| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2018-05-04 | |||
| 16:40:22 | mriedem | \o/ | |
| 16:40:30 | jgwentworth | yeah, seriously | |
| 16:40:44 | mriedem | i of course will require jmccarthy to test the patch in his setup | |
| 16:41:00 | jgwentworth | oh, because you have to use it to do the remove_snap | |
| 16:41:27 | mriedem | yup and that's pointed out in https://bugs.launchpad.net/nova/+bug/1666831 | |
| 16:41:27 | openstack | Launchpad bug 1666831 in OpenStack Compute (nova) ocata "Nova recreates instance directory after migration/resize" [Low,Fix committed] - Assigned to Lee Yarwood (lyarwood) | |
| 16:41:34 | mriedem | "root_disk is used to remove rdb snapshots, but during execution of self.image_backend.by_name() nova recreates instance directory." | |
| 16:41:50 | jgwentworth | so how can we remove_snap without the imagebackend object? | |
| 16:41:53 | mriedem | so in his case, he wasn't even using rbd, he was using qcow2 | |
| 16:42:01 | mriedem | we can, we'll get it if the backend supports clone | |
| 16:42:11 | mriedem | remove_snap is only implemented for image backends that support clone | |
| 16:42:15 | jgwentworth | ohhhh k | |
| 16:42:15 | mriedem | and that's only rb | |
| 16:42:17 | mriedem | *rbd | |
| 16:43:29 | jgwentworth | yeah, makes sense | |
| 16:44:45 | openstackgerrit | Merged openstack/nova master: Convert websocketproxy to use db for token validation https://review.openstack.org/333990 | |
| 16:45:53 | mriedem | and because i added that tempest test, we should be testing a resize of a volume-backed instance on shared local storage since we have the NFS job in nova's experimental queue | |
| 16:45:55 | mriedem | plus CEPH | |
| 16:48:05 | jgwentworth | *mind blown* | |
| 16:54:19 | jmccarthy | back | |
| 17:07:58 | jmccarthy | mriedem: I can test any stuff no probs ! It might take a little while, but hopefully not too long | |
| 17:12:09 | openstackgerrit | Matt Riedemann proposed openstack/nova master: libvirt: check image type before removing snapshots in _cleanup_resize https://review.openstack.org/566367 | |
| 17:12:10 | mriedem | jmccarthy: jgwentworth: ^ | |
| 17:12:21 | jmccarthy | mriedman: Nice ! | |
| 17:12:37 | jmccarthy | mriedman: Let me see about working that in | |
| 17:21:40 | openstackgerrit | Merged openstack/nova master: Add multi-cell negative test for cold migration with target host https://review.openstack.org/524027 | |
| 17:21:49 | openstackgerrit | Merged openstack/nova master: Update layout docs for running console proxies https://review.openstack.org/557489 | |
| 17:37:54 | openstackgerrit | Matt Riedemann proposed openstack/nova master: libvirt: remove old rbd snapshot removal error handling https://review.openstack.org/566369 | |
| 17:38:44 | openstackgerrit | Matt Riedemann proposed openstack/nova master: libvirt: remove old rbd snapshot removal error handling https://review.openstack.org/566369 | |
| 17:39:00 | openstackgerrit | Jay Pipes proposed openstack/nova master: rework how we pass candidate request information https://review.openstack.org/566166 | |
| 17:39:17 | idlemind | jgwentworth thx i'm using lvm via iscsi (simple stuff for now) | |
| 17:49:17 | jmccarthy | mriedman: Ok I thought this might happen - I'm going to need a while longer, unfortunately it's a long weekend here - I'll update the bug asap but it may not be until Tuesday | |
| 17:51:23 | jmccarthy | mriedman: I appreciate your quick efforts ! I'm going to keep at it another while | |
| 18:15:23 | mriedem | jmccarthy: sure np | |
| 18:15:35 | pipesinpain | mriedem, jgwentworth, johnthetubaguy, alex_xu: your eyeballs on https://review.openstack.org/#/c/565565/ would be appreciated. | |
| 18:21:06 | arvindn05 | mriedem: can you look over the spec amendment? https://review.openstack.org/#/c/560718/ | |
| 18:21:36 | arvindn05 | once its approved, hoping the scheduler patch can be upstreamed | |
| 18:53:05 | jgwentworth | pipesinpain: ack, it's on my list | |
| 18:59:15 | openstackgerrit | Chris Dent proposed openstack/nova master: Optional separate database for placement API https://review.openstack.org/362766 | |
| 18:59:16 | openstackgerrit | Chris Dent proposed openstack/nova master: Isolate placement database config https://review.openstack.org/541435 | |
| 18:59:17 | openstackgerrit | Chris Dent proposed openstack/nova master: WIP: Ensure that os-traits sync is attempted only at start of process https://review.openstack.org/553857 | |
| 19:00:17 | openstackgerrit | Chris Dent proposed openstack/nova master: WIP: Add PLACEMENT_DB_ENABLED=True to the nova-next job https://review.openstack.org/564067 | |
| 19:06:45 | pipesinpain | jgwentworth: thx Melanie. | |
| 19:07:27 | openstackgerrit | Merged openstack/nova master: Base test module/class for functional placement db https://review.openstack.org/564590 | |
| 19:19:51 | eandersson | mriedem, I was gonna do a quick pull request to change the log "Successfully synced instances from host '%s'." to DEBUG, but noticed that it's still doing log translation | |
| 19:20:05 | eandersson | Can I remove the log translation in the same commit? | |
| 19:24:12 | mriedem | eandersson: yes | |
| 19:24:18 | mriedem | we don't translate logs anymore | |
| 19:24:38 | eandersson | Can I remove the log translation for all the entries in that file? :D or will that make it too difficult to see what changed | |
| 19:24:51 | eandersson | I can also follow up with a new pull request for that :D | |
| 19:25:39 | mriedem | i'd keep those separate | |
| 19:27:48 | openstackgerrit | Erik Olof Gunnar Andersson proposed openstack/nova master: Changing scheduler sync event from INFO to DEBUG https://review.openstack.org/566392 | |
| 19:29:38 | eandersson | I don't even want to know how much disk space is wasted on that one log line :D | |
| 19:31:13 | mriedem | easiest -1 ever | |
| 19:31:28 | eandersson | haha | |
| 19:31:29 | eandersson | omg | |
| 19:31:39 | eandersson | was too fixated on the translation | |
| 19:31:48 | eandersson | good, get to fix that missing e in computes | |
| 19:32:20 | kashyap | mriedem: This is largely code deletion, should be easy for you: https://review.openstack.org/#/c/565242/ | |
| 19:32:27 | openstackgerrit | Erik Olof Gunnar Andersson proposed openstack/nova master: Changing scheduler sync event from INFO to DEBUG https://review.openstack.org/566392 | |
| 19:33:59 | mriedem | eandersson: off the top of your head, what's the average number of cpus in your compute hosts? | |
| 19:34:01 | mriedem | 16? 32? | |
| 19:35:03 | eandersson | 32 probably | |
| 19:35:23 | mriedem | ok. so you wouldn't run something like 132 concurrent live migrations on a compute host like that would you. | |
| 19:35:28 | mriedem | 32 * 5 | |
| 19:35:53 | eandersson | Yea unlikely | |
| 19:36:06 | mriedem | cool, fyi https://docs.python.org/3.5/library/concurrent.futures.html#concurrent.futures.ThreadPoolExecutor | |
| 19:39:21 | mriedem | your logging change is sane btw, check out a 24 hour CI run for that log message http://logstash.openstack.org/#dashboard/file/logstash.json?query=message%3A%5C%22Successfully%20synced%20instances%20from%20host%5C%22%20AND%20tags%3A%5C%22screen-n-sch.txt%5C%22&from=1d | |
| 19:39:31 | mriedem | ~16K hits | |
| 19:55:26 | openstackgerrit | Jay Pipes proposed openstack/nova master: Add columns for generation to consumers https://review.openstack.org/557958 | |
| 19:55:27 | openstackgerrit | Jay Pipes proposed openstack/nova master: placement: add Project, User and Consumer objects https://review.openstack.org/565403 | |
| 19:55:28 | openstackgerrit | Jay Pipes proposed openstack/nova master: Add create() methods to Project, User and Consumer https://review.openstack.org/565404 | |
| 19:55:29 | openstackgerrit | Jay Pipes proposed openstack/nova master: move consumer ensure to API layer https://review.openstack.org/565405 | |
| 19:55:30 | openstackgerrit | Jay Pipes proposed openstack/nova master: remove Allocation.project_id & Allocation.user_id https://review.openstack.org/565406 | |
| 19:55:31 | openstackgerrit | Jay Pipes proposed openstack/nova master: rework allocation handler _allocations_dict() https://review.openstack.org/565407 | |
| 19:55:32 | openstackgerrit | Jay Pipes proposed openstack/nova master: Add a microversion for consumer generation support https://review.openstack.org/565604 | |
| 20:01:21 | openstackgerrit | Artom Lifshitz proposed openstack/nova master: WIP: Add InstanceNUMATopology to LibvirtLiveMigrateData https://review.openstack.org/566398 | |
| 20:03:50 | openstackgerrit | Artom Lifshitz proposed openstack/nova master: WIP: Add InstanceNUMATopology to LibvirtLiveMigrateData https://review.openstack.org/566398 | |
| 20:15:30 | arvindn05 | mriedem: question on tests cases within conductor | |
| 20:17:11 | arvindn05 | why are we running few tests multiple times? ConductorTaskAPITestCase,ConductorTaskRPCAPITestCase extend from _BaseTaskTestCase and test_compute.BaseTestCase | |
| 20:17:11 | arvindn05 | ||
| 20:18:45 | arvindn05 | they both run test_unshelve_instance_on_host for example...which is defined under the BaseTaskTestCase. i dont see any differences in the test setup either... | |
| 20:19:54 | mriedem | one hits the rpcapi and one doesn't | |
| 20:23:34 | arvindn05 | ahh...noticed the difference in setups now...the difference is very subtle...will add a line comment on top of self.conductor so it draws the difference for future contributors | |
| 20:25:52 | arvindn05 | btw i investigated further on code changes...wanted to confirm something with you | |
| 20:26:26 | arvindn05 | mriedem: i would need to add an else condition here as well correct? to handle rebuild with image remaining the samehttps://github.com/openstack/nova/blob/master/nova/conductor/manager.py#L913 | |
| 20:29:08 | arvindn05 | see http://paste.openstack.org/show/720407/ for how i think the code would need to look....let me know if this is not consistent with your idea and i can rework as needed | |
| 20:31:08 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Use ThreadPoolExecutor for max_concurrent_live_migrations https://review.openstack.org/563505 | |
| 20:31:09 | mriedem | superdan: huzzah ^ that's not too bad | |
| 20:31:29 | mriedem | arvindn05: i said else after the elif here https://github.com/openstack/nova/blob/master/nova/conductor/manager.py#L943 | |
| 20:31:46 | mriedem | arvindn05: idk, do we need to re-validate the host/image for rebuild if the image doesn't change? | |
| 20:32:02 | mriedem | i guess that's just to catch a case that the traits on the compute node resource provider changed even though the image didn't? | |
| 20:32:38 | arvindn05 | yup...melwitt suggested we use the same approach as imageproperties filter....which i beleive has the same behavior | |
| 20:32:46 | arvindn05 | we run even though image hasnt changed | |
| 20:34:14 | arvindn05 | i added the else as well @943 see http://paste.openstack.org/show/720407 - line58 | |
| 20:35:26 | arvindn05 | briefly tested the approach...it works...adding more unit tests | |
| 20:36:38 | superdan | mriedem: looks weird to index by migration uuid instead of instance, is that because of how the cancel migration call works? | |
| 20:37:56 | mriedem | i think it could go either way, live_migration_abort gets the instance and migration | |
| 20:38:22 | mriedem | kevin just doesn't have that patch up yet | |
| 20:39:35 | superdan | okay I'll make a half-assed comment about it | |