| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2017-05-02 | |||
| 17:57:18 | ildikov | dansmith: brand new people, experienced contributors including project decision makers, users and operators and anyone, who's interested | |
| 17:57:19 | mriedem | https://wiki.openstack.org/wiki/Forum | |
| 17:57:43 | dansmith | mriedem: no, there's another one that has a diagram of the split, | |
| 17:57:48 | mriedem | heh "The Forum is not the right place to get started or learn the basics of OpenStack." | |
| 17:58:23 | mriedem | dansmith: this? https://www.openstack.org/ptg/ | |
| 17:58:35 | dansmith | no | |
| 17:59:40 | ildikov | mriedem: dansmith: so the Forum is the place for discussions between the developer and the user/operator community | |
| 18:00:09 | ildikov | mriedem: dansmith: we also have the rest of the Summit happening at the same time and place, so don't forget about that either plz :) | |
| 18:00:32 | ildikov | mriedem: re that sentence you copied here above | |
| 18:00:52 | mriedem | sure, but https://www.openstack.org/summit/boston-2017/summit-schedule/events/18688/nova-project-onboarding | |
| 18:00:57 | mriedem | says "The Forum" on it | |
| 18:01:25 | mriedem | again, anyway, it's just nit picking | |
| 18:01:56 | dansmith | guh, I can't find the thing I'm looking for | |
| 18:01:57 | ildikov | mriedem: thanks for pointing it out, as it's important to have clear descriptions out | |
| 18:02:53 | ildikov | mriedem: but I hope what I wanted to get to went through as well | |
| 18:04:32 | openstackgerrit | Ed Leafe proposed openstack/nova master: Remove the can_host column https://review.openstack.org/461494 | |
| 18:05:53 | mriedem | cdent: your opinions on what might be interesting/good content for a nova project onboarding session are probably valuable as you're maybe the newest nova contributor that's been around awhile and hasn't recently been laid off by osic | |
| 18:06:07 | mriedem | like, 'i wish someone would have explained x to me within the first month' | |
| 18:07:30 | mriedem | unless those things are local delete, bdms, extra specs, quotas, versioned objects, cells (v1 or v2), placement, really anything | |
| 18:11:16 | dansmith | I'm still waiting for someone to explain BDMs to me | |
| 18:12:23 | cdent | mriedem: i'll think on it | |
| 18:12:26 | ildikov | dansmith: let me know when you find the person for that, I would be interested in that session too :) | |
| 18:12:43 | dansmith | It'll be on the 9th green at 9pm | |
| 18:13:06 | openstackgerrit | dane-fichter proposed openstack/nova master: Add trusted certificates to InstanceExtras https://review.openstack.org/457711 | |
| 18:13:19 | cdent | time in nova is weird: it's been > 1.5 year | |
| 18:14:20 | ildikov | dansmith: ok :) | |
| 18:17:25 | mriedem | gibi: doesn't it seems sort of odd that the volume attach/detach notifications don't have a bdm in them? | |
| 18:28:06 | dansmith | melwitt: didn't you implement the multithreaded cross-cell scatter/gather thing for something in quotas? | |
| 18:28:15 | dansmith | melwitt: was thinking I'd shamelessly copy it and start on a few other places | |
| 18:28:59 | mriedem | is it odd that we don't set task_state when doing something like attaching a volume or interface? | |
| 18:30:03 | cdent | mriedem: is there an etherpad where I could or should dump onboarding stuff, or you just hoping for it here? | |
| 18:30:11 | mriedem | cdent: i can create something | |
| 18:30:55 | mriedem | re: you can't delete an instance if it's task_state is set, unless you force delete it, but when attaching a volume we don't set the task_state to anything, so you could delete it while the volume is being attached | |
| 18:57:56 | mriedem | jaypipes: bauwser: cdent: sdague: a start https://etherpad.openstack.org/p/BOS-forum-nova-project-onboarding | |
| 18:58:04 | cdent | roger | |
| 19:04:17 | dansmith | ffs, now I can't get the gate to lose this race to show my debug information | |
| 19:04:34 | dansmith | sdague: did anything change in devstack around the systemd stuff or something since yesterday? | |
| 19:04:44 | dansmith | there was talk about it this morning but I didn't pay attention | |
| 19:06:20 | dansmith | ah, well, one of the other jobs is hitting it at least | |
| 19:08:26 | mriedem | dansmith: the fallout from the systemd thing was the ceph setup in the live migration job | |
| 19:08:28 | mriedem | that's fixed now | |
| 19:08:33 | dansmith | okay | |
| 19:11:44 | cdent | that's me in cyan on the etherpad, that sort of thing useful? | |
| 19:13:31 | mriedem | cdent: i think so | |
| 19:13:33 | mriedem | thanks | |
| 19:14:33 | cdent | I hadn't put backwards compatibility where you just did because the need for the behaviors isn't covered well enough by that phrase. It's a rather special kind of backwards compatibility | |
| 19:14:55 | cdent | I'm trying to think of more "rather special kind" of things | |
| 19:15:12 | mriedem | i don't know what the difference is, or what that means | |
| 19:16:16 | cdent | exactly! | |
| 19:16:51 | mriedem | oh fun functional job timeout | |
| 19:25:38 | openstackgerrit | Chris Dent proposed openstack/nova master: Optional separate database for placement API https://review.openstack.org/362766 | |
| 19:26:49 | ihrachys | someone please reopen the bug https://bugs.launchpad.net/nova/+bug/1669468 for some reason I can't do it. | |
| 19:26:49 | openstack | Launchpad bug 1669468 in OpenStack Compute (nova) "tempest.api.compute.servers.test_novnc.NoVNCConsoleTestJSON.test_novnc fails intermittently in neutron multinode nv job" [Medium,Fix released] - Assigned to Markus Zoeller (markus_z) (mzoeller) | |
| 19:26:54 | ihrachys | it hits master and stable still | |
| 19:27:46 | mriedem | ihrachys: maybe open a new bug? | |
| 19:28:09 | mriedem | i.e. is a regression? | |
| 19:28:11 | ihrachys | mriedem, well I am not sure. seems like same failure. | |
| 19:28:50 | ihrachys | at least superficially it shows same symptom in same test | |
| 19:29:13 | ihrachys | I admit I don't know much about how it's supposed to work to distinguish root causes if they are different. | |
| 19:29:36 | mriedem | checking logstash | |
| 19:30:03 | clarkb | we run novnc in the gate? I thought we didn't because git badness | |
| 19:30:31 | ihrachys | clarkb, seems like we do, at least in ocata for neutron gate | |
| 19:30:42 | mriedem | clarkb: https://github.com/openstack-infra/devstack-gate/commit/f283a62020615f3d266058b9d145b50fe83737db | |
| 19:30:48 | ihrachys | clarkb, and logstash shows hits for masters of some other projects (some jobs are -nv there though) | |
| 19:31:58 | mriedem | ihrachys: logstash doesn't show that hitting on stable | |
| 19:32:08 | mriedem | novnc should only be enabled on master according to that d-g change | |
| 19:32:48 | ihrachys | mriedem, see this just failed for https://review.openstack.org/#/c/461827/ | |
| 19:32:50 | ihrachys | it's ocata | |
| 19:33:03 | ihrachys | the gate-tempest-dsvm-py35-ubuntu-xenial | |
| 19:33:31 | ihrachys | maybe there is something wrong about the job that it picks master, I dunno | |
| 19:34:07 | ihrachys | definitely enabled there: http://logs.openstack.org/27/461827/1/check/gate-tempest-dsvm-py35-ubuntu-xenial/6805498/logs/devstacklog.txt.gz#_2017-05-02_16_15_56_894 | |
| 19:34:08 | mriedem | ihrachys: huh yeah novnc is enabled http://logs.openstack.org/27/461827/1/check/gate-tempest-dsvm-py35-ubuntu-xenial/6805498/logs/local.conf.txt.gz | |
| 19:35:47 | clarkb | I think that actually enabled the feature across the board | |
| 19:35:53 | mriedem | ihrachys: https://review.openstack.org/#/c/344892/5 was ocata so maybe n-novnc is just always enabled, but not used in devstack before ocata | |
| 19:36:02 | clarkb | since its in the base nova list which is used in all the branches | |
| 19:36:08 | mriedem | ok that explains it then | |
| 19:37:04 | mriedem | oh i see logstash won't show the py35 results because of the goofy formatting in the console dump | |
| 19:37:05 | mriedem | b'AssertionError: True is not false : Token must be invalid because the connection closed.' | |
| 19:37:44 | sdague | dansmith: I think I've sorted the systemd issues that crept up, but if you find another one, let me know | |
| 19:37:48 | mriedem | there we go | |
| 19:37:53 | mriedem | ihrachys: (message:"AssertionError: True is not false : Token must be invalid because the connection closed." OR message:"b'AssertionError: True is not false : Token must be invalid because the connection closed.'") AND tags:"console" | |
| 19:38:09 | dansmith | sdague: I'm just grasping at straws about why something is behaving differently all of a sudden | |
| 19:38:18 | mriedem | dansmith: oh, | |
| 19:38:20 | mriedem | the wsgi changes? | |
| 19:38:26 | sdague | dansmith: what kind of different? | |
| 19:38:35 | mriedem | dansmith: did you see the wsgi changes merge? | |
| 19:38:41 | mriedem | i didn't think devstack was using that yet though | |
| 19:38:44 | dansmith | mriedem: ooh, which wsgi changes? | |
| 19:39:00 | dansmith | because I swear that some request id stuff is broken now | |
| 19:39:16 | mriedem | nvm didn't merge yet https://review.openstack.org/#/c/459413/ | |
| 19:39:25 | mriedem | but this one did https://review.openstack.org/#/c/457283/ | |
| 19:39:44 | sdague | mriedem: sure, but that's not getting run in the gate yet | |
| 19:39:55 | mriedem | right because of https://review.openstack.org/#/c/457715/ | |
| 19:40:11 | dansmith | ah, so | |
| 19:40:29 | dansmith | that was kindof what I was wondering.. something that kicked us out of eventlet or something | |
| 19:40:42 | sdague | nope | |
| 19:40:45 | mriedem | ihrachys: re-opened https://bugs.launchpad.net/nova/+bug/1669468 | |
| 19:40:45 | openstack | Launchpad bug 1669468 in OpenStack Compute (nova) "tempest.api.compute.servers.test_novnc.NoVNCConsoleTestJSON.test_novnc fails intermittently in neutron multinode nv job" [Medium,Confirmed] | |
| 19:41:07 | dansmith | mriedem: sdague: so I would love to lead one of you through an exercise to prove that something is broken or that I'm crazy | |
| 19:41:38 | sdague | dansmith: I'm kind of borked today, as the TC meeting is in 20, and I've got a hard stop after that. Tomorrow I'd be fine | |