Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2019-09-18
23:37:00 sean-k-mooney today?
23:37:38 sean-k-mooney what was the failure out of interest
23:38:14 alex_xu yesterday
23:39:54 alex_xu I only saw two bug, one is test_model_sync, another one is db timeout
23:40:40 sean-k-mooney sorry i though you were refering to https://review.opendev.org/#/c/672693/
23:41:07 sean-k-mooney its going to fail i nthe gate beacue devstack failed on the second node in the live migration job
23:42:38 sean-k-mooney looks like the nova compute service db entry for the subnode never got created properly
23:42:49 alex_xu yea, I thought those two in vpmems, just checking what fail in cpu resource
23:43:13 efried_pto mriedem: I'll try to be around for the meeting tomorrow; but if I'm not, you wanna run it?
23:43:32 mriedem sure
23:43:57 mriedem alex_xu: zuul had 2 restarts today
23:44:00 mriedem so that reset everything
23:44:05 efried_pto alex_xu: totally bug 1823251 again.
23:44:06 openstack bug 1823251 in OpenStack Compute (nova) "Spike in TestNovaMigrationsMySQL.test_walk_versions/test_innodb_tables failures since April 1 2019 on limestone-regionone" [High,Confirmed] https://launchpad.net/bugs/1823251
23:44:21 efried_pto I'm trying hard to nail it down, but so far no luck.
23:44:24 alex_xu same bug after reset?
23:44:42 efried_pto dunno, I just rebased and re+Wed like half an hour ago.
23:45:12 efried_pto correction: 1h ago.
23:46:37 efried_pto (only rebased to requeue since it was failing the constraints job bogusly - on the aforementioned bug)
23:46:55 efried_pto now of course it's going to have to wait twelve freaking hours to get a node again.
23:46:58 mriedem you know, it might be worth just skipping TestNovaMigrationsMySQL until all of these things are merged
23:47:04 efried_pto Agreeeeed
23:47:17 mriedem unskip before rc1
23:47:29 efried_pto that would increase our merge percentage by an order of magnitude.
23:47:41 mriedem do it
23:47:44 efried_pto you wanna propose it quick I'll fast approve it.
23:47:51 mriedem merging all of this stuff in the middle of next week is crazy
23:48:07 efried_pto yeah, I'd really rather merge it *this* week.
23:48:16 efried_pto Hell, I would have rather merged it last Thursday
23:48:23 mriedem was just about to say ^
23:48:38 mriedem i'm in the middle of something so you go ahead
23:50:02 alex_xu i can do it
23:50:06 mriedem well let me see here
23:50:38 sean-k-mooney whne the compute service joing the service group i should see a entry in the cell conductor corresponding to the compute node entry correct
23:50:52 openstackgerrit Dustin Cowles proposed openstack/nova master: Provider Config File: YAML file loading and schema validation https://review.opendev.org/673341
23:50:52 openstackgerrit Dustin Cowles proposed openstack/nova master: Provider Config File: Function to further validate and retrieve configs https://review.opendev.org/676029
23:50:53 openstackgerrit Dustin Cowles proposed openstack/nova master: WIP: Provider Config File: Merge provider configs to provider tree https://review.opendev.org/676522
23:50:59 mriedem alex_xu: i've got it
23:51:39 alex_xu mriedem: ok, so I can help to +2
23:51:50 efried_pto alex_xu: ++
23:52:56 sean-k-mooney hum maybe not
23:55:39 sean-k-mooney that compute node does not have the db config so it woul have to go via the conductor but i guess it may not be logged or rabbit could have dropped the message
23:56:25 sean-k-mooney oh never mined i miss read the devstack message
23:56:51 sean-k-mooney it finishe registering the compute servce and then the neutron openvswitch agent did not start
23:58:36 openstackgerrit Matt Riedemann proposed openstack/nova master: Temporarily skip TestNovaMigrationsMySQL https://review.opendev.org/683009
23:59:21 sean-k-mooney wait is this complianing baour the compute agent or neutron openvswitch agent
23:59:23 sean-k-mooney https://storage.gra1.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_5b9/672693/24/gate/nova-live-migration/5b959a8/logs/subnode-2/devstacklog.txt.gz
23:59:42 sean-k-mooney /baour/about/
23:59:51 mriedem alex_xu: efried_pto: there you go https://review.opendev.org/683009
23:59:58 alex_xu mriedem: thanks
#openstack-nova - 2019-09-19
00:02:50 sean-k-mooney ... i think it took ~90 seconds so the 60 second timeout expired but by the time it printed the list the compute service was there so this is just a timeing issue :(
00:06:45 openstackgerrit Matt Riedemann proposed openstack/nova master: Revert "Temporarily skip TestNovaMigrationsMySQL" https://review.opendev.org/683011
00:12:46 openstackgerrit Merged openstack/nova-specs master: Track user_id/project_id for migrations https://review.opendev.org/682511
00:23:42 openstackgerrit Merged openstack/nova master: libvirt: Start reporting PCPU inventory to placement https://review.opendev.org/671793
00:49:43 KeithMnemonic1 mriedem thanks! so how do i proceed, create a qorking branch, apply you the conflicting patches and then try your cherry pick?
00:49:57 KeithMnemonic1 s/qorking/working
00:53:16 melwitt KeithMnemonic1: you can pull down his patch with 'git review -d 683008' and then you just need to resolve the merge conflicts he left in the tests
00:54:06 KeithMnemonic1 and then do a git review once i think they are resolved?
00:54:10 melwitt then --amend the commit and then push a new patch set
00:54:13 melwitt yes
00:54:25 KeithMnemonic1 perfect let me try that, thank you both!
00:55:23 melwitt cool good luck
01:44:07 openstackgerrit Fan Zhang proposed openstack/nova master: Fix exception translation when creating volume https://review.opendev.org/678991
01:51:48 KeithMnemonic1 melwitt quick question, if some of those conflicts are from other patches that cherry pick clean to pike, is the the proper process?
01:52:41 KeithMnemonic1 isnt it better to pull in cleanly any dependencies instead of me trying to shove some stuff in his review?
01:53:41 KeithMnemonic1 and then rebase that patch once they are merged?
01:54:09 melwitt KeithMnemonic1: we don't pull in dependencies that are unrelated to the backport, it can get pretty unwieldy that way. by resolving the conflict you're just making the patch fit within the env of the older release, which shouldn't result in adding a bunch of stuff
01:54:57 melwitt from what I saw when I glanced, the conflicts didn't look too big and mriedem documented all of them in the commit message already. just need to be smoothed out in the .py files
02:29:21 openstackgerrit Takashi NATSUME proposed openstack/nova master: Move Destination object tests to their own test class https://review.opendev.org/683017
03:21:28 openstackgerrit Merged openstack/nova master: Temporarily skip TestNovaMigrationsMySQL https://review.opendev.org/683009
03:21:36 openstackgerrit Merged openstack/nova stable/stein: Fix race in _test_live_migration_force_complete https://review.opendev.org/681744
03:21:45 openstackgerrit Merged openstack/nova stable/stein: Fix a misuse of assertGreaterEqual https://review.opendev.org/682168
05:45:09 openstackgerrit garyk proposed openstack/nova master: Deconstruct the mother of all locks https://review.opendev.org/682242
06:45:45 openstackgerrit hulina proposed openstack/nova master: Nova raise exceptions when extending volume fails https://review.opendev.org/680648
07:41:56 openstackgerrit Takashi NATSUME proposed openstack/nova master: Update keypairs in saving an instance object https://review.opendev.org/683043
07:42:17 gibi sean-k-mooney: thanks for the work on https://review.opendev.org/#/c/669738/ I +Ad it
07:47:42 bauzas gibi: sean-k-mooney: I just left some comments (and I concurrently +Wd, my bad) that honestly I feel we could have defined this test module as a plain top-level function testsuite for configdrivers
07:47:48 bauzas configdrives*
07:49:03 gibi bauzas: ack. I think the reorg for a top level test module can be done separately
07:49:13 bauzas gibi: oh yeah totally
07:49:24 gibi bauzas: and I'm supportive :)
07:49:26 bauzas gibi: just saying this regression testclass is way more than a single check :)
07:49:34 gibi yeah
07:51:33 gibi bauzas: here is a nova-manage heal allocation test addition if you have some review time https://review.opendev.org/#/c/669879/11
07:56:57 openstackgerrit Balazs Gibizer proposed openstack/nova master: Error out interrupted builds https://review.opendev.org/666857
07:56:58 openstackgerrit Balazs Gibizer proposed openstack/nova master: Pull up compute node queries to init_host https://review.opendev.org/682680
07:57:39 openstackgerrit hulina proposed openstack/nova master: Nova raise exceptions when extending volume fails https://review.opendev.org/680648
08:05:55 kashyap mnaser: Hi, reading the scroll.
08:16:21 openstackgerrit Takashi NATSUME proposed openstack/nova master: Fix unit of hw_rng:rate_period https://review.opendev.org/683051
08:16:56 kashyap mnaser: There's a lot scrollback; let's discuss today when you're around. :-)
08:22:46 bauzas gibi: congrats for your visa, I got mine too yesterday by postal service
08:22:58 gibi bauzas: congrats for you too then
08:23:00 bauzas (and I understand why they asked me my passport)
08:24:53 bauzas I have a powerful passport, it was the first time I asked for this but I was a bit afraid my visa application could be not supported because I had some coffee stains
08:25:03 bauzas (on my passport)
08:25:57 kashyap mnaser: But a couple of things to get out of the way:
08:26:15 kashyap mnaser: * When you use 'host-model', you won't actually see 'host-model' in the *live* guest XML; libvirt expands it (see next point)
08:26:26 kashyap mnaser: * To refresh your memory, 'host-model' means: (a) get maximum possible CPU features from the host; (b) make the guest live-migratable; and (c) auto-add critical CPU flags -- provided you have the right underlying bits for microcode, kernel, etc.
08:26:50 kashyap mnaser: * The "IvyBridge-IBRS" model does *not* contain 'vmx'; you have to explicitly add it. (You can `grep` the 'cpu_map.xml'; or if you're running new enough libvirt `grep vmx /usr/share/libvirt/cpu_map/x86_IvyBridge-IBRS.xml`)
08:30:33 kashyap mnaser: And lastly, since you're in the nested setup: make sure that your level-1 guest explicitly has 'vmx' flag set, even if you're using 'host-model'.

Earlier   Later