Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2019-10-02
20:14:05 mriedem more comments
20:14:50 artom dansmith, brainstorming here - but does oslo.concurrency give us a distributed semaphore?
20:15:07 dansmith artom: not without some external service
20:15:11 artom I'm thinking - cast to the computes, but have the grab the semaphore when downloading
20:15:51 bnemec Sounds like a use case for tooz.
20:15:58 mriedem i was just going to make an etcd joke
20:16:08 bnemec oslo.concurrency only works within a single machine.
20:16:24 artom That way the concurrency becomes configurable (just set the semaphore's... err, cardinality?) and you don't ddos the glance server
20:16:26 dansmith artom: yeah, it's just not really necessary when we can threadpool it from the conductor
20:16:49 artom dansmith, for an RPC call though, right?
20:17:07 dansmith artom: yup
20:17:23 artom Yeah, I may be overthinking this
20:17:26 mriedem melwitt: ack i'm cool with that, docs only and regression in train
20:20:26 openstackgerrit Matt Riedemann proposed openstack/nova stable/rocky: Noop CantStartEngineError in targets_cell if API DB not configured https://review.opendev.org/686276
20:26:02 dansmith mriedem: yawna read my response to the parallelism thing and if you're cool with it I can squash those todos in my next rev: https://review.opendev.org/#/c/686232/4/specs/ussuri/approved/image-precache-support.rst@123
20:26:02 openstackgerrit Merged openstack/nova master: Use os-brick locking for volume attach and detach https://review.opendev.org/614190
20:26:12 melwitt mriedem: coolio
20:28:36 mriedem dansmith: replied
20:29:04 dansmith mriedem: thanks
20:30:52 openstackgerrit melanie witt proposed openstack/nova master: Restore console proxy deployment info to cells v2 layout doc https://review.opendev.org/686271
20:35:06 openstackgerrit Dan Smith proposed openstack/nova-specs master: Add image-precache-support spec https://review.opendev.org/686232
20:36:50 efried melwitt, mriedem: +2, good catch
20:38:58 efried aspiers: how's that SEV fixup looking? Would be nice to produce RC2 tomorrow (and not need RC3). Your fix is the last one I know of that's not well on the way to merging.
20:42:12 openstackgerrit Matt Riedemann proposed openstack/nova stable/rocky: Handle get_host_availability_zone error during reschedule https://review.opendev.org/686292
20:44:11 mriedem efried: deadline for RCs is oct 10 right?
20:44:19 efried yes
20:44:27 mriedem i mean, we want time for rc3 for sev issue #12 but sure
20:44:34 efried smy point
20:45:36 efried Oh, I misread, you're actually picking on SEV. I don't think we've had especially more issues for SEV than for other blueprints. Have there been more than one for any?
20:46:08 efried Seems like one for SEV, one for VPMEM, one for console tokens, one (?) for cpu-resources...
20:46:30 efried which is fine, exactly the reason FF happens before the release.
20:51:18 sean-k-mooney efried: i dont think we had any for cpu-resouce but we had/have one for numa migration
20:51:50 sean-k-mooney artom: has your fix for the RT cacheing issue landed?
20:51:58 mriedem yes
20:52:06 sean-k-mooney ok cool
20:52:12 mriedem console proxy was a docs thing
20:52:50 mriedem my point is sev landed late with a ton of churn over 2 releases (speaking as an outside observer to that series), so i wouldn't be surprised if there aren't more hairy issues
20:52:59 mriedem seems like every issue in that series is very hard to resolve
20:53:34 mriedem i also won't be surprised when there are issues with PCPU - we haven't seen issues because we're not using it anywhere yet
20:53:50 mriedem but in 18 months...watch out!
20:54:24 sean-k-mooney well there kind of is one but it was an issue we defeind as out of scope
20:54:31 sean-k-mooney it related to https://review.opendev.org/#/c/686079/
20:55:52 sean-k-mooney you can only have flaoting instnace and pinned instance on the same host and not have mempage issue if you enable the numa aware memory tracking using hw:mem_page_size=<any valid value>
20:56:01 sean-k-mooney but we know it would not address that
20:58:55 efried mriedem: the console token thing we just merged? That wasn't a docs issue.
20:59:12 efried it may not have been related to an actual blueprint either though, not sure.
21:00:02 sean-k-mooney the console issue was passing the path argument in case it was not required right?
21:00:10 dansmith I think he means the cells console proxy doc thing
21:00:16 efried no
21:00:36 efried https://review.opendev.org/#/c/686066/ Reduce scope of 'path' query parameter to noVNC consoles
21:00:50 dansmith this: https://review.opendev.org/#/c/686271/
21:01:14 efried no
21:01:16 efried -----^
21:01:25 dansmith efried: I mean he==mriedem, not you
21:01:30 efried right
21:01:43 efried and confirmed it wasn't a blueprint, it was a bug
21:01:54 efried anywho
21:02:26 sean-k-mooney im going to call it a night o/
21:05:49 mriedem efried: oh i wasn't thinking of the serial console thing
21:06:13 mriedem efried: it was a bug fix for a novnc 1.1.0 regression or something but dug a bit too deep
21:06:20 efried y
21:06:27 mriedem and actually wouldn't have been caught had it not been backported and released on stable
21:07:16 mriedem speaking of, dansmith can you get https://review.opendev.org/#/c/686063/ on stein so we can do a stein release tomorrow?
21:07:22 mriedem i think that's the last thing i was waiting on
21:07:38 dansmith okay
21:25:05 openstackgerrit Eric Fried proposed openstack/nova master: Allow versioned discovery unauthenticated https://review.opendev.org/685181
21:27:12 openstackgerrit Eric Fried proposed openstack/nova master: Modify nova routing table to process null requests https://review.opendev.org/518119
21:46:48 ozzzo Is there an "openstack" command corresponding to every "nova" and "nova-manage" command?
21:47:20 ozzzo For example, can I do "nova-manage service-disable <HV UUID>" via openstack?
21:48:49 ozzzo or "nova service-list" ?
21:49:11 ozzzo The obvious thing to try would be "openstack service list" but that does something different
21:56:23 mriedem yes,
21:56:31 mriedem https://docs.openstack.org/python-openstackclient/latest/cli/command-objects/compute-service.html#compute-service-set
21:56:47 mriedem nova-manage and nova/openstack CLIs are not the same
21:57:18 mriedem the former doesn't use apis, but hits the db directly when the api isn't available, and doesn't require a token, the latter requires a token and uses the api
21:57:40 mriedem you're also looking for this https://docs.openstack.org/python-openstackclient/latest/cli/command-objects/compute-service.html#compute-service-list
21:58:15 ozzzo very nice, thank you!
22:01:59 openstackgerrit Takashi NATSUME proposed openstack/nova master: Fix unit of hw_rng:rate_period https://review.opendev.org/683051
22:02:08 openstackgerrit Takashi NATSUME proposed openstack/nova master: Update keypairs in saving an instance object https://review.opendev.org/683043
22:02:17 openstackgerrit Takashi NATSUME proposed openstack/nova master: Move Destination object tests to their own test class https://review.opendev.org/683017
22:18:56 ozzzo Is there an "openstack" command for "nova-manage cell_v2 discover-hosts" ?
22:19:02 mriedem no
22:20:15 mriedem there is a config option to enable it in a periodic task if you don't want to hit the command line
22:20:30 ozzzo ok ty
22:20:35 mriedem https://docs.openstack.org/nova/latest/configuration/config.html#scheduler.discover_hosts_in_cells_interval
22:27:36 openstack Launchpad bug 1845759 in OpenStack Compute (nova) "Openstack nova-compute service gives cant start new thread with oslo messaging" [Undecided,New]
22:27:36 melwitt mriedem: at first glance I wondered if this is related to the recent eventlet tpool proxy change, but it's reported against ocata https://bugs.launchpad.net/nova/+bug/1845759
22:28:18 mriedem i saw it in triage but didn't have thoughts on it
22:29:32 melwitt me neither. I've never seen anything like it before
22:33:53 ozzzo I was thinking that "openstack compute service delete" would behave similar to "openstack resource provider delete" but the service did not come back when I typed "nova-manage cell_v2 discover_hosts"
22:34:02 ozzzo how can I re-create the hypervisor service that I deleted?
22:38:21 artom sean-k-mooney, https://review.opendev.org/#/c/685387/? Yeah
22:39:19 melwitt ozzzo: I think just create it again https://docs.openstack.org/python-openstackclient/latest/cli/command-objects/service.html#service-create the service command is different than the resource provider command, the resource provider command modifies only the placement service database and the service command modifies both the nova service database and the placement service database (by way of the placement API)
22:44:38 ozzzo ok I'll try that, ty!
22:45:52 melwitt 'openstack compute service delete' will delete the records of the service in the nova database and then it will do an equivalent REST API call as 'openstack resource provider delete' to the placement service
22:50:08 ozzzo that didn't seem to work; I typed "openstack service create --name us01odc-dev1-hv001 --enable nova-compute" but I still only have 2 hypervisor services under "openstack compute service list"
22:51:09 ozzzo It looks like I created a new service called us01odc-dev1-hv001 under "openstack service list"
22:51:28 ozzzo it looks like "service" and "compute service" are 2 different things
22:52:06 melwitt yeah... I'm looking at the doc and the <type> looks like it should be one of compute, identity, volume but I'm not 100% sure
22:52:19 ozzzo on the "compute service" man page I only see delete, list and set, there is no create

Earlier   Later