| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2017-01-25 | |||
| 18:34:36 | openstackgerrit | melanie witt proposed openstack/nova: WIP Count resources to check quota for cells https://review.openstack.org/416521 | |
| 18:34:37 | openstackgerrit | melanie witt proposed openstack/nova: Add online migration to move quotas to API database https://review.openstack.org/410946 | |
| 18:35:02 | dansmith | johnthetubaguy: if you're still around, I'll hit these ^ if you will ack them first | |
| 18:35:08 | dansmith | you clearly understand quotas :) | |
| 19:11:13 | snikitin_ | mriedem: Hi Matt! I have two +2 on my instance diagnostics patches and I need final +W. Could you please look? https://review.openstack.org/#/c/394480/ https://review.openstack.org/#/c/355540/ | |
| 19:21:34 | bauzas | dansmith: still issues with grenade on the scheduler change even with your PS5 http://logs.openstack.org/61/417961/25/check/gate-grenade-dsvm-neutron-multinode-ubuntu-xenial/405b949/logs/new/screen-n-sch.txt.gz?level=TRACE#_2017-01-25_19_01_47_099 | |
| 19:21:44 | bauzas | will look at it later | |
| 19:23:27 | bauzas | fortunatly now the functional job works :) | |
| 19:36:39 | openstackgerrit | Ed Leafe proposed openstack/nova: placement: RT now adds proper Ironic inventory https://review.openstack.org/404472 | |
| 20:03:50 | pumaranikar | johnthetubaguy, https://review.openstack.org/#/c/412634/ keystoneauth with glance client. Addressed your comments | |
| 20:08:28 | bauzas | dansmith: confirmed that grenade doesn't work correctly with http://logs.openstack.org/30/424730/5/check/gate-grenade-dsvm-neutron-ubuntu-xenial/b39c47f/logs/new/screen-n-cpu.txt.gz#_2017-01-25_17_42_37_321 | |
| 20:09:04 | bauzas | dansmith: that's due to auth_url being wrong http://logs.openstack.org/30/424730/5/check/gate-grenade-dsvm-neutron-ubuntu-xenial/b39c47f/logs/etc/nova/nova.conf.txt.gz | |
| 20:09:15 | bauzas | fancy ! | |
| 20:09:27 | bauzas | I think I know the problem | |
| 20:10:01 | bauzas | yup, we're missing some vars https://github.com/openstack-dev/devstack/blob/master/lib/placement#L103 | |
| 20:11:32 | bauzas | I think we just need to source lib/keystone | |
| 20:13:37 | bauzas | ok, updated, let's wait the run... | |
| 20:18:48 | bauzas | wow, just wow | |
| 20:18:59 | bauzas | the gate is late by more than 4 hours | |
| 20:19:02 | bauzas | chilly | |
| 20:26:20 | openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Updated from global requirements https://review.openstack.org/424828 | |
| 21:08:38 | bauzas | damn, now a 401 :( | |
| 21:08:41 | bauzas | dansmith: ^ | |
| 21:08:48 | dansmith | argh | |
| 21:09:08 | bauzas | nova.conf looks correct http://logs.openstack.org/30/424730/6/check/gate-grenade-dsvm-neutron-ubuntu-xenial/16175d7/logs/etc/nova/nova.conf.txt.gz | |
| 21:09:32 | bauzas | but http://logs.openstack.org/30/424730/6/check/gate-grenade-dsvm-neutron-ubuntu-xenial/16175d7/logs/new/screen-n-cpu.txt.gz#_2017-01-25_20_43_30_774 | |
| 21:10:04 | bauzas | and cdent bailed out | |
| 21:10:13 | bauzas | :( | |
| 21:11:14 | bauzas | the creds look correct to me also | |
| 21:14:31 | bauzas | dansmith: okay, so when comparing a tempest nova.conf and a grenade nova.conf, both are similar | |
| 21:14:54 | bauzas | I guess the problem is that we don't correctly have the placement API running | |
| 21:29:47 | mriedem | dansmith: how goes it today? | |
| 21:30:09 | dansmith | mriedem: lots of gate failures and still some discussion over the mysql host change | |
| 21:31:08 | dansmith | mriedem: and sounds like bauzas is still struggling with the grenade change | |
| 21:31:34 | bauzas | indeed | |
| 21:31:38 | mriedem | this https://review.openstack.org/421399 ? | |
| 21:31:43 | mriedem | re: mysql host | |
| 21:31:45 | bauzas | the gate is also 4 hours off | |
| 21:32:11 | dansmith | yeah | |
| 21:32:12 | mriedem | dansmith: we can maybe just drop https://review.openstack.org/421399 | |
| 21:32:16 | bauzas | so, today is probably our last chance to get things done | |
| 21:32:31 | dansmith | mriedem: we can't because we're discovering from the subnode in the devstack multinode job right? | |
| 21:32:33 | dansmith | see my comment just now | |
| 21:32:35 | bauzas | andreykurilin: around ? | |
| 21:32:50 | andreykurilin | bauzas: yes | |
| 21:32:51 | mriedem | i couldn't remember if i needed https://review.openstack.org/421399 for how i was previously doing things in devstack | |
| 21:33:01 | bauzas | andreykurilin: where are we wiht all the client changes ? | |
| 21:33:18 | bauzas | missed a bit of scope since I'm working on another thing | |
| 21:33:44 | dansmith | mriedem: I'm not even sure what variable d-g has that address in | |
| 21:34:38 | andreykurilin | bauzas: all client changes? | |
| 21:34:52 | bauzas | andreykurilin: the fixes for 7.1.0 | |
| 21:35:09 | bauzas | for *releasing | |
| 21:35:20 | mriedem | dansmith: "Because we're calling discover_hosts from the subnode in devstack-gate" isn't actually true | |
| 21:35:21 | andreykurilin | bauzas: it is the last one - https://review.openstack.org/#/c/424256/ | |
| 21:35:29 | dansmith | mriedem: eh? | |
| 21:35:38 | mriedem | we call it from the primary node in d-g | |
| 21:36:08 | mriedem | dansmith: i'm thinking i needed the mysql host fix for something to do with https://review.openstack.org/420976 | |
| 21:36:13 | mriedem | but need to remember quick | |
| 21:36:24 | dansmith | mriedem: isn't this using ansible to run it on the subnode? https://review.openstack.org/#/c/424252/4/devstack-vm-gate.sh@239 | |
| 21:36:48 | mriedem | dansmith: no, | |
| 21:36:49 | mriedem | $ANSIBLE primary | |
| 21:36:52 | mriedem | == primary node | |
| 21:37:12 | dansmith | why via ansible then? | |
| 21:37:52 | mriedem | because that's what runs random shell stuff on the node? | |
| 21:37:56 | mriedem | and what i was told to use | |
| 21:38:08 | dansmith | mriedem: okay | |
| 21:38:23 | dansmith | mriedem: I just removed it, so we'll see | |
| 21:39:14 | openstackgerrit | Dan Smith proposed openstack/nova: WIP: Collected release notes for Ocata CellsV2 https://review.openstack.org/424737 | |
| 21:39:15 | openstackgerrit | Dan Smith proposed openstack/nova: Move instance creation to conductor https://review.openstack.org/319379 | |
| 21:39:42 | mriedem | dansmith: so i just abandoned https://review.openstack.org/#/c/421399/ | |
| 21:40:04 | mriedem | ok you broke the connection to it as a dep, good | |
| 21:40:13 | dansmith | yeah | |
| 21:41:10 | dansmith | I still think the assertion that the subnode shouldn't know the database host is based on nova-centric architecture, but happy to have one fewer thing in the dep chain | |
| 21:41:23 | mriedem | yeah basically f it and we'll sort it out later :) | |
| 21:41:35 | dansmith | yep | |
| 21:43:47 | mriedem | so the devstack change for discover_hosts should be approved at some point today too, just pinged dtroyer | |
| 21:44:07 | dansmith | I pinged him earlier | |
| 21:44:28 | dansmith | the --verbose one failed jenkins in the gate, but should be on its way again soon | |
| 21:44:41 | dansmith | check was over 300 earlier today, but looks like it has calmed down a bit, which is good | |
| 21:45:17 | mriedem | es muy triste | |
| 21:46:26 | dansmith | heh | |
| 21:47:45 | dansmith | melwitt: mriedem: should we remove mriedem's -2 on the bottom quotas patch since we're close? maybe let melwitt hold the -2 onit? | |
| 21:47:50 | dansmith | https://review.openstack.org/#/c/410945/ | |
| 21:48:20 | dansmith | actually can we go ahead and merge the bottom two? | |
| 21:49:39 | mriedem | i thought we weren't going to merge those unless the counting resources one was good | |
| 21:49:44 | mriedem | i've removed the -2 btw | |
| 21:50:17 | mriedem | b/c we don't have reservations in the api db in the bottom changes i thought | |
| 21:50:26 | dansmith | ah | |
| 21:51:19 | dansmith | the counting patch is massive | |
| 21:51:24 | dansmith | which I guess is why it's still WIP | |
| 21:51:41 | mriedem | yeah, and probably too risky at this point | |
| 21:52:00 | mriedem | i.e. i think we need more burn in time with testing for a change like that | |
| 21:52:33 | dansmith | so you're calling it? | |
| 21:52:34 | mriedem | so does that automatically mean the bottom two changes can't get in? i believe they will just fallback to the main db for reservations/usages since those won't be in the api db | |
| 21:52:37 | mriedem | well, | |
| 21:52:44 | mriedem | seems pretty damn risky at this point | |
| 21:52:53 | mriedem | is melwitt around? | |
| 21:53:23 | dansmith | from what I saw of the first two, it doesn't really change anything so I would expect the bottom ones are okay, but.. I forgot about reservations | |
| 21:53:50 | mriedem | and usages | |
| 21:53:56 | mriedem | i think those were the 2 tables that weren't being ported | |