| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2019-10-23 | |||
| 19:11:07 | dansmith | efried: that one is used by others | |
| 19:11:08 | mriedem | oh i didn't realize there were two | |
| 19:11:21 | dansmith | the libvirt-only one should stay in the libvirt group until it's used elsewhere, IMHO | |
| 19:11:22 | efried | which like I said I wouldn't bother to "fix" if it was just that, but IMO makes sense to move, despite being currently libvirt-only, since we're making a special group for image_cache. | |
| 19:11:45 | mriedem | fun that we don't even test remove_unused_resized_minimum_age_seconds | |
| 19:16:55 | mriedem | split the baby in half | |
| 19:18:40 | mriedem | dansmith: i've been thinking about taking a crack at that notification sample thing for your aggregate pre-cache stuff, do you mind if i do or are you already working on it? | |
| 19:19:11 | dansmith | mriedem: I've been meaning to try to corner gibi, but I've been busy in the mornings.. however, I have *no* idea where to start, so please, by all means | |
| 19:19:28 | mriedem | he's all greased up by then, hard to atch | |
| 19:19:29 | mriedem | *catch | |
| 19:19:52 | mriedem | ok i'll see if i can wip something up to start it | |
| 19:30:17 | mriedem | melwitt: fyi there is runway slot room for your host_status patch once that is ready for re-review https://etherpad.openstack.org/p/nova-runways-ussuri | |
| 19:30:36 | melwitt | mriedem: oh cool, thanks | |
| 19:33:08 | mriedem | sean-k-mooney: is https://review.opendev.org/#/q/topic:bp/image-metadata-prefiltering+status:open ready for runways? | |
| 19:34:18 | openstackgerrit | Matt Riedemann proposed openstack/nova stable/rocky: Fix listing deleted servers with a marker https://review.opendev.org/690721 | |
| 19:34:18 | openstack | bug 1849409 in OpenStack Compute (nova) train "openstack server list --deleted --limit -1 hangs" [High,In progress] https://launchpad.net/bugs/1849409 - Assigned to Matt Riedemann (mriedem) | |
| 19:34:18 | openstackgerrit | Matt Riedemann proposed openstack/nova stable/rocky: Add functional regression test for bug 1849409 https://review.opendev.org/690720 | |
| 19:46:10 | openstackgerrit | Eric Fried proposed openstack/nova master: Consolidate [image_cache] conf options https://review.opendev.org/690723 | |
| 19:46:27 | efried | (the baby is mine) | |
| 19:46:27 | efried | mriedem, dansmith: ^ | |
| 19:50:16 | openstackgerrit | Matt Riedemann proposed openstack/nova stable/queens: Fix listing deleted servers with a marker https://review.opendev.org/690725 | |
| 19:50:16 | openstack | bug 1849409 in OpenStack Compute (nova) train "openstack server list --deleted --limit -1 hangs" [High,In progress] https://launchpad.net/bugs/1849409 - Assigned to Matt Riedemann (mriedem) | |
| 19:50:16 | openstackgerrit | Matt Riedemann proposed openstack/nova stable/queens: Add functional regression test for bug 1849409 https://review.opendev.org/690724 | |
| 19:57:45 | openstackgerrit | Elod Illes proposed openstack/nova stable/rocky: cleanup evacuated instances not on hypervisor https://review.opendev.org/687550 | |
| 20:02:15 | elod | mriedem: I've updated the commit message ^^^ (forgot to put me '_off', as today is national holiday here, but tomorrow gibi will be back) | |
| 20:02:29 | mriedem | elod: i saw that, thanks | |
| 20:10:04 | openstackgerrit | Matt Riedemann proposed openstack/nova stable/rocky: cleanup evacuated instances not on hypervisor https://review.opendev.org/687550 | |
| 20:10:46 | openstackgerrit | Matt Riedemann proposed openstack/nova stable/rocky: Add functional test for resize crash compute restart revert https://review.opendev.org/687563 | |
| 20:10:53 | openstackgerrit | Matt Riedemann proposed openstack/nova stable/rocky: Prevent init_host test to interfere with other tests https://review.opendev.org/687862 | |
| 20:11:14 | openstack | bug 1833581 in OpenStack Compute (nova) rocky "instance stuck in BUILD state if nova-compute is restarted" [Low,In progress] https://launchpad.net/bugs/1833581 - Assigned to Balazs Gibizer (balazs-gibizer) | |
| 20:11:14 | openstackgerrit | Matt Riedemann proposed openstack/nova stable/rocky: Functional reproduce for bug 1833581 https://review.opendev.org/687564 | |
| 20:11:21 | openstackgerrit | Matt Riedemann proposed openstack/nova stable/rocky: Error out interrupted builds https://review.opendev.org/687565 | |
| 20:11:53 | openstackgerrit | Eric Fried proposed openstack/nova master: Consolidate [image_cache] conf options https://review.opendev.org/690723 | |
| 20:16:07 | efried | dansmith: support matrix, you want just one section for image caching (where maybe without precaching the support is "partial") or two? | |
| 20:16:44 | dansmith | doesn't matter to me | |
| 20:16:51 | mriedem | i vote for one | |
| 20:17:36 | efried | ack. dansmith, I haven't looked, which drivers does precaching work for today? | |
| 20:17:46 | dansmith | libvirt | |
| 20:18:14 | efried | ight | |
| 20:23:13 | KeithMnemonic | mriedem , melwitt if you use the nova cli --deleted --marker work as expected ;-) | |
| 20:24:08 | mriedem | nova cli A #1 | |
| 20:24:11 | efried | do lxc and xen fall under the same libvirt umbrella? | |
| 20:24:20 | efried | I know they inherit some stuff... | |
| 20:25:16 | mriedem | for image caching....don't know, but i'd think yes? | |
| 20:25:26 | mriedem | at least for xen, not really sure about lxc and how it uses the image cache | |
| 20:26:02 | efried | looks like xen explicitly says nay | |
| 20:26:13 | efried | but there's not a separate driver for lxc afaics | |
| 20:27:53 | efried | powervm with the ssp driver does image caching, but it's not through the ImageCacheManager etc. And it wouldn't respond to any of the conf options. Not sure whether to say that has support or not... | |
| 20:28:19 | efried | oh, a driver note. | |
| 20:28:30 | mriedem | KeithMnemonic: well, with nova list are you using --marker with a server name or id? | |
| 20:28:43 | mriedem | b/c i wouldn't be surprised if nova list doesn't support specifying a server name for the marker like osc does | |
| 20:28:58 | mriedem | efried: i assumed you mean libvirt+xen | |
| 20:29:03 | mriedem | not the xenserver driver | |
| 20:29:08 | efried | uhm | |
| 20:29:13 | mriedem | there are 2 xens | |
| 20:29:20 | mriedem | the one we deprecated in train was xenserver | |
| 20:29:20 | efried | ...oh | |
| 20:29:31 | mriedem | libvirt+xen is also a thing that citrix used to maintain and no longer does | |
| 20:29:35 | efried | o, I see now | |
| 20:29:41 | mriedem | just like libvirt+lxc, libvirt+uml, etc are things | |
| 20:29:51 | efried | but like lxc, libvirt+xen would inherit whatever libvirt has. | |
| 20:29:56 | mriedem | maybe | |
| 20:29:59 | mriedem | usually | |
| 20:30:02 | mriedem | lxc is a weirdo though | |
| 20:30:34 | mriedem | i'd probably mark libvirt+lxc as unknown | |
| 20:32:26 | efried | and libvirt+xen too? | |
| 20:32:29 | openstackgerrit | Matt Riedemann proposed openstack/nova stable/pike: Fix listing deleted servers with a marker https://review.opendev.org/690735 | |
| 20:32:29 | openstack | bug 1849409 in OpenStack Compute (nova) train "openstack server list --deleted --limit -1 hangs" [High,In progress] https://launchpad.net/bugs/1849409 - Assigned to Matt Riedemann (mriedem) | |
| 20:32:29 | openstackgerrit | Matt Riedemann proposed openstack/nova stable/pike: Add functional regression test for bug 1849409 https://review.opendev.org/690734 | |
| 20:37:05 | KeithMnemonic | mriedem an id, just like with osc which confirms it is just osc that is not doing the marker/deleted correctly | |
| 20:41:20 | efried | boo, support matrix doesn't sphinx | |
| 20:43:07 | mriedem | KeithMnemonic: my point is, if you did: nova list --deleted --marker <server name not id> i think it won't work | |
| 20:43:22 | mriedem | nova cli isn't translating the marker from a name to an id, it just assumes id | |
| 20:44:16 | sean-k-mooney | libvirt+lxc will proably work for precaching but until we add support for resizing the filesystem to mage teh resized image the it will still be kind of borked. | |
| 20:44:49 | sean-k-mooney | normally we just assume cloud-init or the use will grow the filecycle to file the root disk itself | |
| 20:44:55 | sean-k-mooney | but that does not work with lxc | |
| 20:45:07 | sean-k-mooney | or at least if it does i dont know how to do it | |
| 20:46:02 | KeithMnemonic | right, i hear you | |
| 20:52:06 | openstackgerrit | Matt Riedemann proposed openstack/nova stable/ocata: Fix listing deleted servers with a marker https://review.opendev.org/690747 | |
| 20:52:06 | openstack | bug 1849409 in OpenStack Compute (nova) train "openstack server list --deleted --limit -1 hangs" [High,In progress] https://launchpad.net/bugs/1849409 - Assigned to Matt Riedemann (mriedem) | |
| 20:52:06 | openstackgerrit | Matt Riedemann proposed openstack/nova stable/ocata: Add functional regression test for bug 1849409 https://review.opendev.org/690746 | |
| 21:02:28 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Require Migration object arg to migrate_instance_finish method https://review.opendev.org/690106 | |
| 21:02:29 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Stop converting Migration objects to dicts for migrate_instance_start https://review.opendev.org/690115 | |
| 21:03:24 | openstackgerrit | Merged openstack/nova master: Add image precaching docs for aggregates https://review.opendev.org/687348 | |
| 21:05:36 | openstackgerrit | Eric Fried proposed openstack/nova master: Add image caching to the support matrix https://review.opendev.org/690748 | |
| 21:05:43 | efried | mriedem, dansmith: ^ | |
| 21:15:30 | melwitt | efried, mriedem: commented on https://review.opendev.org/683988 and while I was looking at the console proxy log I notice the token auth uuid is logged at INFO at different places :( one of which is coming from underlying websockify and not sure how to stop it | |
| 21:17:49 | efried | a websockify bug I guess | |
| 21:18:40 | melwitt | it's not a bug, it's emitting it's 'path' but since noVNC upstream took out the cookie forwarding stuff and had us switch to using the path, the token auth is now in the path | |
| 21:21:15 | mriedem | melwitt: unrelated to this? https://review.opendev.org/#/c/220622/ | |
| 21:21:44 | melwitt | mriedem: that fixes one of the two spots | |
| 21:22:17 | mriedem | so are you talking about this? https://zuul.opendev.org/t/openstack/build/8b011c8d92134f27ab59a8d392267b2e/log/controller/logs/screen-n-novnc-cell1.txt.gz#31 | |
| 21:22:23 | mriedem | INFO nova.console.websocketproxy [-] 213.32.72.99 - - [23/Oct/2019 19:12:55] 213.32.72.99: Path: '?token=badacaffcc2-6d65-4bc4-a2f5-4081318d0433' | |
| 21:22:30 | melwitt | yes that's the other one | |
| 21:22:38 | melwitt | and that's logged from websockify underneath | |
| 21:22:44 | mriedem | i feel like i remember something recently about a bug with logging and the websockify stuff but can't put my finger on it | |
| 21:23:08 | melwitt | orly? ok. it has always emitted the 'Path:' there but now the token auth is in the path | |
| 21:24:56 | mriedem | wonder if there is a way to plumb a flag down to tell it not to log the path or mask certain keys | |
| 21:25:15 | mriedem | it's not a cve since this is just in the logs | |
| 21:25:18 | melwitt | yeah, I have to go take a gander | |