Posted	Nick	Remark
#openstack-nova - 2020-09-10
12:27:26	gibi	I'm hoping that timeout was a one time thing
12:27:36	sean-k-mooney	ya i know
12:27:52	sean-k-mooney	but i was building doc for gmann change a few days ago
12:28:02	sean-k-mooney	and i noticed how slow the release note where
12:28:13	sean-k-mooney	so i tried adding -j auto to it and it was way faster
12:28:53	sean-k-mooney	then i was sad when addign it to all the docs env failed because some had extensions that did not declare support
12:30:24	smcginnis	Some of the local tests I did really sped things up running parallel.
12:31:08	smcginnis	Probably not as much in gate since there are less available cores, but still faster.
12:31:35	sean-k-mooney	i use a dev vm that emulates the gate with 8 threads
12:31:41	sean-k-mooney	and it was noticable
12:32:37	sean-k-mooney	i say emulates because its got 2 numa node and 2 cpu socket with 2 core and 2 threads per core resulting in 8 thread total
12:32:53	sean-k-mooney	so its not exactly the same but still it was faster
12:36:49	stephenfin	weird, looks like the mirrors are not getting updated at the moment
12:37:11	stephenfin	those last few merged patches aren't appearing
12:52:40	openstackgerrit	Stephen Finucane proposed openstack/nova master: functional: Add 'start_compute' helper https://review.opendev.org/746943
12:52:41	openstackgerrit	Stephen Finucane proposed openstack/nova master: functional: Add cold migration tests for PCI servers https://review.opendev.org/746946
12:52:41	openstackgerrit	Stephen Finucane proposed openstack/nova master: functional: Clean up PCI tests https://review.opendev.org/746945
12:52:41	openstackgerrit	Stephen Finucane proposed openstack/nova master: functional: Use tempdir for CONF.instances_path https://review.opendev.org/746944
12:52:42	openstackgerrit	Stephen Finucane proposed openstack/nova master: functional: Add test for SR-IOV neutron ports https://review.opendev.org/746948
12:52:42	openstackgerrit	Stephen Finucane proposed openstack/nova master: conductor: Don't use setattr https://review.opendev.org/746947
12:52:43	openstackgerrit	Stephen Finucane proposed openstack/nova master: functional: Add live migration tests for PCI, SR-IOV servers https://review.opendev.org/746950
12:52:43	openstackgerrit	Stephen Finucane proposed openstack/nova master: fixtures: Handle binding of first port https://review.opendev.org/746949
12:52:44	openstackgerrit	Stephen Finucane proposed openstack/nova master: functional: Expand SR-IOV live migration tests with NUMA https://review.opendev.org/749360
13:08:53	openstackgerrit	Stephen Finucane proposed openstack/nova master: xenapi: Remove driver and tests https://review.opendev.org/749304
13:08:54	openstackgerrit	Stephen Finucane proposed openstack/nova master: xenapi: Remove Xen-only code paths https://review.opendev.org/749306
13:08:54	openstackgerrit	Stephen Finucane proposed openstack/nova master: virt: Remove 'is_xenapi' helper https://review.opendev.org/749305
13:08:55	openstackgerrit	Stephen Finucane proposed openstack/nova master: trivial: Final cleanup https://review.opendev.org/749308
13:08:55	openstackgerrit	Stephen Finucane proposed openstack/nova master: objects: Remove 'XenapiLiveMigrateData' https://review.opendev.org/749307
13:08:56	openstackgerrit	Stephen Finucane proposed openstack/nova master: virt: Remove 'post_interrupted_snapshot_cleanup' API https://review.opendev.org/749313
13:08:56	openstackgerrit	Stephen Finucane proposed openstack/nova master: virt: Remove 'get_console_pool_info' API https://review.opendev.org/749311
13:08:57	openstackgerrit	Stephen Finucane proposed openstack/nova master: virt: Remove various aggregate APIs https://review.opendev.org/749317
13:08:57	openstackgerrit	Stephen Finucane proposed openstack/nova master: virt: Remove 'set_bootable' API https://review.opendev.org/749314
13:08:58	openstackgerrit	Stephen Finucane proposed openstack/nova master: api: Remove 'os-agents' API https://review.opendev.org/749309
13:08:58	openstackgerrit	Stephen Finucane proposed openstack/nova master: virt: Remove 'get_per_instance_usage' API https://review.opendev.org/749318
13:08:59	openstackgerrit	Stephen Finucane proposed openstack/nova master: virt: Remove 'get_all_bw_counters' API https://review.opendev.org/749312
13:08:59	openstackgerrit	Stephen Finucane proposed openstack/nova master: objects: Remove Agent, AgentList https://review.opendev.org/749310
13:09:00	openstackgerrit	Stephen Finucane proposed openstack/nova master: virt: Remove 'change_instance_metadata' API https://review.opendev.org/749316
13:09:00	openstackgerrit	Stephen Finucane proposed openstack/nova master: virt: Remove 'reset_network' API https://review.opendev.org/749315
13:14:39	lyarwood	stephenfin: this is not helping my +ratio ^_^
13:14:44	lyarwood	stephenfin: what happened there?
13:15:20	stephenfin	lyarwood: https://review.opendev.org/#/c/749304/2/nova/exception.py@1293 :(
13:15:39	stephenfin	it should be trivially reapprovable O:)
13:16:09	stephenfin	and for the SR-IOV series, see last comment from me on https://review.opendev.org/#/c/746943/
13:16:48	lyarwood	ACK I'll let CI run on both and reapprove once we are sure it's passing
13:17:17	stephenfin	cool
13:17:50	stephenfin	addressed your comment on https://review.opendev.org/#/c/746946/ too while I was respinning
13:19:35	lyarwood	yup thanks
13:23:38	stephenfin	gibi, lyarwood: One of you want to hit this vTPM doc change real quick? https://review.opendev.org/#/c/750675/
13:23:54	stephenfin	(basically note that resize and cold migrate are supported ops now)
13:24:53	lyarwood	yup looking
13:33:26	openstackgerrit	Artom Lifshitz proposed openstack/nova stable/queens: post live migration: don't call Neutron needlessly https://review.opendev.org/750993
13:38:19	stephenfin	sean-k-mooney: Once this is released, we can turn on parallel builds, afaict https://review.opendev.org/750973
13:39:40	stephenfin	I'm not sure about whether we'll need to bump our lower-constraints or not. On one hand, if we add '-j auto' then we'll no longer be able to build docs using the older versions, but on the other hand, docs could still be build (just omit '-j auto')
13:41:28	sean-k-mooney	yes that is what was failing
13:41:45	sean-k-mooney	the feature classification
13:43:57	sean-k-mooney	i could check i might see if we could make it conditional based on the sphinx version
13:44:26	stephenfin	ah no, I wouldn't go that far
13:44:42	sean-k-mooney	ill see what version of sphinx added it
13:45:05	sean-k-mooney	if its not supper new then i think a min version bump would be fine
13:45:25	stephenfin	sorry, I wasn't clear. I'm talking about lower-constraints for the various oslo packages plus sphinx-feature-classification
13:45:30	sean-k-mooney	otherwise we can make it configurable via an environmnet varable or just postional arg
13:45:37	sean-k-mooney	oh
13:45:43	sean-k-mooney	not of sphinx build
13:45:49	stephenfin	if you tried to build with '-j auto' using sphinx-feature-classification < 1.1.0, it'll fail
13:46:01	sean-k-mooney	yes
13:46:03	stephenfin	because it emits a warning saying this extension doesn't support parallel stuff
13:46:04	sean-k-mooney	it did
13:46:10	sean-k-mooney	yep
13:46:21	sean-k-mooney	release nodes totally works
13:46:23	stephenfin	so does that mean we need to set out lower-constraint for s-f-c to 1.1.0?
13:46:43	stephenfin	or just assume people building docs with older versions will drop the '-j auto'
13:46:48	sean-k-mooney	ya we would in wallaby
13:47:06	sean-k-mooney	be can add -j auto to the other envs
13:47:12	sean-k-mooney	that dont use that already
13:47:13	bauzas	gibi: any notes you'd like to see in the prelude ?
13:47:46	gibi	bauzas: I have nothing specific in my mind right now
13:48:04	gibi	but I will do a scan of merged renos to get a view
13:48:07	stephenfin	bauzas: have you noted that XenAPI is gone?
13:48:13	stephenfin	that's merging as we spak
13:48:15	stephenfin	*speak
13:48:23	bauzas	gibi: I'll use your own cycle highlits as a base
13:48:30	bauzas	stephenfin: ah, no, gtk
13:48:48	bauzas	stephenfin: in general, I take a look at the main reno master page
13:48:58	bauzas	so I'll see it, I guess ;)
13:49:00	gibi	bauzas: cool
13:49:15	sean-k-mooney	stephenfin: i started reviewin the xenapi removeal then stop as i realed it mainly just code deletion and lyarwood and gibi? had alredy reviewed most of it
13:49:49	sean-k-mooney	im glad that is going to make it :)
13:50:33	stephenfin	sean-k-mooney: you might be interested in https://review.opendev.org/#/c/748533/
13:51:08	sean-k-mooney	didnt i reveiew an previous version
13:51:16	sean-k-mooney	i do need to take another look
13:51:22	stephenfin	oh, so you did
13:51:24	stephenfin	my bad
13:51:36	sean-k-mooney	no its good to remind me to look again
13:52:54	stephenfin	https://review.opendev.org/745605 still needs to be respun too :)
13:53:15	sean-k-mooney	oh the filters
13:53:21	sean-k-mooney	actully ill go do that now
13:53:31	lyarwood	sean-k-mooney: some additional review on these would also be good btw, slightly out of my area tbh https://review.opendev.org/#/c/746949/
13:53:40	lyarwood	artom: ^ also if you have bandwidth
13:54:06	sean-k-mooney	oh the fixture change
13:54:24	lyarwood	and the series on top adding the actual tests

Earlier   Later