Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2020-03-20
10:54:19 ygk_12345 kashyap ok sure
10:54:22 kashyap ygk_12345: Can you just run this: `/usr/bin/qemu-system-ppc64` and tell what you get?
10:54:28 nightmare_unreal https://www.irccloud.com/pastebin/KuPFLZ4Q/quota.rst
10:54:37 nightmare_unreal lyarwood: but that file has like no content.
10:54:40 nightmare_unreal see the snippet
10:55:10 ygk_12345 @kashya
10:55:19 ygk_12345 kashyap root@b2a-1:~# /usr/bin/qemu-system-ppc64
10:55:29 ygk_12345 root@b2a-1:~# /usr/bin/qemu-system-ppc64
10:55:54 kashyap ygk_12345: Does that open a QEMU window or not?
10:56:06 ygk_12345 it is commandline
10:56:11 ygk_12345 no GUI
10:57:53 lyarwood nightmare_unreal: ah sorry it builds that list automatically
10:58:18 nightmare_unreal exactly, that's what I was thinking how to add
10:58:19 lyarwood nightmare_unreal: where's your osc change?
10:59:03 kashyap ygk_12345: That's supposed to launch a GUI window; and do you see any error on the command-line?
10:59:30 kashyap ygk_12345: Either way, your installation is broken; please look for Ubuntu's guidance for re-installing virtualization packages properly.
10:59:33 ygk_12345 kashyap /usr/bin/qemu-system-ppc64: relocation error: /usr/bin/qemu-system-ppc64: symbol fdt_check_full version LIBFDT_1.2 not defined in file libfdt.so.1 with link time reference
10:59:50 stephenfin nightmare_unreal: if you're adding this functionality to the code then you don't need to edit the docs - they're generated automatically, as lyarwood notes
11:00:06 stephenfin you need to add the functionality though as it doesn't exist at the moment, as noted on openstack-discuss
11:00:13 nightmare_unreal oh okay
11:00:13 lyarwood yeah just building the docs with https://review.opendev.org/#/c/714062/1 now that I assume fixes this
11:00:38 nightmare_unreal yes
11:01:10 kashyap ygk_12345: Okay, that's broken; as I told above, please re-install Virt Packages from scratch. Read documentation with patience, and take notes of failures.
11:02:46 lyarwood nightmare_unreal: yeah https://review.opendev.org/#/c/714062/ will list --force under quota set now
11:03:31 nightmare_unreal thanks for clarifcation , i just submitted that patch and thought i made mistake by not writing doc
11:03:37 nightmare_unreal first patch :D
11:04:29 lyarwood nightmare_unreal: np, there should be a docs job that builds this in CI, I'll confirm with a link once it does
11:06:17 stephenfin nightmare_unreal: I left a review on that patch. There are a couple of things you need to fix. Feel free to ping me on #openstack-sdks if you have questions/need help
11:06:44 nightmare_unreal sure stephenfin
11:14:03 lyarwood stephenfin: is that the channel for osc now?
11:14:26 stephenfin to the best of my knowledge, yes
11:14:47 stephenfin that's where I've always gone to discuss stuff with mordred et al
11:18:40 openstackgerrit Kevin Zhao proposed openstack/nova master: Add default cpu model for aarch64 https://review.opendev.org/709494
11:20:34 openstackgerrit Merged openstack/nova stable/stein: nova-live-migration: Wait for n-cpu services to come up after configuring Ceph https://review.opendev.org/713837
11:26:42 ygk_12345 @kas
11:27:16 ygk_12345 kashyap now the libvirtd is running fine but still the same nova-compute resource error. I have updated the bug report with my findings
11:34:35 sean-k-mooney lyarwood: it makes total sense to run c-vol on the subnode and we run g-api on the sub node to do multile glance store testing so dont remove either of them
11:35:15 sean-k-mooney c-vol is on the sub node so cinder can do testing with multiple backends too
11:35:19 openstackgerrit Brin Zhang proposed openstack/nova master: libvirt:driver:Disallow AIO=native when 'O_DIRECT' is not available https://review.opendev.org/682772
11:36:11 sean-k-mooney lyarwood: im not sure we we need eitehr in our specific jobs but other project want that for there jobs
11:40:53 lyarwood sean-k-mooney: right, the way we layout the services at the moment isn't granular enough IMHO, we should really have a subnode layout *just* for compute migration testing
11:45:13 sean-k-mooney lyarwood: im not sure i agree but we could
11:45:31 lyarwood sean-k-mooney: https://review.opendev.org/#/c/714057/ - can you hit that regardless
11:45:37 sean-k-mooney lyarwood: in a hyper convered deployment i wound not find it unusal to have cinder vol on a compute
11:45:49 sean-k-mooney glance api is a littel weired but we only have two nodes
11:45:58 lyarwood sean-k-mooney: our default deployment topologies shouldn't be HCI
11:46:05 sean-k-mooney i think it should
11:46:07 lyarwood sean-k-mooney: it's just adding overhead for no reason
11:46:14 sean-k-mooney if we use it
11:46:16 lyarwood duplicating c-vol and g-api is utterly pointless
11:46:22 lyarwood on the computes upstream
11:46:31 lyarwood there's a place for that but it isn't in openstack/nova's gate
11:46:38 lyarwood using devstack
11:46:52 lyarwood as it's no reflective of reality at all
11:46:56 sean-k-mooney right but i think glance or cinder made the change and we just picked it up
11:47:18 lyarwood right and that's fine if they need to test volume migration or multi g-api backends etc
11:47:23 sean-k-mooney e.g. they change the common job definition and we never diabled it again
11:47:29 lyarwood but it shouldn't be the default for all subnode users
11:48:17 sean-k-mooney sure but its jsut that they didnt do it in there job they did it in either the base devstack multinode job or tempest
11:48:36 lyarwood it's in devstack-gate iirc in the feature matrix
11:48:49 lyarwood that builds the enabled services for subnodes etc
11:49:18 sean-k-mooney ok so ya if this was zuulv3 then it would be simpelr
11:49:53 lyarwood yeah indeed
11:50:00 sean-k-mooney with devstack gate you dont have the same level of contol in project repos to alter the job
11:50:17 sean-k-mooney i guess thats is another thing to alter in your migration
11:50:39 lyarwood right sorry my brain has just caught up to where I was 9 weeks ago on this
11:50:57 lyarwood and the need to migrate to control this easily per multinode job
11:51:05 lyarwood friday--
11:55:13 sean-k-mooney but but fridays are good :(
11:55:24 lyarwood after 5:30
11:55:36 lyarwood I'm just burnt out today
11:55:48 sean-k-mooney then have a lazy friday
11:56:34 stephenfin bauzas: You about? Want to rubberstamp this for me, please? https://review.opendev.org/#/c/713611/
12:23:08 lyarwood gibi: https://review.opendev.org/#/c/714057/ if you have bandwidth can you check this fup to the nova-live-migration fixes yesterday.
12:26:10 lyarwood elod: https://review.opendev.org/#/q/topic:bug/1867380+(status:open+OR+status:merged) - would you mind going through the rest of this series so we can unblock stable/pike
12:29:16 elod lyarwood: just commented one
12:30:36 elod lyarwood: but otherwise it would be good to push through these soon :) and thanks for the fixes!
12:30:41 openstackgerrit Lee Yarwood proposed openstack/nova stable/rocky: nova-live-migration: Wait for n-cpu services to come up after configuring Ceph https://review.opendev.org/713840
12:30:41 openstackgerrit Lee Yarwood proposed openstack/nova stable/rocky: Replace ansible --sudo with --become in live_migration/hooks scripts https://review.opendev.org/713839
12:30:44 lyarwood elod: yup thanks, addressed above ^
12:31:15 elod lyarwood: cool, thanks!
12:40:52 artom stephenfin, yep, thanks - lemme coffee up and read that
12:45:59 gibi lyarwood: done. Thanks
13:31:15 luyao dansmith, artom : are you about?
13:31:33 artom lyarwood, I am, dansmith probably isn't, given the time zone
13:32:19 artom Err, luyao ^^
13:33:06 luyao artom: I replied on https://review.opendev.org/#/c/687856, could you look it again
13:35:04 artom luyao, is there a specific question you'd like answered though? Reading the conversations quickly, it looks like the ball is in your court to update the patch
13:35:44 openstackgerrit Kevin Zhao proposed openstack/nova master: Add default cpu model for aarch64 https://review.opendev.org/709494
13:41:34 luyao artom: I don't update my patch since dansmith have some concern on moving cleanup before rpc call and some other details I don't know which solustion is better.
13:42:27 luyao artom: I saw you mentioned functional test, I run your live migration functional test on top of my patch, they runs well
13:58:07 artom luyao, yeah, the fact that the functional test is passing is reassuring (and thanks for the review - I have it fixed locally, will push a new version soon)
14:13:28 luyao artom: If you still have questions or concerns about my reply on vpmem live migration , please leave comments there. Or you can ping me in IRC, I'll be online for about two hours I can response promply during this time.
14:13:42 artom luyao, ok, thanks
14:26:46 openstackgerrit Merged openstack/nova master: libvirt: Correctly resize encrypted LUKSv1 volumes https://review.opendev.org/706900
14:34:07 openstackgerrit Artom Lifshitz proposed openstack/nova master: Increase code reuse in test_numa_servers https://review.opendev.org/714125
14:34:31 artom stephenfin, ^^ quick side-fix for you
14:34:48 artom (as I work on NUMA LM func tests)
14:44:00 luyao dansmith: Are you about?
14:44:18 dansmith luyao: yes, but I don't really want to argue about moving that rpc call

Earlier   Later