Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2019-05-29
20:52:10 sean-k-mooney fun fact. resize between flavor with different size hugepages is broken
20:53:01 sean-k-mooney mainly becasue the nova.virt.hardware modeuld does not know what a rebuild or resize is and we end up using the page size form the embed instance numa toplogy instead of the new flavor/image
21:14:41 melwitt sean-k-mooney: reading through the full comments on https://bugs.launchpad.net/nova/+bug/1829062 it doesn't look like it's the same issue as the pause/resume rabbit reconnecting thing we were talking about
21:14:43 openstack Launchpad bug 1829062 in StarlingX "nova placement api non-responsive due to eventlet error" [Critical,In progress] - Assigned to Gerry Kopec (gerry-kopec)
21:15:30 mriedem melwitt: replies to your replies in https://review.opendev.org/#/c/638073/31 but +W - some of the stuff in there could go into your FUP
21:15:31 melwitt I would expect their wsgi config is default threads=1
21:16:39 melwitt mriedem: yaaaaaay thanks. will add to the FollowUP
21:16:53 sean-k-mooney melwitt: the fact that its complaining about resuming on a different thread implies they have treads!=1
21:17:55 sean-k-mooney but part of the annoyance it that placmeent has never used eventlet and we only started monkeypatching it in stien after extracting placment
21:18:06 melwitt sean-k-mooney: o. yeah
21:18:30 sean-k-mooney so people running runnign placment under the non extract version are running a config that we really have not tested well at all
21:18:41 sean-k-mooney since the gate uses extracted placement
21:19:04 melwitt right
21:19:25 melwitt ok, I'll throw another comment onto that bug to clarify the wsgi config thing
21:19:47 melwitt I'm trying to write the reno also
21:19:52 sean-k-mooney placement is written assuming eventlet is not used so there might be dragons there wen run under eventlet in general
21:20:03 sean-k-mooney ack
21:20:22 melwitt this whole issue makes me :*****(
21:21:02 melwitt have two paragraphs so far
21:21:09 sean-k-mooney there are too many ways to fix it all of which suck for different reasons
21:25:19 mriedem tonyb: this is all yours https://review.opendev.org/#/c/655692/
21:26:06 tonyb[m] mriedem: Okay I
21:26:14 tonyb[m] 'll look after the school run
21:27:03 aspiers efried: you still around?
21:27:10 efried y
21:27:22 aspiers turns out that placement's inventory policy hardcodes a minimum of 1, not 0, for resource class inventory :-/
21:27:35 aspiers for MEM_ENCRYPTION_CONTEXT, this is clearly a problem
21:28:14 efried why?
21:28:23 aspiers there are even explicit tests for this https://opendev.org/openstack/placement/src/branch/master/placement/tests/functional/gabbits/allocations.yaml#L107
21:28:37 aspiers because hosts which can't do SEV need to be inventoried as 0 units
21:29:02 efried the test you pointed to is for an *allocation* requesting zero units of a thing.
21:29:15 aspiers yes, and it expects a 400 failure
21:29:23 efried I don't see the issue here.
21:29:56 aspiers the issue is that if I lower the minimum from 1 to 0 then these tests fail
21:30:04 efried "lower the minimum" - of what?
21:30:06 aspiers in the schema
21:30:14 efried argh, in *what* schema?
21:30:23 aspiers https://opendev.org/openstack/placement/src/branch/master/placement/schemas/inventory.py#L29
21:30:49 aspiers without that, MEM_ENCRYPTION_CONTEXT can't be inventoried as 0
21:31:01 efried If you can't do SEV, you don't put up any inventory of MEM_ENC_CTX. If you don't want SEV, you don't include MEM_ENC_CTX in your GET /a_c request or consequently in your allocation.
21:31:06 aspiers maybe we can tweak the schema to treat this resource class differently
21:31:08 efried You never inventory a thing with 0 total
21:31:09 efried no
21:31:13 efried do not muck with the schema
21:31:13 aspiers ohhh I see
21:31:16 efried the schema is right
21:31:28 aspiers OK, so just don't include it in the inventory
21:31:30 efried right
21:31:37 aspiers sounds good, I'll try that. thanks!
21:31:40 efried ++
21:32:01 sean-k-mooney aspiers: ya you only create an invetory if there is capasity to consume
21:32:16 sean-k-mooney so if a host does not support sev dont create an sev inventory on the RP
21:34:06 openstackgerrit melanie witt proposed openstack/nova stable/stein: Add reno about nova-api eventlet monkey-patching and rabbitmq https://review.opendev.org/662095
21:35:08 openstackgerrit Matt Riedemann proposed openstack/nova master: Fix hard-delete of instance with soft-deleted referential constraints https://review.opendev.org/661398
21:35:09 openstackgerrit Matt Riedemann proposed openstack/nova master: Update usage in RT.drop_move_claim during confirm resize https://review.opendev.org/641806
21:35:09 openstackgerrit Matt Riedemann proposed openstack/nova master: Robustify attachment tracking in CinderFixtureNewAttachFlow https://review.opendev.org/658904
21:35:10 openstackgerrit Matt Riedemann proposed openstack/nova master: Add Migration.cross_cell_move and get_by_uuid https://review.opendev.org/614012
21:35:10 openstackgerrit Matt Riedemann proposed openstack/nova master: Add InstanceAction/Event create() method https://review.opendev.org/614036
21:35:11 openstackgerrit Matt Riedemann proposed openstack/nova master: Add Instance.hidden field https://review.opendev.org/631123
21:35:11 openstackgerrit Matt Riedemann proposed openstack/nova master: Add TargetDBSetupTask https://review.opendev.org/627892
21:35:12 openstackgerrit Matt Riedemann proposed openstack/nova master: Add CrossCellMigrationTask https://review.opendev.org/631581
21:35:12 openstackgerrit Matt Riedemann proposed openstack/nova master: Execute TargetDBSetupTask https://review.opendev.org/633853
21:35:13 openstackgerrit Matt Riedemann proposed openstack/nova master: Add prep_snapshot_based_resize_at_dest compute method https://review.opendev.org/633293
21:35:13 openstackgerrit Matt Riedemann proposed openstack/nova master: Add PrepResizeAtDestTask https://review.opendev.org/627890
21:35:14 openstackgerrit Matt Riedemann proposed openstack/nova master: Add prep_snapshot_based_resize_at_source compute method https://review.opendev.org/634832
21:35:14 openstackgerrit Matt Riedemann proposed openstack/nova master: Add nova.compute.utils.delete_image https://review.opendev.org/637605
21:35:15 openstackgerrit Matt Riedemann proposed openstack/nova master: Add PrepResizeAtSourceTask https://review.opendev.org/627891
21:35:15 openstackgerrit Matt Riedemann proposed openstack/nova master: Refactor ComputeManager.remove_volume_connection https://review.opendev.org/642183
21:35:16 openstackgerrit Matt Riedemann proposed openstack/nova master: Add power_on kwarg to ComputeDriver.spawn() method https://review.opendev.org/642590
21:35:16 openstackgerrit Matt Riedemann proposed openstack/nova master: Add finish_snapshot_based_resize_at_dest compute method https://review.opendev.org/635080
21:35:17 openstackgerrit Matt Riedemann proposed openstack/nova master: Add FinishResizeAtDestTask https://review.opendev.org/635646
21:35:28 openstackgerrit Matt Riedemann proposed openstack/nova master: Add archive_deleted_rows wrinkle to cross-cell functional test https://review.opendev.org/651650
21:35:29 openstackgerrit Matt Riedemann proposed openstack/nova master: Add CrossCellWeigher https://review.opendev.org/614353
21:35:29 openstackgerrit Matt Riedemann proposed openstack/nova master: Add functional test for anti-affinity cross-cell migration https://review.opendev.org/661859
21:35:30 openstackgerrit Matt Riedemann proposed openstack/nova master: Support cross-cell moves in external_instance_event https://review.opendev.org/658478
21:35:30 openstackgerrit Matt Riedemann proposed openstack/nova master: Add cross-cell resize policy rule and enable in API https://review.opendev.org/638269
21:35:31 openstackgerrit Matt Riedemann proposed openstack/nova master: WIP: Enable cross-cell resize in the nova-multi-cell job https://review.opendev.org/656656
21:36:37 openstackgerrit melanie witt proposed openstack/nova stable/stein: Add reno about nova-api eventlet monkey-patching and rabbitmq https://review.opendev.org/662095
21:37:36 openstackgerrit melanie witt proposed openstack/nova stable/stein: Add reno about nova-api eventlet monkey-patching and rabbitmq https://review.opendev.org/662095
21:44:52 openstackgerrit Merged openstack/nova master: Fix the server group "policy" field type in api-ref https://review.opendev.org/661869
21:45:05 openstackgerrit Merged openstack/nova master: docs: Don't version links to reno docs https://review.opendev.org/661967
22:23:42 openstackgerrit Adam Spiers proposed openstack/nova master: WIP: Provide HW_CPU_X86_AMD_SEV trait when SEV is supported https://review.opendev.org/638680
22:24:03 openstackgerrit Adam Spiers proposed openstack/nova master: Track inventory for new MEM_ENCRYPTION_CONTEXT resource class https://review.opendev.org/662105
22:33:11 aspiers efried: brand new patch added to the top of the SEV series ^^^
22:33:23 efried yup, looking at the series now.
22:34:35 aspiers This should be the last of the detection / inventory changes. The remaining ones are all about actually consuming the results of that and booting an SEV guest.
22:35:48 aspiers Looking at http://specs.openstack.org/openstack/nova-specs/specs/train/approved/amd-sev-libvirt-support.html#work-items, there are now Gerrit reviews (some merged) for the first 6 out of 11 MVP work items. So halfway ... kinda :)
22:36:40 aspiers #7 is gonna be the tough one
22:36:51 aspiers The others are mostly polish
22:37:24 aspiers Alright, I'm done for today
22:41:49 openstackgerrit Merged openstack/nova master: cleanup evacuated instances not on hypervisor https://review.opendev.org/512623
#openstack-nova - 2019-05-30
01:18:24 openstackgerrit Guo Jingyu proposed openstack/nova master: Define and catch QemuGuestAgentNotRunning for set-password failure https://review.opendev.org/661466
01:35:12 openstackgerrit zhaixiaojun proposed openstack/nova master: Blacklist python-cinderclient 4.0.0 https://review.opendev.org/661578
01:58:18 openstackgerrit Takashi NATSUME proposed openstack/nova master: Add TODO note for mox removal https://review.opendev.org/576758
02:23:58 openstackgerrit melanie witt proposed openstack/nova master: Follow up for counting quota usage from placement https://review.opendev.org/662056
02:38:07 openstackgerrit Merged openstack/nova stable/stein: docs: Don't version links to reno docs https://review.opendev.org/662025
02:58:21 openstackgerrit Merged openstack/os-vif stable/rocky: Prevent "qbr" Linux Bridge from replying to ARP messages https://review.opendev.org/655692
03:32:26 openstackgerrit melanie witt proposed openstack/nova-specs master: Amend count-quota-usage-from-placement to reflect implementation https://review.opendev.org/662130
05:33:42 openstackgerrit Merged openstack/nova master: Add get_counts() to InstanceMappingList https://review.opendev.org/638072
05:55:09 openstackgerrit Merged openstack/nova master: Fix hard-delete of instance with soft-deleted referential constraints https://review.opendev.org/661398

Earlier   Later