amotoki has joined #openstack-infra | 05:21 | |
| jhesketh | greghaynes: just looking at 165682 and, please excuse my ignorance, but why would instanciating at import time be a bigger issue than using a global? | 05:21 |
| *** Sukhdev has quit IRC | 05:22 | |
| *** mriedem has joined #openstack-infra | 05:22 | |
| *** mriedem_away has quit IRC | 05:23 | |
| greghaynes | jhesketh: It causes lots of python headaches, it causes lots of issues with debugging errors (since they are going to happen as the result of an import rather than as a result of a routine) | 05:25 |
| *** otter768 has quit IRC | 05:25 | |
| *** mrmartin has joined #openstack-infra | 05:25 | |
| greghaynes | theres also major performance issues | 05:25 |
| *** deepakcs has joined #openstack-infra | 05:25 | |
| greghaynes | its pretty univerally accepted as a python antipattern | 05:26 |
| *** hdd has joined #openstack-infra | 05:26 | |
| *** Longgeek has quit IRC | 05:26 | |
| jhesketh | greghaynes: sure, I understand it's not a good idea, just wanted your opinion on why it's any worse than a global. I think your points are valid though | 05:27 |
| greghaynes | jhesketh: yes, globals dont have that issue | 05:27 |
| jhesketh | greghaynes: did you want to turn it into a function to reduce code duplication, or shall I re-review | 05:27 |
| greghaynes | jhesketh: Yea, im up for that, I just wanted to make sure that was acceptable before doing it :) | 05:27 |
| *** yamahata has quit IRC | 05:28 | |
| jhesketh | greghaynes: it should put the global in one place which is much more preferable. We can also place it next to where fake_list = None for clarity | 05:28 |
| greghaynes | Yep | 05:28 |
| greghaynes | Ill hack that out tomorrow | 05:29 |
| jhesketh | thanks greghaynes :-) | 05:29 |
| *** tsg has quit IRC | 05:29 | |
| greghaynes | np, ty for the reviews | 05:29 |
| *** BharatK has joined #openstack-infra | 05:31 | |
| jhesketh | and to you too ;-) | 05:31 |
| *** mrmartin has quit IRC | 05:32 | |
| *** james_li has quit IRC | 05:35 | |
| *** unicell has joined #openstack-infra | 05:37 | |
| *** emagana has joined #openstack-infra | 05:49 | |
| *** jamielennox is now known as jamielennox|away | 05:49 | |
| *** mrmartin has joined #openstack-infra | 05:51 | |
| openstackgerrit | Spencer Krum proposed openstack-infra/system-config: Print the node definition before apply test https://review.openstack.org/167346 | 05:52 |
| *** dims has joined #openstack-infra | 05:54 | |
| *** tkelsey has joined #openstack-infra | 05:54 | |
| *** tkelsey has quit IRC | 05:59 | |
| *** pal_ has joined #openstack-infra | 06:00 | |
| *** asettle has quit IRC | 06:00 | |
| *** ddieterly has joined #openstack-infra | 06:04 | |
| *** salv-orlando has quit IRC | 06:07 | |
| *** ddieterly has quit IRC | 06:08 | |
| *** markvoelker has quit IRC | 06:17 | |
| *** achanda has quit IRC | 06:18 | |
| *** markus_z has joined #openstack-infra | 06:18 | |
| *** markus_z has quit IRC | 06:18 | |
| *** achanda has joined #openstack-infra | 06:21 | |
| *** rlandy has joined #openstack-infra | 06:24 | |
| *** esker has quit IRC | 06:25 | |
| *** esker has joined #openstack-infra | 06:26 | |
| *** dims has quit IRC | 06:28 | |
| *** sushilkm has joined #openstack-infra | 06:28 | |
| *** sushilkm has left #openstack-infra | 06:28 | |
| *** TheJulia has quit IRC | 06:29 | |
| *** sarob has joined #openstack-infra | 06:33 | |
| *** vigneshvar has joined #openstack-infra | 06:35 | |
| *** sarob has quit IRC | 06:38 | |
| *** achanda has quit IRC | 06:44 | |
| *** hdd has quit IRC | 06:44 | |
| *** achanda has joined #openstack-infra | 06:45 | |
| *** markvoelker has joined #openstack-infra | 06:47 | |
| *** achanda has quit IRC | 06:49 | |
| *** achanda has joined #openstack-infra | 06:50 | |
| *** achanda has quit IRC | 06:51 | |
| *** markvoelker has quit IRC | 06:52 | |
| *** saggi has quit IRC | 06:54 | |
| *** afazekas is now known as __afazekas | 07:01 | |
| seeg | hello, please review https://review.openstack.org/#/c/167152 :) | 07:03 |
| *** ddieterly has joined #openstack-infra | 07:05 | |
| *** _nadya_ has joined #openstack-infra | 07:06 | |
| *** ddieterly has quit IRC | 07:10 | |
| *** asettle has joined #openstack-infra | 07:11 | |
| *** Longgeek has joined #openstack-infra | 07:12 | |
| *** ujuc has quit IRC | 07:13 | |
| *** fabo_ is now known as fabo | 07:15 | |
| *** stevemar has quit IRC | 07:17 | |
| *** Guest17738 has quit IRC | 07:18 | |
| *** salv-orlando has joined #openstack-infra | 07:19 | |
| *** sergsh has joined #openstack-infra | 07:21 | |
| *** Longgeek has quit IRC | 07:22 | |
| *** otter768 has joined #openstack-infra | 07:26 | |
| *** scheuran has joined #openstack-infra | 07:30 | |
| *** otter768 has quit IRC | 07:31 | |
| *** afazekas has joined #openstack-infra | 07:32 | |
| *** Ala has joined #openstack-infra | 07:32 | |
| *** mrunge has joined #openstack-infra | 07:32 | |
| *** doude has quit IRC | 07:32 | |
| *** HeOS has quit IRC | 07:32 | |
| *** Ala has quit IRC | 07:33 | |
| *** Ala has joined #openstack-infra | 07:33 | |
| *** doude has joined #openstack-infra | 07:33 | |
| openstackgerrit | Matthias Runge proposed openstack-infra/project-config: Force to install test deps on the gate https://review.openstack.org/167515 | 07:37 |
| *** Hal has joined #openstack-infra | 07:37 | |
| *** Hal is now known as Guest41796 | 07:37 | |
| *** emagana has quit IRC | 07:40 | |
| *** Guest41796 has quit IRC | 07:41 | |
| *** ibiris_away is now known as ibiris | 07:43 | |
| *** amotoki has quit IRC | 07:43 | |
| *** markvoelker has joined #openstack-infra | 07:48 | |
| *** _nadya_ has quit IRC | 07:50 | |
| *** Bsony has joined #openstack-infra | 07:50 | |
| *** chlong has quit IRC | 07:52 | |
| *** markvoelker has quit IRC | 07:53 | |
| *** ibiris is now known as ibiris_away | 07:55 | |
| *** mrda is now known as mrda-away | 07:55 | |
| *** mpaolino has joined #openstack-infra | 07:56 | |
| *** Virgil_Schiev has joined #openstack-infra | 07:56 | |
| *** tnovacik has joined #openstack-infra | 08:05 | |
| *** ddieterly has joined #openstack-infra | 08:06 | |
| *** mpavone has joined #openstack-infra | 08:07 | |
| nibalizer | nt | 08:09 |
| *** tkelsey has joined #openstack-infra | 08:09 | |
| *** ddieterly has quit IRC | 08:11 | |
| *** dims has joined #openstack-infra | 08:13 | |
| *** _nadya_ has joined #openstack-infra | 08:17 | |
| *** YorikSar has quit IRC | 08:21 | |
| *** scheuran has quit IRC | 08:26 | |
| *** nilasae has joined #openstack-infra | 08:26 | |
| *** _nadya_ has quit IRC | 08:28 | |
| *** e0ne has joined #openstack-infra | 08:30 | |
| *** jistr has joined #openstack-infra | 08:38 | |
| *** Hal has joined #openstack-infra | 08:39 | |
| *** Hal is now known as Guest39881 | 08:39 | |
| *** sobersabre has joined #openstack-infra | 08:40 | |
| *** jlanoux has joined #openstack-infra | 08:41 | |
| *** ibiris_away is now known as ibiris | 08:41 | |
| *** YorikSar has joined #openstack-infra | 08:43 | |
| *** HeOS has joined #openstack-infra | 08:43 | |
| *** ominakov has joined #openstack-infra | 08:43 | |
| *** dims has quit IRC | 08:45 | |
| *** yfried__ has quit IRC | 08:45 | |
| *** ildikov has joined #openstack-infra | 08:46 | |
| openstackgerrit | Christian Berendt proposed openstack-infra/jeepyb: Replace 'javascript' method/parameternames https://review.openstack.org/127748 | 08:47 |
| openstackgerrit | Christian Berendt proposed openstack-infra/jeepyb: Rename upload_rss and move non-swift-relevant parts into the main method https://review.openstack.org/127749 | 08:48 |
| *** dtantsur|afk is now known as dtantsur | 08:48 | |
| *** markvoelker has joined #openstack-infra | 08:49 | |
| *** dizquierdo has joined #openstack-infra | 08:50 | |
| *** dhritishikhar has joined #openstack-infra | 08:50 | |
| *** Bsony_ has joined #openstack-infra | 08:51 | |
| *** jgrimm is now known as zz_jgrimm | 08:51 | |
| *** dhritishikhar_ has joined #openstack-infra | 08:52 | |
| *** andreykurilin_ has joined #openstack-infra | 08:52 | |
| *** markvoelker has quit IRC | 08:54 | |
| *** Bsony has quit IRC | 08:54 | |
| openstackgerrit | Christian Berendt proposed openstack-infra/jeepyb: Rename upload_rss and move non-swift-relevant parts into the main method https://review.openstack.org/127749 | 08:56 |
| *** _nadya_ has joined #openstack-infra | 08:57 | |
| openstackgerrit | Sebastian Kalinowski proposed openstack-infra/project-config: Rename fuel-tasks-validator to fuel-tasklib https://review.openstack.org/162952 | 08:57 |
| *** juggler has quit IRC | 08:58 | |
| *** ibiris is now known as ibiris_away | 08:58 | |
| *** dhritishikhar has quit IRC | 08:59 | |
| *** dhritishikhar_ has quit IRC | 08:59 | |
| openstackgerrit | Sergey Kraynev proposed openstack/requirements: Bump last version of ceilometer-client https://review.openstack.org/167527 | 08:59 |
| *** ibiris_away is now known as ibiris | 09:01 | |
| *** notnownikki has joined #openstack-infra | 09:02 | |
| *** hashar has joined #openstack-infra | 09:02 | |
| *** hashar is now known as hasharConfcall | 09:02 | |
| *** Longgeek has joined #openstack-infra | 09:07 | |
| *** e0ne has quit IRC | 09:10 | |
| openstackgerrit | yolanda.robla proposed openstack-infra/project-config: Add stackforge/puppet-nscld https://review.openstack.org/165922 | 09:14 |
| *** ibiris is now known as ibiris_away | 09:16 | |
| *** ssam2 has joined #openstack-infra | 09:17 | |
| *** mattt has joined #openstack-infra | 09:19 | |
| *** arxcruz has joined #openstack-infra | 09:20 | |
| *** Bsony_ has quit IRC | 09:22 | |
| *** Longgeek has quit IRC | 09:22 | |
| *** hichihara has quit IRC | 09:24 | |
| *** andreykurilin_ has quit IRC | 09:25 | |
| *** andreykurilin__ has joined #openstack-infra | 09:25 | |
| *** otter768 has joined #openstack-infra | 09:27 | |
| *** wznoinsk has quit IRC | 09:27 | |
| *** otter768 has quit IRC | 09:31 | |
| *** dizquierdo has quit IRC | 09:33 | |
| *** wznoinsk has joined #openstack-infra | 09:33 | |
| *** ibiris_away is now known as ibiris | 09:35 | |
| *** amotoki has joined #openstack-infra | 09:39 | |
| *** amotoki has quit IRC | 09:41 | |
| *** dims has joined #openstack-infra | 09:43 | |
| *** ociuhandu has joined #openstack-infra | 09:44 | |
| *** dims_ has joined #openstack-infra | 09:46 | |
| *** asettle has quit IRC | 09:49 | |
| *** markvoelker has joined #openstack-infra | 09:50 | |
| *** dims has quit IRC | 09:50 | |
| openstackgerrit | yolanda.robla proposed openstack-infra/puppet-jenkins: Add extension settings in jjb https://review.openstack.org/162152 | 09:52 |
| *** markvoelker has quit IRC | 09:54 | |
| *** liusheng has joined #openstack-infra | 10:00 | |
| *** liusheng has quit IRC | 10:01 | |
| openstackgerrit | yolanda.robla proposed openstack-infra/puppet-zuul: Add fetch_url and poll parameters to zuul.conf https://review.openstack.org/162591 | 10:02 |
| *** ociuhandu has quit IRC | 10:02 | |
| openstackgerrit | Sergey Kraynev proposed openstack/requirements: Bump last version of ceilometer-client https://review.openstack.org/167527 | 10:04 |
| *** hasharConfcall is now known as hashar | 10:04 | |
| *** LinuxJed_ has joined #openstack-infra | 10:04 | |
| *** scheuran has joined #openstack-infra | 10:04 | |
| *** pelix has joined #openstack-infra | 10:04 | |
| *** e0ne has joined #openstack-infra | 10:07 | |
| *** LinuxJedi has quit IRC | 10:07 | |
| *** ddieterly has joined #openstack-infra | 10:07 | |
| openstackgerrit | Sebastian Kalinowski proposed openstack-infra/project-config: Add CI jobs for fuel-tasklib https://review.openstack.org/167539 | 10:10 |
| *** sarob has joined #openstack-infra | 10:10 | |
| *** markus_z has joined #openstack-infra | 10:10 | |
| *** mrmartin has quit IRC | 10:11 | |
| *** pc_m has joined #openstack-infra | 10:12 | |
| *** ddieterly has quit IRC | 10:13 | |
| *** sarob has quit IRC | 10:16 | |
| *** marcusvrn has joined #openstack-infra | 10:16 | |
| *** kmartin has quit IRC | 10:17 | |
| *** Longgeek has joined #openstack-infra | 10:21 | |
| *** tnovacik has quit IRC | 10:28 | |
| *** cdent has joined #openstack-infra | 10:28 | |
| *** LinuxJed_ has quit IRC | 10:30 | |
| *** Virgil_Schiev has quit IRC | 10:30 | |
| *** enikanorov has joined #openstack-infra | 10:30 | |
| openstackgerrit | Christian Schwede proposed openstack-infra/reviewstats: Sync Swift core group https://review.openstack.org/167549 | 10:32 |
| *** derekh has joined #openstack-infra | 10:33 | |
| *** enikanorov__ has quit IRC | 10:33 | |
| *** bknudson has quit IRC | 10:33 | |
| *** Longgeek has quit IRC | 10:36 | |
| *** ominakov has quit IRC | 10:36 | |
| *** ihrachyshka has joined #openstack-infra | 10:39 | |
| *** ajo_ has joined #openstack-infra | 10:44 | |
| *** ajo has quit IRC | 10:44 | |
| *** asettle has joined #openstack-infra | 10:45 | |
| openstackgerrit | Ricardo Carrillo Cruz proposed openstack-infra/project-config: Add stackforge/puppet-autossh https://review.openstack.org/167553 | 10:46 |
| *** asettle has quit IRC | 10:46 | |
| *** vigneshvar has quit IRC | 10:47 | |
| *** dtantsur is now known as dtantsur|brb | 10:47 | |
| *** Bsony has joined #openstack-infra | 10:48 | |
| *** asettle has joined #openstack-infra | 10:48 | |
| *** samueldmq-away is now known as samueldmq- | 10:49 | |
| *** samueldmq- is now known as samueldmq | 10:49 | |
| *** _nadya_ has quit IRC | 10:50 | |
| *** markvoelker has joined #openstack-infra | 10:50 | |
| *** e0ne is now known as e0ne_ | 10:54 | |
| *** sdake has joined #openstack-infra | 10:55 | |
| *** markvoelker has quit IRC | 10:55 | |
| *** zz_johnthetubagu is now known as johnthetubaguy | 10:58 | |
| *** sdake__ has joined #openstack-infra | 10:59 | |
| *** ihrachyshka has quit IRC | 11:02 | |
| *** ildikov has quit IRC | 11:02 | |
| *** sdake has quit IRC | 11:02 | |
| *** yamamoto has quit IRC | 11:03 | |
| *** e0ne_ has quit IRC | 11:04 | |
| *** e0ne has joined #openstack-infra | 11:07 | |
| *** dizquierdo has joined #openstack-infra | 11:08 | |
| *** spredzy is now known as spredzy|afk | 11:10 | |
| *** pelix has quit IRC | 11:13 | |
| *** hashar is now known as hasharLunch | 11:13 | |
| *** yfried__ has joined #openstack-infra | 11:14 | |
| *** ildikov has joined #openstack-infra | 11:15 | |
| *** BharatK has quit IRC | 11:18 | |
| *** sobersabre has quit IRC | 11:18 | |
| *** ajo_ is now known as ajo | 11:18 | |
| *** mrmartin has joined #openstack-infra | 11:19 | |
| *** asettle has quit IRC | 11:19 | |
| kaisers | Hi! Possibly a bit off topic: Can somebody pls give me a hint how/where to contact an admin for the openstack wiki? | 11:19 |
| *** pelix has joined #openstack-infra | 11:20 | |
| *** tnovacik has joined #openstack-infra | 11:20 | |
| *** sobersabre has joined #openstack-infra | 11:23 | |
| *** ominakov has joined #openstack-infra | 11:25 | |
| AJaeger | kaisers, why do you need an admin? | 11:27 |
| AJaeger | reed is an admin as weel as some of the infra roots - better check back later today for that | 11:27 |
| *** otter768 has joined #openstack-infra | 11:28 | |
| kaisers | AJaeger: Ok, thanks. My colleague created some too many wiki pages (REALLY MY COLLEAGUE ;-) ) that we want to remove. Mediawiki seems to allow this only for admins... | 11:28 |
| AJaeger | kaisers, indeed - so come back at US time, please ;) | 11:29 |
| kaisers | AJaeger: will do | 11:30 |
| *** pal_ has quit IRC | 11:31 | |
| *** otter768 has quit IRC | 11:32 | |
| *** yamamoto has joined #openstack-infra | 11:33 | |
| *** mpaolino has quit IRC | 11:33 | |
| *** doude has quit IRC | 11:34 | |
| *** ldnunes has joined #openstack-infra | 11:35 | |
| *** doude has joined #openstack-infra | 11:35 | |
| *** asettle has joined #openstack-infra | 11:36 | |
| *** yamamoto has quit IRC | 11:37 | |
| *** asettle has quit IRC | 11:37 | |
| *** ricolin has joined #openstack-infra | 11:40 | |
| *** mase_x200 has joined #openstack-infra | 11:42 | |
| openstackgerrit | Anthony Michon proposed openstack-infra/project-config: Add cerberus related projects https://review.openstack.org/163876 | 11:42 |
| openstackgerrit | Merged openstack-infra/project-config: Add new puppet-puppet module https://review.openstack.org/165442 | 11:43 |
| *** Somay has joined #openstack-infra | 11:45 | |
| BobBall | sdague: Just to check, regarding the CI reporting, the format that you've specified in the email seems quite different to how Jenkins' results appear in the comments (i.e. the Jenkins-posted-comments have a hyperlink and no colon); does gerrit parse the Jenkins comments and format it differently for display? | 11:49 |
| *** woodster_ has joined #openstack-infra | 11:50 | |
| *** chlong has joined #openstack-infra | 11:50 | |
| Somay | Hello all, I am a new developer here. Can anyone review my patch https://review.openstack.org/#/c/163137/ related to Jenkins Job Builder? | 11:51 |
| *** markvoelker has joined #openstack-infra | 11:51 | |
| *** mattt has left #openstack-infra | 11:55 | |
| *** markvoelker has quit IRC | 11:56 | |
| *** sdake__ has quit IRC | 11:59 | |
| *** dprince has joined #openstack-infra | 11:59 | |
| *** sdake has joined #openstack-infra | 11:59 | |
| *** xyang1 has joined #openstack-infra | 12:03 | |
| BobBall | sdague: I've found the gerrit command and it's actually "- test-name-no-spaces" not "* test-name-no-spaces"; will update to that message. | 12:03 |
| *** mfink__ has quit IRC | 12:04 | |
| *** ociuhandu has joined #openstack-infra | 12:05 | |
| *** ghostpl_ has joined #openstack-infra | 12:05 | |
| *** sc68cal has quit IRC | 12:05 | |
| *** ildikov has quit IRC | 12:06 | |
| *** openstackgerrit has quit IRC | 12:06 | |
| *** sc68cal has joined #openstack-infra | 12:06 | |
| *** openstackgerrit has joined #openstack-infra | 12:06 | |
| *** pelix has quit IRC | 12:07 | |
| *** dims_ has quit IRC | 12:07 | |
| *** dims has joined #openstack-infra | 12:08 | |
| *** ddieterly has joined #openstack-infra | 12:09 | |
| *** markvoelker has joined #openstack-infra | 12:09 | |
| *** andreykurilin__ has quit IRC | 12:09 | |
| seeg | could https://review.openstack.org/#/c/167152/1 be merged? :) | 12:13 |
| *** ddieterly has quit IRC | 12:13 | |
| *** pelix has joined #openstack-infra | 12:14 | |
| *** rfolco has quit IRC | 12:14 | |
| *** spredzy|afk is now known as spredzy | 12:14 | |
| *** ociuhandu has quit IRC | 12:14 | |
| *** rfolco has joined #openstack-infra | 12:15 | |
| *** ociuhandu has joined #openstack-infra | 12:15 | |
| *** ZZelle has quit IRC | 12:15 | |
| *** ZZelle has joined #openstack-infra | 12:16 | |
| *** mpavone has quit IRC | 12:17 | |
| *** aysyd has joined #openstack-infra | 12:18 | |
| *** Somay has quit IRC | 12:20 | |
| *** Longgeek has joined #openstack-infra | 12:21 | |
| *** bradjones has quit IRC | 12:21 | |
| *** bradjones has joined #openstack-infra | 12:21 | |
| *** bradjones has quit IRC | 12:21 | |
| *** bradjones has joined #openstack-infra | 12:21 | |
| *** jcoufal has joined #openstack-infra | 12:22 | |
| *** sushilkm has joined #openstack-infra | 12:22 | |
| *** sushilkm has left #openstack-infra | 12:22 | |
| *** dkliban_afk is now known as dkliban | 12:22 | |
| *** mase_x200 has quit IRC | 12:23 | |
| *** Longgeek has quit IRC | 12:26 | |
| *** macjack has joined #openstack-infra | 12:28 | |
| *** _nadya_ has joined #openstack-infra | 12:28 | |
| *** e0ne is now known as e0ne_ | 12:29 | |
| *** macjack has quit IRC | 12:31 | |
| *** pal_ has joined #openstack-infra | 12:32 | |
| *** mase_x200 has joined #openstack-infra | 12:34 | |
| *** gordc has joined #openstack-infra | 12:34 | |
| fungi | kaisers: how long is the list of pages you need removed? | 12:36 |
| *** yfried__ has quit IRC | 12:36 | |
| *** mbacchi-pi has joined #openstack-infra | 12:36 | |
| *** aysyd has quit IRC | 12:37 | |
| *** e0ne_ is now known as e0ne | 12:37 | |
| *** mpaolino has joined #openstack-infra | 12:38 | |
| *** dhritishikhar has joined #openstack-infra | 12:39 | |
| *** dhritishikhar_ has joined #openstack-infra | 12:39 | |
| *** doug-fish has joined #openstack-infra | 12:41 | |
| *** spredzy is now known as spredzy|afk | 12:41 | |
| *** dprince has quit IRC | 12:42 | |
| *** mfink_ has joined #openstack-infra | 12:42 | |
| fungi | seeg: see inline comment. there's one error you'll need to correct, then i can approve it | 12:43 |
| seeg | ok | 12:44 |
| *** aysyd has joined #openstack-infra | 12:44 | |
| kaisers | fungi: Hi! 3 pages | 12:44 |
| *** TheJulia has joined #openstack-infra | 12:45 | |
| *** yfried__ has joined #openstack-infra | 12:45 | |
| fungi | kaisers: okay, that's really not a lot. just link them in here and i'll take a look | 12:45 |
| *** dkranz has joined #openstack-infra | 12:46 | |
| kaisers | fungi: ok, it's : https://wiki.openstack.org/wiki/Cinder_Driver_Configuration_with_Quobyte , https://wiki.openstack.org/wiki/Nova_Driver_Configuration_with_Quobyte , https://wiki.openstack.org/wiki/Quobyte_Manila_Driver_Configuration | 12:46 |
| *** ominakov has quit IRC | 12:47 | |
| fungi | kaisers: thanks, i'll check them out... should those redirect to other pages with equivalent information on them? | 12:47 |
| kaisers | fungi: No need to redirect, they where just created. All the stuff is now reachable via https://wiki.openstack.org/wiki/Quobyte . Thanks for looking into this! :) | 12:48 |
| fungi | kaisers: no problem. i'll clean those up in just a few moments | 12:49 |
| openstackgerrit | Przemyslaw Kaminski proposed openstack-infra/project-config: Add fuel-dev-tools project to Stackforge https://review.openstack.org/167152 | 12:49 |
| seeg | fungi, done :) | 12:50 |
| *** tnovacik has quit IRC | 12:50 | |
| *** Bsony has quit IRC | 12:50 | |
| *** Somay has joined #openstack-infra | 12:51 | |
| *** bknudson has joined #openstack-infra | 12:51 | |
| *** mfink has quit IRC | 12:53 | |
| samueldmq | morning | 12:54 |
| samueldmq | anteaya, hi, you around? | 12:54 |
| fungi | kaisers: okay, those three are gone. let me know if they need to be restored for any reason | 12:54 |
| kaisers | fungi: Thanks a lot !! | 12:55 |
| *** jyuso1 has joined #openstack-infra | 12:55 | |
| *** Bsony has joined #openstack-infra | 12:56 | |
| *** rwsu has quit IRC | 12:56 | |
| samueldmq | anteaya, actually I am having an introduction to puppet, before getting deeper on project-config/system-config | 12:57 |
| *** ominakov has joined #openstack-infra | 12:57 | |
| samueldmq | anteaya, a see a puppet background as a pre-req to clearly understand things :) | 12:58 |
| *** tnovacik has joined #openstack-infra | 12:58 | |
| *** craigbr has joined #openstack-infra | 12:59 | |
| *** craigbr has quit IRC | 13:00 | |
| fungi | seeg: looks like you removed that line from the wrong project. see my comment on the latest patchset | 13:00 |
| *** dkranz has quit IRC | 13:00 | |
| seeg | ah, right :) | 13:00 |
| *** dprince has joined #openstack-infra | 13:01 | |
| openstackgerrit | Przemyslaw Kaminski proposed openstack-infra/project-config: Add fuel-dev-tools project to Stackforge https://review.openstack.org/167152 | 13:01 |
| *** dhritishikhar_ has quit IRC | 13:02 | |
| *** julim has joined #openstack-infra | 13:02 | |
| *** arxcruz has quit IRC | 13:04 | |
| *** james_li has joined #openstack-infra | 13:04 | |
| *** pblaho has joined #openstack-infra | 13:05 | |
| james_li | clarkb: fungi: could you please review https://review.openstack.org/#/c/167413/2 ? we are switching solum gate to centos7. | 13:06 |
| *** rfolco has quit IRC | 13:07 | |
| *** baoli has joined #openstack-infra | 13:07 | |
| *** yamamoto has joined #openstack-infra | 13:08 | |
| *** dhritishikhar has quit IRC | 13:08 | |
| *** radez_g0n3 is now known as radez | 13:09 | |
| *** ddieterly has joined #openstack-infra | 13:09 | |
| mordred | fungi: so - when you're trying to test uploading glance images and then booting nova nodes based on them ... | 13:11 |
| mordred | fungi: it helps if you upload the images to the same account that you're running nova commands in | 13:11 |
| fungi | mordred: plenty of times have i booted a node with one tenant and then spent too much time trying to figure out why it wasn't in the nova list for another tenant | 13:12 |
| mordred | fungi: just in case you didn't know | 13:12 |
| mordred | fungi: ++ | 13:12 |
| *** radez is now known as radez_g0n3 | 13:12 | |
| mordred | fungi: that said - both centos and ubuntu worked in hp cloud last night - so I _think_ I may have working content | 13:12 |
| *** yamahata has joined #openstack-infra | 13:12 | |
| pabelanger | morning | 13:13 |
| jogo | http://logs.openstack.org/32/167432/1/check/gate-tempest-lib-requirements/fc7a154/console.html#_2015-03-24_23_17_48_088 | 13:14 |
| jogo | project-requirements-change.py is now stacktracing? | 13:14 |
| *** ddieterly has quit IRC | 13:14 | |
| fungi | james_li: lgtm, though deepakcs found a memory leak/resource exhaustion condition running tempest jobs on centos7 when trying to get the glusterfs integration tests running. you might want to find out if that's been resolved | 13:14 |
| fungi | jogo: shouldn't be since my latest image rebuilds last night | 13:15 |
| fungi | jogo: we had several patches to it yesterday to get support for feature branches implemented correctly | 13:15 |
| *** Bsony has quit IRC | 13:16 | |
| jogo | ahh | 13:16 |
| * jogo rechecks | 13:16 | |
| *** dkranz has joined #openstack-infra | 13:16 | |
| fungi | since it was blocking swift merging their ec branch back into master in preparation for rc | 13:16 |
| *** dhritishikhar has joined #openstack-infra | 13:17 | |
| *** dhritishikhar_ has joined #openstack-infra | 13:17 | |
| james_li | fungi: good to know, thanks! | 13:17 |
| deepakcs | fungi, james_li lgtm, I don't think its resolved, as I switched to ubuntu job and centos-devel folks didn't follow up. ianw was trying to write a new tool to capture each process' data to figure where what is gettign consumed, but i don't think that tool yet madeit into the infra upstream | 13:17 |
| *** zz_jgrimm is now known as jgrimm | 13:17 | |
| *** tiswanso has joined #openstack-infra | 13:18 | |
| *** craigbr has joined #openstack-infra | 13:18 | |
| fungi | deepakcs: thanks! james_li will probably want to check in with ianw on that when he's around | 13:18 |
| deepakcs | james_li, lgtm, per my observation, the centos7 issue happened only on the rax provider VMs , hpcloud VMs ran fine, and i believe rax is Xen and hpcloud is KVM | 13:18 |
| sdague | BobBall: cool, thank you! | 13:18 |
| deepakcs | fungi, james_li lgtm, FWIW, the resource exhaustion issue happened even without using gluster as backend for cinder. | 13:19 |
| fungi | yep | 13:20 |
| fungi | which is why i brought it up | 13:20 |
| openstackgerrit | Merged openstack-infra/project-config: Add fuel-dev-tools project to Stackforge https://review.openstack.org/167152 | 13:20 |
| *** Bsony has joined #openstack-infra | 13:21 | |
| fungi | seems likely to resurface for solum jobs too depending on how tempest or other things are exercising their environment on centos7 | 13:21 |
| openstackgerrit | Monty Taylor proposed openstack-infra/project-config: Add elements for Infra servers https://review.openstack.org/140840 | 13:22 |
| *** spzala has joined #openstack-infra | 13:22 | |
| *** dustins has joined #openstack-infra | 13:22 | |
| *** Longgeek has joined #openstack-infra | 13:22 | |
| *** rfolco has joined #openstack-infra | 13:22 | |
| *** peristeri has joined #openstack-infra | 13:23 | |
| james_li | deepakcs: fungi: so the resource exhaustion issue is only with centos7? how about centos6? | 13:23 |
| *** ildikov has joined #openstack-infra | 13:24 | |
| deepakcs | james_li, didn't try with centos6, so can't be sure | 13:24 |
| deepakcs | james_li, as i said above, it is not sure if its issue with centos7 or rax provider (read Xen) specific env. because centos7 on hpcloud worked fine. FYI | 13:24 |
| jerryz | fungi: could you take a look at this patch https://review.openstack.org/#/c/156178/, i can't figure out why pep8 complained undefined name 'time' | 13:25 |
| *** craigbr has quit IRC | 13:25 | |
| james_li | deepakcs: thanks! | 13:25 |
| deepakcs | james_li, wc | 13:25 |
| fungi | james_li: not really sure since the vast majority of our devstack jobs run under ubuntu | 13:26 |
| deepakcs | fungi, do u want to ignite this issue in centos-devel ? they needed a rax and hpcloud VM to debug, the last i know | 13:27 |
| *** weshay has joined #openstack-infra | 13:27 | |
| *** e0ne is now known as e0ne_ | 13:28 | |
| *** Bsony has quit IRC | 13:29 | |
| *** otter768 has joined #openstack-infra | 13:29 | |
| fungi | jerryz: that's really strange, but the unit tests trip over it too http://logs.openstack.org/78/156178/4/check/gate-nodepool-python27/d99f859/console.html#_2015-03-25_12_50_20_483 | 13:31 |
| *** bradjones has quit IRC | 13:31 | |
| jerryz | fungi: i can't reproduce it on my local machine with tox -epep8 or tox -epy27 | 13:32 |
| *** bradjones has joined #openstack-infra | 13:32 | |
| fungi | deepakcs: i'll let whoever's interested in debugging it reach out when they're ready | 13:32 |
| *** bhunter71 has joined #openstack-infra | 13:32 | |
| *** bradjones has quit IRC | 13:32 | |
| *** bradjones has joined #openstack-infra | 13:32 | |
| deepakcs | fungi, ok :) | 13:32 |
| *** dhritishikhar has quit IRC | 13:33 | |
| *** otter768 has quit IRC | 13:33 | |
| *** dhritishikhar_ has quit IRC | 13:33 | |
| *** radez_g0n3 is now known as radez | 13:34 | |
| *** krtaylor has quit IRC | 13:35 | |
| *** e0ne_ is now known as e0ne | 13:35 | |
| *** ddieterly has joined #openstack-infra | 13:36 | |
| *** ddieterly has quit IRC | 13:36 | |
| *** openstackgerrit has quit IRC | 13:36 | |
| fungi | jerryz: are you trying on ubuntu 14.04? (that's what's running those jobs in our infrastructure) | 13:36 |
| *** openstackgerrit has joined #openstack-infra | 13:36 | |
| jerryz | fungi: yes | 13:36 |
| *** amitgandhinz has joined #openstack-infra | 13:37 | |
| *** Longgeek has quit IRC | 13:37 | |
| *** craigbr has joined #openstack-infra | 13:38 | |
| *** Bsony has joined #openstack-infra | 13:38 | |
| fungi | jerryz: i'll see if i can reproduce it shortly when i get a moment | 13:39 |
| *** yfried__ has quit IRC | 13:40 | |
| jerryz | fungi: thanks! just for your reference: http://paste.openstack.org/show/196561/ | 13:41 |
| *** yfried__ has joined #openstack-infra | 13:41 | |
| *** krtaylor has joined #openstack-infra | 13:41 | |
| *** annegentle has joined #openstack-infra | 13:42 | |
| *** tnovacik has quit IRC | 13:43 | |
| *** dboik has joined #openstack-infra | 13:44 | |
| *** dboik has quit IRC | 13:44 | |
| *** dboik has joined #openstack-infra | 13:45 | |
| *** r-daneel has joined #openstack-infra | 13:45 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 13:47 | |
| *** james_li has quit IRC | 13:47 | |
| *** sarob has joined #openstack-infra | 13:48 | |
| *** yamamoto has quit IRC | 13:49 | |
| *** Somay has quit IRC | 13:49 | |
| *** Somay has joined #openstack-infra | 13:50 | |
| mrmartin | fungi hi | 13:50 |
| mrmartin | fungi: the pgsql puppet module is here: https://github.com/mkissam/puppet-pgsql_backup | 13:50 |
| mrmartin | can we move it under openstack-infra ? | 13:51 |
| *** mtanino has joined #openstack-infra | 13:51 | |
| mordred | mrmartin: i think that's a great idea | 13:52 |
| mrmartin | so it is surprisingly very similar to the puppet-mysql_backup :) | 13:52 |
| mordred | hehe | 13:53 |
| mordred | good. I like things that are similar | 13:53 |
| *** sarob has quit IRC | 13:53 | |
| mrmartin | ok, I prepare the required infra patch | 13:54 |
| mrmartin | I just discovered, that an ask-staging.o.o server exists, and regular ask.o.o backups going there daily. | 13:55 |
| *** mfink_ has quit IRC | 13:55 | |
| mrmartin | mordred, don't you know what is the criteria to get a free ATC pass for Vancouver? Do we have an expected commit level? | 13:57 |
| AJaeger | mrmartin, there's a question on ask.o.o that answers this AFAIU | 13:59 |
| mrmartin | awesome | 13:59 |
| mrmartin | I don't ask.o.o :D | 13:59 |
| AJaeger | https://ask.openstack.org/en/question/45531/atc-pass-for-the-openstack-summit/ | 13:59 |
| AJaeger | mrmartin, reed sent out the last bunch yesterday - if you didn't get one, ask him | 14:00 |
| *** mugsie has quit IRC | 14:00 | |
| *** mugsie has joined #openstack-infra | 14:00 | |
| *** nilasae has quit IRC | 14:01 | |
| mrmartin | AJaeger, thanks I've already have at least two. | 14:01 |
| mrmartin | so i got two | 14:02 |
| *** sputnik13 has joined #openstack-infra | 14:02 | |
| *** yfried__ has quit IRC | 14:02 | |
| anteaya | samueldmq: there is plenty of puppet lying around, learning it would be a good thing | 14:04 |
| *** craigbr has quit IRC | 14:04 | |
| *** sputnik13 has quit IRC | 14:05 | |
| *** mjturek1 has joined #openstack-infra | 14:05 | |
| *** dhritishikhar has joined #openstack-infra | 14:05 | |
| *** dhritishikhar_ has joined #openstack-infra | 14:05 | |
| *** ominakov has quit IRC | 14:05 | |
| *** yfried has joined #openstack-infra | 14:06 | |
| *** ddieterly has joined #openstack-infra | 14:09 | |
| *** kgiusti has joined #openstack-infra | 14:09 | |
| *** ChuckC has joined #openstack-infra | 14:09 | |
| *** james_li has joined #openstack-infra | 14:10 | |
| *** craigbr has joined #openstack-infra | 14:10 | |
| *** yfried has quit IRC | 14:10 | |
| *** jgrimm is now known as zz_jgrimm | 14:11 | |
| *** dhritishikhar_ has left #openstack-infra | 14:12 | |
| *** prad has joined #openstack-infra | 14:12 | |
| *** craigbr has quit IRC | 14:13 | |
| *** sabeen1 has joined #openstack-infra | 14:13 | |
| *** _nadya_ has quit IRC | 14:17 | |
| *** arxcruz has joined #openstack-infra | 14:17 | |
| *** dtantsur|brb is now known as dtantsur | 14:17 | |
| *** _nadya_ has joined #openstack-infra | 14:17 | |
| *** dhritishikhar has quit IRC | 14:18 | |
| *** spredzy|afk is now known as spredzy | 14:18 | |
| *** mase_x200 has quit IRC | 14:19 | |
| fungi | jerryz: i can reproduce it on an ubuntu 14.04 system by running `tox -e pep8` | 14:19 |
| mordred | fungi: I thought our puppetforge account was openstackinfra not openstackci, right? | 14:20 |
| fungi | mordred: right | 14:20 |
| mordred | k | 14:20 |
| mordred | we just haven't updated all the metadata.json yet | 14:20 |
| *** hdd has joined #openstack-infra | 14:20 | |
| *** zz_jgrimm is now known as jgrimm | 14:20 | |
| *** craigbr has joined #openstack-infra | 14:21 | |
| fungi | mordred: right. i think jeblair and nibalizer were working through puppet-httpd first to see if there's anything else we need to change | 14:21 |
| *** timcline has joined #openstack-infra | 14:21 | |
| zxiiro | So I think I'm pretty close to getting my nodepool setup working. I'm currently failing with jenkins_manager http://pastebin.com/RcHN1C6N | 14:21 |
| fungi | mordred: once we have a success there, we can update the metadata in all the others to be consistent with what we have known-working | 14:21 |
| zxiiro | Any ideas on what the problem might be? I confirmed that my jenkins user and apikey configured are working so I'm not sure why nodepool says it can't authenticate | 14:22 |
| *** Bsony has quit IRC | 14:22 | |
| zxiiro | I'm assuming at this step it's trying to login to Jenkins API to create a node. | 14:22 |
| fungi | zxiiro: sounds like either a jenkins configuration issue or the credentials you put in your nodepool config aren't correct | 14:23 |
| *** zz_dimtruck is now known as dimtruck | 14:23 | |
| *** mugsie has quit IRC | 14:23 | |
| *** thingee has joined #openstack-infra | 14:23 | |
| zxiiro | fungi: I'm pretty confident the credentials are right. On the Jenkins side what permissions should the user have for nodepool to work? I assume create/manage slaves but is there any others? | 14:23 |
| fungi | zxiiro: ours is set for administrator privs in jenkins | 14:24 |
| *** mugsie has joined #openstack-infra | 14:24 | |
| *** ajo_ has joined #openstack-infra | 14:25 | |
| *** ajo has quit IRC | 14:25 | |
| fungi | zxiiro: you've got the apikey in nodepool.yaml set to what you generated in jenkins for your user, and then obtained the credentials-id uuid from the jenkins xml configuration and added that to nodepool.yaml as well? | 14:25 |
| *** craigbr has quit IRC | 14:25 | |
| zxiiro | fungi: no i didn't set credentials-id since it said it was optional and would use the sshkey configured in the image section | 14:26 |
| fungi | zxiiro: ahh, got it. you're right it does appear to be optional | 14:26 |
| zxiiro | fungi: i guess i might as well try the credentials-id way. What is that credential for? is it to login to the slave? or login to jenkins? | 14:26 |
| *** hdd has quit IRC | 14:27 | |
| *** garyh has joined #openstack-infra | 14:27 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/project-config: Add puppet-diskimage-builder module https://review.openstack.org/167607 | 14:27 |
| *** dkranz has quit IRC | 14:27 | |
| dhellmann | fungi: there's a copyright configuration field for sphinx copyright notices that is included in the footer by default | 14:27 |
| *** doug-fish has quit IRC | 14:28 | |
| fungi | zxiiro: we create an ssh credential in the jenkins webui associated with the ~jenkins/.ssh/id_rsa on the master (it corresponds to the rsa public key we put in ~jenkins/.ssh/authorized_keys on the slaves via puppet) | 14:28 |
| openstackgerrit | Monty Taylor proposed openstack-infra/system-config: Add puppet-diskimage-builder to our module list https://review.openstack.org/167608 | 14:28 |
| fungi | dhellmann: in the conf.py or in each individual rst file? | 14:28 |
| *** unicell1 has joined #openstack-infra | 14:28 | |
| dhellmann | fungi: conf.py, let me find an example | 14:29 |
| zxiiro | ok that's what i thought | 14:29 |
| *** BharatK has joined #openstack-infra | 14:29 | |
| fungi | dhellmann: oh, looks like we use that already, we just don't have license terms mentioned in it | 14:29 |
| *** yfried has joined #openstack-infra | 14:29 | |
| dhellmann | fungi: http://git.openstack.org/cgit/openstack/oslo-specs/tree/doc/source/conf.py#n65 | 14:29 |
| dhellmann | right | 14:29 |
| fungi | dhellmann: so you're saying license information is traditionally added to the copyright notice? | 14:29 |
| zxiiro | so something must be wrong with my apikey but i'm pretty sure i'm using the right one... I guess i'll keep poking at this hopefully something will work | 14:29 |
| openstackgerrit | Monty Taylor proposed openstack-infra/puppet-nodepool: Consume puppet-diskimage-builder https://review.openstack.org/167610 | 14:29 |
| fungi | dhellmann: easy enough. thanks! i'll see about getting a slew of patches proposed for that | 14:30 |
| dhellmann | fungi: I'm not sure there's a tradition, I'm just saying we could use that existing field and add a link to the relevant license file | 14:30 |
| zxiiro | FWIW I got some notes which I hope to turn it into better docs to contribute back to nodepool once i get htis working | 14:30 |
| dhellmann | fungi: sounds good, thanks | 14:30 |
| fungi | dhellmann: okay, good enough. just didn't know if there was something more standard for that purpose | 14:30 |
| *** unicell has quit IRC | 14:31 | |
| dhellmann | fungi: we could also force the value via oslosphinx | 14:31 |
| dhellmann | that would require fewer patches | 14:31 |
| fungi | zxiiro: awesome, thanks. the only other thing i can think of is that (at least at one point in time) csrf protection needed to be disabled on our jenkins masters to be able to work for this. not sure if that's still the case | 14:31 |
| *** hasharLunch is now known as hashar | 14:31 | |
| yolanda | hi, anyone having trouble using pypi under a proxy? | 14:31 |
| fungi | dhellmann: though at the risk of possibly being less correct | 14:31 |
| dhellmann | fungi: true | 14:31 |
| *** sputnik13 has joined #openstack-infra | 14:32 | |
| yolanda | having errors like Could not fetch URL https://pypi.python.org/simple/pbr/: connection error: hostname 'xxxx' doesn't match either of 'www.python.org', 'python.org', ... | 14:32 |
| yolanda | i have a pip.conf with proxy and trusted hosts | 14:32 |
| yolanda | but still same error | 14:32 |
| *** deepakcs has quit IRC | 14:32 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/project-config: Add puppet-diskimage_builder module https://review.openstack.org/167607 | 14:32 |
| openstackgerrit | Monty Taylor proposed openstack-infra/system-config: Add puppet-diskimage_builder to our module list https://review.openstack.org/167608 | 14:34 |
| *** ChuckC has quit IRC | 14:35 | |
| *** nelsnelson has joined #openstack-infra | 14:36 | |
| *** ociuhandu has quit IRC | 14:37 | |
| *** dkranz has joined #openstack-infra | 14:38 | |
| *** carl_baldwin has joined #openstack-infra | 14:38 | |
| *** ociuhandu has joined #openstack-infra | 14:40 | |
| yolanda | fungi, mordred, any experience with that? looks as i cannot get into pypi under a proxy | 14:41 |
| fungi | yolanda: i get to use a real network instead, so no experience at all with being stuck behind a proxy | 14:41 |
| yolanda | that looks as a bug, but even forcing trusted hosts, or no ssl check, it's failing all time | 14:42 |
| fungi | yolanda: right, the error message suggests that your connection is ending up at a webserver which doesn't have the requested vhost | 14:43 |
| *** james_li has quit IRC | 14:43 | |
| yolanda | that webserver is the proxy | 14:43 |
| yolanda | i have a pip.conf with proxy=xxx | 14:43 |
| fungi | well, i mean that error looks like the proxy is requesting the pypi vhost from a server which doesn't serve it | 14:44 |
| fungi | so it could be a problem with the proxy | 14:44 |
| *** carl_baldwin has quit IRC | 14:44 | |
| *** mugsie has quit IRC | 14:44 | |
| mordred | yolanda: also make sure that your pip.conf is inside of the dib context | 14:44 |
| mordred | and not just on the host machine | 14:44 |
| *** tnovacik has joined #openstack-infra | 14:44 | |
| *** wenlock has joined #openstack-infra | 14:45 | |
| yolanda | mordred, i just created a /root/.pip/pip.conf prior to the install | 14:45 |
| yolanda | and also http_proxy and https_proxy vars are set | 14:45 |
| *** nelsnelson has quit IRC | 14:45 | |
| yolanda | they are declared with an export just before doing the pip install | 14:45 |
| yolanda | and then doing the install with | 14:46 |
| yolanda | sudo -E -H /usr/zuul-env/bin/pip install /opt/git/openstack-infra/zuul-cloner | 14:46 |
| fungi | yolanda: does the "doesn't match either of 'www.python.org', 'python.org', ..." also mention pypi.python.org in the list? | 14:46 |
| *** smccully has joined #openstack-infra | 14:46 | |
| mordred | AJaeger, jeblair: can we talk about project-config/governance repo depends-on lines? | 14:46 |
| *** reed has joined #openstack-infra | 14:46 | |
| mordred | neither thing that we're doing there makes any sense to me | 14:47 |
| mordred | basically - I think we should stop having depends-on lines | 14:47 |
| AJaeger | mordred, we discussed that in an infra meeting a couple of weeks ago and wrote it down in the infra manual | 14:47 |
| mordred | really? | 14:47 |
| mordred | sigh | 14:47 |
| mordred | sorry | 14:47 |
| mordred | well, I disagree, for the record | 14:47 |
| AJaeger | http://docs.openstack.org/infra/manual/creators.html#add-project-to-the-governance-repository | 14:47 |
| mordred | neither should depend on the other | 14:47 |
| mordred | you need both | 14:47 |
| mordred | but blocking a new infra repo on landing a governance patch is crazy | 14:48 |
| AJaeger | mordred, let's include anteaya in the discussion, plesae | 14:48 |
| *** mugsie has joined #openstack-infra | 14:48 | |
| fungi | mordred: it was an attempt to prevent our reviewers from accidentally approving addition of "official" projects which were not yet approved by the tc, since we've had instances in the past where that came into question. it also forces us to make sure we keep the governance project list current | 14:48 |
| mordred | gotcha | 14:48 |
| mordred | I understand now | 14:48 |
| mordred | infra is a weird case | 14:48 |
| yolanda | going to try updating pip version, i'd say we are on 1.5.2 | 14:48 |
| mordred | ok - I remove my objection | 14:49 |
| *** mpaolino has quit IRC | 14:49 | |
| mordred | I am annoyed that I'm going to be blocked on next steps on dib-nodepool in some ways until next tuesday | 14:49 |
| *** jaypipes_ is now known as jaypipes | 14:49 | |
| *** samueldmq_ has joined #openstack-infra | 14:49 | |
| fungi | mordred: and while i agree that it probably gets in our own way for infra projects, it's sort of nice to the rest of the community if we don't exempt ourselves from a requirement we put on everyone else just because we can | 14:49 |
| notnownikki | is anyone here able to +2A this now the governance part is merged? https://review.openstack.org/#/c/163402/ | 14:49 |
| mordred | yes - makes sense for everyone else - I'd actually kinda like an openstack-infra/* in the governance repo tbh | 14:49 |
| clarkb | mordred why would that block anything? nodepool is perfectly capable of installing dib as a dep today | 14:50 |
| mordred | because the rubberstamping of infra projects is really a waste of the TC's time, process consistency that it is | 14:50 |
| *** carl_baldwin has joined #openstack-infra | 14:50 | |
| AJaeger | clarkb, fungi, I already voted on 163402 with +2... | 14:50 |
| mordred | clarkb: sure. it just means I need to duplicate the work in the puppet-dib repo in the nodepool repo now | 14:50 |
| fungi | AJaeger: notnownikki: yep, looking it over now | 14:50 |
| clarkb | mordred what work is that? | 14:51 |
| mordred | clarkb: because it turns out we need more things installed | 14:51 |
| mordred | it's not a BIG deal | 14:51 |
| notnownikki | mordred, i was also thinking we might want -core groups for some of these puppet modules, as we have people with expertise in them who aren't infra-core, sound sensible? (case by case basis of course) | 14:51 |
| mordred | it's just annoying | 14:51 |
| clarkb | oh is this because of ya deboostrap? | 14:51 |
| *** nelsnelson has joined #openstack-infra | 14:51 | |
| mordred | and yum,and yum-utils, and python-lzma | 14:51 |
| jerryz | fungi: do you know where the problem is? | 14:51 |
| mordred | like I said -it's fine | 14:52 |
| mordred | ignore me | 14:52 |
| clarkb | mordred that doesnt block anything you just get bigger images for a week :) | 14:52 |
| mordred | nope | 14:52 |
| mordred | I will absolutely not do that | 14:52 |
| mordred | I will make another nodepool patch | 14:52 |
| *** ajo_ has quit IRC | 14:52 | |
| mordred | bigger images will not work on rackspace | 14:52 |
| clarkb | why not? | 14:52 |
| *** Somay has quit IRC | 14:53 | |
| mordred | it's not just that I'm chasing small images - it's that I'm chasing rackspace working | 14:53 |
| *** ajo has joined #openstack-infra | 14:53 | |
| mordred | clarkb: can we forget that I complained and just let me go back to making the nodepool patch? | 14:53 |
| *** esker has quit IRC | 14:53 | |
| clarkb | sure? I am genuinely curious how these things are related. because well they shouldnt be | 14:53 |
| *** tsg_ has joined #openstack-infra | 14:54 | |
| *** mugsie has quit IRC | 14:54 | |
| fungi | jerryz: not yet, though i did confirm that if i import time again inside that method it passes | 14:54 |
| *** mugsie has joined #openstack-infra | 14:54 | |
| *** ChuckC has joined #openstack-infra | 14:55 | |
| *** claudiub has joined #openstack-infra | 14:55 | |
| jerryz | fungi: i also reproduced it on another clean 14.04 box | 14:55 |
| *** erikmwil_ has joined #openstack-infra | 14:55 | |
| *** erikmwilson is now known as Guest11500 | 14:55 | |
| *** erikmwil_ is now known as erikmwilson | 14:55 | |
| *** erikmwilson_ has joined #openstack-infra | 14:56 | |
| fungi | jerryz: i see the problem | 14:56 |
| zxiiro | fungi: do you know if i need to open JNLP / CLI ports | 14:56 |
| openstackgerrit | Monty Taylor proposed openstack-infra/puppet-nodepool: Consume puppet-diskimage-builder https://review.openstack.org/167610 | 14:56 |
| openstackgerrit | Monty Taylor proposed openstack-infra/puppet-nodepool: Install additional depends for dib https://review.openstack.org/167625 | 14:56 |
| *** tonytan4ever has joined #openstack-infra | 14:56 | |
| *** craigbr has joined #openstack-infra | 14:56 | |
| mordred | clarkb: they are related | 14:57 |
| mordred | clarkb: because the cloud images are cloud-init based | 14:57 |
| *** erikmwilson has quit IRC | 14:57 | |
| mordred | clarkb: and the rackspace work is not | 14:57 |
| mordred | clarkb: so, unless I want to construct an element for a few weeks taht removes cloud-init and doesn't screw something else up | 14:57 |
| *** Somay has joined #openstack-infra | 14:57 | |
| *** erikmwilson has joined #openstack-infra | 14:57 | |
| mordred | clarkb: the thing I want to do is actually use the code that i've been working on and testing | 14:57 |
| mordred | that requires some new packages | 14:57 |
| mordred | I agree, the new repo is NOT a blocker | 14:57 |
| mordred | I was overreacting and for that I apologize | 14:58 |
| mordred | the patch to puppet-nodepool above should suffice, and I have updated the "port to new repo | 14:58 |
| fungi | jerryz: see inline comment, but you need to rebase and then rework that patch. we no longer import time in that file | 14:58 |
| mordred | patch to fix it | 14:58 |
| fungi | zxiiro: i'll get you a link to the iptables rules we have for our jenkins masters | 14:59 |
| *** armax has joined #openstack-infra | 14:59 | |
| openstackgerrit | Merged openstack-infra/project-config: Project for puppet-bandersnatch https://review.openstack.org/163402 | 15:00 |
| *** nelsnelson has quit IRC | 15:00 | |
| openstackgerrit | Brian Hunter proposed openstack-infra/nodepool: Pretty Age in nodepool cli image/node lists https://review.openstack.org/161994 | 15:01 |
| *** kmartin has joined #openstack-infra | 15:01 | |
| fungi | zxiiro: per http://git.openstack.org/cgit/openstack-infra/system-config/tree/modules/openstack_project/manifests/jenkins.pp#n24 we open 80/tcp, 443/tcp for webui+api access and (from nodepool) 8888/tcp for the zeromq-publisher plugin | 15:02 |
| *** mattt has joined #openstack-infra | 15:02 | |
| zxiiro | hmm ok so maybe jnlp isn't the issue then | 15:03 |
| mordred | clarkb: in good news - the "process config drive" stuff is working on HP both with and without config-drive attached | 15:03 |
| jerryz | fungi: got it. first rule about git. rebase constantly:) | 15:03 |
| mordred | clarkb: and for both ubuntu and centos | 15:03 |
| mordred | clarkb: I'm working on testing rax now | 15:03 |
| hogepodge | quick question, how can we get gerritbot to post activity on the openstack/defcore repository to the #defcore room? | 15:03 |
| fungi | jerryz: well, just mainly remember that even though gerrit has you reviewing in the context of the parent commit, we test in the context of the branch tip | 15:03 |
| fungi | hogepodge: http://ci.openstack.org/irc.html mentions the access controls we'll need in place | 15:04 |
| mordred | clarkb: unfortunately, the debug cycle involves uploading images to clouds ... :) | 15:05 |
| *** nelsnelson has joined #openstack-infra | 15:05 | |
| fungi | hogepodge: then see http://docs.openstack.org/infra/manual/creators.html#configure-gerritbot-to-announce-changes | 15:05 |
| clarkb | mordred: ya thats the worst part | 15:06 |
| openstackgerrit | Marton Kiss proposed openstack-infra/project-config: Add puppet-pgsql_backup as a separate puppet module https://review.openstack.org/167631 | 15:06 |
| clarkb | mordred: also how closely to deboostrap/yum based distros does this make our images. Put another way how hard would it be to support SUSE or Arch or Gentoo? | 15:06 |
| mordred | clarkb: I don't think it would be too hard to support them | 15:06 |
| mordred | clarkb: we'll need to make a thing that knows how to install them into a chroot - but pretty much everyone has one of those | 15:07 |
| mordred | clarkb: I mean, I dont' even know redhat well and it took me all of 2 hours to make the main yum-minimal element | 15:07 |
| *** nelsnelson has quit IRC | 15:07 | |
| mordred | clarkb: now that you've mentioned it, I'll probably just rage-support them one weekend $because | 15:08 |
| *** nelsnelson has joined #openstack-infra | 15:08 | |
| mordred | AJaeger: does suse have anything similar to debootstrap? | 15:08 |
| *** doude has quit IRC | 15:08 | |
| *** doude has joined #openstack-infra | 15:08 | |
| *** openstackgerrit has quit IRC | 15:08 | |
| clarkb | mordred: I don't know that we need to rage support them :) mostly curious how flexible the underlying "build from scratch" bits in dib are | 15:09 |
| *** openstackgerrit has joined #openstack-infra | 15:09 | |
| AJaeger | mordred, what's debootstrap? | 15:09 |
| AJaeger | image building tool? SUSE uses kiwi | 15:09 |
| openstackgerrit | Michael Krotscheck proposed openstack-infra/storyboard: Remove page_size_maximum https://review.openstack.org/167633 | 15:09 |
| krotscheck | jeblair: ^^ I think that supersedes yours? | 15:10 |
| mordred | AJaeger: it's a thing that can install debian into a chroot | 15:10 |
| mordred | AJaeger: basically - is there a way to install suse into something that does not use an ISO and an installer | 15:10 |
| * mordred goes to read about kiwi | 15:10 | |
| openstackgerrit | Fabien Boucher proposed openstack-infra/puppet-ssh: Add the trusted source as class parameter https://review.openstack.org/167635 | 15:11 |
| clarkb | gentoo should be simple since all their installs are basically that iirc | 15:11 |
| mordred | AJaeger: another way to ask the question would be- how is the openSUE-13.2.tar.xz file in https://github.com/openSUSE/docker-containers-build/tree/openSUSE-13.2/docker created? | 15:11 |
| krotscheck | Well, maybe. | 15:11 |
| * krotscheck pondres. | 15:11 | |
| AJaeger | mordred, I think with kiwi but can figure out later | 15:12 |
| *** ominakov has joined #openstack-infra | 15:12 | |
| mordred | https://marc.waeckerlin.org/computer/blog/bootstrap_opensuse <-- this seems to indicate that it's possible | 15:12 |
| mattt | mordred: remember i mentioned the memcached installation issue yesterday or the day before? | 15:12 |
| mordred | mattt: vaguely - it's possible I was drunk | 15:12 |
| mattt | mordred: ha! | 15:13 |
| *** YorikSar has quit IRC | 15:13 | |
| krotscheck | jeblair: Added a comment. There's probably some discussion to be had on the default behavior. | 15:13 |
| mattt | mordred: so we're seeing the problem again ... i don't have a bunch of data on hand but it may be limited to devstack-trusty-hpcloud-b4 | 15:13 |
| * krotscheck will put that up as a separate patch | 15:13 | |
| mattt | mordred: i see success in devstack-trusty-hpcloud-b1 and devstack-trusty-rax-iad | 15:14 |
| mordred | mattt: AWESOME | 15:14 |
| *** jaypipes has quit IRC | 15:14 | |
| mordred | clarkb, fungi: ^^ this one from mattt will make you happy | 15:14 |
| *** sushilkm has joined #openstack-infra | 15:14 | |
| *** sushilkm has left #openstack-infra | 15:14 | |
| clarkb | mattt: mordred is that current since yesterday afternoon? because I just rebuilt devstack-trusty in hpcloud yesterday after I cleaered up the disk space and all hpcloud should use the same image | 15:14 |
| mattt | clarkb: these are failures from today | 15:15 |
| mattt | clarkb: rather, i do have one failure from today! | 15:15 |
| clarkb | ok so likely not an image problem if its also passing in hpcloud today | 15:15 |
| openstackgerrit | Luigi Toscano proposed openstack-infra/project-config: Add functional test job for python-saharaclient https://review.openstack.org/167641 | 15:16 |
| mattt | clarkb: yeah i'm not sure what b1 and b4 denote but my failures are all w/ b4 | 15:16 |
| fungi | clarkb: unless devstack-trusty is stale in hpcloud-b4 and it only ran there once | 15:16 |
| anteaya | AJaeger: thanks for trying to include me, I had to complete some manditory training about Ontario laws for inclusivity for people with disablilities | 15:16 |
| clarkb | fungi: I manually updated it though :/ shouldn't be stale | 15:16 |
| anteaya | mordred: and I think others have made the point about the importance of keeping the governance repo up to date for electoral officials | 15:17 |
| anteaya | you can't have it both way | 15:17 |
| anteaya | s | 15:17 |
| anteaya | you want the big tent you got it | 15:17 |
| mattt | clarkb fungi : let me try to gather a bit more data here | 15:17 |
| mordred | anteaya: yah - well - I understand the point for sure | 15:17 |
| anteaya | others have jobs and responsiblitiles they ahve to do as well | 15:17 |
| anteaya | great thanks | 15:17 |
| anteaya | this really isn't going to be an easy election season at all | 15:17 |
| clarkb | mattt: ok let us know, also you can make that image locally if you have a bit of bandwidth, disk space, and time | 15:17 |
| fungi | clarkb: oh, you did devstack-trusty? i figured you'd only done bare-trusty | 15:18 |
| anteaya | I feel terrible for dumpting this on tristan | 15:18 |
| *** ghostpl_ has quit IRC | 15:18 | |
| mordred | anteaya: I thought it was an infra policy, not an openstack policy | 15:18 |
| mordred | anteaya: thus my confusion | 15:18 |
| mordred | anteaya: I have been understandinged | 15:18 |
| clarkb | mattt: ma not help in this case but https://git.openstack.org/cgit/openstack-infra/project-config/tree/tools/build-image.sh should get you going if you need it | 15:18 |
| anteaya | well to be honest I might have started out as my policy | 15:18 |
| mordred | anteaya: it makes perfect sense in theh context that it exists | 15:18 |
| anteaya | because I really saw no way of doing the job of election official withoout it | 15:18 |
| mordred | since it's about knowing that it's approved and ok to put there | 15:18 |
| clarkb | fungi: I did devstack-trusty and devstack-precise as soon as I had the disk space freed up | 15:19 |
| *** ghostpl_ has joined #openstack-infra | 15:19 | |
| anteaya | right | 15:19 |
| fungi | ahh | 15:19 |
| anteaya | it has to be tracked first, not afterward | 15:19 |
| mordred | this is not generally a problem with infra projects, since we are infra and know that a new repo is approved to be in infra - BUT - fungi rightly pointed out that we should not be exempt | 15:19 |
| anteaya | as the being tracked afterward just lead to having to run around and clean things up | 15:19 |
| anteaya | yes, thank you | 15:20 |
| openstackgerrit | Fabien Boucher proposed openstack-infra/system-config: Use "$puppetmaster_server" as parameter to the ssh class https://review.openstack.org/167646 | 15:20 |
| anteaya | if anything infra needs to be the role model for expected behaviour | 15:20 |
| anteaya | not the exception | 15:20 |
| mordred | yup | 15:20 |
| mordred | in any case- I misunderstood and have been corrected | 15:20 |
| anteaya | thanks | 15:20 |
| anteaya | just wanted to let you know taht I am sorry I missed the conversation and glad that it happened | 15:21 |
| openstackgerrit | Joe Gordon proposed openstack-infra/project-config: Make tox-queries job voting https://review.openstack.org/167647 | 15:22 |
| *** Longgeek has joined #openstack-infra | 15:23 | |
| clarkb | mattt: fungi but I am double checking all of that now | 15:24 |
| openstackgerrit | Michael Krotscheck proposed openstack-infra/storyboard: Remove page_size_default https://review.openstack.org/167652 | 15:25 |
| krotscheck | There we go | 15:25 |
| clarkb | yup all of the glance images are newer than the qcow2 | 15:25 |
| clarkb | * for devstack-trusty | 15:25 |
| openstackgerrit | Merged openstack-infra/elastic-recheck: Functional test should not require a config file https://review.openstack.org/167248 | 15:26 |
| openstackgerrit | Marton Kiss proposed openstack-infra/project-config: Add puppet-pgsql_backup as a separate puppet module https://review.openstack.org/167631 | 15:27 |
| anteaya | asselin: are you available to add this ci account to the filtered group please? https://review.openstack.org/#/q/owner:fwaas.ci%2540cisco.com+status:open,n,z | 15:27 |
| anteaya | teach me to review a neutron patch | 15:28 |
| *** ryanpetrello has quit IRC | 15:28 | |
| *** Longgeek has quit IRC | 15:28 | |
| openstackgerrit | Merged openstack-infra/elastic-recheck: Remove resolved bugs https://review.openstack.org/167237 | 15:29 |
| *** stevemar has joined #openstack-infra | 15:29 | |
| *** otter768 has joined #openstack-infra | 15:29 | |
| AJaeger | mordred, the openSUSE image was created by kiwi | 15:29 |
| mattt | clarkb: thanks! here is what i have so far http://paste.openstack.org/show/196635/ | 15:31 |
| zxiiro | fungi: is it possible to launch a node using foreground tools rather than nodepoold? | 15:31 |
| mordred | AJaeger: ok. that's unfortunatley probably a bit heavy-weight for what I'd be looking for ... I'l give the blog post above a try and see where it gets | 15:31 |
| *** sputnik13 has quit IRC | 15:32 | |
| AJaeger | mordred, I can bring you together with the engineer that created the image | 15:32 |
| *** ryanpetrello has joined #openstack-infra | 15:32 | |
| *** samueldmq_ has quit IRC | 15:33 | |
| anteaya | could someone please disable account 11103: http://paste.openstack.org/show/196636/ in accordance with this ml post: http://lists.openstack.org/pipermail/openstack-infra/2015-March/002582.html | 15:33 |
| mordred | AJaeger: it's likely he and I will see the world differently, which is ok - I believe the redhat and ubuntu people all think I'm crazy too | 15:33 |
| *** flavio has joined #openstack-infra | 15:34 | |
| *** otter768 has quit IRC | 15:34 | |
| AJaeger | mordred, meet flavio - flavio, meet mordred ;) | 15:35 |
| flavio | hi mordred, AJaeger told me you asked some questions about the Docker image I built for openSUSE | 15:35 |
| AJaeger | mordred, you better tell flavio what you want, he just joined for the conversation. | 15:35 |
| mattt | clarkb: http://paste.openstack.org/show/196645/, found another failure | 15:35 |
| clarkb | mattt: mordred where is memcache being installed from? | 15:36 |
| clarkb | default ubuntu package list? | 15:36 |
| mattt | clarkb: yep | 15:36 |
| clarkb | mattt: any chance you can get ansible to expose the underlying apt-get errors? I think that will help debug quite a bit | 15:36 |
| fungi | zxiiro: not sure what you mean. if i understand your question, i use `nova boot ...` for that | 15:36 |
| mordred | flavio: hi! | 15:37 |
| *** mpaolino has joined #openstack-infra | 15:37 | |
| mordred | flavio: so - I'm looking for a way to do a lightweight installing of suse into a chroot - with content kinda-similar to what I'd get from docker | 15:37 |
| *** yamahata has quit IRC | 15:37 | |
| mordred | flavio: preferrably without needing an installation iso to start with | 15:37 |
| zxiiro | fungi: kind of like "nodepool image-update" but for spinning up a node. I figure it'd be more convenient to be able to tell it to just spin up one to troubleshoot than to have to run nodepoold and wait for it to fail | 15:38 |
| *** Sukhdev has joined #openstack-infra | 15:38 | |
| *** e0ne is now known as e0ne_ | 15:38 | |
| zxiiro | fungi: I can confirm that my credentials are correct though because i get a much different error if i purposely set the wrong credentials | 15:38 |
| mordred | flavio: a) kiwi can clearly make docker .xz files - and I could untar those into a chroot pretty easily ... will it involved downloading an iso first? | 15:38 |
| mattt | clarkb: we have more verbose ansible logging at http://logs.openstack.org/86/166986/8/check/os-ansible-deployment-dsvm-check-commit/1a06bd7/logs/ansible-logging/ansible.log (for example), but that's not showing anything useful | 15:38 |
| mordred | flavio: b) can I run kiki on a non-suse host system? | 15:38 |
| flavio | mordred: do you know the Docker image is missing stuff like the kernel? | 15:40 |
| fungi | zxiiro: got it. no nodepool doesn't have a launch subcommand for its cli (yet anyway) | 15:40 |
| flavio | mordred: I don't know if you can run kiwi on a non-suse system, I guess it would be possible. You should contact the kiwi developers to have the right answer | 15:41 |
| mordred | flavio: that's fine - I don't actually need the docker image itself- mostly using that as an example of the size of thign I'm looking for | 15:41 |
| *** thedodd has joined #openstack-infra | 15:41 | |
| flavio | mordred: or you could run kiwi inside of Docker, there's a opensuse/kiwi docker image for that | 15:41 |
| flavio | mordred: about question a: you just have to download the tar.xz file and extract that - pay attention while doing that otherwise you risk to overwrite your host files | 15:42 |
| mordred | flavio: is that tar.xz uploaded anywhere that's not atht git repo? like an official web server or anything? | 15:43 |
| *** amotoki has joined #openstack-infra | 15:43 | |
| jogo | jeblair: for your gate-tempest-dsvm-full-trove job | 15:44 |
| jogo | jeblair: why not make it run the subset of tempest that tests trove? | 15:44 |
| flavio | mordred: the Docker image is built using kiwi, the kiwi project is https://github.com/openSUSE/docker-containers | 15:45 |
| *** ssam2 has quit IRC | 15:45 | |
| flavio | mordred: you can take a look at the Makefiles to learn how kiwi is used | 15:45 |
| flavio | mordred: the output produced by kiwi is then uploaded to the other git repository you have seen. And from there the Docker Inc. automated system takes it and creates the official opensuse image | 15:46 |
| *** hdd has joined #openstack-infra | 15:46 | |
| mordred | flavio: so the git repo is the only place it's uploaded once it's built? | 15:46 |
| flavio | mordred: unfortunately we have to ship the tar.xz to git, that's how Docker's automated build works | 15:46 |
| mordred | clarkb: WOOT! rackspace works | 15:46 |
| mordred | greghaynes, SpamapS: ^^ | 15:46 |
| jeblair | jogo: i think it should do that and i probably just hadn't gotten around to it; is it easy for you to add that? | 15:46 |
| flavio | mordred: yes, then the automated system of Docker Inc. downloads it on their build service and creates the Docker image, which is then uploaded to the Docker Inc servers | 15:46 |
| flavio | mordred: and from there it's available to the world | 15:47 |
| jeblair | jogo: if not, i'll try to figure it out and push that up today | 15:47 |
| jogo | jeblair: yeah, I can tackle that if you want. | 15:47 |
| mordred | flavio: gothcha. cool! well, I now have several new things to go poke at and learn about. | 15:47 |
| mordred | flavio: thanks! | 15:47 |
| jeblair | jogo: cool | 15:47 |
| jogo | jeblair: thanks | 15:48 |
| flavio | mordred: feel free to ping me if you need help. You can always find me on freenode as "flavio" | 15:48 |
| jogo | jeblair: so instead of going for keystone, I was going to start from the other end and other things like trove | 15:48 |
| *** e0ne_ has quit IRC | 15:48 | |
| mordred | flavio: sweet. I really appreciate it | 15:49 |
| jeblair | jogo: yeah, trove has not been tested by other projects for, i think, 2 weeks now | 15:49 |
| jogo | jeblair: yeah I just read https://review.openstack.org/#/q/topic:devstack-trove,n,z | 15:49 |
| *** doug-fish has joined #openstack-infra | 15:50 | |
| jeblair | jogo: yep, that's my WIP to sever the tie with devstack, and then it needs tempest-lib stuff to happen to gets its tests out of tempest | 15:50 |
| jeblair | jogo: i haven't started on the last thing at all, and probably won't do that work since i have no idea about it | 15:50 |
| krotscheck | nibalizer, clarkb: I am on a train. I will shortly be in portlandia. Are there any good places-to-work downtown that you can recommend? | 15:50 |
| *** e0ne has joined #openstack-infra | 15:51 | |
| jeblair | jogo: but when both of those are complete, it can move out of the integrated change queue | 15:51 |
| *** annegentle has quit IRC | 15:51 | |
| clarkb | krotscheck: hogepodge likes Sisters Coffee in the pearl | 15:51 |
| *** gokrokve has joined #openstack-infra | 15:51 | |
| clarkb | krotscheck: floyds coffee in old town is good too | 15:51 |
| jogo | jeblair: nice | 15:52 |
| clarkb | krotscheck: mcmenamins are also good options, tend to have power and terrible service so you can hang out all day | 15:52 |
| clarkb | krotscheck: ringlers is a good option for that | 15:52 |
| *** emagana has joined #openstack-infra | 15:53 | |
| krotscheck | clarkb: That sounds like a really good list, awesome. | 15:53 |
| krotscheck | clarkb: Are you going to be at nibalizer's hack day tomorrow? | 15:53 |
| *** rwsu has joined #openstack-infra | 15:53 | |
| clarkb | krotscheck: yup | 15:53 |
| krotscheck | clarkb: Cool, will see you tomorrow then :) | 15:53 |
| clarkb | awesome | 15:53 |
| *** Swami has joined #openstack-infra | 15:54 | |
| clarkb | krotscheck: and you can always do the powells thing but that may be quite a bit more crowded | 15:54 |
| clarkb | krotscheck: also you may end up just reading books all day | 15:54 |
| *** ddieterly has quit IRC | 15:55 | |
| *** jordanP has joined #openstack-infra | 15:55 | |
| krotscheck | clarkb: I've done worse. | 15:55 |
| *** spzala has quit IRC | 15:56 | |
| *** aarefiev has quit IRC | 15:56 | |
| *** Bsony has joined #openstack-infra | 15:56 | |
| *** annegentle has joined #openstack-infra | 15:57 | |
| jogo | jeblair: you already set up the job https://review.openstack.org/#/c/163521/ | 15:57 |
| clarkb | krotscheck: if you have not done powells before I recommend wandering through for an hour or 5 | 15:57 |
| *** erikmwilson has quit IRC | 15:57 | |
| asselin | anteaya, added fwaas.ci@cisco.com | 15:57 |
| jogo | and its voting on tempest | 15:58 |
| *** yamahata has joined #openstack-infra | 15:58 | |
| *** aarefiev has joined #openstack-infra | 15:58 | |
| krtaylor | sweston, I finished the comments on the monitoring dashboard spec, ping me when you get a chance | 15:59 |
| *** patrickeast has joined #openstack-infra | 15:59 | |
| mattt | clarkb: added another entry http://paste.openstack.org/show/196652/, looks like all our failures are limited to that b4 | 16:00 |
| *** ddieterly has joined #openstack-infra | 16:00 | |
| mattt | clarkb: i've modified the gate script to do an explicit apt-get of memcached outside of ansible, just waiting to hit a b4 :) | 16:00 |
| clarkb | mattt: ok, good to know that it is likely something in that region. I am almost positive it isn't the image though, my hunch is its something odd with networking there | 16:00 |
| krotscheck | clarkb: I have. I may have forgotten part of my wallet there. | 16:01 |
| *** erikmwilson has joined #openstack-infra | 16:01 | |
| mattt | clarkb: it's always that memcached package tho, we install packages before that and they go through fine | 16:01 |
| *** tiswanso has quit IRC | 16:01 | |
| mattt | clarkb: i'll keep you posted w/ what i find | 16:01 |
| clarkb | mattt: thanks | 16:01 |
| *** tiswanso has joined #openstack-infra | 16:01 | |
| clarkb | mattt: I am going to go ahead and hold a devstack-trusty node in -b4 as well and poke around on it | 16:02 |
| *** Bsony has quit IRC | 16:02 | |
| *** unicell1 has quit IRC | 16:02 | |
| mattt | clarkb: awesome, thank you! | 16:02 |
| *** pal_ has quit IRC | 16:02 | |
| jogo | jeblair: so recent trove patch https://review.openstack.org/#/c/166845/ | 16:03 |
| jogo | jeblair: it runs tempest-dsvm-trove | 16:03 |
| jogo | but also has -full,-postgres-full,-neutron-full | 16:03 |
| *** ssam2 has joined #openstack-infra | 16:03 | |
| jogo | none of which have trove enabled anymore | 16:03 |
| anteaya | asselin: thank you | 16:04 |
| anteaya | asselin: you are doing a great job on the mailing list | 16:04 |
| anteaya | asselin: it is awesome to witness | 16:04 |
| asselin | anteaya, np | 16:04 |
| clarkb | yolanda: are you still around? 160885, jhesketh makes a good point, any particular reason to only toggle those two apache options? | 16:04 |
| anteaya | asselin: :) | 16:04 |
| yolanda | clarkb, let me look , haven't checked | 16:05 |
| *** ricolin has quit IRC | 16:05 | |
| jogo | jeblair: do we want to run a trove testing with postgres? | 16:05 |
| mordred | BobBall: hey - did you ever get firewall issues sorted to be able to follow up on https://review.openstack.org/#/c/155770 ? | 16:05 |
| mrmartin | fungi: https://review.openstack.org/#/c/167631/ the puppet-pgsql_backup patch for project-config | 16:06 |
| fungi | mrmartin: thanks! i'll take a look today | 16:06 |
| *** pal_ has joined #openstack-infra | 16:06 | |
| clarkb | mrmartin: fungi I think I am -2 on that | 16:07 |
| mrmartin | clarkb: why? | 16:07 |
| clarkb | mrmartin: fungi why not just add it to the existing backup module like we have done with mysql? | 16:07 |
| yolanda | clarkb, well, i just added those because i needed honestly, i don't have any objection on updating more | 16:07 |
| clarkb | I am not sure this deserves a separate module | 16:07 |
| BobBall | Still waiting mordred :( I've escalated them now | 16:08 |
| mrmartin | because it is called puppet-mysql_backup ? | 16:08 |
| mordred | BobBall: JOY | 16:08 |
| mordred | BobBall: that must be eversomuch fun | 16:08 |
| BobBall | Let me see if I can use a proxy or smt to upload the change | 16:08 |
| clarkb | mrmartin: wait we what let me figure out what we did there | 16:08 |
| mordred | BobBall: no rush - just curious | 16:08 |
| BobBall | Oh yes. I'm sure you know what corporate firewalls are like. | 16:08 |
| clarkb | mrmartin: I thought this was all part of bup | 16:08 |
| BobBall | Well, rush is useful since I'm on vacation from tomorrow :) | 16:08 |
| yolanda | clarkb, in that case, we'd better switch to some data structure, one for prefork, one for mpm, what do you think? | 16:08 |
| mordred | BobBall: ah - yes - rush then! | 16:08 |
| mrmartin | clarkb: we have this: https://github.com/openstack-infra/puppet-mysql_backup | 16:08 |
| clarkb | mrmartin: I am corrected | 16:08 |
| *** hdd has quit IRC | 16:08 | |
| mrmartin | clarkb: is that module deprecated? | 16:09 |
| clarkb | mrmartin: fungi nevermind then I guess this is the established pattern | 16:09 |
| fungi | clarkb: i assumed it was just the module to do the equivalent of puppet-mysql_backup but for postgres | 16:09 |
| clarkb | mrmartin: no, I was mistaken in how it was organized. I thought it was all part of the larger bug module | 16:09 |
| *** belmoreira has joined #openstack-infra | 16:09 | |
| *** Mithrandir has quit IRC | 16:09 | |
| clarkb | fungi: ya, basically I don't think we should have a mysql_backup module but that has sailed | 16:09 |
| anteaya | could someone please disable account 11103: http://paste.openstack.org/show/196636/ in accordance with this ml post: http://lists.openstack.org/pipermail/openstack-infra/2015-March/002582.html | 16:09 |
| clarkb | since I believe the upstream mysql and postgres module both offer similar | 16:10 |
| mrmartin | clarkb: I agree, it could be better to refactor this to a single bup module, but I suggest to go on what we have, and don't delay too much the askbot migration | 16:10 |
| clarkb | the benefit is when tied to bup | 16:10 |
| clarkb | mrmartin: ya its fine | 16:10 |
| mrmartin | ok. | 16:10 |
| clarkb | https://github.com/puppetlabs/puppetlabs-mysql/tree/master/manifests/backup | 16:10 |
| clarkb | I don't thin kwe should compete with ^ instead we should offer something complementary that works well with bup. But we have already made the mysql_backup module so postgres module is fine | 16:11 |
| *** sarob has joined #openstack-infra | 16:12 | |
| clarkb | though now that I read more of those manifest I remember the other reason we didn't use it is that it assumes local management of the mysql server which breaks with trove | 16:12 |
| *** rlandy has quit IRC | 16:12 | |
| mrmartin | and the backup is missing from here: https://github.com/puppetlabs/puppetlabs-postgresql | 16:13 |
| fungi | right, the alternative we discussed was to have db credentials on the bup servers themselves and run the database dumps locally there | 16:13 |
| mrmartin | yeap, that's not nicest solution we have, but works | 16:13 |
| openstackgerrit | Joe Gordon proposed openstack-infra/project-config: Stop running tempest-dsvm-neutron in juno https://review.openstack.org/167674 | 16:14 |
| jogo | mtreinish: thanks for explaining it to me ^ | 16:14 |
| jeblair | jogo: oh, wow, yeah that branch selecetor never should have merged | 16:15 |
| jeblair | jogo: we try to do 'stable/(A|B)' always, so that when we stop supporting B we can change it to 'stable/A' and when we stop supporting A we can drop it | 16:16 |
| jeblair | jogo: we might want to ask SlickNik about whether we need a postgres tempest job for trove | 16:16 |
| jeblair | jogo: but knowing mtreinish, i bet he knows the answer too :) | 16:17 |
| jogo | jeblair: makes sense (on both counts) | 16:17 |
| openstackgerrit | Bob Ball proposed openstack-infra/nodepool: Add support for requsting that config drive be used. https://review.openstack.org/155770 | 16:17 |
| *** mpaolino has quit IRC | 16:17 | |
| jogo | jeblair: well we don't test postgres with trove at the moment | 16:17 |
| *** ildikov has quit IRC | 16:18 | |
| clarkb | 161695 brings up the question of where do we as openstack have our configs vs what things are reconsumable | 16:18 |
| jeblair | jogo: then i reckon we can leave it for another day -- but _adding_ a job like which makes a lot of sense for trove that without worrying about adding it to nova changes is a big plus of this rationalization :) | 16:18 |
| clarkb | I am not sure openstack-infra/system-config is anything that should be reconsumable | 16:18 |
| BobBall | mordred: ^^ | 16:18 |
| clarkb | this is why we are making openstack-infra/puppet-openstackci for example | 16:18 |
| *** Mithrandir has joined #openstack-infra | 16:19 | |
| anteaya | could someone please disable account 11103: http://paste.openstack.org/show/196636/ in accordance with this ml post: http://lists.openstack.org/pipermail/openstack-infra/2015-March/002582.html | 16:19 |
| jeblair | clarkb: i agree, in the long term; it's worth considering whether we're willing to make some compromises before we get completely there | 16:19 |
| jeblair | clarkb: also, the jenkins puppet stuff is just a mess anyway :) | 16:20 |
| jeblair | it's like the oldest cruftiest thing we have in there :) | 16:20 |
| ctlaugh | I am working on getting nodepool to build an image for arm64 and the first "problem" I ran into is the installation of puppet. apt.puppetlabs.com does not have arm64 packages, and install_puppet.sh currently pulls from there. I see that Ubuntu has 3.4.3 available -- is there any known reason why I might have a problem using this vs. packages from puppetlabs.com? | 16:21 |
| mordred | BobBall: woot! | 16:21 |
| jeblair | ctlaugh: we mostly use it in order to have a consistent version across all of our systems/images which have various operating systems | 16:21 |
| *** chlong has quit IRC | 16:22 | |
| jeblair | ctlaugh: i think you might be okay with the plan you suggest | 16:22 |
| clarkb | jeblair: ctlaugh yup, its a convenience thing for us since the various distros all ship different versions | 16:22 |
| clarkb | but since we use puppet apply for image builds (no master involved) I expect 3.4.3 to just work | 16:22 |
| jeblair | ctlaugh: at least partially because the puppet used to build an image is pretty basic and should be compatible with a wide range of versions | 16:22 |
| ctlaugh | great, thank you clarkb, jeblair | 16:23 |
| *** belmoreira has quit IRC | 16:23 | |
| anteaya | can someone help me understand why it is so difficult for me to get a ci account disabled today? | 16:23 |
| *** carl_baldwin has quit IRC | 16:23 | |
| anteaya | I'm asking for it to be disabled at the ci account's request | 16:23 |
| jeblair | anteaya: there's too much scrolling by for me to read everything. feel free to pick one of us and ping us directly :) | 16:23 |
| fungi | heading out for a lunch walk, back well before 1800 utc | 16:24 |
| anteaya | jeblair: that feels like shouting to me | 16:24 |
| *** YorikSar has joined #openstack-infra | 16:24 | |
| anteaya | jeblair: which is contrary to my comfort level | 16:24 |
| anteaya | :/ | 16:24 |
| anteaya | but thank you | 16:24 |
| anteaya | fungi: enjoy your lunch and walk | 16:25 |
| jeblair | anteaya: well, i mean, try what you've been doing since that usually works, but if you're being ignored on something time sensitive, i'm saying it's okay to ping me directly and i won't think you're shouting at me :) | 16:25 |
| anteaya | jeblair: oh okay thanks | 16:25 |
| anteaya | more time sensitive for me since I don't want to move on to do other things and then forget about it | 16:25 |
| jeblair | anteaya: also, you just taught me how to look up an account id number, thanks :) | 16:26 |
| jeblair | anteaya: done | 16:26 |
| jeblair | 11103 is inactive | 16:26 |
| *** carl_baldwin has joined #openstack-infra | 16:27 | |
| *** ctlaugh has quit IRC | 16:27 | |
| *** ctlaugh has joined #openstack-infra | 16:27 | |
| openstackgerrit | Joe Gordon proposed openstack-infra/project-config: Stop running the integrated-gate template on trove https://review.openstack.org/167679 | 16:27 |
| openstackgerrit | Nikki Heald proposed openstack-infra/puppet-bandersnatch: Initial import of puppet-bandersnatch module https://review.openstack.org/167680 | 16:27 |
| jogo | jeblair: I think that is the last bit to remove trove from the integrated gate | 16:28 |
| jeblair | jogo: that's a phrase whose meaning i do not know :). i do agree that change removes many jobs that do not test trove. | 16:29 |
| *** imcsk8 is now known as imcsk8|afk | 16:30 | |
| clarkb | nibalizer: reading the diff on this base into template.pp change is making my brain melt. have you found an particular differ to do a better job? /me tries gerrit proper | 16:30 |
| jeblair | jogo: i'm sorry to be such a stickler about terminology, but i don't want to inadvertently cause any more confusion. | 16:30 |
| jogo | jeblair: no worries, which part should I adjust? | 16:31 |
| *** emagana has quit IRC | 16:31 | |
| *** davide has joined #openstack-infra | 16:31 | |
| *** mfink has joined #openstack-infra | 16:31 | |
| clarkb | nibalizer: gerrit proper does a little better. Looks like the move of the redhat bind-utils confused gertty | 16:31 |
| *** jordanP has quit IRC | 16:31 | |
| jeblair | jogo: this change stops running tests of other 'integrated' projects on changes to trove (which are clearly unecessary). | 16:32 |
| anteaya | jeblair: thank you | 16:32 |
| anteaya | jeblair: oh you're welcome | 16:32 |
| jeblair | jogo: notably, trove and the rest of the integrated projects will still share a change queue because of the transitive link through tempest. | 16:33 |
| *** chlong has joined #openstack-infra | 16:33 | |
| *** sushilkm has joined #openstack-infra | 16:34 | |
| jeblair | jogo: so if 'integrate gate' means 'set of jobs shared by projects', i think your statement was correct. if it means 'shared change queue', your change is a good next step but we still have some work to do. | 16:34 |
| *** sushilkm has left #openstack-infra | 16:34 | |
| *** scheuran has quit IRC | 16:35 | |
| * anteaya is afk for a bit | 16:36 | |
| zxiiro | got my nodepool setup working finally :) | 16:36 |
| jeblair | jogo: (i think some people use that term to mean both of those things in some circumstances and it's hard for me to tell which thing is meant sometimes; similarly with "co-gate"). | 16:37 |
| jogo | jeblair: to be clear, my patch does not pull trove changes out of the 'integrated' gate queue in zuul? | 16:37 |
| jogo | jeblair: pushing a new commit message soon | 16:37 |
| jeblair | jogo: your commit message is correct | 16:37 |
| zxiiro | I got another question about nodepool. If I have 2 jenkins masters and I set minimum ready images to 1. Will it boot 1 of that image on each jenkins master or only one in total? | 16:38 |
| jeblair | jogo: changes to trove will still share a change queue in the gate pipeline in zuul even after your patch lands. that is because trove and tempest still share a job, and tempest and nova,etc., share jobs | 16:38 |
| *** hdd has joined #openstack-infra | 16:39 | |
| *** unicell has joined #openstack-infra | 16:39 | |
| jeblair | jogo: which is why the final step for removing it from the shared change queue is moving the trove tests out of tempest | 16:39 |
| *** emagana has joined #openstack-infra | 16:39 | |
| jeblair | jogo: however, for the past 2 weeks, a failure of a trove test has been unable to cause a nova change to fail to merge because trove has not been tested in nova changes | 16:39 |
| *** emagana has quit IRC | 16:39 | |
| *** chlong has quit IRC | 16:39 | |
| clarkb | nibalizer: comments on 162830 for followup change ideas but +2 | 16:39 |
| clarkb | zxiiro: one in total | 16:40 |
| jogo | jeblair: pushing a new patch to better document that distinction | 16:40 |
| openstackgerrit | Joe Gordon proposed openstack-infra/project-config: Stop running jobs that don't install trove on trove changes. https://review.openstack.org/167679 | 16:40 |
| clarkb | zxiiro: the max servers should reflect your cloud provider limits | 16:40 |
| jogo | jeblair: exciting | 16:40 |
| jogo | jeblair: any other projects that you have on your hit list? | 16:40 |
| openstackgerrit | Nikki Heald proposed openstack-infra/puppet-bandersnatch: Initial import of puppet-bandersnatch module https://review.openstack.org/167680 | 16:40 |
| jogo | err decouple list* | 16:40 |
| *** emagana has joined #openstack-infra | 16:40 | |
| clarkb | zxiiro: sorry I read it as max, but minimum works the same way | 16:40 |
| *** gokrokve_ has joined #openstack-infra | 16:40 | |
| jeblair | jogo: i asked SergeyLukjanov about sahara, but i think he's been busy; i was going to finish up trove then take another look at that | 16:41 |
| clarkb | its a minimum for that label regardless of jenkins masters | 16:41 |
| zxiiro | clarkb: good ot know, thanks! | 16:41 |
| jogo | jeblair: I can take on doing this for another project or two | 16:42 |
| zxiiro | I'll submit some docs patches later now that I've sat through the whole process of setting this up. Hopefully it'll help the next guy | 16:42 |
| *** gokrokve has quit IRC | 16:43 | |
| jeblair | jogo: i think it makes sense to stop at trove and sahara for now. i think the rest of the 'compute layer' projects are tightly coupled enough that it warrants more thinking before going further. | 16:43 |
| jeblair | jogo: i figured those were easy, low hanging fruit that pretty much everyone agreed there was no tight coupling. | 16:43 |
| *** yolanda has quit IRC | 16:43 | |
| *** yolanda has joined #openstack-infra | 16:44 | |
| openstackgerrit | Jerry Zhao proposed openstack-infra/nodepool: add option to use ipv6 for image update and node launching https://review.openstack.org/156178 | 16:45 |
| *** sobersabre has quit IRC | 16:46 | |
| *** markus_z has quit IRC | 16:46 | |
| clarkb | cinerama: one thing I realized I should mention is that our older puppet moduels tend to have a single monolithic init.pp that installs all the things in a super opinionated way. But as we move to make them more reconsumable I think its a good idea to support mysql and apache for our use cases but not force anyone to use them if they want to layer their own webserver and db around an application | 16:46 |
| jogo | jeblair: yeah sahara sounds like the only other low hanging fruit right now | 16:47 |
| clarkb | cinerama: which is why I have commented about splitting things up. Hopefully that makes sense | 16:47 |
| jogo | jeblair: if SergeyLukjanov and you don't mind I can start the process for it | 16:47 |
| clarkb | zxiiro: awesome, thanks for taking notes :) | 16:47 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack-infra/devstack-gate: [DO-NOT-MERGE] Enable libvirt debug logs and core dumps https://review.openstack.org/167691 | 16:47 |
| openstackgerrit | Michael Krotscheck proposed openstack-infra/storyboard: Remove page_size_default https://review.openstack.org/167652 | 16:47 |
| cinerama | clarkb: yeah i agree, though at first i was confused cause i was trying to make it like how the others were | 16:47 |
| jeblair | jogo: maybe go aheand and write the first project-config patch for it, and we can wait for SergeyLukjanov to give it an okay before proceeding? | 16:48 |
| cinerama | clarkb: the main thing i guess is where things have to be done in a certain order, it makes deps that live outside of the puppetsphere or outside of the module somewhat more challenging | 16:48 |
| jogo | jeblair: sounds good to me, thanks | 16:48 |
| clarkb | cinerama: ya, it does | 16:48 |
| *** marun has quit IRC | 16:48 | |
| jeblair | jogo: did you find my old patches for trove? should be able to see the sequence i did for that... | 16:48 |
| jogo | jeblair: yeah I found most, not all had the topic set to devstack-trove | 16:48 |
| clarkb | cinerama: but you can do really high level deps like foo:db -> foo::app -> foo::webserver | 16:49 |
| *** gyee has joined #openstack-infra | 16:49 | |
| *** mfink_ has joined #openstack-infra | 16:49 | |
| zaro | morning | 16:49 |
| *** mfink_ has quit IRC | 16:49 | |
| *** marun has joined #openstack-infra | 16:49 | |
| jeblair | jogo: heh, can you change the topic after merge? :) | 16:50 |
| cinerama | clarkb: yeah that might be the way to go. do we have a preference on using the arrow-style ordering in modules? i was wondering if that was considered cheaty | 16:50 |
| jogo | I wonder if the author can | 16:50 |
| clarkb | cinerama: yes, don't actually use the arrows :) its just handy for one liner puppet illustrations | 16:50 |
| mtreinish | jeblair: sounds like a gerrit bug if you can't :) | 16:51 |
| cinerama | also we should have a puppet style guide with our cultural conventions. the linter doesn't cover higher level things i know we do or prefer | 16:51 |
| cinerama | would make for more consistent reviewing, too | 16:51 |
| *** BharatK has quit IRC | 16:52 | |
| clarkb | yolanda: please see comment on 163376 | 16:53 |
| *** sputnik13 has joined #openstack-infra | 16:55 | |
| yolanda | so clarkb, the intention there is to make that clear and more reusable, and avoid hardcoding things. So it can be easily switched to another mirror if needed | 16:55 |
| clarkb | yolanda: yes that is why we have module.env | 16:55 |
| yolanda | i know you can install modules from another sources, but just this duplication of git.openstack.org everywhere, while you can have that and env var that defines it... | 16:56 |
| clarkb | yolanda: we already put a system in place to make this clear and reusable and not hardcode things | 16:56 |
| yolanda | but that module.env is hardcoded with git.openstack.org | 16:56 |
| yolanda | what will you do if you need to change mirror? replace that everywhere? | 16:56 |
| *** melwitt has joined #openstack-infra | 16:56 | |
| clarkb | yolanda: yes, because that is where we expect people to install our code from. If however you don't want to install our environment then you can provide a different modules.env file that is sourced. Maybe all it does is update the array in our module.env | 16:57 |
| yolanda | yes, i know, that i could have it as an alternative, but even with that i found code looked much clear. If i need to fork that, and have my own modules.env, i may only need to change one line | 16:57 |
| *** jerryz has quit IRC | 16:58 | |
| *** jistr has quit IRC | 16:58 | |
| mordred | yeah - it's harder to track for a downstream if they need to change every line and to also track each new module we add and change those lines too | 16:58 |
| clarkb | mordred: you don't have to do that is what I am saying... | 16:58 |
| yolanda | clarkb, i suppose it's a matter of preferences there, but i thought it was a good change to make things look clear | 16:58 |
| clarkb | just update our array in a supplied module.env | 16:59 |
| *** amotoki has quit IRC | 16:59 | |
| yolanda | clarkb, yes, i understand you, but why have to change 60 lines if we can just change one? | 16:59 |
| clarkb | my concern here is we already have one method of making this change, it is tested and supposedly works well because forj was happy with it | 16:59 |
| clarkb | yolanda: you don't need to change 60 lines... | 16:59 |
| clarkb | you need maybe 3 lines | 16:59 |
| yolanda | clarkb, i don't point to git.openstack to clone the modules | 17:00 |
| yolanda | i point to git.gozer | 17:00 |
| clarkb | yes I know | 17:00 |
| yolanda | so i need to replace everything, right? | 17:00 |
| *** sarob has quit IRC | 17:00 | |
| clarkb | hold on one second let me write what I am thinking | 17:00 |
| *** flavio has quit IRC | 17:00 | |
| yolanda | ok | 17:00 |
| *** patrickeast has quit IRC | 17:00 | |
| yolanda | also, not even thinking on downstream, but for local testing as exampel | 17:00 |
| *** tnovacik has quit IRC | 17:00 | |
| yolanda | example | 17:00 |
| *** tim_o has quit IRC | 17:01 | |
| *** BobBall is now known as BobBall_AWAY | 17:01 | |
| openstackgerrit | Aleksey Ripinen proposed openstack-infra/storyboard-webclient: Support for tags https://review.openstack.org/166847 | 17:02 |
| clarkb | for k in ${!array[@]} ; do array[$k]=${array[$k]/git.openstack.org/git.gozer/} | 17:02 |
| clarkb | done | 17:03 |
| openstackgerrit | Joe Gordon proposed openstack-infra/project-config: Remove tempest-dsvm-ceilolmeter-mongodb job https://review.openstack.org/167695 | 17:03 |
| yolanda | well, yes, that's an option, but from my point of view it's cleaner to update an env var | 17:03 |
| *** tqtran has joined #openstack-infra | 17:04 | |
| wenlock | question, is anyone having errors running this command? git clone https://review.openstack.org/p/openstack-infra/zuul --depth 1 | 17:04 |
| wenlock | or maybe i missed the chatter | 17:04 |
| clarkb | I would rather we didn't have 2 layers of indirection here | 17:04 |
| mordred | wenlock: yes | 17:04 |
| wenlock | :D | 17:04 |
| mordred | wenlock: it's not supported because git on trusty is broken | 17:04 |
| wenlock | O.o | 17:04 |
| mordred | wenlock: you'll want to change to pointing at git.openstack.org instead | 17:04 |
| clarkb | or don't shallow clone | 17:04 |
| wenlock | ahhh | 17:04 |
| *** dalgaaf has joined #openstack-infra | 17:04 | |
| mordred | wenlock: the git people decided that it would be a good diea to have shallow clones want to have WRITE access to the repo | 17:04 |
| yolanda | clarkb, what do you mean with 2 level? | 17:04 |
| wenlock | mordred, clarkb, cool.... ty! | 17:05 |
| clarkb | yolanda: we (well mostly wenlock actually) devised modules.env as a way to extend the module lists arbitrarily | 17:05 |
| jeblair | wenlock, mordred: we decided it would be a good idea for them not to :) | 17:05 |
| clarkb | yolanda: that is one level of indirection that we support. You can tell it to load file from over there | 17:05 |
| *** jlanoux has quit IRC | 17:05 | |
| clarkb | yolanda: this proposal adds a second leve lwhere another env var will update contents of a file loaded from $place | 17:05 |
| mordred | clarkb: it's shell - env vars overriding env vars is kinda how it works, no? | 17:06 |
| yolanda | clarkb, ok, i understand, although i'm not 100% convinced, i really don't like to see repeating and hardcoded settings on any puppet manifest | 17:07 |
| mordred | clarkb, yolanda: it's possible I may be missing something - can you point me to yolanda's proposal? | 17:07 |
| yolanda | https://review.openstack.org/#/c/163376/3/modules.env | 17:08 |
| wenlock | clarkb , mordred, found out this works too: git clone https://review.openstack.org/openstack-infra/zuul --depth 1 | 17:08 |
| wenlock | which is better? | 17:08 |
| mordred | wenlock: I'd clone from git.o.o always | 17:08 |
| wenlock | don't do --depth 1 or remove the p? or got to mirror? | 17:08 |
| clarkb | wenlock: git.o.o, ^ clones against gerrit itself and we try to avoid that | 17:08 |
| mordred | it's the scalable git farm | 17:08 |
| wenlock | perfect | 17:08 |
| wenlock | ok :D ty guys | 17:08 |
| clarkb | mordred: yes, except this particular file is us directly asserting where the code we intsall our systems come from | 17:08 |
| mordred | clarkb: I disagree with that | 17:09 |
| clarkb | mordred: its fine for people to assert other locations for that and we have a method to do so | 17:09 |
| clarkb | but the env file for openstack-infra's system configs don't need to support arbitrary locations | 17:09 |
| mordred | clarkb: I think this file is part of the system that people are reconsuming - there is nothing infra specific about it | 17:09 |
| clarkb | the module installer does that | 17:09 |
| mordred | clarkb: then why are we making anything in our system reconsumable? | 17:09 |
| clarkb | mordred: it is also not supposed to be reconsumable | 17:09 |
| mordred | clarkb: why | 17:09 |
| mordred | why would you want to make it not reconsumable/ | 17:10 |
| clarkb | because system config is the glue we use to run all the things with our bits | 17:10 |
| wenlock | mordred we consume it from zuul-layout.yaml.erb, it's a key file for us | 17:10 |
| *** ildikov has joined #openstack-infra | 17:10 | |
| clarkb | modules like openstackci would be reconsumable, and maybe we eventually have system-config just wrap that thinly | 17:10 |
| yolanda | clarkb, but put on the situation where you want to deploy an infra environment for testing | 17:10 |
| jeblair | clarkb: yeah, that's what i'm imagining | 17:10 |
| wenlock | zuul-layout.yaml.erb is generic IMO to zuul checking | 17:10 |
| yolanda | you may need to update lots of system-config settngs | 17:10 |
| yolanda | why not put things as parameterized and easy as possible? | 17:11 |
| mordred | I don't understand the pushback at all - even without the reconsumable argument ... | 17:11 |
| jeblair | yolanda: yeah, we're not there yet | 17:11 |
| mordred | i'd consider that pach to be an improvement | 17:11 |
| clarkb | mordred: the pushback is the change is not necesary, we already have a way to do this | 17:11 |
| mordred | over having a hardcoded string 100 times | 17:11 |
| yolanda | jeblair, but there is a spirit to be, right? | 17:11 |
| jeblair | mordred: which patch? | 17:11 |
| mordred | jeblair: https://review.openstack.org/#/c/163376/3/modules.env | 17:11 |
| clarkb | so my ask is that we try to use the method that exists and if that doesn't work fix the existing method | 17:11 |
| *** hashar has quit IRC | 17:11 | |
| clarkb | or replace it | 17:11 |
| *** annegentle has quit IRC | 17:11 | |
| *** Somay has quit IRC | 17:11 | |
| clarkb | don't have multiple methods in place that effect the same change in the system | 17:11 |
| mordred | I'm just saying - there is a string that is repeated 100x - it's a normal shell pattern to put that in a variable | 17:12 |
| *** annegentle has joined #openstack-infra | 17:12 | |
| mordred | so - in general - why would that cause this much conversation even/ | 17:12 |
| mordred | ? | 17:12 |
| jeblair | mordred: yeah, i thought the intent with modules.env was that downstreams would replace it | 17:12 |
| jeblair | i mean, that's what the thinking was when we added it | 17:12 |
| wenlock | jeblair, thats correct | 17:12 |
| yolanda | jeblair, yes, we have our own one | 17:12 |
| wenlock | jeblair we wanted install_modules.sh to be generic so we can use it in other cases. | 17:12 |
| *** Longgeek has joined #openstack-infra | 17:13 | |
| yolanda | but even with that, is easier to be based on yours, just have a different env var for the base path, and add other bits | 17:13 |
| yolanda | it makes code more cleaner as well | 17:13 |
| wenlock | jeblair and sometimes our ci piplines have module version hell that has to be resolved in that env. not upstream | 17:13 |
| yolanda | instead of having 100 repeated git.openstack.org there | 17:13 |
| clarkb | yolanda: you don't need that... | 17:13 |
| clarkb | my example above should mostly work if you replace the name of arra | 17:14 |
| clarkb | and fix the gozer git host | 17:14 |
| clarkb | anyways it has enough +2's to merge I did not -2 | 17:14 |
| jeblair | yolanda: it's true, but we do now have a repeated "$OPENSTACK_GIT_ROOT" | 17:14 |
| clarkb | I am merely pointing out that there is an existing system in place to do this. It would be better if we fixed that if it needs fixing | 17:14 |
| yolanda | well, we have, but it's an env var name, not an url | 17:14 |
| jeblair | so honestly, the change seems like a no-op. i don't think it substantially changes how downstreams consume this | 17:15 |
| jeblair | i could take it or leave it. | 17:15 |
| *** gokrokve_ has quit IRC | 17:15 | |
| clarkb | its sounds like the problem is we made module.env too configurable | 17:16 |
| clarkb | maybe we should consider scaling it back? | 17:16 |
| *** ibiris is now known as ibiris_away | 17:16 | |
| clarkb | wenlock: ^ though you likely need the fullfunctionality there | 17:17 |
| *** dannywilson has joined #openstack-infra | 17:17 | |
| *** sergsh has quit IRC | 17:17 | |
| wenlock | clarkb whats the change id on that? | 17:17 |
| yolanda | clarkb, what functionality do you want to revert? | 17:17 |
| clarkb | wenlock: 163376 | 17:17 |
| clarkb | yolanda: not necessarily revert. The point I am trying to make is that we seem to have decided we need to exteremes of customization here | 17:18 |
| *** harlowja_away is now known as harlowja | 17:18 | |
| clarkb | yolanda: maybe we can improve the system by determining the real needs and have one implementation | 17:18 |
| clarkb | one extereme is customize everything from the ground up arbitrarily | 17:18 |
| yolanda | clarkb, well, from our side, we need that openstack modules, we also need private ones, and different branches | 17:18 |
| clarkb | the other is only customize what openstack-infra is doing | 17:19 |
| yolanda | that's our need | 17:19 |
| wenlock | yolanda, when we have private ones, we just call install_modules.sh and point it to a different modules.env | 17:19 |
| yolanda | but i'd like to know about other use cases | 17:19 |
| wenlock | so we might have a module.private.env for that case | 17:19 |
| yolanda | so wenlock, do you run install_modules.sh several times? | 17:20 |
| wenlock | yolanda we have that often with private pipelines using this same code, that have different module needs | 17:20 |
| wenlock | yolanda, you got it | 17:20 |
| mattt | clarkb: any luck w/ that memcached issue? i've not had any luck landing on a -b4 | 17:20 |
| clarkb | mattt: sorry no, was waiting for job to finish on the node I grabbed, will look at it soon | 17:20 |
| *** tqtran is now known as tqtran_afk | 17:20 | |
| yolanda | wenlock, but i'm normally reusing what upstream has, just switching the git base and the branch | 17:20 |
| mattt | clarkb: no worries, i'm afk for a bit but feel free to ping me if you need anything from me and i'll get it once i'm back at keyboard | 17:20 |
| *** masayukig_ has quit IRC | 17:21 | |
| yolanda | we just have the same, with updated bits and few lines | 17:21 |
| wenlock | yolanda, yes, upstream for that makes sense to simplify it like you had in that patch to me.... | 17:21 |
| yolanda | few extra lines | 17:21 |
| wenlock | not sure i understand that push back to variablize it when modules.env in fact is just another shell script that has customizations | 17:21 |
| pabelanger | modules.env.d ? | 17:21 |
| wenlock | lol | 17:21 |
| *** ghostpl_ has quit IRC | 17:22 | |
| pabelanger | too much? | 17:22 |
| wenlock | pabelanger, your basically right | 17:22 |
| clarkb | pabelanger: lifeless will tell you to only do that if its run parts :P | 17:22 |
| wenlock | but i wasn't ready to jump there | 17:22 |
| *** ibiris_away is now known as ibiris | 17:22 | |
| openstackgerrit | Doug Hellmann proposed openstack-infra/project-config: Add oslo-vmware-stable-maint group https://review.openstack.org/167702 | 17:22 |
| wenlock | simple is always better | 17:22 |
| *** masayukig_ has joined #openstack-infra | 17:23 | |
| *** pal__ has joined #openstack-infra | 17:23 | |
| dhellmann | fungi, anteaya, AJaeger: if there's another way to manage the oslo-vmware-stable-maint group, or if I need to do something else to create it, please let me know ^^ | 17:23 |
| *** Sukhdev has quit IRC | 17:23 | |
| yolanda | wenlock, well, when i see that modules.env i always see that git.openstack.org everywhere so for me it looks much nicer with an env var for the url, and i don't think it adds extra complexity because the env var is defined in same file as well | 17:23 |
| yolanda | but it's a preference | 17:23 |
| jeblair | yolanda: does that change help you in any way? | 17:24 |
| *** dannywilson has quit IRC | 17:24 | |
| yolanda | jeblair, well, if i do a diff from my modules.env to yours, it looks simler | 17:24 |
| yolanda | simpler | 17:24 |
| *** sarob has joined #openstack-infra | 17:24 | |
| jeblair | yolanda: okay, that's good to know | 17:25 |
| yolanda | and i'm regulary following how do we difer from upstream, so that's a reason from my side | 17:25 |
| *** pal_ has quit IRC | 17:25 | |
| clarkb | mrmartin: still around for 166882 did you want to address fungi's comment? | 17:25 |
| clarkb | mrmartin: not sure if you hvae a preference there but thought I would ask before approving the change as is | 17:25 |
| jeblair | yolanda: so the point isn't so much to make the file itself more reconsumable -- it's that you are in fact creating a local version as we intended, but you just want to easily be able to see the difference between your local version and upstream? | 17:25 |
| *** ajmiller has joined #openstack-infra | 17:25 | |
| mrmartin | clarkb checking | 17:26 |
| yolanda | that's one reason, the other is code style preference, i really don't like hardcoded things anywhere | 17:26 |
| yolanda | so i prefer one hardcoded line that 100 | 17:26 |
| *** carl_baldwin has quit IRC | 17:27 | |
| clarkb | mattt: to try and reproduce just do an `apt-get update && apt-get install memcached`? | 17:27 |
| mrmartin | clarkb, I suggest to go with the actual form. | 17:27 |
| jeblair | yolanda: okay. i think that's a borderline case here. but your point about the diff has convinced me this is worth the code churn. | 17:27 |
| clarkb | mrmartin: go with that is currently there? | 17:28 |
| mrmartin | yeap | 17:28 |
| jeblair | yolanda: that might be a good thing to include in the commit message next time | 17:28 |
| clarkb | mrmartin: ok approving now | 17:28 |
| yolanda | jeblair, ok | 17:28 |
| mrmartin | it was tested before | 17:28 |
| yolanda | it was a simple thing so i didn't expect so much discussion :) | 17:28 |
| jeblair | yolanda: so that reviewers can figure out how this makes things better for you | 17:28 |
| *** tiswanso has quit IRC | 17:28 | |
| yolanda | ok, i'll be more detailed in the commit messages for the future, that's a good point | 17:28 |
| jeblair | yolanda: yeah, that's true. i think there are two things that prompted the discussion -- one is the concern that the system may not function like we thought -- that there was a design decision that you were trying to work around | 17:29 |
| *** emagana has quit IRC | 17:29 | |
| clarkb | yolanda: you and nibalizer should figure out that order that 162830 and 166252 should go in | 17:29 |
| clarkb | I think they conflict massively around base.pp :) | 17:30 |
| yolanda | mm, let me check | 17:30 |
| openstackgerrit | Aleksey Ripinen proposed openstack-infra/storyboard-webclient: Support for tags https://review.openstack.org/166847 | 17:30 |
| jeblair | yolanda: and the other is that we have limited review bandwidth, and we want to make sure our time is spent going in the right direction, and we don't want to merge even an apparently simple change if we can't see how it advances what we want to do | 17:30 |
| *** otter768 has joined #openstack-infra | 17:30 | |
| jeblair | yolanda: (i think most of us believe we can make the code prettier once it works a lot better than it does now :) | 17:30 |
| yolanda | clarkb, i don't mind if nibalizer change lands first, i know he's been rebasing several times | 17:31 |
| clarkb | yolanda: ok | 17:31 |
| yolanda | jeblair, yes, things can always be better | 17:31 |
| clarkb | jeblair: yes, in particular I think a lot of the "just make it work" changes historically haven't looked at the larger system as a whole | 17:32 |
| clarkb | jeblair: both from up and down stream | 17:32 |
| *** achanda has joined #openstack-infra | 17:32 | |
| yolanda | jeblair, so as we are really users of the downstream-puppet effort, we can provide feedback on how things affect us and how to improve them | 17:32 |
| *** esker has joined #openstack-infra | 17:32 | |
| clarkb | so if the intent is to move towards better reconsumable bits with clear api boundaries and testing and so on we need to be a bit more critical especially around changes that seem to go against previous design decisisons | 17:32 |
| jeblair | clarkb: i agree; and this is going to be a rough process, especially before we actually have openstackci in place | 17:33 |
| jeblair | and after too, as we try to figure out how to use it :) | 17:33 |
| yolanda | clarkb, one positive thing of isolate things, and make things parameterizable, is that testing is far easier | 17:33 |
| yolanda | so if we have small bits of modules, and you can send parameters, you can test locally, all the options, and be aware of the problems | 17:34 |
| *** carl_baldwin has joined #openstack-infra | 17:34 | |
| *** otter768 has quit IRC | 17:35 | |
| openstackgerrit | Merged openstack-infra/puppet-askbot: Added missing askbot cron jobs https://review.openstack.org/166882 | 17:35 |
| *** ghostpl_ has joined #openstack-infra | 17:35 | |
| *** Somay has joined #openstack-infra | 17:35 | |
| *** emagana has joined #openstack-infra | 17:35 | |
| *** claudiub has quit IRC | 17:35 | |
| pabelanger | since I'm mostly playing catch up, what was the downside of using existing puppet functionality, like the Puppetfile and something like librarian-puppet-simple for module dependencies? | 17:35 |
| clarkb | pabelanger: puppet librarian installs puppet via gem so breaks any puppet installation you might already have | 17:36 |
| pabelanger | that's been my tool of choice for managing puppet dependencies for a while | 17:36 |
| clarkb | pabelanger: its asically the most terrible thing ever | 17:36 |
| clarkb | because it breaks the specific tool it claims to fix | 17:36 |
| pabelanger | clarkb, librarian-puppet or librarian-puppet-simple? | 17:36 |
| pabelanger | I know librarian-puppet is a hog | 17:36 |
| pabelanger | that's what I changed to -simple | 17:37 |
| clarkb | pabelanger: librarian-puppet. I don't think librarian-puppet-simple existed at the time | 17:37 |
| pabelanger | Ya | 17:37 |
| clarkb | pabelanger: so instead we wrote a relatively tiny shell script that just worked | 17:37 |
| openstackgerrit | David Shrewsbury proposed openstack-infra/shade: Custom exception needs str representation https://review.openstack.org/167711 | 17:37 |
| jeblair | yeah, my thought was "there's a -simple" ? | 17:37 |
| jeblair | i'm sure we would have been all over something with simple in the name if it existed :) | 17:37 |
| pabelanger | I too had problems with librarian-puppet, especially when calculating dependencies.... like hours long. | 17:37 |
| *** ajmiller has quit IRC | 17:37 | |
| jeblair | WOW | 17:38 |
| clarkb | pabelanger: I have also looked at r10k but we already have a thing that just works | 17:38 |
| clarkb | pabelanger: also I don't know that librarian (in its many forms) or r10k solve the problem yolanda is trying to address | 17:38 |
| pabelanger | /me nods | 17:38 |
| pabelanger | haven't checked out r10k yet | 17:38 |
| clarkb | since they all have hard coded url paths iirc | 17:38 |
| *** tiswanso has joined #openstack-infra | 17:38 | |
| zaro | jeblair: i've been looking at fixing the shallow clone thing you told me about yester, http://logs.openstack.org/72/151472/9/check/jenkins-job-builder-compare-xml/d359b0a/console.html | 17:39 |
| *** Longgeek has quit IRC | 17:39 | |
| zaro | jeblair: but actually i don't really understand the problem. | 17:39 |
| pabelanger | Cool, since need to consume the modules.env stuff myself. Should be a fun trip | 17:40 |
| zaro | jeblair: don't see why doing a zuul-clone is different then git clone. | 17:40 |
| zaro | jeblair: for the run-compare-xml.sh script | 17:40 |
| *** ivar-lazzaro has joined #openstack-infra | 17:40 | |
| Shrews | mordred: looks like, with a v1 glance api endpoint, uploading a changed image produces a separate image (not updating the current one). This is what you were seeing? | 17:41 |
| Shrews | with shade, that is | 17:41 |
| mordred | Shrews: I'm not sure I tracked it down that far ... what I was seeing is that in the ansible playbook it seemed to have done something | 17:42 |
| *** ivar-lazzaro has quit IRC | 17:42 | |
| mordred | but then the image wasn't updated | 17:42 |
| mordred | I think | 17:42 |
| mordred | Shrews: whatever you said is probably right | 17:42 |
| *** ominakov has quit IRC | 17:43 | |
| openstackgerrit | Merged openstack-infra/system-config: Parameterize url to clone puppet modules https://review.openstack.org/163376 | 17:43 |
| jeblair | zaro: the problem is that review.openstack.org can't handle shallow clones anymore | 17:43 |
| *** achanda has quit IRC | 17:43 | |
| *** ivar-lazzaro has joined #openstack-infra | 17:43 | |
| Shrews | mordred: manual testing with shade shows the behavior i described. can't get the v2 endpoint to work, so not sure what that does | 17:43 |
| *** ajmiller has joined #openstack-infra | 17:43 | |
| jeblair | zaro: you could simply change it from "git clone https://review.openstack.org/p/openstack-infra/project-config --depth 1" to "git clone https://git.openstack.orgopenstack-infra/project-config --depth 1" | 17:43 |
| jeblair | zaro: but it will be even more efficient if you use zuul cloner | 17:43 |
| jeblair | zaro: because it will use the local git cache on disk | 17:44 |
| *** bhuvan has joined #openstack-infra | 17:44 | |
| openstackgerrit | Jordan Pittier proposed openstack-infra/system-config: Third party CI: document how to properly format message with Gerrit Trigger https://review.openstack.org/167716 | 17:44 |
| *** spzala has joined #openstack-infra | 17:44 | |
| wenlock | zaro, specifically it can't handel the shallow clones on the /p/ url | 17:45 |
| jeblair | oh jordanp isn't here? | 17:45 |
| zaro | jeblair: the script is doing this "git clone https://git.openstack.org/openstack-infra/jenkins-job-builder --depth 1", no 'p' there | 17:45 |
| jeblair | that's cool that it looks like there's a way to configure gerrit trigger with the format we want :) | 17:45 |
| jeblair | git clone https://review.openstack.org/p/openstack-infra/project-config --depth 1 | 17:46 |
| jeblair | zaro: ^ | 17:46 |
| clarkb | jeblair: yup, salv-orlando has apparently been doing it for a while now | 17:46 |
| jeblair | zaro: it clones project-config with /p/ | 17:46 |
| jeblair | zaro: but while we're touching it, we should just use zuul-cloner everywhere | 17:46 |
| openstackgerrit | Jordan Pittier proposed openstack-infra/system-config: Third party CI: document how to properly format message with Gerrit Trigger https://review.openstack.org/167716 | 17:46 |
| *** tnovacik has joined #openstack-infra | 17:47 | |
| jeblair | clarkb: neato | 17:47 |
| * zaro still confused. | 17:47 | |
| jeblair | zaro: how can i help? | 17:47 |
| zaro | jeblair: yes, i get it, use zuul-coner but you claim that the its using '/p/'. i don't see any evidence of that | 17:48 |
| zaro | in the run-compare-xml.sh sript. that's wehre it's failing right? | 17:48 |
| jeblair | zaro: http://git.openstack.org/cgit/openstack-infra/jenkins-job-builder/tree/tools/run-compare-xml.sh#n25 | 17:49 |
| zaro | jeblair: ok. i see. i have an old checkout. sorry about that. | 17:50 |
| *** e0ne has quit IRC | 17:50 | |
| *** achanda has joined #openstack-infra | 17:51 | |
| *** Somay has quit IRC | 17:51 | |
| *** ajmiller has quit IRC | 17:52 | |
| *** sushilkm has joined #openstack-infra | 17:52 | |
| *** sushilkm has left #openstack-infra | 17:52 | |
| *** Ala has quit IRC | 17:53 | |
| clarkb | jeblair: mordred 167426 is a good one to look at as far as reconsumable bits, our config, and what they might look like together | 17:53 |
| wenlock | zaro, do you think that is something that will be fixable on gerrit ? | 17:54 |
| clarkb | wenlock: zaro its not gerrit that is breaking, its trusty's version of git | 17:54 |
| wenlock | clarkb but we see that on 12.04 too | 17:54 |
| wenlock | are you saying it's git server side? | 17:55 |
| clarkb | wenlock: yes | 17:55 |
| wenlock | gotcha | 17:55 |
| jeblair | (the local git replicas that we run on the gerrit server) | 17:55 |
| *** derekh has quit IRC | 17:56 | |
| clarkb | jeblair: mordred I think eventually the goal should be to have just a site.pp, but that isn't an immediate thing. 167426 sort of points out some of the gaps around making that leap immediately | 17:56 |
| wenlock | jeblair , clarkb thanks that wasn't clear.... so i assume canonical has some bug sitting against trusty somewhere on that | 17:56 |
| clarkb | *just have a site.pp for the things implemented in openstackci | 17:57 |
| jeblair | wenlock: there's a git upstream patch that is languishing | 17:57 |
| clarkb | mattt: so on this node in hpcloud-b4 memcached was already installed I believe by devstack | 17:57 |
| *** ssam2 has quit IRC | 17:57 | |
| clarkb | mattt: which makes this even more mysterious | 17:57 |
| wenlock | jeblair, cool, ty | 17:57 |
| *** ajmiller has joined #openstack-infra | 17:58 | |
| zaro | jeblair: actually i did have the latest checkout. There are two 'run-compare-xml.sh' files. one in project-config and one in jjb. the one in project-config does not have '/p' and that's the one i was looking at. | 17:59 |
| wenlock | clarkb, jeblair, whats funny is this must be something fairly recent no? is it that a new patch of git on ubuntu in the last day or so broke it? | 18:00 |
| *** gokrokve has joined #openstack-infra | 18:00 | |
| clarkb | wenlock: we just upgraded review.openstack.org to trusty | 18:00 |
| wenlock | oh doh | 18:00 |
| wenlock | cool | 18:00 |
| fungi | okay, back, catching up now | 18:02 |
| *** thingee has quit IRC | 18:02 | |
| *** nikhil_k is now known as nikhil_k|afk | 18:02 | |
| *** crc32 has joined #openstack-infra | 18:03 | |
| *** nikhil_k|afk is now known as nikhil_k|eto | 18:03 | |
| mordred | clarkb: well, it would need to be a site.pp and a modules.env - no? | 18:04 |
| mordred | clarkb: but I agree with the principle of what that repo wants to be | 18:05 |
| clarkb | mordred: yes, you still need the module management to install your dependencies | 18:05 |
| clarkb | mordred: but the point is that system-config should evolve away from anything people reuse since the ideal end point for it in our case is mostly a site.pp | 18:05 |
| clarkb | mordred: it won't happen overnight though | 18:06 |
| *** gokrokve has quit IRC | 18:06 | |
| mordred | nope | 18:06 |
| mordred | one small step ata time | 18:06 |
| *** Hefeweizen has quit IRC | 18:06 | |
| *** stevemar has quit IRC | 18:06 | |
| *** stevemar2 has joined #openstack-infra | 18:06 | |
| clarkb | mordred: your puppet-diskimage_builder change has a -1 that needs to be addressed | 18:07 |
| zaro | ahh looks like the run-compare-xml.sh in project-config is working fine but i'm guessing the git clone in there is should use zuul-cloner instead. | 18:07 |
| clarkb | zaro: out of curiousity how was the orange circle chosen for WIP icon? | 18:07 |
| *** Guest39881 has quit IRC | 18:08 | |
| *** ChuckC has quit IRC | 18:08 | |
| *** james_li has joined #openstack-infra | 18:08 | |
| mordred | clarkb: yes - I was waiting on a dib build :) | 18:08 |
| zaro | clarkb: it's actually called a donut. one of the standard shapes in ms word :) | 18:08 |
| *** carl_baldwin has quit IRC | 18:08 | |
| zaro | clarkb: actual answer is that it was easy | 18:09 |
| openstackgerrit | Monty Taylor proposed openstack-infra/project-config: Add puppet-diskimage_builder module https://review.openstack.org/167607 | 18:09 |
| zaro | and no license repricusions, *i think*. | 18:09 |
| *** marun has quit IRC | 18:09 | |
| *** ChuckC has joined #openstack-infra | 18:09 | |
| *** gokrokve has joined #openstack-infra | 18:09 | |
| *** carl_baldwin has joined #openstack-infra | 18:10 | |
| *** patrickeast has joined #openstack-infra | 18:10 | |
| *** pblaho has quit IRC | 18:10 | |
| *** erlon has joined #openstack-infra | 18:11 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/project-config: Move os-client-config from Stackforge to OpenStack https://review.openstack.org/167213 | 18:11 |
| *** sslypushenko has quit IRC | 18:11 | |
| *** sslypushenko has joined #openstack-infra | 18:12 | |
| mordred | AJaeger, anteaya, jeblair: I have a question about governance/project-config sequencing when it concerns a rename | 18:12 |
| pleia2 | elarson: sorry for the delay, busy morning :) left comments in 165224 about what you need to get those test passing | 18:13 |
| mordred | we have landed the governance patch to make os-client-config part of the openstackclient project | 18:13 |
| mordred | but it's in the governance projects.yaml as stackforge/os-client-config since that's where it is now | 18:13 |
| jeblair | mordred: i think that meets our requirements | 18:13 |
| mordred | clearly we'll need to clean up the governance repo to show it as openstack/os-client-config | 18:13 |
| mordred | do we do hat patch as a patch that depends on the rename patch in project-config? | 18:13 |
| *** arxcruz has quit IRC | 18:14 | |
| jeblair | mordred: we should probably be specific and suggest people write governance patches either one way or the other | 18:14 |
| mordred | yah | 18:14 |
| clarkb | zaro: gotcha, if it came straight from Word it may not be reusable? but you can likely ask someone that hasn't seen the icon yet to make a 128x128 (or whatever the size is ) 2 dimentional orange donut icon and that would be whiteroom enough :) depends on how paranoid you want to be I guess | 18:14 |
| jeblair | mordred: but the main thing is that the TC didn't reject it | 18:14 |
| mordred | totes | 18:14 |
| jeblair | mordred: so yeah, i'd go ahead and write the cleanup patch and depend it on the project-config rename | 18:14 |
| mordred | jeblair: perhaps the governance repo should not mention the git namespace | 18:14 |
| *** pal__ has quit IRC | 18:14 | |
| mordred | jeblair: its project list should just have os-client-config - not {stackforge,openstack}/os-client-config | 18:14 |
| jeblair | mordred: hrm, that might be tricky for automated systems that consume that yaml | 18:15 |
| mordred | jeblair: possibly so :) | 18:15 |
| jeblair | mordred: though in principle i agree with you :) | 18:15 |
| mordred | I mean, we do have a single second level namespace because of jjb | 18:15 |
| mordred | so there would not be name conflicts | 18:15 |
| openstackgerrit | Monty Taylor proposed openstack-infra/project-config: Move os-client-config from Stackforge to OpenStack https://review.openstack.org/167213 | 18:15 |
| *** jaypipes has joined #openstack-infra | 18:16 | |
| zaro | clarkb: no the icon didn't come straight from word. it one of those shape things that you can draw, like a rectangle, circle, line, etc.. so i had to create the donut and put the color in, make it transparent, etc.. | 18:16 |
| openstackgerrit | Ramy Asselin proposed openstack-infra/puppet-openstackci: Create a log server like http://logs.openstack.org https://review.openstack.org/167425 | 18:16 |
| *** melwitt has quit IRC | 18:16 | |
| clarkb | zaro: eprfect | 18:16 |
| jeblair | omg my priority review list is huge | 18:17 |
| clarkb | mattt: even clearing out the cache for memcached package then doing an apt-get install --reinstall memcached is working | 18:17 |
| mordred | ok. | 18:17 |
| mordred | jeblair: done | 18:17 |
| *** esker has quit IRC | 18:18 | |
| *** ddieterly has quit IRC | 18:18 | |
| *** ddieterly has joined #openstack-infra | 18:19 | |
| *** afazekas has quit IRC | 18:19 | |
| openstackgerrit | yolanda.robla proposed openstack-infra/project-config: Add stackforge/puppet-nslcd https://review.openstack.org/165922 | 18:20 |
| *** SumitNaiksatam has joined #openstack-infra | 18:20 | |
| *** ddieterly has quit IRC | 18:20 | |
| *** ddieterly has joined #openstack-infra | 18:20 | |
| *** sarob has quit IRC | 18:21 | |
| fungi | jeblair: at least for the scripts i'm using that reconsume reference/projects.yaml i could work around it fairly trivially. the governance.o.o site publisher might be trickier | 18:21 |
| *** melwitt has joined #openstack-infra | 18:22 | |
| fungi | for my use cases i'm already also pulling info from gerrit at the same time, so it's easy to cross-reference to get the namespace | 18:22 |
| *** notnownikki has quit IRC | 18:26 | |
| *** pelix has quit IRC | 18:27 | |
| *** HeOS has quit IRC | 18:27 | |
| openstackgerrit | Doug Hellmann proposed openstack-infra/project-config: Add stable-compat jobs for some Oslo libs https://review.openstack.org/167734 | 18:28 |
| *** salv-orlando has quit IRC | 18:28 | |
| *** krtaylor has quit IRC | 18:29 | |
| *** dtantsur is now known as dtantsur|afk | 18:29 | |
| *** carl_baldwin has quit IRC | 18:30 | |
| clarkb | yolanda: let me know what you think about m comment on 160885, I think it addresses jheskeths concern in a way common to some of the other changes being made | 18:30 |
| clarkb | yolanda: and in general having a decent way to tune that stuff might eventually end up in our apache module | 18:31 |
| clarkb | yolanda: so I think its win win | 18:31 |
| *** carl_baldwin has joined #openstack-infra | 18:32 | |
| *** e0ne has joined #openstack-infra | 18:33 | |
| *** MarkAtwood has joined #openstack-infra | 18:34 | |
| *** salv-orlando has joined #openstack-infra | 18:35 | |
| *** spzala has quit IRC | 18:35 | |
| *** ddieterly has quit IRC | 18:36 | |
| *** dizquierdo has quit IRC | 18:37 | |
| *** ibiris is now known as ibiris_away | 18:40 | |
| *** craigbr has quit IRC | 18:41 | |
| *** johnthetubaguy is now known as zz_johnthetubagu | 18:42 | |
| *** ddieterly has joined #openstack-infra | 18:43 | |
| *** stevemar2 has quit IRC | 18:44 | |
| *** ddieterly has quit IRC | 18:44 | |
| *** ddieterly has joined #openstack-infra | 18:44 | |
| *** stevemar2 has joined #openstack-infra | 18:45 | |
| *** ibiris_away is now known as ibiris | 18:45 | |
| *** thedodd has quit IRC | 18:48 | |
| dhellmann | fungi, clarkb, jeblair : Is there a way for me to export the history of a module from oslo-incubator into a repo for an existing oslo library like oslo.log? | 18:48 |
| clarkb | dhellmann: you can git filter branch it out of oslo-incubator then push those commits as changes to oslo.log | 18:48 |
| dhellmann | clarkb: is there a way to do it that doesn't involve re-reviewing all of the changes? maybe with a feature branch? | 18:49 |
| *** thedodd has joined #openstack-infra | 18:49 | |
| clarkb | dhellmann: there is also the thing that jelair did for keystone way back when but I think he really doesn't ever want to do that again (but I should let him speak for himself on that) | 18:49 |
| dhellmann | I don't know how many changes there are, but I expect it to be quite a few | 18:49 |
| *** sarob has joined #openstack-infra | 18:49 | |
| jeblair | clarkb, dhellmann: it may be _slightly_ easier if the existing oslo.log shares a commit in the ancestry of oslo-incubator | 18:50 |
| dhellmann | clarkb, jeblair: to be clear, I'm looking for a way for *me* to do this, not for you guys | 18:50 |
| dhellmann | hmm, I don't know if they'll share a commit after the history has been filtered | 18:50 |
| morganfainberg | dhellmann could you do something shady and maek them share a commit by keeping a common file between them? | 18:51 |
| clarkb | dhellmann: the first thing I describe you can do without any help from us | 18:51 |
| morganfainberg | dhellmann, s/shady/clever | 18:51 |
| jeblair | dhellmann: if you want to bypass code review, you'll need one of us; at least to do the actual push | 18:51 |
| *** Sukhdev has joined #openstack-infra | 18:51 | |
| *** krtaylor has joined #openstack-infra | 18:52 | |
| *** craigbr has joined #openstack-infra | 18:52 | |
| dhellmann | jeblair: hmm, ok. I thought a branch didn't necessarily require code reviews? maybe I am conflating some things I've seen though | 18:52 |
| jeblair | dhellmann: but if you want to see if you can set up an (out-of-gerrit) oslo.log repo with a branch that looks the way you want.... | 18:52 |
| dhellmann | if I got everything into a branch and then had to review the merge, that would be fine | 18:52 |
| jeblair | dhellmann: we can push that branch into gerrit (as a non-master branch), and then do a merge commit for that | 18:52 |
| dhellmann | ok, yeah, that's what I was thinking of, but I thought I could do the push to the branch myself. I guess not? | 18:53 |
| jeblair | dhellmann: nah, pushing a whole branch is an admin-only bypassing code review thing | 18:53 |
| dhellmann | ah, ok | 18:53 |
| jeblair | dhellmann: (which i imagine we would be okay doing here since these commits actually _have_ been code reviewed) | 18:53 |
| dhellmann | I can make branches, but not change their content I guess :-) | 18:53 |
| dhellmann | right, that was my reasoning, too | 18:54 |
| jeblair | dhellmann: yeah, and push changes to them to be reviewed of course | 18:54 |
| dhellmann | ok, let me see how many commits we're actually talking about then | 18:54 |
| *** thedodd has quit IRC | 18:54 | |
| jeblair | dhellmann: which, as clarkb says, if you're deaing with a smallish number of changes, is easier | 18:54 |
| dhellmann | hmm, 16 commits | 18:54 |
| *** _nadya_ has quit IRC | 18:55 | |
| dhellmann | I might just ptl-approve those | 18:55 |
| jeblair | dhellmann: yeah, i'd seriously consider it for that number | 18:55 |
| dhellmann | yeah, ok, we'll go that route then | 18:55 |
| jeblair | i was imagining a cutoff would be something like 20-30 | 18:55 |
| dhellmann | thanks, jeblair & clarkb | 18:55 |
| jeblair | dhellmann: no prob | 18:55 |
| * jeblair feeds | 18:56 | |
| clarkb | let us know if you run into an trouble with it, curious to know how it goes as a general "this may solve the problem" solution | 18:56 |
| *** dhritishikhar_ has joined #openstack-infra | 18:56 | |
| dhellmann | clarkb, jeblair : can we set up a branch where tests won't be run? that way we can preserve the old commits intact, and just fix the tests and stuff in the merge commit | 18:57 |
| *** dangers_away is now known as dangers | 18:57 | |
| dhellmann | clarkb: yeah, if this turns into too much of a hassle we'll just make an oslo.versionutils library with one module in it, but that feels kinda silly if we can avoid it | 18:58 |
| *** Rockyg has joined #openstack-infra | 18:59 | |
| *** Bsony has joined #openstack-infra | 18:59 | |
| clarkb | asselin: responded to your comments on the logserver change hopefully that clears up what I meant to say | 19:00 |
| asselin | clarkb, thanks | 19:00 |
| anteaya | mordred: okay here is my problem, where is the patch that ensures the governance repo is the canonical repo for electoral officials to reference at any point in the release cycle? | 19:01 |
| anteaya | mordred: if you create a rename patch that references a merged governance patch, the governance repo is no longer the canonical source | 19:01 |
| clarkb | asselin: basically I mean in your metadata.json file you should list those deps. not puppet require => type deps | 19:01 |
| anteaya | as it is now out of sync with the reality as reflected in git.o.o | 19:01 |
| mordred | anteaya: yes - this is the inherent problem with two sources of truth | 19:02 |
| mordred | anteaya: because if we land the governance repo change first | 19:02 |
| asselin | clarkb, ok..that was just the question I was going to ask | 19:02 |
| mordred | it will also be out of sync with reality | 19:02 |
| anteaya | why do we have two sources of truth | 19:02 |
| anteaya | mordred: the point was to have one source of truth, the governance repo | 19:02 |
| mordred | anteaya: because project-config is the one that drives what's in git.o.o | 19:02 |
| anteaya | right | 19:02 |
| *** thedodd has joined #openstack-infra | 19:02 | |
| mordred | anteaya: there is no sequnece where we can do this where something will not be broken for some period of time | 19:02 |
| *** harlowja is now known as harlowja_away | 19:02 | |
| mordred | it's physically impossible | 19:03 |
| anteaya | sure there is | 19:03 |
| mordred | anteaya: please tell me the sequence | 19:03 |
| anteaya | the governace repo first then the project-config repo | 19:03 |
| *** spzala has joined #openstack-infra | 19:03 | |
| mordred | then you're broken | 19:03 |
| anteaya | why? | 19:03 |
| mordred | because the governance repo references a non-existent location | 19:03 |
| anteaya | what problem are you trying to solve? | 19:03 |
| mordred | because project config hasn't moved it yet | 19:03 |
| mordred | anteaya: project rename | 19:03 |
| fungi | tools consuming the data in the governance repo will look to/point at the wrong repo names | 19:03 |
| anteaya | well I'm trying to solve the election official issue | 19:03 |
| clarkb | greghaynes: were you still planning on updating the nodepool test monkeypatch change today? /me wants to get that one in :) | 19:03 |
| mordred | so, it will be wrong for them too | 19:03 |
| mordred | anteaya: because their list will say "openstack/os-client-config" | 19:04 |
| mordred | and that repo will not exist | 19:04 |
| anteaya | and I have yet to see a governace repo change that wasn't closely followed by action in project config | 19:04 |
| mordred | a rename necessarily has a lag | 19:04 |
| anteaya | until the rename | 19:04 |
| anteaya | right | 19:04 |
| mordred | because we don't land the rename patch until we do the rename | 19:04 |
| fungi | it's less that there are two sources of truth and more that we've asserted some lockstep relationship between git namespace and official project status. if we flattened the entire git hierarchy this would cease to be an issue (lots of things would in fact) | 19:04 |
| *** marun has joined #openstack-infra | 19:04 | |
| anteaya | because we need a gerrit downtime | 19:04 |
| anteaya | why is this a problem today? | 19:04 |
| mordred | anteaya: I am working on a rename patch | 19:04 |
| *** marun has quit IRC | 19:04 | |
| anteaya | I see that | 19:04 |
| mordred | which also needs a governance rename patch | 19:04 |
| anteaya | right | 19:05 |
| mordred | because we accepted the governance patch for stackforge | 19:05 |
| anteaya | why are your needs today greater than the electoral official needs | 19:05 |
| mordred | they are not | 19:05 |
| anteaya | okay good | 19:05 |
| mordred | I'm saying I will break them either way | 19:05 |
| mordred | I will break the election officials | 19:05 |
| mordred | there is no way to not | 19:05 |
| mordred | that is what I'm pointing out | 19:05 |
| anteaya | I am not seeing breakage | 19:05 |
| mordred | if I land the governance change right now | 19:05 |
| mordred | doing the rename | 19:05 |
| mordred | the governance repo will tell the election officials incorrect information | 19:05 |
| mordred | until the rename is performed | 19:06 |
| mordred | that will, by the definition you have described, be breaking them | 19:06 |
| mordred | and I don't want to do that | 19:06 |
| anteaya | do you think that election officials have access to knowing what projects are standing by for a rename window? | 19:06 |
| *** marun has joined #openstack-infra | 19:06 | |
| fungi | either the project name in the governance repo is wrong between when it's renamed in the gerrit maintenance and when it's renamed in the governance repo by approval of the tc, or the other way around. but it's still wrong between the two changes in either order | 19:06 |
| mordred | yes. what fungi said | 19:06 |
| anteaya | as opposed to knowing which projects got changed in project-config and aren't in the governace repo? | 19:06 |
| anteaya | sure | 19:06 |
| anteaya | but there is a list of known projects waiting for a rename window | 19:07 |
| mordred | anteaya: I think when the project is in the governance repo, it's likely the case that election officials will know what's up | 19:07 |
| mordred | as opposed to things not being there in the first palce | 19:07 |
| anteaya | yes | 19:07 |
| anteaya | look it isn't like i get my jollies having to have this in place | 19:07 |
| mordred | anteaya: I'm not saying you do | 19:07 |
| mordred | I'm trying to figure out how to nagivate things appropriately | 19:08 |
| anteaya | but in a world where eveyone is supposed to be able to do whatever they want | 19:08 |
| pleia2 | as long as we document this somewhere so I remember (my brain knows all about renames, but I travel a lot and things fall out when I context switch), I shall not break | 19:08 |
| anteaya | it is getting harder and harder to accomplish requisite tasks | 19:08 |
| fungi | also, i'm still the one generating atc lists/electorate rolls, at least for now, and i keep tabs on this stuff going on | 19:08 |
| anteaya | fungi: well it is all yours then | 19:08 |
| anteaya | fungi: set it up however you like | 19:08 |
| fungi | so can exercise human judgement to make corrections when needed | 19:08 |
| pleia2 | thanks fungi | 19:08 |
| anteaya | I tried and keep getting flak for it | 19:08 |
| pleia2 | anteaya: it's good to know about your experience here, thanks for bringing it up | 19:09 |
| *** hashar has joined #openstack-infra | 19:09 | |
| *** achanda has quit IRC | 19:10 | |
| anteaya | pleia2: sorry I did try to support a manageable system | 19:10 |
| pleia2 | anteaya: knowing the pain points is the first step :) | 19:10 |
| fungi | also, if we generate rolls and people are left out, it's not a crime. it's better avoided, but if someone speaks up and points out that they didn't get a ballot we'll figure out why pretty quickly | 19:11 |
| anteaya | well knowing your electorate is sort of important | 19:11 |
| anteaya | good luck | 19:11 |
| *** ddieterly has quit IRC | 19:11 | |
| *** tsg_ has quit IRC | 19:12 | |
| *** mpavone has joined #openstack-infra | 19:14 | |
| *** ibiris is now known as ibiris_away | 19:15 | |
| *** ddieterly has joined #openstack-infra | 19:16 | |
| *** hdd has quit IRC | 19:16 | |
| *** hdd has joined #openstack-infra | 19:16 | |
| *** tsg has joined #openstack-infra | 19:17 | |
| mordred | fungi: I have proposed a patch to the governance repo to remove the top level namespace from the file | 19:19 |
| *** radez is now known as radez_g0n3 | 19:19 | |
| mordred | fungi: so that there will be no confusion and hopefully life will be easier for everyone | 19:20 |
| mordred | anteaya: ^^ | 19:20 |
| greghaynes | clarkb: ah, yep | 19:20 |
| greghaynes | clarkb: ill do that over lunch | 19:20 |
| greghaynes | or maybe right now | 19:20 |
| *** tkelsey has quit IRC | 19:20 | |
| mordred | greghaynes: lunch. right now. what's the difference? :) | 19:20 |
| greghaynes | those two really should be the same thing, shouldnt they | 19:21 |
| fungi | mordred: cool. it will require some tool adjustments on my part, but i think it's worth the initial pain to implement as long as things like the governance.o.o publication toolchain can cope | 19:21 |
| *** dimtruck is now known as zz_dimtruck | 19:21 | |
| openstackgerrit | Merged openstack-infra/system-config: Remove duplicate _version_ definition from askbot's Chinese solr schema https://review.openstack.org/167345 | 19:22 |
| *** radez_g0n3 is now known as radez | 19:22 | |
| mordred | I read that ^^ and thought it said "solr system" at the end, and i was going to giggle | 19:24 |
| *** Longgeek has joined #openstack-infra | 19:24 | |
| *** thedodd has quit IRC | 19:25 | |
| reed | is this the reason of the failure? 2015-03-25 19:22:34.904 | InsecurePlatformWarning | 19:25 |
| anteaya | mordred: it is up to fungi | 19:25 |
| fungi | reed: that's noise coming from calls into urllib3 by our log uploader | 19:26 |
| fungi | reed: what log are you looking at? | 19:26 |
| reed | maybe this is the error: 2015-03-25 19:22:25.164 | ERROR: InvocationError: could not find executable 'doc8' | 19:26 |
| reed | http://logs.openstack.org/95/132395/11/check/gate-training-guides-tox-slides/e6049a1/console.html | 19:26 |
| fungi | reed: yep, i think so | 19:28 |
| *** sdake__ has joined #openstack-infra | 19:28 | |
| *** otter768 has joined #openstack-infra | 19:28 | |
| * reed dives | 19:28 | |
| fungi | reed: looks like you should be able to reproduce by running `tox -v -e slides` | 19:29 |
| *** emagana has quit IRC | 19:29 | |
| *** Longgeek has quit IRC | 19:29 | |
| reed | it's happening outside of my dir | 19:29 |
| reed | I'm starting to think that it was a mistake to merge upstream training with training guides :) | 19:30 |
| nibalizer | jesusaurus: what do you think of https://review.openstack.org/#/c/167346/ | 19:30 |
| nibalizer | is the added complexity worh the clarity? | 19:31 |
| *** sdake has quit IRC | 19:32 | |
| mordred | fungi: nope. doc generation breaks | 19:32 |
| mordred | fungi: and all of teh solutions to that are bad | 19:32 |
| fungi | mordred: i suspected as much | 19:33 |
| clarkb | nibalizer: I think you can do that simpler, just add an echo | 19:33 |
| mordred | fungi: oh well | 19:33 |
| *** thedodd has joined #openstack-infra | 19:33 | |
| clarkb | nibalizer: maybe xargs has a print option too | 19:33 |
| fungi | mordred: it would almost certainly need to cross-reference another repo or a live system to do the necessary lookup | 19:33 |
| fungi | mordred: OR we set up redirects on the git servers ;) | 19:34 |
| jesusaurus | nibalizer: yeah, i think its great to know which nodes the warnings/errors are coming from | 19:34 |
| mordred | fungi: it would need to be live - because the docs aren't going to get re-built | 19:34 |
| mordred | after a gerrit move | 19:34 |
| fungi | yep | 19:34 |
| pleia2 | cinerama: any objections to merging 147947 now and continuing on fixups as clarkb suggests? (I'm inclined to agree, massive patch hanging out there is hard to review) | 19:34 |
| clarkb | oh wait you can't && in the xargs command :/ | 19:34 |
| openstackgerrit | greghaynes proposed openstack-infra/nodepool: Monkeypatch Fake Clients for tests https://review.openstack.org/165682 | 19:34 |
| mordred | so there would really need to be a live redirector that queried the gerrit API with a project name, did the hash mapping and then redirected you to the right place | 19:34 |
| cinerama | pleia2: absolutely not. i was hoping for that about 5 patchsets ago tbh | 19:35 |
| mordred | clarkb: once you get to && in your xargs, it's time for a for loop | 19:35 |
| pleia2 | cinerama: fair enough, hitting the +A button then :) | 19:35 |
| fungi | mordred: yeah. if we flattened all our git namespaces then the solution would be easier, but the getting there would be much, much more work | 19:35 |
| nibalizer | clarkb: yea i actually banged on it quite a bit | 19:35 |
| cinerama | pleia2: thank you | 19:35 |
| mordred | fungi: so much more work | 19:35 |
| reed | fungi, now I see what's wrong: when I rebased I have overlooked the fact that someone added a test-requirement.txt to tox.ini, not what I have in my patch requirements.txt | 19:35 |
| clarkb | mordred: maybe, for loops don't solve a lot of problems that xargs solves in this case. namely output collection in parallel testing | 19:35 |
| nibalizer | i think that is the simplest i can get it to work | 19:35 |
| mordred | clarkb: hrm. good point | 19:35 |
| *** tonytan4ever has quit IRC | 19:35 | |
| nibalizer | but it ended up being kidnof -_- complex | 19:35 |
| mordred | clarkb: it does not, in fact, do that | 19:35 |
| openstackgerrit | Merged openstack-infra/project-config: Add oslo-vmware-stable-maint group https://review.openstack.org/167702 | 19:36 |
| fungi | reed: that would explain why it's complaining about a missing tool | 19:36 |
| mordred | clarkb: maybe write it in go? ;) | 19:36 |
| nibalizer | clarkb: mordred i actaully did sortof rewrite the output section | 19:36 |
| *** Rockyg has quit IRC | 19:36 | |
| *** zz_dimtruck is now known as dimtruck | 19:36 | |
| nibalizer | we could dump log-per-node in logs/* then have devstack-logs (or similar) run to collect them | 19:36 |
| nibalizer | but then the admin would have to paw through finding whre the error was | 19:37 |
| clarkb | nibalizer: ya I am about to -1 and tell you to not collect the logs that way | 19:37 |
| clarkb | nibalizer: xargs does it for you | 19:37 |
| nibalizer | clarkb: nope doesn't work right | 19:37 |
| *** rlandy has joined #openstack-infra | 19:37 | |
| clarkb | nibalizer: are you sure? that was one of the things that everone thought would break in that test but it worked | 19:37 |
| nibalizer | it splits the node deff from the puppet run | 19:37 |
| nibalizer | so you get 10 puppet node defs then 10 puppet runs | 19:38 |
| nibalizer | you can see a patchset where I tried | 19:38 |
| sdague | so, what's up with nodepool and centos7 images? It's putting an extra 3+ hours delay into our devstack patches | 19:38 |
| sdague | which is somewhat problematic for the sprint | 19:38 |
| clarkb | nibalizer: looking | 19:38 |
| *** reazem has joined #openstack-infra | 19:38 | |
| *** harlowja_away is now known as harlowja | 19:38 | |
| *** emagana has joined #openstack-infra | 19:39 | |
| *** thedodd has quit IRC | 19:39 | |
| zxiiro | Is nodepoold typically run as root? I tried running it as a nodepool user and it exits immediately with no output, running it as "nodepoold -d" works though. | 19:40 |
| clarkb | nibalizer: I think thats a buffering thing | 19:40 |
| *** mrmartin has quit IRC | 19:40 | |
| clarkb | zxiiro: no it is not typically run as root, we run it as a 'nodepool' user which does have sudo privs to build disk images with diskimage-builder | 19:40 |
| clarkb | sdague: let me see | 19:40 |
| zxiiro | hmm ok i wonder why in my case it quites immediately then. Guess i'll keep poking at it | 19:41 |
| *** melwitt has quit IRC | 19:41 | |
| clarkb | zxiiro: it will daemonize itself, so you will need to make sure it is happy doing that | 19:41 |
| *** mikeit has joined #openstack-infra | 19:42 | |
| zxiiro | yeah i suspect permissions related somewhere | 19:42 |
| *** sarob has quit IRC | 19:42 | |
| nibalizer | clarkb: so i dont know hwo to get the jenkins logs for patchset9 but that is where the problem is | 19:42 |
| clarkb | nibalizer: http://logs.openstack.org/46/167346/9/check/gate-infra-puppet-apply-trusty/f4281b4/ | 19:42 |
| clarkb | nibalizer: I think the cat is flushing stdout so you get the cat output then the puppet apply output | 19:43 |
| *** annegentle has quit IRC | 19:43 | |
| clarkb | nibalizer: oh also its stdout vs stderr, maybe you need to be writing to the same buffer | 19:43 |
| nibalizer | clarkb: http://logs.openstack.org/46/167346/9/check/gate-infra-puppet-apply-trusty/f4281b4/console.html#_2015-03-25_03_23_04_449 | 19:43 |
| lifeless | morning | 19:43 |
| clarkb | nibalizer: maybe try 1>&2 in patchset 9's cat | 19:43 |
| samueldmq | anteaya, trying 'Running your own CI infrastructure' (http://ci.openstack.org/running-your-own.html) could be a good idea to start ? | 19:44 |
| nibalizer | ok | 19:44 |
| clarkb | nibalizer: ya I am pretty sure its because one set is using stdout and the other set is using stderr | 19:44 |
| samueldmq | anteaya, I am concerned because there is a warning saying: 'Parts of this file are out of date' | 19:44 |
| lifeless | clarkb: pabelanger damn straight | 19:44 |
| fungi | samueldmq: yeah, it needs someone to work out what's out of date in it and propose updates | 19:44 |
| lifeless | clarkb: actually its more .d implies 'processses all' | 19:44 |
| lifeless | clarkb: I don't care about the underlying iterator | 19:45 |
| anteaya | samueldmq: yes, umm, running your own ci is a lot of work | 19:45 |
| openstackgerrit | Merged openstack-infra/puppet-zanata: Initial commit of Zanata-specific sections https://review.openstack.org/147947 | 19:45 |
| anteaya | samueldmq: and an admirable goal | 19:45 |
| fungi | samueldmq: which can be an interesting, if at time potentially frustrating, way to get familiar | 19:45 |
| openstackgerrit | Spencer Krum proposed openstack-infra/system-config: Print the node definition before apply test https://review.openstack.org/167346 | 19:45 |
| * nibalizer -> lunch | 19:45 | |
| *** imcsk8|afk is now known as imcsk8 | 19:45 | |
| anteaya | samueldmq: however it might be easier to find another way to contribute | 19:45 |
| anteaya | samueldmq: at least to understand how our files are organized | 19:45 |
| anteaya | samueldmq: I understand it is tough to find a way in | 19:46 |
| anteaya | samueldmq: there really is no easy way | 19:46 |
| clarkb | sdague: Deficit: devstack-centos7: 0 (start: 0 min-ready: 1 ready: 1 capacity: 187) | 19:46 |
| clarkb | sdague: so nodepool thinks it has met demand /me goes to look at zuul | 19:46 |
| anteaya | samueldmq: I think yesterday fungi suggested you might start by reviewing some documentation patches? | 19:46 |
| samueldmq | fungi, anteaya yes I agree it is interesting | 19:46 |
| sdague | clarkb: ok, I pushed a patch 3.5 hrs ago | 19:46 |
| sdague | it's still running the centos job | 19:46 |
| sdague | it took 3hrs to get started on that | 19:46 |
| samueldmq | fungi, anteaya I think I will try it out .. and see what happens :) | 19:47 |
| *** sarob has joined #openstack-infra | 19:47 | |
| anteaya | samueldmq: as you see fit | 19:47 |
| sdague | https://review.openstack.org/#/c/167669 | 19:47 |
| openstackgerrit | Paul Belanger proposed openstack-infra/nodepool: Correct diskimage-builder name usage https://review.openstack.org/167771 | 19:47 |
| fungi | samueldmq: have at it! let us know when you have questions | 19:47 |
| *** melwitt has joined #openstack-infra | 19:48 | |
| fungi | samueldmq: we may not always have time to answer questions, but we often at least have time to point you at more updated bits of documentation or source code which might answer it for you | 19:48 |
| samueldmq | anteaya, fungi k thanks :) will try and see if something needs an update | 19:48 |
| samueldmq | fungi, ack | 19:48 |
| pabelanger | samueldmq, it is a great way to learn -infra too! | 19:49 |
| *** hdd has quit IRC | 19:51 | |
| *** sarob has quit IRC | 19:51 | |
| *** hdd has joined #openstack-infra | 19:51 | |
| *** mtanino has quit IRC | 19:51 | |
| *** erikmwilson has quit IRC | 19:52 | |
| clarkb | sdague: it looks like we had almost no capacity for devstack-centos7 node when your change was pushed http://paste.openstack.org/show/196730/ but we did have capacity and despite that the deficit calculation never went above 3 during that time. I wonder if we have actually run that test against your change >1 time | 19:53 |
| sdague | jeblair's patch from yesterday is still in the check queue for similar reasons. | 19:53 |
| *** radez is now known as radez_g0n3 | 19:53 | |
| sdague | If this is going to remain a delay issue, I'd really like to get this out of our check queue again | 19:53 |
| sdague | especially for this week | 19:54 |
| sdague | as the point of putting all the +2ers in a room was that we could move fast on landing things | 19:54 |
| clarkb | sdague: yup thats it | 19:54 |
| sdague | clarkb: why did it get run multiple times? | 19:54 |
| clarkb | http://logs.openstack.org/69/167669/1/check/check-tempest-dsvm-centos7/74fd092/console.html http://logs.openstack.org/69/167669/1/check/check-tempest-dsvm-centos7/893160d/console.html http://logs.openstack.org/69/167669/1/check/check-tempest-dsvm-centos7/af86530/console.html | 19:55 |
| *** mtanino has joined #openstack-infra | 19:55 | |
| clarkb | sdague: because the test is exploding in a way that normally indicates a failure that zuul should retry | 19:55 |
| fungi | clarkb: sdague: afaik nobody has yet figured out why tempest runs on centos7 oom and disconnect the worker from jenkins/internet | 19:55 |
| clarkb | sdague: "FATAL: hudson.remoting.RequestAbortedException: java.io.IOException: Unexpected termination of the channel" | 19:56 |
| sdague | fungi: ok, then I think that needs to go back to experimental | 19:56 |
| samueldmq | pabelanger, yeah I agree it is :-) | 19:56 |
| clarkb | fungi: oh right the issue glusterfs was running into | 19:56 |
| fungi | this was the same issue which caused the glusterfs devs to switch their job to ubuntu | 19:56 |
| dhellmann | fungi, jeblair : could one of you add me to https://review.openstack.org/#/admin/groups/648,members please so I can populate the rest of its members? | 19:56 |
| fungi | dhellmann: doing now | 19:56 |
| dhellmann | fungi: cool, thanks (I just noticed that merge, and then the discussion went in a different direction...) | 19:56 |
| *** rlandy has quit IRC | 19:57 | |
| fungi | dhellmann: done | 19:57 |
| *** Rockyg has joined #openstack-infra | 19:57 | |
| openstackgerrit | Sean Dague proposed openstack-infra/project-config: move centos7 back to experimental https://review.openstack.org/167773 | 19:57 |
| * clarkb goes looking for where it was added so we can revert it | 19:57 | |
| dhellmann | fungi: thanks! | 19:57 |
| sdague | clarkb: oh, I can find that | 19:57 |
| clarkb | sdague: I don't think we want it in experimental either as it will cause the same problem there | 19:57 |
| cody-somerville | Isn't "java.io.IOException: Unexpected termination of the channel" like a network issue? | 19:58 |
| cody-somerville | We see it ever now and again too. | 19:58 |
| fungi | right, if someone runs check experimental, they'll never get results | 19:58 |
| sdague | fungi: ok | 19:58 |
| sdague | so then we just delete it? | 19:58 |
| jeblair | cody-somerville: are you seeing that on changes that report back to gerrit? | 19:58 |
| jeblair | cody-somerville: or rather, jobs that report back to gerrit | 19:59 |
| clarkb | cody-somerville: yes usually it means the jenkins master and slave lost connectivity for transient reasons so zuul reruns. In this case its a OOM specific to centos7 that happens consistently | 19:59 |
| sdague | experimental will at least let patche flow forward | 19:59 |
| fungi | cody-somerville: often yes. in this case the worker becomes completely unresponsive at the console and unreachable over the network for a while, and eventually the kernel oom killer terminates enough things that it becomes reachable again | 19:59 |
| nibalizer | looking at java logs... i've seen things... no one should ever see.... | 19:59 |
| cody-somerville | ah, interesting. | 19:59 |
| jeblair | oh, oom during run, missed that | 19:59 |
| clarkb | I71e08fb511f7f524c7d72a9f308dbe3890b620a4 thats the change I think | 19:59 |
| *** wenlock has quit IRC | 19:59 | |
| clarkb | sdague: ^ | 19:59 |
| *** MarkAtwood has quit IRC | 20:00 | |
| sdague | clarkb: ok, I'll do the revert there | 20:00 |
| *** pblaho has joined #openstack-infra | 20:01 | |
| fungi | if you hold one of the workers running the job (or manually run the same commands the job does on one) you lose contact with it late in the tempest tests, then when it becomes reachable again the kernel log often mentions the oom killer taking out mysqld, apache, et cetera | 20:01 |
| fungi | i didn'y analyze the process dumps to try and figure out what's sucking up all the available memory | 20:01 |
| *** davide has quit IRC | 20:02 | |
| fungi | figured once i confirmed it was reproducible someone else with a vested interest in the platform would work out the details | 20:02 |
| openstackgerrit | Sean Dague proposed openstack-infra/project-config: Revert "Promote devstack centos7 to non-voting check job" https://review.openstack.org/167775 | 20:02 |
| sdague | ok, there is the propper revert | 20:02 |
| clarkb | sdague: change needs a rebase? | 20:03 |
| clarkb | oh the parent is the change it reverts | 20:03 |
| fungi | yep | 20:03 |
| fungi | clarkb: sdague: i think where it seemed stable was that the problem so far has only been observed in rax, so if you get lucky and it runs in hpcloud a few times in a row it seems like it's just fine | 20:04 |
| clarkb | I am going to manually kill the running jobs now, to get those changes to move while we get the revert in | 20:04 |
| *** sabeen3 has joined #openstack-infra | 20:05 | |
| fungi | and yeah, those three current examples are all in rax-iad specifically | 20:05 |
| fungi | which supports that theory | 20:05 |
| openstackgerrit | Sean Dague proposed openstack-infra/project-config: Revert "Promote devstack centos7 to non-voting check job" https://review.openstack.org/167775 | 20:06 |
| sdague | ok, the revert is rebased | 20:06 |
| *** Bsony has quit IRC | 20:06 | |
| *** carl_baldwin has quit IRC | 20:06 | |
| mattt | clarkb: that is super odd | 20:07 |
| *** sabeen1 has quit IRC | 20:07 | |
| *** tkelsey has joined #openstack-infra | 20:09 | |
| openstackgerrit | Joe Gordon proposed openstack-infra/project-config: Add gate-tempest-dsvm-sahara https://review.openstack.org/167779 | 20:10 |
| openstackgerrit | Joe Gordon proposed openstack-infra/project-config: Add sahara tempest job to sahara and tempest https://review.openstack.org/167780 | 20:10 |
| *** erikmwilson has joined #openstack-infra | 20:10 | |
| *** carl_baldwin has joined #openstack-infra | 20:11 | |
| *** annegentle has joined #openstack-infra | 20:12 | |
| *** tkelsey has quit IRC | 20:13 | |
| *** stevemar2 has quit IRC | 20:15 | |
| *** sarob has joined #openstack-infra | 20:15 | |
| *** tqtran_afk is now known as tqtran | 20:16 | |
| arosen | mordred: quick question you said that nodepool can output images as raw? Do you know where this configuration setting needs to be done? | 20:16 |
| jogo | jeblair: ^ will wait for SergeyLukjanov on those two patches before going further | 20:16 |
| jogo | jeblair: I added you as a co-author too since I cut and paste your commit message | 20:17 |
| *** rlandy has joined #openstack-infra | 20:18 | |
| *** tnovacik has quit IRC | 20:19 | |
| *** dboik_ has joined #openstack-infra | 20:20 | |
| *** sarob has quit IRC | 20:20 | |
| *** emagana has quit IRC | 20:21 | |
| *** sarob has joined #openstack-infra | 20:22 | |
| *** MarkAtwood has joined #openstack-infra | 20:22 | |
| krotscheck | Anyone know off the top of their head what the openstackid base URL is for openID? | 20:22 |
| arosen | mordred: it looks like disk-image-create only suports qcow2 or tar from what i can see. | 20:23 |
| arosen | -t qcow2,tar -- set the image types of the output image files (default qcow2) | 20:23 |
| arosen | File types should be comma separated | 20:23 |
| jeblair | krotscheck: no but i think you could look in system-config for the config of groups.o.o if you need to dig for it | 20:23 |
| jeblair | krotscheck: (also, apparently it has documentation in progress for that sort of thing) | 20:24 |
| fungi | there's also https://review.openstack.org/165199 which might provide a clue | 20:24 |
| fungi | yeah, that | 20:24 |
| clarkb | arosen thats just an example | 20:24 |
| *** dboik has quit IRC | 20:24 | |
| clarkb | not an exhaustive list | 20:24 |
| krotscheck | hrm. | 20:24 |
| krotscheck | Ok, seems like it uses rds. | 20:25 |
| *** dannywilson has joined #openstack-infra | 20:25 | |
| *** james_li has quit IRC | 20:25 | |
| arosen | clarkb: do i need to edit this file https://github.com/openstack-infra/nodepool/blob/master/nodepool/nodepool.py#L778 and add -t raw then or is there a way to do it via a config? | 20:25 |
| clarkb | arosen you can set the type in nodepool | 20:26 |
| arosen | clarkb: where is this set? | 20:26 |
| *** tonytan4ever has joined #openstack-infra | 20:29 | |
| *** dannywilson has quit IRC | 20:29 | |
| *** gyee has quit IRC | 20:30 | |
| greghaynes | hrm, we should document that image type feature | 20:31 |
| greghaynes | in dib we basically pass types through to qemu-img convert so anything that it supports as a target dib supports | 20:32 |
| *** baoli has quit IRC | 20:32 | |
| *** dkliban is now known as dkliban_brb | 20:35 | |
| *** carl_baldwin has quit IRC | 20:36 | |
| *** yfried is now known as yfried|afk | 20:36 | |
| *** med_ has quit IRC | 20:36 | |
| clarkb | arosen http://ci.openstack.org/nodepool/configuration.html#diskimages qemu-img-options? I would have to go look at code to confirm | 20:39 |
| *** achanda has joined #openstack-infra | 20:39 | |
| *** gokrokve has quit IRC | 20:39 | |
| *** gokrokve has joined #openstack-infra | 20:40 | |
| *** annegentle has quit IRC | 20:40 | |
| *** bhuvan has quit IRC | 20:41 | |
| *** bhuvan has joined #openstack-infra | 20:41 | |
| arosen | clarkb: i'll try setting: qemu-img-options: -f raw | 20:42 |
| *** mrmartin has joined #openstack-infra | 20:42 | |
| *** wenlock has joined #openstack-infra | 20:45 | |
| *** gokrokve has quit IRC | 20:47 | |
| *** gokrokve has joined #openstack-infra | 20:48 | |
| *** dboik_ has quit IRC | 20:48 | |
| *** dboik has joined #openstack-infra | 20:49 | |
| *** annegentle has joined #openstack-infra | 20:50 | |
| *** mjturek1 has quit IRC | 20:51 | |
| *** Rockyg has quit IRC | 20:51 | |
| *** Sukhdev has quit IRC | 20:51 | |
| sigmavirus24 | clarkb: do you need anything else from #openstack-ansible to debug that hpcloud-b4 issue mattt and I are seeing? | 20:52 |
| clarkb | sigmavirus24: yes, I have nothing else to go on. hpcloud-b4 devstack-trusty node is perfectly capable of manually installing memcached | 20:53 |
| clarkb | I don't know what I need but we need something more | 20:53 |
| openstackgerrit | Merged openstack-infra/project-config: Revert "Promote devstack centos7 to non-voting check job" https://review.openstack.org/167775 | 20:53 |
| openstackgerrit | Joe Gordon proposed openstack-infra/project-config: Add gate-tempest-dsvm-sahara https://review.openstack.org/167779 | 20:53 |
| openstackgerrit | Joe Gordon proposed openstack-infra/project-config: Add sahara tempest job to sahara and tempest https://review.openstack.org/167780 | 20:53 |
| *** thingee has joined #openstack-infra | 20:53 | |
| *** alexpilotti has joined #openstack-infra | 20:54 | |
| openstackgerrit | Michael Krotscheck proposed openstack-infra/storyboard: Removed Nickname https://review.openstack.org/165194 | 20:54 |
| openstackgerrit | Michael Krotscheck proposed openstack-infra/storyboard: WIP- Switched StoryBoard to openstackid.org https://review.openstack.org/167792 | 20:54 |
| *** sdake has joined #openstack-infra | 20:55 | |
| *** sdake has quit IRC | 20:55 | |
| krotscheck | mrmartin: Hey, if you get a chance, can you help me out with this WIP patch? Only the SREG email is coming back, and none of the ax. parameters. | 20:55 |
| *** sdake has joined #openstack-infra | 20:55 | |
| mrmartin | hi krotscheck | 20:55 |
| *** erikmwilson has quit IRC | 20:55 | |
| *** erikmwilson_ is now known as erikmwilson | 20:55 | |
| mrmartin | krotscheck, do you have a local openstackid somewhere? | 20:56 |
| *** carl_baldwin has joined #openstack-infra | 20:56 | |
| sigmavirus24 | clarkb: I'll see if I can get more details | 20:56 |
| *** yfried|afk is now known as yfried | 20:56 | |
| sigmavirus24 | clarkb: oh we're doing it inside of lxc if you think that would make a difference | 20:56 |
| *** jgrimm is now known as zz_jgrimm | 20:56 | |
| fungi | mrmartin: oh, you're back. i've reindexed chinese on the new server if you want to test again | 20:57 |
| mrmartin | fungi: thnx. | 20:57 |
| mrmartin | I'll look that tomorrow | 20:57 |
| clarkb | sigmavirus24: it may affect your ubuntu mirror and incrase the chances of network failures to that location | 20:57 |
| *** mrda-away is now known as mrda | 20:57 | |
| *** tnovacik has joined #openstack-infra | 20:58 | |
| *** sdake__ has quit IRC | 20:58 | |
| fungi | mrmartin: also not urgent, but when i update puppet on the server i get: Error: /Stage[main]/Openstack_project::Ask/File[/usr/share/solr/WEB-INF/lib/lucene-analyzers-smartcn-4.7.2.jar]: Could not evaluate: Could not retrieve information from environment production source(s) file:/tmp/solr-4.7.2/contrib/analysis-extras/lucene-libs/lucene-analyzers-smartcn-4.7.2.jar | 20:58 |
| mrmartin | strange | 20:59 |
| fungi | probably a download failure or something, i haven't had time to look into it... but i got it twice in a row so it doesn't seem random | 21:00 |
| *** e0ne has quit IRC | 21:00 | |
| *** kgiusti has left #openstack-infra | 21:01 | |
| mrmartin | does the /tmp/solr-4.7.2 exists? | 21:01 |
| *** hashar has quit IRC | 21:02 | |
| fungi | mrmartin: nope | 21:02 |
| mrmartin | it is extracting this file, because that required for the Chinese indexing, it is a spec. analyzer | 21:02 |
| mrmartin | ok, I recorded that, will check tomorrow | 21:02 |
| fungi | mrmartin: thanks. have a good evening | 21:03 |
| *** Bsony has joined #openstack-infra | 21:03 | |
| *** marun has quit IRC | 21:03 | |
| *** marun has joined #openstack-infra | 21:04 | |
| *** dkliban_brb is now known as dkliban | 21:04 | |
| *** tiswanso has quit IRC | 21:05 | |
| *** yfried is now known as yfried|afk | 21:06 | |
| *** pc_m has quit IRC | 21:06 | |
| *** mattfarina has quit IRC | 21:07 | |
| fungi | clarkb: oh, yesterday i updated https://review.openstack.org/165660 based on your suggestion | 21:07 |
| *** ociuhandu has quit IRC | 21:10 | |
| *** dprince has quit IRC | 21:12 | |
| openstackgerrit | Jeremy Stanley proposed openstack-infra/infra-manual: Dynamically update copyright year at publication https://review.openstack.org/167795 | 21:12 |
| *** davide has joined #openstack-infra | 21:13 | |
| *** rlandy has quit IRC | 21:13 | |
| *** Longgeek has joined #openstack-infra | 21:13 | |
| *** julim has quit IRC | 21:14 | |
| *** wenlock has quit IRC | 21:14 | |
| *** dkranz has quit IRC | 21:15 | |
| *** ldnunes has quit IRC | 21:15 | |
| *** wenlock has joined #openstack-infra | 21:17 | |
| *** mrmartin has quit IRC | 21:18 | |
| *** ivar-laz_ has joined #openstack-infra | 21:19 | |
| openstackgerrit | Thai Tran proposed openstack-infra/project-config: Create a repository for stackforge/xstatic-angular-lrdragndrop https://review.openstack.org/167798 | 21:21 |
| *** achanda has quit IRC | 21:22 | |
| *** mjturek1 has joined #openstack-infra | 21:22 | |
| *** ivar-lazzaro has quit IRC | 21:22 | |
| *** Ryan_Lane has joined #openstack-infra | 21:23 | |
| reed | damn, i am going nuts trying to fix something that is broken upstream | 21:24 |
| *** Ryan_Lane has quit IRC | 21:24 | |
| reed | and when I say upstream I mean in master | 21:24 |
| krotscheck | mrmartin: Sorry, was meeting gothicmindfood to cowork. | 21:24 |
| krotscheck | mrmartin: Will ping you in #storyboard | 21:24 |
| *** rfolco has quit IRC | 21:24 | |
| openstackgerrit | Jeremy Stanley proposed openstack-infra/infra-manual: Mention license in document footer https://review.openstack.org/167799 | 21:24 |
| fungi | reed: any details? | 21:24 |
| *** Longgeek has quit IRC | 21:25 | |
| *** achanda has joined #openstack-infra | 21:25 | |
| *** dustins has quit IRC | 21:25 | |
| reed | fungi, not your fault and such a minor thing but https://review.openstack.org/#/c/142259/ proves that this slides thing is broken everywhere | 21:25 |
| fungi | reed: sure, mostly just wondering if i can help figure it out | 21:25 |
| reed | 2015-03-02 16:36:13.836 | ERROR: InvocationError: '/home/jenkins/workspace/gate-training-guides-tox-slides/.tox/slides/bin/doc8 doc' | 21:26 |
| fungi | reed: did they maybe add a job which was already broken and never worked? | 21:26 |
| *** ddieterly has quit IRC | 21:26 | |
| reed | no, it's tox.ini that is wrong in master | 21:26 |
| *** sarob has quit IRC | 21:26 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/puppet-nodepool: Install additional depends for dib https://review.openstack.org/167625 | 21:27 |
| openstackgerrit | Monty Taylor proposed openstack-infra/puppet-nodepool: Consume puppet-diskimage-builder https://review.openstack.org/167610 | 21:27 |
| jeblair | krotscheck: the plan is (perhaps was?) to eventually move storyboard to openstackid -- however, i think it's important that our core set of tools not change haphazardly. so we plan on switching gerrit, wiki, and $bugtracker together. | 21:27 |
| krotscheck | jeblair: That's fair. What's the schedule on that. | 21:28 |
| reed | jeblair, probably the wiki is safer to move by itself, earlier... but like with Ask OpenStack, the migration of existing users is still to be defined | 21:28 |
| *** ivar-laz_ has quit IRC | 21:28 | |
| reed | since the wiki is less crucial than gerrit | 21:28 |
| *** ivar-lazzaro has joined #openstack-infra | 21:28 | |
| jeblair | krotscheck: replace bugtracker then move bugtracker and gerrit together | 21:28 |
| jeblair | krotscheck: (since launchpad has no other authn options, replacing it is a pre-req) | 21:29 |
| *** rlandy has joined #openstack-infra | 21:29 | |
| krotscheck | jeblair: So by 'bugtracker' to be replaced is launchpad. | 21:29 |
| krotscheck | kk | 21:29 |
| fungi | reed: i see the issue... they're installing doc8 into a virtualenv, then trying to pipe output through doc8 in the normal executable path | 21:30 |
| *** esker has joined #openstack-infra | 21:30 | |
| fungi | reed: just wondering how it got that way, other than adding an already-broken job to gate on it | 21:30 |
| *** Ryan_Lane has joined #openstack-infra | 21:30 | |
| jeblair | krotscheck: yeah, sorry that was unclear. replace launchpad-as-a-bugtracker with storyboard, then move storyboard+gerrit+others to use openstackid. | 21:30 |
| jeblair | krotscheck: (except "storyboard" is now a variable) | 21:31 |
| krotscheck | jeblair: Right, I figured as much. | 21:31 |
| jeblair | reed: that's true, but i also want to avoid developers using multiple single-sign-on systems (which somewhat defeats the point) | 21:32 |
| sigmavirus24 | clarkb: what mirror did you use to install memcached? We just had another failure on b4 | 21:32 |
| openstackgerrit | Joe Gordon proposed openstack-infra/project-config: Add tempest-dsvm-sahara https://review.openstack.org/167779 | 21:33 |
| openstackgerrit | Joe Gordon proposed openstack-infra/project-config: Add sahara tempest job to sahara and tempest https://review.openstack.org/167780 | 21:33 |
| reed | fungi, putting " doc8 doc/training-guides" in tox.ini instead of " doc8 doc" as before, seems to pass on my machine | 21:34 |
| openstackgerrit | Joe Gordon proposed openstack-infra/project-config: Add check-tempest-dsvm-heat https://review.openstack.org/167802 | 21:34 |
| openstackgerrit | Joe Gordon proposed openstack-infra/project-config: Add heat tempest job to heat and tempest https://review.openstack.org/167803 | 21:34 |
| fungi | reed: yeah it looks like it is finding the actual doc8 according to the log | 21:35 |
| greghaynes | clarkb: https://review.openstack.org/#/c/167804/ | 21:35 |
| fungi | reed: actually, running it on a machine i have handy, it's passing for me. that error doesn't occur | 21:36 |
| reed | oh, that's exactly what I didn't need: a magic bug :) | 21:37 |
| fungi | reed: yeah, tox -e slides on a held bare-trusty worker in rackspace | 21:37 |
| *** tnovacik has quit IRC | 21:37 | |
| fungi | reed: with tip of master for that repo | 21:38 |
| clarkb | sigmavirus the one hpcloud provides | 21:39 |
| clarkb | er I think would need to double check the image | 21:39 |
| reed | fungi, you get no invocation error? | 21:39 |
| fungi | reed: nope, and tox exits 0 (success) | 21:40 |
| reed | maybe a different version? | 21:40 |
| fungi | reed: wondering if we have something nondeterministic on our hands | 21:40 |
| reed | .tox/slides/bin/doc8 --version says > 0.5.0 | 21:40 |
| fungi | reed: .tox/slides/bin/pip list says i have doc8 (0.5.0) | 21:41 |
| *** achanda has quit IRC | 21:41 | |
| *** esker has quit IRC | 21:41 | |
| fungi | --version agrees | 21:41 |
| *** MarkAtwood has quit IRC | 21:42 | |
| *** ivar-laz_ has joined #openstack-infra | 21:42 | |
| fungi | and the invocation is still the same | 21:42 |
| fungi | slides runtests: commands[0] | doc8 doc | 21:42 |
| *** carl_baldwin has quit IRC | 21:42 | |
| *** ivar-laz_ has quit IRC | 21:43 | |
| SpamapS | oy | 21:43 |
| fungi | running it a few more times to see if i can get it to fail ever | 21:43 |
| *** mikeit has quit IRC | 21:43 | |
| SpamapS | I just realized I have done approximately 5 minutes of work to prepare for presentation tomorrow night | 21:43 |
| *** ivar-laz_ has joined #openstack-infra | 21:44 | |
| fungi | SpamapS: sounds about right. it's a 60-minute presentation? | 21:44 |
| openstackgerrit | Stephanie Miller proposed openstack-infra/puppet-zanata: Add OpenID login provider support to Zanata config https://review.openstack.org/166073 | 21:44 |
| fungi | "wing it" | 21:44 |
| *** ivar-lazzaro has quit IRC | 21:44 | |
| reed | if I run it as `.tox/slides/bin/doc8 doc` it passes | 21:44 |
| fungi | reed: maybe it's a tox version problem. what does tox --version say? i have 1.9.0 here | 21:45 |
| *** gokrokve has quit IRC | 21:45 | |
| reed | 1.8.0 | 21:45 |
| reed | i'm on ubuntu somethingold | 21:46 |
| *** doug-fish has left #openstack-infra | 21:46 | |
| fungi | reed: if you don't want to bother with replacing your distro version of tox, you can use it to bootstrap tox into another virtualenv | 21:47 |
| fungi | tox newtox && newtox/bin/pip install -U tox && newtox/bin/tox -e slides | 21:47 |
| jhesketh | Morning | 21:47 |
| *** samueldmq is now known as samueldmq-away | 21:48 | |
| fungi | reed: nevermind. i meant: virtualenv tox && newtox/bin/pip install -U tox && newtox/bin/tox -e slides | 21:48 |
| * fungi needs to detox now | 21:48 | |
| *** annegentle has quit IRC | 21:48 | |
| reed | it fails in the gate though | 21:49 |
| reed | so even if I upgrade, that's not going to solve the issue in gate | 21:49 |
| reed | right? | 21:50 |
| fungi | reed: yep, trying to figure out if it's tox version dependent execution behavior | 21:50 |
| fungi | i'm testing your 1.8.0 to see if that fails, and also trying what is on the gate workers | 21:50 |
| SpamapS | fungi: yes but I need at least 15 minutes to add jokes to these slides. ;) | 21:50 |
| *** dims has quit IRC | 21:50 | |
| fungi | SpamapS: your jokes must be way better than mine | 21:50 |
| openstackgerrit | Stephanie Miller proposed openstack-infra/puppet-zanata: Add apache configuration for zanata https://review.openstack.org/164011 | 21:51 |
| reed | fungi, I think I solved it by specifying the exact paths | 21:51 |
| reed | fungi, it works better this way | 21:51 |
| SpamapS | fungi: they're in Esperanto.. so... | 21:52 |
| *** achanda has joined #openstack-infra | 21:52 | |
| fungi | SpamapS: that's my problem. jokes just don't translate well into lojban | 21:52 |
| fungi | reed: yeah, seems not to be tox causing it. latest (1.9.2) passes for me as well as yours (1.8.0) | 21:52 |
| jeblair | SpamapS: are you using presentty? | 21:53 |
| fungi | reed: curious why, but if you're able to recreate the problem then patch away | 21:53 |
| SpamapS | jeblair: so, thats the 5 minutes I spent | 21:53 |
| SpamapS | jeblair: I MAY add a feature which pops up eog for screenshots. ;) | 21:54 |
| SpamapS | eog or sensible-browser | 21:54 |
| SpamapS | or something like that, not entirely sure yet | 21:54 |
| jeblair | "Presentty: prepare 60 minute presentations in 5 minutes!" | 21:54 |
| *** dims has joined #openstack-infra | 21:54 | |
| SpamapS | "Presentty: prepare 60 minute presentations in 5 minutes!*" | 21:54 |
| SpamapS | * actual prep time does not include 6 hours ratholing on features for presentty | 21:55 |
| *** dims has quit IRC | 21:55 | |
| fungi | SpamapS: i suppose trying to render your screenshots with libcaca won't be sufficient | 21:55 |
| *** dims has joined #openstack-infra | 21:55 | |
| jeblair | SpamapS: yeah, i mean, was more like two months for me | 21:55 |
| SpamapS | fungi: it kind of defeats the purpose. :) | 21:55 |
| SpamapS | jeblair: amateur | 21:55 |
| jeblair | SpamapS: and those pieces of high ansi art i created were a couple hours each | 21:56 |
| jeblair | ansi art is not easy | 21:56 |
| SpamapS | What was the program that made it easy "back in the day?" I forget now. | 21:56 |
| jeblair | i used 'thedraw' | 21:56 |
| jeblair | back in the day | 21:56 |
| jeblair | i used tetradraw recently | 21:57 |
| jeblair | it is not nearly as easy to use as thedraw | 21:57 |
| fungi | i just used a text editor and two xterms to constantly cat the file in a delayed loop | 21:57 |
| fungi | back in the day | 21:57 |
| SpamapS | http://en.wikipedia.org/wiki/List_of_text_editors#ASCII_and_ANSI_art hmm.. not a very complete list | 21:57 |
| SpamapS | that was it. | 21:57 |
| SpamapS | TheDraw! | 21:57 |
| fungi | and had the ansi escapes i needed to know memorized | 21:58 |
| jeblair | but it was usuable enough that i did not have to write an editor before i could write presentty :) | 21:58 |
| SpamapS | http://en.wikipedia.org/wiki/TheDraw hah | 21:58 |
| jeblair | so dodged that rathole | 21:58 |
| fungi | i think the beer has claimed most of those braincells over the years | 21:58 |
| SpamapS | umsdos is a thing. ;) | 21:58 |
| *** timcline has quit IRC | 21:58 | |
| openstackgerrit | Timur Nurlygayanov proposed openstack-infra/project-config: Add HAOS project into Stackforge https://review.openstack.org/167813 | 21:58 |
| SpamapS | www.syaross.org/thedraw/ hahaha | 21:58 |
| jeblair | SpamapS: wow, TheDraw and the space shuttle on the same page. the world has changed for the worse. :( | 21:58 |
| *** mriedem is now known as mriedem_away | 21:59 | |
| SpamapS | jeblair: aw that does make me :( | 21:59 |
| *** mrunge has quit IRC | 22:01 | |
| *** tsg_ has joined #openstack-infra | 22:04 | |
| *** tsg has quit IRC | 22:04 | |
| *** ociuhandu has joined #openstack-infra | 22:04 | |
| BobBall_AWAY | Who would be able to remove the email address from the XenProject CI gerrit account? :) I want to redirect the email to a list for ongoing monitoring but not get any emails from the changes. | 22:06 |
| *** Swami has quit IRC | 22:06 | |
| *** asettle has joined #openstack-infra | 22:08 | |
| *** jamielennox|away is now known as jamielennox | 22:08 | |
| *** hodos has joined #openstack-infra | 22:08 | |
| *** peristeri has quit IRC | 22:09 | |
| *** rlandy has quit IRC | 22:09 | |
| openstackgerrit | Thai Tran proposed openstack-infra/project-config: Create a repository for stackforge/xstatic-angular-lrdragndrop https://review.openstack.org/167798 | 22:09 |
| *** weshay has quit IRC | 22:10 | |
| openstackgerrit | Jeremy Stanley proposed openstack-infra/infra-manual: Mention license in document footer https://review.openstack.org/167799 | 22:11 |
| *** crc32 has quit IRC | 22:11 | |
| openstackgerrit | Khai Do proposed openstack-infra/project-config: Use zuul-cloner instead of git clone https://review.openstack.org/167816 | 22:11 |
| *** annegentle has joined #openstack-infra | 22:12 | |
| openstackgerrit | greghaynes proposed openstack-infra/nodepool: Add node deletion tests https://review.openstack.org/165582 | 22:13 |
| zigo_ | fungi: Could you please approve this one? https://review.openstack.org/#/c/167798/ | 22:13 |
| *** HeOS has joined #openstack-infra | 22:13 | |
| *** sdake__ has joined #openstack-infra | 22:14 | |
| zigo_ | fungi: This wrong naming of the lib is blocking me hard for packaging Murano-dashboard (because Horizon is broken for me right now...). | 22:14 |
| *** dboik has quit IRC | 22:15 | |
| fungi | zigo_: we have to schedule maintenance windows to rename projects. it's not a supported feature in gerrit, so we actually have to take it offline and update its database and filesystem (as well as make accompanying changes on other related servers) | 22:15 |
| *** amitgandhinz has quit IRC | 22:15 | |
| zigo_ | fungi: Would it be easier to just add a new project, then delete the old one? | 22:15 |
| *** amitgandhinz has joined #openstack-infra | 22:16 | |
| arosen | is there an easy way to kill nodepoold -d ? so that i can just control C it as i test things? | 22:16 |
| *** jcoufal has quit IRC | 22:16 | |
| fungi | zigo_: gerrit also doesn't support deleting projects, we have to retire them | 22:16 |
| zigo_ | fungi: Ok. So basically, I need to be patient? :) | 22:16 |
| *** Bsony has quit IRC | 22:17 | |
| *** prad has quit IRC | 22:17 | |
| *** sdake has quit IRC | 22:17 | |
| BobBall_AWAY | arosen: You have to kill -9 it or wait for the ctrl+c to take effect (effectively when it's finished with the currently-running processes of node building or image building) | 22:17 |
| zigo_ | Please have it happen asap, it's a real major blocker for me. | 22:17 |
| *** sdake has joined #openstack-infra | 22:17 | |
| arosen | BobBall_AWAY: yup that's what i've been doing. | 22:17 |
| *** achanda has quit IRC | 22:17 | |
| fungi | zigo_: we've got a maintenance window coming up in two days (i need to go ahead and send the maintenance announcement for that now), but the change will need to look similar to https://review.openstack.org/162952 | 22:18 |
| zigo_ | tqtran: Did you read? ^ | 22:18 |
| BobBall_AWAY | I imagine it's a safety feature since nodepool can't resume building nodes or images | 22:18 |
| *** HeOS has quit IRC | 22:18 | |
| zigo_ | tqtran: Your change doesn't look like #162952 | 22:19 |
| openstackgerrit | Thai Tran proposed openstack-infra/project-config: Create a repository for stackforge/xstatic-angular-lrdragndrop https://review.openstack.org/167798 | 22:19 |
| *** amitgandhinz has quit IRC | 22:20 | |
| tqtran | zigo__: just corrected it | 22:20 |
| *** sdake__ has quit IRC | 22:21 | |
| fungi | tqtran: zigo_: in this case the acl is not tied to the project (there seems to be one acl file for all the stackforge/xstatic-.* packages) but you'll want to make sure it's properly renamed throughout all files in the project-config repo (gerrit/projects.yaml, gerritbot/channels.yaml and zuul/layout.yaml) and link it at | 22:21 |
| fungi | https://wiki.openstack.org/wiki/Meetings/InfraTeamMeeting#Upcoming_Project_Renames | 22:21 |
| tqtran | fungi: would it be easier to create a new repo for it and then retire the old one? | 22:22 |
| *** davide has quit IRC | 22:23 | |
| fungi | tqtran: we'd really like to not accumulate more cruft repos than we already do, so renaming is best | 22:23 |
| *** achanda has joined #openstack-infra | 22:23 | |
| ianw | sdague / fungi: the problem with turning centos7 back to experimental is that if i try re-running it a considerable number of times to try and replicate the issue, then i'm also running a ton of other stuff and increasing general load | 22:23 |
| fungi | ianw: so better to remove it entirely i guess? | 22:24 |
| *** smccully has quit IRC | 22:24 | |
| fungi | ianw: looks like you only need to run it in rackspace to recreate the problem | 22:24 |
| openstackgerrit | Khai Do proposed openstack-infra/jenkins-job-builder: Fix jenkins-job-builder-compare-xml job by using zuul-cloner https://review.openstack.org/167821 | 22:24 |
| fungi | ianw: leaving it in experimental also has the added problem that check experimental will never return a result if that job hits a devstack-centos7 worker in rackspace | 22:25 |
| fungi | since it will run indefinitely | 22:25 |
| *** Alexandra_ has joined #openstack-infra | 22:25 | |
| openstackgerrit | Khai Do proposed openstack-infra/jenkins-job-builder: Fix jenkins-job-builder-compare-xml job by using zuul-cloner https://review.openstack.org/167821 | 22:27 |
| *** asettle has quit IRC | 22:27 | |
| openstackgerrit | David Lyle proposed openstack-infra/project-config: Create a repository for stackforge/xstatic-angular-lrdragndrop https://review.openstack.org/167798 | 22:27 |
| david-lyle | tqtran: added the new one to the gerritbot | 22:28 |
| david-lyle | in your patch | 22:28 |
| *** hdd has quit IRC | 22:28 | |
| openstackgerrit | Thai Tran proposed openstack-infra/project-config: Create a repository for stackforge/xstatic-angular-lrdragndrop https://review.openstack.org/167798 | 22:28 |
| fungi | ianw: i also think i just remembered that job switching from tempest smoke to full recently? i think smoke doesn't trigger the issue | 22:29 |
| tqtran | david-lyle, fungi: done patch uploaded | 22:29 |
| *** nelsnelson has quit IRC | 22:29 | |
| tqtran | ops another error | 22:30 |
| fungi | ianw: though maybe not. i can't find evidence to back up that recollection | 22:30 |
| openstackgerrit | melanie witt proposed openstack-infra/project-config: Adjust regression exceptions for Nova Cells V1 job https://review.openstack.org/166396 | 22:30 |
| sigmavirus24 | clarkb: idea. on hpcloud-b4 Can you reach mirror.rackspace.com? | 22:31 |
| *** jcoufal has joined #openstack-infra | 22:31 | |
| ianw | fungi: it's running smoke tests serial i think | 22:33 |
| fungi | ianw: but yeah, we had at least 3 changes hit it today | 22:34 |
| fungi | and all were rackspace workers | 22:34 |
| *** Alexandra_ has quit IRC | 22:34 | |
| *** asettle has joined #openstack-infra | 22:35 | |
| ianw | fungi: that correlates to what was seen with the glusterfs job, but i do think this coming back to the base devstack job is a relatively new thing | 22:35 |
| fungi | ianw: sounds likely, but i doubt having more devstack changes hang in the check pipeline is going to do much to help us track down the cause | 22:36 |
| BobBall_AWAY | sdague: FYI I didn't get a chance to test + deploy the message change. Will have to do this after my vacation. | 22:37 |
| *** unicell has left #openstack-infra | 22:37 | |
| *** andreykurilin_ has joined #openstack-infra | 22:37 | |
| *** dimtruck is now known as zz_dimtruck | 22:38 | |
| fungi | ianw: though you do have access to a tenant in rackspace where you can run this to reproduce it without having to rerun the job, right? | 22:38 |
| ianw | fungi: no i'm not advocating to hold up the queue | 22:39 |
| arosen | clarkb: if i specify qemu-img-options: -t raw, nodepool says: ERROR: Element 'raw' not found in '/root/project-config/nodepool/elements:/usr/local/bin/../share/diskimage-builder/elements' | 22:39 |
| arosen | any ideas? | 22:39 |
| openstackgerrit | Monty Taylor proposed openstack-infra/shade: WIP Create a swift container if it doesn't exist https://review.openstack.org/167823 | 22:39 |
| *** dkliban is now known as dkliban_afk | 22:40 | |
| ianw | fungi: i do have a personal account; so far others have not replicated the issue outside the gate using that | 22:40 |
| *** jcoufal has quit IRC | 22:40 | |
| greghaynes | arosen: I dont think qemu-img-options is what you want | 22:41 |
| openstackgerrit | Merged openstack-infra/jenkins-job-builder: Convert all inline publisher examples to tests https://review.openstack.org/166064 | 22:41 |
| *** dboik has joined #openstack-infra | 22:42 | |
| *** ajmiller has quit IRC | 22:42 | |
| *** e0ne has joined #openstack-infra | 22:43 | |
| arosen | greghaynes: any ideas of what I need instead? | 22:45 |
| openstackgerrit | Thai Tran proposed openstack-infra/project-config: Create a repository for stackforge/xstatic-angular-lrdragndrop https://review.openstack.org/167798 | 22:45 |
| greghaynes | arosen: AFAICT there isnt support for specifying non qcow2 output from dib for nodepool | 22:45 |
| *** HeOS has joined #openstack-infra | 22:46 | |
| nibalizer | clarkb: no dice http://logs.openstack.org/46/167346/12/check/gate-infra-puppet-apply-precise/6e7fda8/console.html#_2015-03-25_19_46_43_547 | 22:47 |
| nibalizer | :/ | 22:47 |
| *** annegentle has quit IRC | 22:48 | |
| mordred | arosen: it's in works | 22:49 |
| *** baoli has joined #openstack-infra | 22:49 | |
| greghaynes | mordred: are there patches? | 22:49 |
| clarkb | jeblair: one issue with zuul cloner compared to ggp is it doesn't print the change under test | 22:49 |
| clarkb | sigmavirus24: http://archive.ubuntu.com/ubuntu is what the node I tested on used | 22:50 |
| *** HeOS has quit IRC | 22:51 | |
| clarkb | nibalizer: boo | 22:52 |
| openstackgerrit | Spencer Krum proposed openstack-infra/project-config: Run rspec-puppet unit tests on puppet-elasticsearch https://review.openstack.org/167827 | 22:52 |
| *** dboik_ has joined #openstack-infra | 22:53 | |
| openstackgerrit | Thai Tran proposed openstack-infra/project-config: Renaming stackforge/xstatic-angular-irdragndrop https://review.openstack.org/167798 | 22:53 |
| mordred | greghaynes: well, we haven't done them yet because it's pointless for us until we get rackspace working on our image content and also working for uploads | 22:53 |
| *** gyee has joined #openstack-infra | 22:54 | |
| greghaynes | ah, ok. Just checking in case I poked at that | 22:54 |
| mordred | greghaynes: that said - we'll necessarily have to support it as part of the patches to nodepool in support of dib-nodepool | 22:54 |
| clarkb | I am pretty sure nodepool does support it, it does not support >1 image output though | 22:54 |
| greghaynes | clarkb: The code I read doesnt allow setting the -t arg | 22:54 |
| *** bhunter71 has quit IRC | 22:55 | |
| clarkb | oh wow and it hard appends a .qcow2 suffix :/ | 22:55 |
| *** gordc has quit IRC | 22:55 | |
| greghaynes | well, the input validation isnt great so tecnically you can set it, but not in an intended way | 22:55 |
| clarkb | I must've confused code I wrote long ago that never merged with something we actually supported | 22:55 |
| greghaynes | yes | 22:55 |
| *** andreykurilin_ has quit IRC | 22:56 | |
| *** dboik has quit IRC | 22:56 | |
| *** achanda has quit IRC | 22:57 | |
| *** tsg_ has quit IRC | 22:57 | |
| marun | clarkb: hi! | 22:57 |
| clarkb | hello | 22:57 |
| marun | clarkb: is there a way to set an arbitrary devstack env variable from a job definition in project-config? | 22:58 |
| marun | clarkb: I'm assuming no, since stack.sh is run as a different user (stack) | 22:58 |
| *** baoli has quit IRC | 22:58 | |
| clarkb | sigmavirus24: I can ping mirror.rackspace.com from hpcloud too | 22:58 |
| *** markvoelker has quit IRC | 22:58 | |
| marun | clarkb: HenryG wants to create an ipv4+6 experimental job for neutron and I'm not sure there's a way to set IP_VERSION without adding a DEVSTACK_GATE variable in devstack-gate | 22:59 |
| clarkb | marun: there is, you have to put it in the localrc which devstack will source | 22:59 |
| clarkb | marun: and there is a d-g variable whose contents are appended to the localrc | 22:59 |
| sigmavirus24 | clarkb: interesting. Is b4 in a different region than b1? | 22:59 |
| *** Alexandra_ has joined #openstack-infra | 22:59 | |
| marun | clarkb: ah, ok! | 22:59 |
| jeblair | clarkb: yeah, though "change under test" might be "changes under test". but yeah. if we switch the ggp macro out, we can throw in an echo for that | 22:59 |
| clarkb | marun: DEVSTACK_LOCAL_CONFIG appears to be the variable | 23:00 |
| marun | clarkb: cool, thank you! | 23:00 |
| clarkb | sigmavirus24: nope | 23:00 |
| marun | HenryG: ^^ | 23:00 |
| clarkb | sigmavirus24: b1-b5 are all the same region but are possibly different AZs | 23:00 |
| HenryG | marun, clarkb, jeblair: thanks! | 23:00 |
| *** asettle has quit IRC | 23:01 | |
| marun | HenryG: so set DEVSTACK_LOCAL_CONFIG in the job def to include the ip version you want | 23:01 |
| sigmavirus24 | So we use mirror.rackspace.com for all of our packages so we can pin dependencies and such. It seems to be an issue between that and b4 but I'm at a loss for what would be the cause | 23:01 |
| ianw | 2015-03-25 15:07:40,575 INFO nodepool.image.build.devstack-fedora21-dib: Cloning into '/opt/dib_cache/source-repositories/puppet_bandersnatch_ff7683c13b27b4f004f99455e3101388fffb1960.tmp'... | 23:01 |
| ianw | 2015-03-25 15:07:40,576 INFO nodepool.image.build.devstack-fedora21-dib: fatal: Could not read from remote repository. | 23:01 |
| *** e0ne has quit IRC | 23:01 | |
| clarkb | marun: HenryG however | 23:01 |
| ianw | ^ guessing this was just a temporary failure at the time? | 23:01 |
| clarkb | marun: HenryG I have encouraged others to try a devstack change to enable ipv6 | 23:01 |
| *** Alexandra_ has quit IRC | 23:01 | |
| clarkb | marun: HenryG and aiui there were only a few things that needed correcting. So its possible you don't even need a new job just self testing devstack + other changes | 23:02 |
| *** Alexandra_ has joined #openstack-infra | 23:02 | |
| clarkb | marun: HenryG because ideally we would just test ipv6 as a first class citizen on every job run | 23:02 |
| *** Alexandra_ has quit IRC | 23:02 | |
| *** asettle has joined #openstack-infra | 23:02 | |
| marun | clarkb, HenryG: Ideally, yes. | 23:03 |
| clarkb | last I saw there were ~2 tests that failed | 23:03 |
| clarkb | and that just had to do with how tempest asserted ip addrs showing up on VMs | 23:03 |
| clarkb | but that was quite a few weeks ago | 23:03 |
| *** achanda has joined #openstack-infra | 23:03 | |
| marun | HenryG: Does that make sense? Get it working on a devstack patch and then the default neutron tempest job to use ipv6 | 23:03 |
| marun | then the -> then change the | 23:03 |
| openstackgerrit | Wayne Warren proposed openstack-infra/infra-specs: (WIP) Initial work on gitub-integration spec. https://review.openstack.org/163637 | 23:04 |
| HenryG | marun: OK, that sounds like a one-liner I am willing to try :) | 23:04 |
| alexpilotti | fungi clarkb: a patch landed on DevStack that broke the Hyper-V CI and angela-s2’s CI (see conversation on -dev) | 23:05 |
| *** cdent has quit IRC | 23:05 | |
| fungi | alexpilotti: that's unfortunate. let's see if we can find some devstack core reviewers | 23:06 |
| clarkb | HenryG: marun curious to know how that goes | 23:06 |
| *** mark_gloshen has joined #openstack-infra | 23:06 | |
| alexpilotti | tried to contact the autor on IRC, no reply yet. I wonder if it wouldnt be the case to revert it as there might be others involved | 23:06 |
| fungi | alexpilotti: they usually hang around in #openstack-qa but also lurk in here a lot of the time | 23:06 |
| fungi | (contrary to popular belief, devstack is a qa repo, not an infra repo) | 23:07 |
| mark_gloshen | Is this a good place to ask for help with review.openstack.org (gerrit)? | 23:07 |
| *** hdd has joined #openstack-infra | 23:07 | |
| alexpilotti | fungi: patch: https://review.openstack.org/#/c/158424/ | 23:07 |
| clarkb | mark_gloshen: yup ask away | 23:08 |
| jeblair | mark_gloshen: it's the best place to do that | 23:08 |
| alexpilotti | sdague and dtroyer approved it | 23:08 |
| *** ivar-lazzaro has joined #openstack-infra | 23:08 | |
| mark_gloshen | Ok, I read the docs and setup my accounts, keys, etc. but am having problems with git-review -s | 23:08 |
| jeblair | ianw: probaably a race with repo creation (unless repo creation broke) | 23:08 |
| *** hichihara has joined #openstack-infra | 23:08 | |
| *** HeOS has joined #openstack-infra | 23:08 | |
| anteaya | alexpilotti: usually the way forward for a patch that broke your ci is to offer a patch to fix both the intention of the original patch and your ci | 23:08 |
| harlowja | qq; something that i've always wondered | 23:09 |
| harlowja | why aren't nova, glance, ironic... published on pypi | 23:09 |
| clarkb | mark_gloshen: can you paste the entire output of git-review -s to http://paste.openstack.org and share the link here? | 23:09 |
| anteaya | alexpilotti: paying attention to what broke your ci so you can fix it is rather the point of having it | 23:09 |
| *** bhuvan has quit IRC | 23:09 | |
| harlowja | like shells exist for nova (ironic and others it seems); but why not a package | 23:09 |
| alexpilotti | anteaya: we’re still testing to see if we got the workaround right | 23:09 |
| mordred | harlowja: we used to not do it on purpose | 23:09 |
| alexpilotti | anteaya: was wondering if we could just revert it, fix it and re-merge it | 23:09 |
| *** hashirabdi has joined #openstack-infra | 23:09 | |
| *** e0ne has joined #openstack-infra | 23:09 | |
| harlowja | mordred any idea why? | 23:10 |
| mark_gloshen | http://paste.openstack.org/show/196756/ | 23:10 |
| anteaya | alexpilotti: that's fine but you are asking the wrong folks | 23:10 |
| anteaya | alexpilotti: sdague and dtroyer approve lots of things | 23:10 |
| mordred | harlowja: yeah - because one could not in any way reasonably expect "pip install nova" to get you a working nova | 23:10 |
| *** Longgeek has joined #openstack-infra | 23:10 | |
| mordred | harlowja: however ... | 23:10 |
| fungi | chmouel: dtroyer: ianw: sdague: alexpilotti is asking for a revert of https://review.openstack.org/158424 | 23:10 |
| *** chlong has joined #openstack-infra | 23:10 | |
| anteaya | alexpilotti: and I tend to agree with their judgement | 23:10 |
| harlowja | mordred meh; that seems like a cop-out | 23:10 |
| mordred | harlowja: no - it was a very active choice, but one taht came from a distro-centric worldview | 23:10 |
| clarkb | mark_gloshen: it looks like whereever you are running that from is not able to talke to review.openstack.org over port tcp 29418 | 23:11 |
| mordred | it was very explicitly about not publishing something known to be "broken" | 23:11 |
| fungi | alexpilotti: those are the devstack core reviewers who happen to be in channel anyway https://review.openstack.org/#/admin/groups/50,members | 23:11 |
| mordred | harlowja: however, we've also had conversations recently about changing that and publishing server projects too | 23:11 |
| harlowja | hmmm, doesn't seem like any software would ever be published then, lol | 23:11 |
| mordred | harlowja: for all the reasons | 23:11 |
| mark_gloshen | I was able to connect via ssh though | 23:11 |
| *** ivar-laz_ has quit IRC | 23:11 | |
| mordred | harlowja: it just hasn't been high enough priority for anyone to actually work on it | 23:11 |
| clarkb | mark_gloshen: over port 29418? you can try `ssh -p29418 mark.gloshen@review.openstack.org gerrit ls-projects` | 23:11 |
| harlowja | mordred thx; interseting to know | 23:12 |
| alexpilotti | fungi: tx! | 23:12 |
| mark_gloshen | http://paste.openstack.org/show/196757/ | 23:12 |
| fungi | harlowja: though also at least a couple of our more popular server projects' names are already taken by other projects on pypi | 23:12 |
| mordred | there is also that issue | 23:12 |
| harlowja | https://pypi.python.org/pypi/ironic seems like reserved, nova seems https://pypi.python.org/pypi/nova ok | 23:12 |
| ianw | alexpilotti: ok, did you already propose the revert (ianw checks queue...) | 23:12 |
| mark_gloshen | @clarkb I just tried the ls-projects and it worked | 23:12 |
| harlowja | https://pypi.python.org/pypi/neutron seems reserved/ok | 23:13 |
| alexpilotti | ianw: right away | 23:13 |
| jeblair | harlowja: tempest is not us | 23:13 |
| *** HeOS has quit IRC | 23:13 | |
| fungi | mark_gloshen: try adding -v to your git review -s | 23:13 |
| fungi | mark_gloshen: for more verbose debug output | 23:13 |
| fungi | there might be some clues in there about what it's trying to do | 23:14 |
| harlowja | jeblair ya, good catch on that one, so some issues there i guess with naming | 23:14 |
| *** hodos has quit IRC | 23:15 | |
| mark_gloshen | http://paste.openstack.org/show/196758/ | 23:15 |
| *** mpavone has quit IRC | 23:15 | |
| alexpilotti | ianw: on it! | 23:16 |
| fungi | clarkb: mark_gloshen: i think i've seen this crop up a couple times now, where if the repo requires a cla and the account hasn't agreed to the icla or filed contact info, git-review fails its test push | 23:16 |
| harlowja | let's see, maybe i can get someone to work on getting things published to pypi; and figuring out what to do about the name stuffs | 23:16 |
| alexpilotti | ianw: https://review.openstack.org/#/c/167835/ | 23:17 |
| clarkb | fungi: interesting | 23:17 |
| mark_gloshen | I did agree to to the icla | 23:17 |
| mark_gloshen | let me check my contact info | 23:17 |
| clarkb | https://review.openstack.org/#/admin/projects/stackforge/cookbook-openstack-ops-database does require the cla | 23:17 |
| *** sabeen3 has quit IRC | 23:17 | |
| ianw | alexpilotti: can you just add to the changelog some analysis of what's wrong? i guess this patch is really about laying groundwork for neutron as default | 23:17 |
| fungi | clarkb: if we can confirm that's what's happening, i consider that a git-review bug | 23:17 |
| alexpilotti | ianw: was doing it already, tx | 23:18 |
| clarkb | fungi: ya, ideally it could spit out the you need valid cla error message from gerrit | 23:18 |
| mark_gloshen | Aha | 23:18 |
| *** ChuckC has quit IRC | 23:18 | |
| mark_gloshen | That was it | 23:18 |
| fungi | clarkb: agreed, that's the easy fix which comes to mind anyway | 23:18 |
| mark_gloshen | Thank you fungi and clarkb | 23:18 |
| fungi | clarkb: harder fix is to find something else it can test which won't tickle cla enforcement | 23:18 |
| clarkb | sigmavirus24: fwiw our test nodes all come with the packages devstack needs precached for your tests. It may be worth mounting /var/cache/apt/archive into your containers | 23:19 |
| fungi | mark_gloshen: great! apologies for the confusion there | 23:19 |
| clarkb | sigmavirus24: I don't know that that will ifx this particular problem but it will likely make your jobs faster and more reliable in general | 23:19 |
| sigmavirus24 | clarkb: I didn't know that | 23:19 |
| sigmavirus24 | Thank you! | 23:19 |
| fungi | clarkb: sigmavirus24: soon i also intend to have a mechanism in place where individual projects can list additional distro packages they want cached as well | 23:20 |
| sigmavirus24 | fungi: that's awesome | 23:20 |
| alexpilotti | ianw: done | 23:20 |
| JayF | fungi: clarkb: This could reduce the time in the agent_ssh-src job and the coreos builder *significantly* | 23:20 |
| *** Longgeek has quit IRC | 23:20 | |
| * JayF makes a note to follow up on mounting that into his build containers in devstack if possible | 23:20 | |
| *** krtaylor has quit IRC | 23:21 | |
| ianw | alexpilotti: and can you point to or open a bug please | 23:21 |
| fungi | sigmavirus24: basic plan is to have nodepool scrape all branches of the git repos we host including in stackforge for a specific file which contains a cross-distro format list of packages used by their ci jobs | 23:21 |
| sigmavirus24 | not another project-config like repo containing them all? | 23:22 |
| fungi | sigmavirus24: then we can both use that as info on what packages we'll benefit from caching and also jobs can reference that list to be able to install things at run-time | 23:22 |
| * sigmavirus24 isn't complaining, just curious | 23:22 | |
| clarkb | fungi: I was thinking about that a little more and would it make any sense to have a bindep list in openstack/requirements just to simplify what is going on? I don't think we need any hard enforcemnt on that stuff though since its all already in the distros | 23:22 |
| clarkb | so maybe thats not worth it | 23:22 |
| *** Bsony has joined #openstack-infra | 23:23 | |
| fungi | clarkb: nah, having a bindep list in each project allows those projects to self-test changes to their lists. can't really cause conflicts there since the distro has non-conflicting package sets anyway | 23:23 |
| openstackgerrit | Ian Wienand proposed openstack-infra/system-config: Split diskimage-builder built images out https://review.openstack.org/167837 | 23:23 |
| clarkb | fungi: right the packages already being in the distros simplify some parts of this | 23:24 |
| clarkb | but also make the use of a tool like bindep mandatory | 23:25 |
| fungi | we'll be able to just union the set of bindep outputs from those plus whatever devstack wants | 23:25 |
| alexpilotti | ianw: done! | 23:25 |
| krotscheck | clarkb: You up for baileys tonight? | 23:25 |
| clarkb | jhesketh: did you see my comment about updating the dg tempest jobs to use your new devstack log publisher with footers thing? | 23:25 |
| *** achanda has quit IRC | 23:25 | |
| fungi | clarkb: well, not mandatory. i have a fallback list going into project-config which contains all the stuff we already preinstall on bare-.* nodes | 23:25 |
| clarkb | krotscheck: I just escaped downtown. I5 NB is shutdown and going to make transit a compelte mess today | 23:25 |
| clarkb | krotscheck: probably best to avoid getting back in car/bus | 23:26 |
| openstackgerrit | James E. Blair proposed stackforge/gertty: Add held changes https://review.openstack.org/167838 | 23:26 |
| krotscheck | clarkb: You got it. See you tomorrow :) | 23:26 |
| fungi | clarkb: so jobs can just use that unless a project provides their own bindep file, and then that file addition even becomes self-testing | 23:26 |
| clarkb | krotscheck: but do go to baileys and go upstairs for a slightly different vbe | 23:26 |
| clarkb | krotscheck: you go down the alley and its a door on your left | 23:26 |
| ianw | alexpilotti: sorry to be annoying; but it's just "Partial-bug" because the patch will need to be reproposed with a fix to the problem | 23:26 |
| clarkb | fungi: gotcha | 23:26 |
| jeblair | markmcclain: by your request: https://review.openstack.org/167838 | 23:26 |
| *** yfried|afk is now known as yfried | 23:26 | |
| *** achanda has joined #openstack-infra | 23:27 | |
| jeblair | mordred: ^ you may also be interested | 23:27 |
| jhesketh | clarkb: er, I don't think so sorry | 23:27 |
| jhesketh | clarkb: any pointers to where to find it, or happy to restate? | 23:27 |
| clarkb | jhesketh: ya let me find the change number | 23:27 |
| alexpilotti | ianw: good point. Done :-) | 23:27 |
| jhesketh | cheers | 23:27 |
| *** dhritishikhar_ has quit IRC | 23:27 | |
| clarkb | jhesketh: 167378 | 23:27 |
| krotscheck | clarkb: btw, If I can get some time with you early tomorrow to ask about zuul, I'd appreciate it. | 23:27 |
| clarkb | krotscheck: sure | 23:28 |
| jhesketh | clarkb: thanks, looking | 23:28 |
| *** Bsony has quit IRC | 23:28 | |
| krotscheck | clarkb: Thanks | 23:28 |
| mordred | jeblair: nice | 23:28 |
| jhesketh | clarkb: ah right, yes we should. Should we do that in another change though? | 23:28 |
| clarkb | jhesketh: either way works. I didn't think it deserved a -1 | 23:28 |
| clarkb | jhesketh: but feel free to update existing change or push new change depending on how you want to do it | 23:29 |
| jhesketh | clarkb: well that's nearly ready to merge, so I'll do it as a new one since it's also kinda changing something different | 23:29 |
| markmcclain | jeblair: awesome | 23:29 |
| jeblair | markmcclain: i started with the minimal heuristic i thought was needed, but let me know if you want to change it, or if you think it's worth making configurable. that part's easy now that the framework is there | 23:31 |
| *** ChuckC has joined #openstack-infra | 23:32 | |
| *** dboik_ has quit IRC | 23:32 | |
| *** wenlock has quit IRC | 23:32 | |
| *** dboik has joined #openstack-infra | 23:33 | |
| clarkb | jhesketh: and I think 165582 addresses your comments now | 23:33 |
| alexpilotti | ianw: sorry, I uploaded a new patch straight after you +2a’d it :-) | 23:34 |
| *** hdd has quit IRC | 23:34 | |
| ianw | alexpilotti: what's the difference? | 23:34 |
| alexpilotti | ianw: changed a word in the commit message | 23:34 |
| alexpilotti | s/can/might/ | 23:34 |
| *** Alexandra_ has joined #openstack-infra | 23:35 | |
| markmcclain | jeblair: the behavior you wrote feels correct… I'll try it out and see how it matches up | 23:35 |
| jhesketh | clarkb: cool, will re-review | 23:35 |
| *** dboik_ has joined #openstack-infra | 23:35 | |
| *** e0ne has quit IRC | 23:36 | |
| *** asettle has quit IRC | 23:36 | |
| alexpilotti | ianw: thanks for your help! :-) | 23:36 |
| *** yfried is now known as yfried|afk | 23:36 | |
| *** Alexandra_ has quit IRC | 23:37 | |
| stevebaker | anteaya: hey, this smaller time limit is causing a timeout on a new test that I really want to run https://review.openstack.org/#/c/166320/ | 23:37 |
| *** chlong has quit IRC | 23:37 | |
| *** chlong has joined #openstack-infra | 23:37 | |
| ianw | alexpilotti: as a courtesy when it commits i'll put a revert-of-revert back into the queue so it can get fixed up, but you might like to keep an eye on that review and maybe test it out first to ensure it's ok | 23:37 |
| stevebaker | anteaya: https://review.openstack.org/#/c/165287/ | 23:37 |
| *** dboik has quit IRC | 23:37 | |
| alexpilotti | ianw: we were already thinking about submitting a revert + fix | 23:38 |
| *** hdd has joined #openstack-infra | 23:38 | |
| ianw | alexpilotti: ok, that's great too! we can just let the neutron experts review that as usual | 23:39 |
| jeblair | markmcclain: oops, it conflicts with HEAD; about to upload a new patchset | 23:40 |
| clarkb | mordred: have yu seen nibalizer's -1 on 165157? | 23:40 |
| *** HeOS has joined #openstack-infra | 23:41 | |
| *** mescanef has quit IRC | 23:41 | |
| *** mescanef has joined #openstack-infra | 23:41 | |
| mordred | clarkb: yes - haven't come back aroudn to it yet | 23:41 |
| mordred | nibalizer: I think we should align with your other thing | 23:42 |
| anteaya | stevebaker: how long does the new test take to run? | 23:42 |
| *** tsg has joined #openstack-infra | 23:42 | |
| anteaya | stevebaker: having heat functional tests take 2 hours presents a problem for us | 23:42 |
| anteaya | or it did anyway | 23:42 |
| clarkb | anteaya: I think stevebaker fixed the underlying slowness issue and got it back to near an hour | 23:43 |
| anteaya | he did | 23:43 |
| stevebaker | anteaya: I don't know, its timing out ;) Locally it is running in about 2 minutes | 23:43 |
| anteaya | now he wants a new test | 23:43 |
| openstackgerrit | James E. Blair proposed stackforge/gertty: Add held changes https://review.openstack.org/167838 | 23:43 |
| stevebaker | anteaya: actually the test is likely failing but the job times out before the test times out. What I really need is the test timeout error | 23:43 |
| anteaya | stevebaker: how can something that runs locally in 2 minutes not finish in 70? | 23:43 |
| nibalizer | mordred: right but move mine to be like that one or this one to be like that one | 23:44 |
| anteaya | stevebaker: if the test takes 2 minutes to run locally can you lower the test timeout on that test? | 23:44 |
| anteaya | to get the error returned | 23:44 |
| nibalizer | i like the directory separation since it will keep the main repo holding the .yml files saner | 23:44 |
| anteaya | having to increase a test timeout beyond 70 minutes for a 2 minutes test doesnt' make sense to me | 23:44 |
| jeblair | stevebaker, anteaya: we usually set the jenkins timeout to 5 minutes longer than what we tell devstack in order to get full log collection | 23:44 |
| stevebaker | yeah, I might have to. | 23:44 |
| anteaya | jeblair: and I did in this case, yes | 23:45 |
| anteaya | jenkins timeout is 75 minutes | 23:45 |
| anteaya | or that was my intention | 23:45 |
| anteaya | https://review.openstack.org/#/c/166320/ | 23:45 |
| anteaya | wiht this patch | 23:45 |
| jeblair | anteaya: cool, just emitting information into the channel randomly :) | 23:45 |
| stevebaker | related, the test should in theory pass when this review is approved https://review.openstack.org/#/c/165227/ | 23:45 |
| *** HeOS has quit IRC | 23:46 | |
| openstackgerrit | Joshua Hesketh proposed openstack-infra/project-config: Update devstack log publisher to use new macro https://review.openstack.org/167842 | 23:46 |
| clarkb | fungi: 167799 lgtm, is that something you want a lot of input on prior to approval? | 23:46 |
| jhesketh | clarkb: ^ | 23:47 |
| clarkb | fungi: it also looks fine when I scrunch up my browser into a tiny window but you probabl tested that well before I did :) | 23:47 |
| openstackgerrit | Merged openstack-infra/infra-manual: Dynamically update copyright year at publication https://review.openstack.org/167795 | 23:47 |
| *** markvoelker has joined #openstack-infra | 23:47 | |
| fungi | clarkb: couldn't hurt. licenses are important | 23:47 |
| anteaya | jeblair: thanks for checking up on me so that I am working with accurate information, thank you | 23:47 |
| nibalizer | jeblair: so in https://review.openstack.org/#/c/164908/ nan added puppet unit tests which i think is significant because we've not done that before on the infra modules | 23:47 |
| nibalizer | as long as information is being radiated out into channels | 23:48 |
| fungi | clarkb: my first patchset there ended up being a but too busy in the footer, hence striving for brevity but perhaps at the expense of readability | 23:48 |
| stevebaker | anteaya: I think our tests could use better test and build timeout tuning, I'll look into it | 23:48 |
| openstackgerrit | Merged stackforge/gertty: Add change list options to configuration https://review.openstack.org/161032 | 23:48 |
| jeblair | nibalizer: thanks | 23:49 |
| clarkb | nibalizer: the first ~30 lines of https://review.openstack.org/#/c/164908/5/spec/classes/elastic_search_spec.rb are why I dislike rspec | 23:49 |
| anteaya | stevebaker: thank you | 23:49 |
| clarkb | nibalizer: we've just written the module twice... | 23:49 |
| anteaya | stevebaker: in the meantime I am much more comfortable helping you get a test working than extending a timeout | 23:50 |
| anteaya | stevebaker: all you need is one more project-config core review on https://review.openstack.org/#/c/165227/ | 23:50 |
| clarkb | nibalizer: the second chunk is a bit more useful since it checks parameter things | 23:50 |
| stevebaker | clarkb: any chance? https://review.openstack.org/#/c/165227/ | 23:50 |
| bknudson | is there a git hook for tox -e pep8? | 23:51 |
| nibalizer | clarkb: yea i've never really done the rspec-puppet thing | 23:51 |
| *** markvoelker has quit IRC | 23:52 | |
| nibalizer | clarkb: also could land this if you're feeling button-pressy https://review.openstack.org/#/c/164842/ | 23:52 |
| jeblair | nibalizer, clarkb: i agree, that's not the kind of testing i think is useful... | 23:52 |
| *** r-daneel has quit IRC | 23:52 | |
| openstackgerrit | Ian Wienand proposed openstack-infra/nodepool: Automate reload of logging configuration https://review.openstack.org/167844 | 23:52 |
| *** alexpilotti has quit IRC | 23:53 | |
| clarkb | stevebaker: mostly looks fine, just curious about dib. Is it completely unused in that test at this point? | 23:53 |
| nibalizer | jeblair: yea if we're not gonna write it and require new features to be covered by it we probably dont want to be in the rspec-puppet business | 23:53 |
| stevebaker | clarkb: it has been completely expunged from devstack | 23:53 |
| clarkb | nibalizer: is that hiera behavior documented somewhere I can read up on | 23:53 |
| nibalizer | jeblair: how would you feel about a test that runs puppet for real then uses https://pypi.python.org/pypi/envassert to inspect the system? | 23:54 |
| clarkb | nibalizer: I am assuming that setting $group globally is doing magic somewhere | 23:54 |
| openstackgerrit | Merged openstack-infra/nodepool: Add node deletion tests https://review.openstack.org/165582 | 23:54 |
| *** ghostpl_ has quit IRC | 23:54 | |
| nibalizer | clarkb: https://review.openstack.org/#/c/164845/5/modules/openstack_project/files/puppetmaster/hiera.yaml | 23:55 |
| nibalizer | follow on change | 23:55 |
| *** garyh has quit IRC | 23:55 | |
| nibalizer | basically when both of those are in then variables set in group/groupname.yaml will ovveride whatever is in common.yaml | 23:55 |
| clarkb | I see thanks | 23:55 |
| nibalizer | this is only required because regexes on node definitions is supported in site.pp but not in hiera lookups | 23:56 |
| jeblair | nibalizer: that's more interesting to me (heck, even the noop tests catch most of our errors -- what we're left with are errors that are mostly "just didn't do what we thought it would") so envassert wolud be helpful there | 23:56 |
| jeblair | nibalizer: baby steps if we want to try that though :) | 23:56 |
| nibalizer | jeblair: yea i think a test for system-config that applies only o_p::server and checks some basics is a good way to start | 23:57 |
| *** annegentle has joined #openstack-infra | 23:57 | |
| nibalizer | can evaluate usefulness then | 23:57 |
| jeblair | nibalizer, clarkb: let's ask around and make sure we don't want to be in the rspec business | 23:57 |
| nibalizer | jeblair: *nod* | 23:57 |
| jeblair | nibalizer, clarkb: that's a first-patch, and i want to be careful about making that decision | 23:58 |
| jeblair | nibalizer, clarkb: but if we agree, i can ask the author to drop the rspec tests | 23:58 |
| clarkb | nibalizer: comment on 164842 | 23:58 |
| nibalizer | jeblair: i like the idea of saying 'should infra use rspec-puppet tests on its modles' to the community | 23:58 |
| nibalizer | and seeing what shakes out | 23:59 |
| openstackgerrit | Ramy Asselin proposed openstack-infra/system-config: Add puppet-openstackci to included modules https://review.openstack.org/167444 | 23:59 |
| clarkb | I personally will not write any rspec tests that rewrite the module | 23:59 |
| clarkb | DRY | 23:59 |
| clarkb | if others feel that that is actually a beneficial task I won't get in the way but it seems like a lot of not so useful busy work to me | 23:59 |
| openstackgerrit | Ramy Asselin proposed openstack-infra/system-config: WIP: Use common log server implementation https://review.openstack.org/167426 | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!