| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2017-08-31 | |||
| 13:56:40 | mriedem | so backporting to ocata and newton will be difficult if we don't decouple those | |
| 13:56:50 | mriedem | it would be nice if the code fix patch is self-contained | |
| 14:01:49 | gibi | mriedem: would you like to have an independent regression test for that bug under tests/functional/regressions ? | |
| 14:01:59 | gibi | mriedem: that could be backportable | |
| 14:02:37 | mriedem | it's not really a regression | |
| 14:02:46 | mriedem | but yes a separate standalone test would be nice | |
| 14:02:59 | mriedem | it's just a latent bug from what i can tell | |
| 14:03:19 | mriedem | well, it was maybe regressed when the migration record stuff was changed to track evacuations, but that's pretty old now (liberty or mitaka i think) | |
| 14:03:46 | mriedem | so you could put it in tests/functional/regressions if you wanted, that's fine | |
| 14:03:50 | mriedem | just decouple from ServerMovingTests | |
| 14:06:33 | gibi | mriedem: OK, cool | |
| 14:08:03 | mriedem | i also need to look at splitting https://review.openstack.org/#/c/499399/ up a bit later today | |
| 14:08:08 | mriedem | since it's a pretty large change atm | |
| 14:10:50 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Remove deprecated 'null_kernel' opt https://review.openstack.org/499611 | |
| 14:10:51 | openstackgerrit | Stephen Finucane proposed openstack/nova master: WIP! conf: Remove deprecated 'multi_instance_display_name_template' opt https://review.openstack.org/499612 | |
| 14:17:14 | efried | where da source for the placement API reference doc? https://developer.openstack.org/api-ref/placement/ | |
| 14:17:24 | efried | found typo, wanna fix. | |
| 14:17:54 | sdague | new ironic issue - https://bugs.launchpad.net/nova/+bug/1714248 ? | |
| 14:17:55 | openstack | Launchpad bug 1714248 in OpenStack Compute (nova) "Compute node HA for ironic doesn't work due to the name duplication of Resource Provider " [Undecided,New] | |
| 14:18:59 | efried | nova project, placement-api-ref/source/parameters.yaml? | |
| 14:19:09 | sdague | efried: yep | |
| 14:19:14 | efried | thx | |
| 14:19:25 | efried | starts_with => startswith, according to my playing around. | |
| 14:19:50 | efried | ...for GET /traits | |
| 14:21:24 | sdague | efried: cool | |
| 14:21:45 | sdague | the names of parameters is in the rst, so depending you might only need to change that side, not that parameters.yaml | |
| 14:22:23 | efried | sdague rgrep showed starts_with only in that yaml file, so... | |
| 14:23:00 | openstackgerrit | Matt Riedemann proposed openstack/nova stable/pike: [placement] Add test for empty resources in allocation https://review.openstack.org/499615 | |
| 14:23:01 | openstackgerrit | Matt Riedemann proposed openstack/nova stable/pike: [placement] Require at least one resource class in allocation https://review.openstack.org/499616 | |
| 14:23:21 | bauzas | mriedem: I'm still not at full steam, but how can I help by reviewing changes ? | |
| 14:23:38 | bauzas | mriedem_away: the above gibi's changes ? | |
| 14:24:23 | gibi | bauzas: that change need some work based on the above discussion to make it backportable | |
| 14:24:32 | bauzas | k | |
| 14:26:12 | efried | The list of links returned by GET /resource_providers doesn't have one for .../allocations. I think it should (because that URI responds). LP bug? | |
| 14:30:30 | gibi | bauzas: mriedem_away has a refactor here https://review.openstack.org/#/c/498596/ that needs a +2 | |
| 14:31:04 | gibi | bauzas: and some more test coverage here https://review.openstack.org/#/c/496202 that also needs some core love | |
| 14:31:30 | bauzas | gibi: I need to look at what you added during my 2.5wks PTO :) | |
| 14:32:01 | gibi | bauzas: I had my one week PTO during that time so not that much but others were busy | |
| 14:37:52 | edleafe | efried: that sounds reasonable, but I would wait for cdent to chime in. It may be an oversight, or there might be a good reason. | |
| 14:38:03 | edleafe | efried: my money's on an oversight | |
| 14:38:05 | edleafe | :) | |
| 14:38:49 | efried | edleafe Rgr. Found the spot in the code where it would be added. I think the per-microversion link splitup was done before the allocations microversion. | |
| 14:39:12 | edleafe | efried: that would make sense | |
| 14:39:26 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Remove deprecated 'allow_instance_snapshots' opt https://review.openstack.org/499621 | |
| 14:39:29 | efried | Will confirm, open bug, and if it's intentional, cdent can just close the sucker. | |
| 14:43:53 | efried | nope, https://review.openstack.org/#/c/366789/ merged 20160909; https://review.openstack.org/#/c/468923/ merged 20170606. Both authored by cdent. So... maybe intentional. | |
| 14:45:34 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Move additional nova-net opts to 'network' https://review.openstack.org/499168 | |
| 14:47:47 | efried | edleafe FYI: https://bugs.launchpad.net/nova/+bug/1714275 | |
| 14:47:49 | openstack | Launchpad bug 1714275 in OpenStack Compute (nova) "GET /resource_providers: "links" doesn't include "allocations"" [Undecided,New] | |
| 14:49:54 | edleafe | efried: cool. I'm sure cdent will contribute his 2 cents when he is back on the ground | |
| 14:51:12 | dansmith | edleafe: are you going to be in denver? know about cdent? | |
| 14:51:53 | edleafe | dansmith: we both are | |
| 14:51:59 | dansmith | okay cool | |
| 14:55:03 | gibi | mriedem_away: I added my subteam update to the nova meeting agenda as I cannot joint today | |
| 15:02:21 | kashyap | melwitt: Following up from yesterday, and as your test proved, re-defining from live domain is expected to work - Eric, the author of the API confirms | |
| 15:10:35 | mriedem | gibi: ok | |
| 15:11:58 | openstackgerrit | Eric Fried proposed openstack/nova master: Update placement docs: api-ref link, startswith https://review.openstack.org/499635 | |
| 15:12:07 | efried | sdague FYI ^ | |
| 15:12:37 | efried | mriedem backport potential? ^ | |
| 15:12:46 | efried | (test out that gerritbot update :) | |
| 15:13:39 | openstackgerrit | Bob Ball proposed openstack/nova master: XenAPI: Unit tests must mock os_xenapi calls https://review.openstack.org/499573 | |
| 15:13:59 | sdague | efried: api-ref doesn't need backporting | |
| 15:14:04 | sdague | only master is published | |
| 15:14:22 | efried | right, okay | |
| 15:14:53 | efried | sdague But the rst... | |
| 15:15:12 | sdague | efried: what about it? | |
| 15:15:21 | efried | could be backported | |
| 15:15:49 | efried | to provide the link for readers @pike level | |
| 15:16:44 | sdague | efried: oh, in that case you probably want to split the 2 up | |
| 15:16:49 | sdague | do the parameter fix first | |
| 15:16:55 | sdague | then do the rst separate | |
| 15:17:01 | sdague | because we don't backport api-ref | |
| 15:17:18 | efried | And we can't have a separate thing that just drops to pike? | |
| 15:17:29 | efried | has to be a backport? | |
| 15:17:37 | sdague | efried: you could modify the backport | |
| 15:17:43 | sdague | and drop that field | |
| 15:17:48 | sdague | sorry, that file | |
| 15:17:57 | sdague | it's just a cleaner backport if it was 2 patches | |
| 15:17:58 | efried | Yeah, wfm. | |
| 15:18:12 | efried | I'll split it up. And open a bug, cause that's required to backport, yah? | |
| 15:19:04 | sdague | efried: sure | |
| 15:25:03 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Add recreate test for forced host evacuate not setting dest allocations https://review.openstack.org/499678 | |
| 15:28:46 | openstackgerrit | Eric Fried proposed openstack/nova master: [placement] api-ref GET /traits name:startswith https://review.openstack.org/499682 | |
| 15:29:05 | mriedem | gibi: replied in https://review.openstack.org/#/c/499399/ - basically, you're correct, but this is for backporting to pike, and we don't really support things like shared resource providers yet anyway | |
| 15:29:25 | mriedem | i think when we do, we remove this code and have conductor call the scheduler to sort that all out during the move operation | |
| 15:31:34 | openstackgerrit | Eric Fried proposed openstack/nova master: [placement] Update user doc with api-ref link https://review.openstack.org/499635 | |
| 15:37:11 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Create allocations against forced dest host during evacuate https://review.openstack.org/499399 | |
| 15:45:51 | mriedem | gah, why do we have nova.tests.unit.scheduler.test_utils and nova.tests.unit.scheduler.test_scheduler_utils?! | |
| 15:47:49 | artom | Back when I worked web hosting support, I reeeaaally wanted to answer those type of why questions ("why is server down?") with "because god hates you." | |
| 15:49:37 | vdrok | good morning folks! in ironic, we sometimes hit the issue with one of the smoke tests in tempest not being able to boot an instance on a specific host as RescheduledException happens. the reason seems to be this piece of code https://review.openstack.org/#/c/499545/2/nova/scheduler/utils.py. Is there a better way to achieve this kind of thing? as this does not look pretty | |
| 15:50:13 | openstackgerrit | Lucian Petrut proposed openstack/nova master: HyperV: Perform proper cleanup after failed instance spawns https://review.openstack.org/499690 | |
| 15:51:32 | vdrok | basically, for ironic it would be good to distinguish cases when len(force_hosts)==1 and still do reschedules in this case, as there are multiple nodes assigned to the host | |
| 15:54:14 | mriedem | 1:M :( | |
| 15:54:35 | mriedem | plus forcing anything | |
| 15:55:07 | vdrok | basically the failing test is https://github.com/openstack/tempest/blob/master/tempest/scenario/test_server_multinode.py#L40, we could skip it. but it kinda helps to ensure that hash ring works | |
| 15:57:06 | bauzas | mriedem: FWIW, I'm writing a new spec for changing the boot request to be using the requested_destination flag | |
| 15:57:16 | bauzas | mriedem: and no "force" flag to be used | |
| 15:57:39 | bauzas | mriedem: so we could just send force_hosts in hell | |
| 15:57:45 | bauzas | that said, there is a flaw | |
| 15:58:19 | bauzas | since live-migrate and evacuate only accept a single string for passing a target, the usual tuple (host,node) isn't possible | |