| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2021-10-19 | |||
| 14:56:49 | gibi | that will redirect | |
| 14:56:53 | gibi | to a zoom call | |
| 14:57:13 | dansmith | thanks, can we put that in the etherpad or is it secret? | |
| 14:57:16 | gibi | sure | |
| 14:57:17 | gibi | I can | |
| 14:57:27 | dansmith | thanks | |
| 14:57:30 | gibi | but it is already on the https://ptg.opendev.org/ptg.html page | |
| 14:57:54 | dansmith | okay, it's just easier to jump back and forth if the etherpads have them | |
| 14:58:42 | gibi | sure added it L49 | |
| 14:59:10 | dansmith | sweet thanks | |
| 15:00:01 | hgy | Is it begin? | |
| 15:00:20 | gibi | hgy: you are in the good place if you are for the rbac discussion | |
| 15:00:30 | gibi | hgy: we will switch to that topic soon | |
| 15:01:13 | hgy | I almost missed it. I thought it was 16:00 | |
| 15:01:16 | hgy | Thank you | |
| 15:25:06 | hgy | 1 | |
| 15:26:57 | gibi | .11 | |
| 15:31:53 | bauzas | I'm lost in translation too | |
| 15:32:14 | bauzas | wondering what to start with the rbac thing :) | |
| 15:51:40 | stephenfin | bauzas: can you mute? | |
| 15:51:44 | dansmith | bauzas: mute | |
| 15:52:28 | bauzas | done and sorry | |
| 15:52:33 | stephenfin | ta | |
| 15:52:43 | bauzas | I was about to mention we have 8 mins and we somehow need to wrap up | |
| 15:52:48 | bauzas | 8 min left* | |
| 15:53:28 | dansmith | bauzas: yeah I have to go to another session at the top of the hour now | |
| 15:53:45 | bauzas | I'll use my stick then | |
| 15:54:54 | artom | "super interesting discussion" | |
| 15:55:00 | artom | gdi the contempt is just dripping | |
| 15:55:12 | artom | :D | |
| 16:18:34 | belmoreira | bauzas it's getting late for me... I will need to leave in 5 min. I see that melwitt is OK with the change in the vncproxy. Thank you all. | |
| 16:18:46 | bauzas | belmoreira: cool | |
| 16:19:24 | belmoreira | regarding the instance move between projects, I guess it will be difficult... there is a lot of complexity | |
| 16:19:29 | belmoreira | see you tomorrow | |
| 16:21:07 | ralonsoh | sean-k-mooney, hi! https://review.opendev.org/c/openstack/nova/+/807359 | |
| 16:21:12 | ralonsoh | if you have time, trivial patch | |
| 16:22:09 | sean-k-mooney | ya i can take a look it looks correct | |
| 16:22:14 | ralonsoh | thanks! | |
| 16:25:52 | artom | It's going to be hard to concentrate on the substance with that backdrop! | |
| 16:36:24 | gibi | bauzas: I lost zoom | |
| 16:36:26 | stephenfin | bauzas: did you kill Zoom again? | |
| 16:36:26 | bauzas | NOT ME ! | |
| 16:36:37 | sean-k-mooney | sure it wasnt :P | |
| 16:36:39 | stephenfin | oh, it's gmann | |
| 16:36:40 | artom | bauzas, there are better ways to censor me ;) | |
| 16:36:42 | stephenfin | he was host | |
| 16:36:46 | melwitt | it said "the host left" | |
| 16:36:59 | bauzas | nah it was gmann :) | |
| 16:37:11 | sean-k-mooney | so can we restart the same one | |
| 16:37:33 | gibi | it is back | |
| 16:37:40 | stephenfin | https://www.openstack.org/ptg/rooms/newton | |
| 16:37:59 | stephenfin | for anyone rejoining (atom, melwitt, ...?) ^ | |
| 16:40:28 | gmann | oh, did i end meeting? | |
| 16:41:13 | gmann | bauzas: so sorry about that. i mean to leave meeting and join horizon. sorry for mesh up | |
| 16:41:25 | bauzas | gmann: heh no worries at all :p | |
| 16:41:46 | bauzas | gmann: you helped me shutting artom's mouth when he was having an argument with me, appreciated :p | |
| 16:41:56 | stephenfin | bauzas: you're unmuted again | |
| 16:41:56 | bauzas | (just kidding) | |
| 16:42:00 | stephenfin | and smiling like a maniac :P | |
| 16:42:00 | gmann | :) | |
| 16:42:13 | stephenfin | stop teasing gmann :P | |
| 17:31:29 | opendevreview | Jan Hartkopf proposed openstack/nova master: ensure samples folder exists for microversion https://review.opendev.org/c/openstack/nova/+/813672 | |
| 17:33:48 | melwitt | sean-k-mooney: sorry if I missed it but do the review priority labels prevent someone from +1ing or +2ing their own patch? I'm thinking of gibi's point that +1 shouldn't be used as a ping | |
| 17:34:43 | melwitt | (that's how the old manual etherpad process worked, you can only raise attention to someone else's patch) | |
| 17:38:16 | gmann | melwitt: sean-k-mooney i am wondering how to do that in project-config acl. I have not seen exclusive list we can n any label. or you are saying just to document that as guidlines ? | |
| 17:38:34 | gmann | *can add in any label | |
| 17:41:27 | melwitt | gmann: I don't know if it's possible to exclude self in a label, that's what I was asking sean :) is (1) is it possible? and (2) would we want to do that? | |
| 17:42:15 | gmann | ok | |
| 19:36:12 | opendevreview | Ilya Popov proposed openstack/nova master: Fix to use NUMA cell with more free memory first https://review.opendev.org/c/openstack/nova/+/805649 | |
| #openstack-nova - 2021-10-20 | |||
| 00:01:39 | opendevreview | Merged openstack/nova stable/ussuri: Fix request path to query a resource provider by uuid https://review.opendev.org/c/openstack/nova/+/805004 | |
| 00:31:30 | melwitt | stephenfin: I noticed that when I try to run a subset of unit tests for db like "tox -epy38 nova.tests.unit.db" the tests fail with "oslo_db.sqlalchemy.enginefacade.AlreadyStartedError: this TransactionFactory is already started". I assume it's related to the db migration stuff but I don't understand how. any ideas? | |
| 00:32:27 | sean-k-mooney[m] | isnt that the error we fixed in the past with the run once decorator | |
| 00:32:53 | sean-k-mooney[m] | when it was reinitalised after SIG_HUP | |
| 00:33:19 | sean-k-mooney[m] | im not really sure why you would see that in the unit tests however | |
| 00:34:30 | melwitt | sean-k-mooney[m]: yeah but this is unit tests? when you run all tests everything passes, when you run only nova.tests.unit.db you get 70+ failures ¯\_(ツ)_/¯ | |
| 00:34:47 | sean-k-mooney[m] | melwitt regarding self +1 ill check i think it might be possible to do yes by defining on +0 as allowed for the owner | |
| 00:35:24 | sean-k-mooney[m] | so the owner of the patch would be allowed to clear the review priorty but not set it to +1 or +2 | |
| 00:35:41 | melwitt | sean-k-mooney[m]: ack. I brought it up as a possible option for making sure the +1 doesn't get used as a "ping" since some were concerned about that | |
| 00:36:31 | sean-k-mooney[m] | ill see if i can create a draft patch tomorow and include that and we can discuss it in the review | |
| 00:37:03 | melwitt | cool | |
| 00:37:19 | sean-k-mooney[m] | regarding the db failure i have not run them in a few days but i can try it in the morning | |
| 00:37:27 | sean-k-mooney[m] | is it only when you filter | |
| 00:37:33 | sean-k-mooney[m] | or do they fail if you run all them | |
| 00:37:34 | melwitt | yes only when filtering | |
| 00:37:41 | sean-k-mooney[m] | ok that is odd | |
| 00:37:44 | melwitt | everything passes if you run all | |
| 00:37:57 | sean-k-mooney[m] | im not sure why filtering would break them | |
| 00:38:28 | melwitt | me neither. it's weird | |
| 00:38:48 | sean-k-mooney[m] | maybe stephenfin will spot something | |
| 00:39:01 | melwitt | it might only be just me but I tried wiping everything, upgrading tox, re-cloning the nova repo | |
| 00:39:26 | sean-k-mooney[m] | i wonder if there was an sqlalchemy release or soemthing that has chnaged behaivor | |
| 00:52:41 | sean-k-mooney[m] | so ya just quickly ssh to my home server cloned nova and ran the db test and ya it failed for me too | |
| 00:53:01 | sean-k-mooney[m] | i only got 16 failure but it proably not deterministic | |
| 00:53:19 | sean-k-mooney[m] | its likely differnt based on the number of cores/parallel tests we have | |
| 00:53:52 | sean-k-mooney[m] | the server i ran it on had 48 threads so there will be less tests running in the same process since it will use more of them | |
| 00:54:31 | melwitt | oh, yeah ok | |
| 00:55:21 | sean-k-mooney[m] | looks like its coming direclty form the db fixture which is od | |
| 00:55:24 | sean-k-mooney[m] | *odd | |
| 00:55:30 | melwitt | yeah | |
| 00:56:12 | melwitt | I have run tests like this subset a lot of times in the past and this is the first time I see it fail like this. so I figure it must be a fairly recent change | |