Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2018-07-13
01:46:14 mriedem and do the logic there?
01:46:35 mriedem then your acceptable status values are based on (1) the microversion and (2) the rpcapi version
01:46:57 mriedem allowed_statuses = ['running']
01:46:59 mriedem if support_abort_in_queue:
01:47:16 Kevin_Zheng and can_send_version?
01:47:20 mriedem yeah,
01:47:29 mriedem if support_abort_in_queue and can_send_version(5.1):
01:47:41 mriedem allowed_statuses.extend(['queued', 'preparing'])
01:47:51 gmann microversion check is on APOI controller side not enough ?
01:47:52 Kevin_Zheng Then should I merge patch 2 and 4?
01:47:53 mriedem if migration.status not in allowed_statuses:
01:47:54 gmann API
01:47:58 mriedem raise exception.InvalidMigrationState(
01:48:12 mriedem gmann: no because if the compute is queens,
01:48:20 mriedem it doesn't know how to cancel a migration in queued or preparing status
01:49:11 gmann that what compute version or rpc version check will handle right
01:49:19 mriedem Kevin_Zheng: i don't think you merge them
01:49:56 mriedem Kevin_Zheng: i think patch 2 has the rpcapi version bump b/c of the signature change
01:50:12 mriedem and then in the same change that adds the microversion, you can move the status checking from api into rpcapi
01:50:26 mriedem based and pass in the support_abort_in_queue flag
01:50:35 Kevin_Zheng move it in patch 4?
01:50:36 mriedem *pass in the support_abort_in_queue
01:51:21 gmann this one will be passed all way till rpc right? - https://review.openstack.org/#/c/573136/6/nova/api/openstack/compute/server_migrations.py@149
01:51:45 openstackgerrit Chen proposed openstack/nova master: fix cellv2 delete_host https://review.openstack.org/582171
01:52:06 mriedem gmann: that's what i'm suggesting,
01:52:13 gmann ohk. i got confuse when you mentioned microversion checks on compute.apy.py
01:52:13 mriedem i'll leave a comment on patch 2 and then i need to call it a day
01:52:19 mriedem and i'll talk to dansmith about my idea in the morning
01:52:47 Kevin_Zheng Ok, good night
01:53:55 openstackgerrit Julia Kreger proposed openstack/nova master: ironic: provide facilities to gracefully navigate versions https://review.openstack.org/582416
01:54:45 TheJulia mriedem: looks like I'll need to tune some job timeouts with that rescue test :\ Something for tomorrow since we seem to have a job that now wants to time out independently as well :\
01:56:13 openstackgerrit Julia Kreger proposed openstack/nova master: virt/ironic: Implement rescue and unrescue https://review.openstack.org/416487
01:56:23 TheJulia mriedem: parent update rebase ^
01:57:24 mriedem Kevin_Zheng: gmann: dansmith: left my idea in https://review.openstack.org/#/c/568542/
01:58:03 Kevin_Zheng Thanks I think it will work, lets collect more ideas.
01:58:59 mriedem TheJulia: thanks, lgtm
01:59:56 TheJulia mriedem: no, thank you! :)
02:02:31 gmann mriedem: thanks. ll check
02:19:32 openstackgerrit Takashi NATSUME proposed openstack/nova master: Transform instance.live_migration_force_complete notification https://review.openstack.org/480561
02:32:47 openstackgerrit Yikun Jiang (Kero) proposed openstack/nova master: Microversion 2.64 - Use new format policy in server group https://review.openstack.org/567534
02:44:17 openstackgerrit Yikun Jiang (Kero) proposed openstack/nova master: Microversion 2.64 - Use new format policy in server group https://review.openstack.org/567534
02:45:25 openstackgerrit Zhenyu Zheng proposed openstack/nova master: Use ThreadPoolExecutor for max_concurrent_live_migrations https://review.openstack.org/563505
02:45:45 openstackgerrit Merged openstack/nova master: Follow up for Ie49d605c66062d2548241d7e04f5a2a6b98c011e https://review.openstack.org/582292
02:45:57 openstackgerrit Merged openstack/nova master: Add policy field to ServerGroup notification object https://review.openstack.org/563401
02:59:45 openstackgerrit Merged openstack/nova master: Change the ServerGroupAntiAffinityFilter to adapt to new policy https://review.openstack.org/571166
03:32:18 openstackgerrit Eric Fried proposed openstack/nova master: WIP: Compute: Handle reshaped provider trees https://review.openstack.org/576236
03:44:40 openstackgerrit Yikun Jiang (Kero) proposed openstack/nova master: Fix all invalid obj_make_compatible test case https://review.openstack.org/574240
04:03:14 openstackgerrit Yikun Jiang (Kero) proposed openstack/python-novaclient master: Microversion 2.64 - Use new format policy in server group https://review.openstack.org/578261
04:35:09 openstackgerrit Hironori Shiina proposed openstack/nova master: ironic: Log an error when API version is not available https://review.openstack.org/582447
06:26:15 openstackgerrit Yikun Jiang (Kero) proposed openstack/nova master: Fix all invalid obj_make_compatible test case https://review.openstack.org/574240
06:33:06 tssurya /set weechat.look.prefix_join "▬▬▶"
06:40:29 tssurya oops, sorry ^^
06:40:44 openstackgerrit zhaolihui proposed openstack/python-novaclient master: Fix flavor keyerror when nova boot vm https://review.openstack.org/582147
06:43:47 openstackgerrit Yikun Jiang (Kero) proposed openstack/nova master: Address nits in server group policy series https://review.openstack.org/581616
06:45:10 openstackgerrit Hironori Shiina proposed openstack/nova master: ironic: Log an error when API version is not available https://review.openstack.org/582447
06:45:12 openstackgerrit Zhenyu Zheng proposed openstack/nova master: Compute: add support to abort queued live migration https://review.openstack.org/568542
06:58:28 openstackgerrit OpenStack Proposal Bot proposed openstack/nova master: Imported Translations from Zanata https://review.openstack.org/578019
06:59:00 openstackgerrit Zhenyu Zheng proposed openstack/nova master: Fix ServerMigrationSampleJsonTestsV2_24 to use its own sample file https://review.openstack.org/581562
07:41:10 openstackgerrit Zhenyu Zheng proposed openstack/nova master: Compute: add support to abort queued live migration https://review.openstack.org/568542
07:53:41 openstackgerrit Zhenyu Zheng proposed openstack/nova master: Fix ServerMigrationSampleJsonTestsV2_24 to use its own sample file https://review.openstack.org/581562
08:26:51 Miouge As an admin, can I add a VM to a server group after it’s creation?
08:27:16 Miouge I don’t find it in the API, so is the only option to go an edit the DB and re-schedule the instances?
08:35:24 giblet_off I'm taking the day off
08:50:47 gmann giblet_off: take rest do not be always too hard working :)
09:03:36 openstackgerrit tianhui proposed openstack/nova master: Change default video model from cirrus to vga https://review.openstack.org/582471
09:14:46 eumel8 Hello Nova Cores! Maybe it's possible to import Zanata translations: https://review.openstack.org/#/c/578019/
09:43:53 openstackgerrit Chen proposed openstack/nova master: Clean up group members in db when delete https://review.openstack.org/581403
09:57:06 openstackgerrit huanhongda proposed openstack/nova-specs master: Return "deleted" and "estimated_reclaim" time of instance for admin https://review.openstack.org/581638
10:07:19 openstackgerrit Chen proposed openstack/nova master: fix cellv2 delete_host https://review.openstack.org/582171
10:47:31 stephenfin eumel8: Sure, done
11:16:36 openstackgerrit Merged openstack/nova master: Imported Translations from Zanata https://review.openstack.org/578019
12:10:09 eumel8 thx stephenfin
12:54:28 mriedem sahid: i replied in https://review.openstack.org/#/c/515423/ - are you aware of certain vif types that will return None for port binding profile and/or vif_details?
12:54:37 mriedem b/c otherwise those should be coming back as dicts from the neutron api
13:12:40 openstackgerrit Surya Seetharaman proposed openstack/nova master: Online migration tool for populating queued-for-delete https://review.openstack.org/582536
13:30:59 openstackgerrit Merged openstack/nova master: Adapt _validate_instance_group_policy to new policy model https://review.openstack.org/571465
13:36:51 mriedem oh most illustrious, handsome and just dansmith, i draw your attention to https://review.openstack.org/#/c/568542/19/nova/compute/rpcapi.py@693 for discussion once i'm back from delivering my child to the quarry hill day camp
13:37:22 mriedem tl;dr i think we should drop the 5.1 rpc api bump since the interface doesn't change, and just check the compute service version in the api iff new microversion and status is queued/preparing
13:37:31 dansmith oye that's a lot of reading for a friday
13:37:52 dansmith I said that was one option yeah
13:37:58 dansmith I'll read for your reasoning
13:59:34 mriedem yup 3am rambling
13:59:36 melwitt efried, leakypipes, stephenfin: reminder that non-client library release freeze (os-traits and os-vif) is next thurs July 19 so if there's anything you want to make the release, get them in by the end of next thurs
14:00:06 mriedem yikun: ^ you should start working on the advanced policy changes to novaclient
14:03:11 melwitt novaclient release freeze is same as r-3 July 26
14:29:52 mriedem dansmith: thanks; i think i'll clean this change up and then rebase the api change on top with the service version check i'm talking about, then Kevin_Zheng can flesh out the rest
14:30:04 dansmith ack
14:30:04 melwitt dansmith: I wanted to get your thoughts on this approach for a bug fix before you are out next week https://review.openstack.org/582408
14:32:07 mriedem melwitt: btw, i've had a functional test up for one of those forever https://review.openstack.org/#/c/562041/
14:32:39 melwitt mriedem: oh, sorry, I didn't know
14:32:40 dansmith melwitt: we probably don't want to filter on deleted for the join, but I would think we should sort by deleted and pick the first service in the list, which would be the non-deleted one, if there is one
14:33:07 dansmith would that solve the bug without this patch?
14:33:40 melwitt dansmith: I think it would because it would also avoid generating a UUID for a deleted service
14:34:02 dansmith so that might be a better approach
14:34:09 melwitt Service.save()ing a deleted service will always raise ServiceNotFound, for example
14:35:09 dansmith because of _check_minimum_version() yeah?
14:35:27 melwitt no, ServiceNotFound will raise if version checks out
14:35:40 melwitt if version is too old, we raise ServiceTooOld first before we get to the db.service_update call
14:36:15 dansmith aight, well, anyway
14:36:24 melwitt to be clear, _check_minimum_version raises ServiceTooOld if the deleted service has too old a version, db.service_update raises ServiceNotFound for the deleted service

Earlier   Later