| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2018-06-20 | |||
| 17:10:11 | dansmith | tssurya: are you opposed to making this unified or are you okay with either? | |
| 17:10:12 | tssurya | dansmith: ack | |
| 17:10:15 | dansmith | and which is your preference? | |
| 17:10:25 | tssurya | I am okay with either | |
| 17:10:27 | dansmith | all things equal, I mean, which do you think is better? | |
| 17:10:33 | mordred | dansmith: would we have the addresses dict since those are coming from neutron? or do those get stored in the cells db? | |
| 17:10:40 | tssurya | as long as we give at least the uuid back | |
| 17:10:57 | dansmith | mordred: we would have to hit neutron for it since our _cache_ of those values are in the cell db | |
| 17:11:07 | tssurya | like keeping the current 500 behavior if 1 out of 70 goes down is wrong for sure | |
| 17:11:19 | dansmith | tssurya: yeah okay | |
| 17:11:21 | mordred | nod. I mean - I already have code to hit neutron api if the addresses dict is empty - so I'm fine if it's not | |
| 17:11:43 | tssurya | 1) we fix this with skipping down cells as first step 2) return just the uuids with rest as empty | |
| 17:11:48 | mordred | tssurya: uuid and name (filtering on name client-side is frequent/important) | |
| 17:11:50 | tssurya | makes sense ? | |
| 17:12:17 | mordred | if it's possible | |
| 17:12:18 | dansmith | mordred: we don't have easy access to the name, but we can get it from the reqspec if you think it's important | |
| 17:12:21 | tssurya | mordred: we were thinking of skipping a cell if filters are used | |
| 17:12:40 | dansmith | yeah, if you're filtering, and we don't know the value, we assume no match | |
| 17:12:51 | tssurya | dansmith: we don't have the name in req_spec right ? | |
| 17:12:57 | dansmith | tssurya: the display name? | |
| 17:13:05 | tssurya | yes | |
| 17:13:32 | tssurya | meaning we don't have the display_name | |
| 17:13:36 | tssurya | in req_spec | |
| 17:13:52 | dansmith | oh, I guess I'm thinking of buildreq, which stores it while we're building | |
| 17:13:56 | dansmith | so yeah maybe not | |
| 17:14:31 | mordred | ah - we don't have the name? hrm. | |
| 17:15:15 | tssurya | mordred: no :( the only info we can get is uuid, flavor, image, avz, project_id | |
| 17:15:49 | dansmith | the point here is to not store the same thing in multiple places where it gets out of sync (like in cellsv1) | |
| 17:16:19 | stephenfin | zcorneli|lunch: One nit left in that that should be fixed but not blocking on it. If you could address that (and anything else you fancy fixing) in a follow up, that would be appreciated. | |
| 17:16:20 | tssurya | dansmith: yea and I agree with not adding info into top API, but we should make use of the existing info | |
| 17:16:37 | dansmith | tssurya: yeah | |
| 17:16:42 | tssurya | infact just the uuid would also be good enough with the state as UNKNOWN like you said | |
| 17:16:58 | tssurya | and we would anyways need a "odd microversion" bump | |
| 17:17:23 | mordred | hrm. well - it would make things a bit awkward for ansible dynamic inventory without name ... but if that's life, I guess that's life | |
| 17:17:56 | tssurya | mordred: :) | |
| 17:18:19 | mordred | basically it'll make operations that are working with servers that are up not able to find the server if part of the api control is down if/when those servers are being managed by name and not uuid | |
| 17:18:22 | mordred | BUT | |
| 17:18:53 | mordred | maybe that's just life - the existing system would mean you can't deal with ANY of your servers - so it's still an improvement | |
| 17:19:08 | dansmith | yeah | |
| 17:19:15 | dansmith | if we do the 1a option then it gets better | |
| 17:19:18 | mordred | yup | |
| 17:19:28 | tssurya | yea | |
| 17:19:40 | mordred | so yeah - I think 1 followed by 1a would be great | |
| 17:20:23 | dansmith | mordred: thanks for the discussion! | |
| 17:20:26 | tssurya | mordred: totally appreciate your view and opinion on this :) thanks! | |
| 17:20:49 | tssurya | dansmith: so you will write a comment on yours and mordred's behalf in the spec ? | |
| 17:20:58 | dansmith | tssurya: am now yeah | |
| 17:21:04 | tssurya | and then we can discuss it with mriedem and gibi next week | |
| 17:21:18 | tssurya | dansmith: thanks! | |
| 17:21:29 | dansmith | yup | |
| 17:29:10 | openstackgerrit | Merged openstack/nova master: Optimize member_of check for nested providers https://review.openstack.org/573203 | |
| 17:45:55 | zcorneli | stephenfin: Ack. I'll fix that up. | |
| 18:00:39 | dansmith | zcorneli: bonus points if you fix up the stuff in the devstack patch so we can merge that | |
| 18:00:45 | dansmith | if not, I'll try to get around to it at some point | |
| 18:01:13 | zcorneli | Sure, IIRC it looked pretty simple/quick to fix up the stuff mriedem was looking for. | |
| 18:01:27 | dansmith | yup | |
| 18:01:28 | zcorneli | dansmith: I've also got the discard patch almost ready to review as well. | |
| 18:01:34 | dansmith | zcorneli: excellent | |
| 18:02:05 | zcorneli | dansmith: Going to finish the fix up patch, rebase onto it, then post, I think. | |
| 18:02:14 | zcorneli | Then I can take a crack at the devstack fixup. | |
| 18:02:19 | dansmith | cool | |
| 18:11:09 | openstackgerrit | Chris Dent proposed openstack/nova master: WIP: [placement] Add /reshaper handler for POST https://review.openstack.org/576927 | |
| 18:23:02 | openstackgerrit | Merged openstack/nova master: Ensure that os-traits sync is attempted only at start of process https://review.openstack.org/553857 | |
| 18:39:45 | mnaser | so mriedem was working on a patch adding `nova-status upgrade check as part of openstack ansible deployment which is really sweet, but one of the issues we're having is detecting what action we're missing | |
| 18:39:56 | mnaser | is it possible to perhaps have a different exit code depending on which 'issue' | |
| 18:40:13 | mnaser | in this case, upgrading from queens to master (rocky), we need to restart placement | |
| 18:41:15 | mnaser | like maybe some sort of .. cli friendly output so we can run `nova-status upgrade check` once | |
| 18:41:35 | mnaser | then have conditionals depending if it returns something like out_of_date_placement => restart placement, etc | |
| 18:42:21 | dansmith | returning different statuses per check would make it unstable across releases, which is probably undesirable | |
| 18:42:38 | dansmith | how about just some sort of robot-friendly output? | |
| 18:42:57 | mnaser | yeah i was just thinking how complicated that would become at some point | |
| 18:43:10 | mnaser | but robot friendly output, something like one line per 'check' .. only if there is something | |
| 18:43:25 | mnaser | that way, we can run it once, ignore exit code, and then have tasks after it conditional if output matches that string | |
| 18:43:26 | dansmith | like --only-critical | |
| 18:43:36 | dansmith | and one per line, csv-separated or something | |
| 18:43:41 | mnaser | yes | |
| 18:43:48 | dansmith | --only-critical --robot-friendly | |
| 18:44:10 | cdent | --with-ponies | |
| 18:44:17 | mnaser | --emojis | |
| 18:44:18 | cdent | --and-pudding | |
| 18:44:23 | cdent | exactly | |
| 18:44:26 | dansmith | too far. | |
| 18:45:05 | mnaser | there is 6 warnings and 11 failures | |
| 18:45:15 | mnaser | UpgradeCheckResult | |
| 18:45:17 | mnaser | er | |
| 18:45:17 | mnaser | https://github.com/openstack/nova/blob/master/nova/cmd/status.py | |
| 18:45:34 | mnaser | UpgradeCheckResult has 'user friendly' messages but we can just add another value to it | |
| 18:46:40 | mnaser | i think ill push up a patch for that | |
| 19:05:52 | openstackgerrit | Lee Yarwood proposed openstack/nova master: Add regression test for bug #1764883 https://review.openstack.org/562072 | |
| 19:05:53 | openstack | bug 1764883 in OpenStack Compute (nova) "Evacuation fails if the source host returns while the migration is still in progress" [Undecided,In progress] https://launchpad.net/bugs/1764883 - Assigned to Lee Yarwood (lyarwood) | |
| 19:05:54 | openstackgerrit | Lee Yarwood proposed openstack/nova master: compute: Ensure pre-migrating instances are destroyed during init_host https://review.openstack.org/562284 | |
| 19:21:47 | openstackgerrit | Mohammed Naser proposed openstack/nova master: Add machine output for nova-status upgrade checks https://review.openstack.org/576944 | |
| 19:21:49 | mnaser | dansmith: ^ my first attempt.. not sure who might be best to look over it too :) | |
| 19:21:52 | mnaser | we can use that in openstack ansible | |
| 19:26:40 | mordred | mnaser: are there any projects you aren't hacking on? | |
| 19:27:37 | mnaser | mordred: :D i'm sure there is a quite a few | |
| 19:31:54 | dansmith | mnaser: mriedem is the right person :) | |
| 19:33:12 | mnaser | dansmith: cool, thanks | |
| 19:40:32 | openstackgerrit | Carlos Camacho proposed openstack/nova stable/queens: Be graceful about vif plugging in early ironic driver startup https://review.openstack.org/576948 | |
| 20:41:43 | openstackgerrit | Merged openstack/nova master: Fix nits from change Id609789ef6b4a4c745550cde80dd49cabe03869a https://review.openstack.org/574324 | |