Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2022-09-07
09:52:10 sean-k-mooney cryptic?
09:52:36 sean-k-mooney oh thet fact we sometime have the rows and somethimes have named tuples (dicts before)
09:53:01 sean-k-mooney we shoudl fix mixing the type in the cache but that a spereate latent bug
09:55:06 bauzas sean-k-mooney: I said cryptic because you really need to understand SQLA internals
09:55:24 bauzas this is no longer really an ORM
09:55:43 sean-k-mooney you dont really
09:55:51 sean-k-mooney they are issuing deprecation warnings
09:55:57 sean-k-mooney that tell you exactly what to fix
09:56:05 sean-k-mooney because they are trying to do a 2.0 release
09:56:23 sean-k-mooney so the are getting ready to drop legacy interfaces
09:56:54 bauzas well, I don't want to discuss that much about it, but I understand your point
11:15:06 opendevreview Merged openstack/placement master: Make us compatible with oslo.db 12.1.0 https://review.opendev.org/c/openstack/placement/+/855862
13:50:31 artom Do we have a antelope PTG etherpad already?
13:50:50 gibi yes
13:50:58 gibi https://etherpad.opendev.org/p/nova-antelope-ptg
18:10:42 JayF Heya, only one more patch left in the Ironic driver fixes stuff (I still have the CI one that's getting backported... but when these land that becomes less important). https://review.opendev.org/c/openstack/nova/+/821352 this puts the last fix in stable/train
18:14:18 elodilles JayF: nova-tox-validate-backport job failed -> can you re-apply the cherry pick from the stable/ussuri patch? o:)
18:14:53 JayF oh, whoops, I didn't realize that was non-voting, I was just checking ironic job and voting jobs
18:14:57 JayF I'll absolutely fix, ty
18:15:15 opendevreview Jay Faulkner proposed openstack/nova stable/train: Ignore plug_vifs on the ironic driver https://review.opendev.org/c/openstack/nova/+/821352
18:15:58 JayF I'll be out of your hair soon ;) this is the last ironic driver fix in Nova in really a long time, until the design issue currently at the spec stage gets fixed
18:21:06 elodilles JayF: no problem & thanks for the update ;) +2'd
18:26:47 melwitt we need sean-k-mooney to come back to ^ as well, I haven't been reviewing it bc I was the one who uploaded it (the backport) originally
20:02:07 fungi finest nova folk... further reviews of https://review.opendev.org/850003 would be much appreciated, in order to avoid releasing zed with a known security bug and creating more backport work
20:06:22 melwitt I can take a look at it later today, thanks for the link
21:54:49 sean-k-mooney fungi: melwitt +wd
21:55:01 sean-k-mooney also reviewed the ironic one
21:56:20 sean-k-mooney fungi: we are going to bring up that issue with neutron in the PTG too hopefully we can block vnic_type chages for bound port going forward in there api but we will start the backport of that once RC1 is out
21:57:10 JayF thanks :D
21:59:16 sean-k-mooney no worries :) i needed to do one thing i had left before i finsihed for the night got to drop again soon but hopefully that will land soon
22:17:14 opendevreview Merged openstack/nova stable/train: Ignore plug_vifs on the ironic driver https://review.opendev.org/c/openstack/nova/+/821352
22:22:44 fungi sean-k-mooney: thanks for the update!
22:43:35 sean-k-mooney fungi: unfortunetly that failed https://review.opendev.org/c/openstack/nova/+/849985 i think its conflicted with the pci serise which merged since it was writen
22:44:21 fungi oh, oof
22:44:24 sean-k-mooney its proably not that hard to reslove so ill see if we can still get this in zed
22:44:44 fungi there's still time yes. and then we'll want backports of course, but one step at a time
22:45:06 fungi at least it's already public, so the only real looming deadline is not releasing zed with it if we can help that
22:45:25 sean-k-mooney yep
22:45:48 sean-k-mooney ill check with gibi when they are onlien and pull this locally in the morning and see whast happening
22:48:33 sean-k-mooney actuly its failiing in a test that its not modifying
22:54:25 sean-k-mooney fungi: so those tests pass locally for me
22:55:13 sean-k-mooney im going to recheck it and see but we might have some uninited shareing i dont think this is related to this patch
22:55:21 fungi interesting. i'll try to keep an eye on it, just wanted to make sure it was on someone's radar
#openstack-nova - 2022-09-08
06:59:36 gibi sean-k-mooney, fungi: I will check that secu fix now
07:19:49 opendevreview Balazs Gibizer proposed openstack/nova master: Reproduce bug 1981813 in func env https://review.opendev.org/c/openstack/nova/+/849985
07:19:49 opendevreview Balazs Gibizer proposed openstack/nova master: Gracefully ERROR in _init_instance if vnic_type changed https://review.opendev.org/c/openstack/nova/+/850003
07:21:21 gibi sean-k-mooneym, stephenfin, fungi: ^^ needed to rebase and adapt the test to the changes on master. It should be green again
07:36:53 fungi thanks!
09:21:55 opendevreview ribaudr proposed openstack/nova master: Attach Manila shares via virtiofs (objects) https://review.opendev.org/c/openstack/nova/+/839401
09:21:55 opendevreview ribaudr proposed openstack/nova master: Attach Manila shares via virtiofs (manila abstraction) https://review.opendev.org/c/openstack/nova/+/831194
09:21:56 opendevreview ribaudr proposed openstack/nova master: Attach Manila shares via virtiofs (drivers and compute manager part) https://review.opendev.org/c/openstack/nova/+/833090
09:21:56 opendevreview ribaudr proposed openstack/nova master: Attach Manila shares via virtiofs (api) https://review.opendev.org/c/openstack/nova/+/836830
09:21:57 opendevreview ribaudr proposed openstack/nova master: Bump compute version and check shares support https://review.opendev.org/c/openstack/nova/+/850499
09:21:57 opendevreview ribaudr proposed openstack/nova master: Add metadata for shares https://review.opendev.org/c/openstack/nova/+/850500
09:21:58 opendevreview ribaudr proposed openstack/nova master: Add instance.share_attach notification https://review.opendev.org/c/openstack/nova/+/850501
09:21:58 opendevreview ribaudr proposed openstack/nova master: Add instance.share_detach notification https://review.opendev.org/c/openstack/nova/+/851028
09:21:59 opendevreview ribaudr proposed openstack/nova master: Add shares to InstancePayload https://review.opendev.org/c/openstack/nova/+/851029
09:22:00 opendevreview ribaudr proposed openstack/nova master: Add instance.power_on_error notification https://review.opendev.org/c/openstack/nova/+/852084
09:22:00 opendevreview ribaudr proposed openstack/nova master: Add instance.power_off_error notification https://review.opendev.org/c/openstack/nova/+/852278
09:22:02 opendevreview ribaudr proposed openstack/nova master: Add helper methods to attach/detach shares https://review.opendev.org/c/openstack/nova/+/852085
09:22:02 opendevreview ribaudr proposed openstack/nova master: Add libvirt test to ensure metadata are working. https://review.opendev.org/c/openstack/nova/+/852086
09:22:04 opendevreview ribaudr proposed openstack/nova master: Add virt/libvirt error test cases https://review.opendev.org/c/openstack/nova/+/852087
09:22:04 opendevreview ribaudr proposed openstack/nova master: Add share_info parameter to reboot method for each driver (driver part) https://review.opendev.org/c/openstack/nova/+/854823
09:22:06 opendevreview ribaudr proposed openstack/nova master: Support rebooting an instance with shares (compute and API part) https://review.opendev.org/c/openstack/nova/+/854824
09:22:06 opendevreview ribaudr proposed openstack/nova master: Change microversion to 2.94 https://review.opendev.org/c/openstack/nova/+/852088
10:17:46 sean-k-mooney gibi: cool will re reivew. i forgto tthat while this passes locally that ws because it was not rebased which will be done by zuul when its testing
10:17:56 gibi yepp
10:18:06 gibi I also tend to forget that fact
10:26:30 sean-k-mooney its rare that that rebase succced but brakes something with out causing a merge conflict
10:26:39 sean-k-mooney so mostly it does not matter
10:34:04 sean-k-mooney gibi: they look fine, i rehecked the second patch as it failed due to a vm segfault
10:34:17 gibi ahh, thanks
11:25:29 opendevreview Merged openstack/nova master: Reproduce bug 1981813 in func env https://review.opendev.org/c/openstack/nova/+/849985
12:04:02 Uggla gibi, looking at that comment. https://review.opendev.org/c/openstack/nova/+/833090/16/nova/compute/manager.py#3903 the idea is to get a regular list instead of a ShareMappingList. Is there another way to do that ?
12:47:08 Uggla gibi, forget ^
14:51:50 whoami-rajat stephenfin, hey, would it still be viable to get this merged? the nova and novaclient changes have merged https://review.opendev.org/c/openstack/python-openstackclient/+/831014
15:09:18 stephenfin whoami-rajat: Sure. It won't be in the initial Zed release but we can backport. Bit of work needed on it though.
15:34:52 whoami-rajat stephenfin, ack, one question, do you mean place it above the --hostname or below it? https://review.opendev.org/c/openstack/python-openstackclient/+/831014/5/openstackclient/compute/v2/server.py#3104
15:35:06 whoami-rajat I think options are be in sequence of microversion?
16:12:08 stephenfin whoami-rajat: sorry, below
16:29:40 whoami-rajat stephenfin, ack, updated the patch, thanks
16:29:53 stephenfin I just replied :)
16:30:21 stephenfin --confirm-reimage just isn't descriptive enough, IMO, and we generally need pairs for boolean options
16:43:31 stephenfin whoami-rajat: and I left a few more comments on there in reply
16:46:05 whoami-rajat stephenfin, I left a comment midway of your comment mentioning a case where the else would not be appropriate
16:46:22 whoami-rajat stephenfin, so we do allow rebuilding volume backed instance if the old and new image is same
16:46:38 whoami-rajat and if we don't put the elif microversion >=2.91 that case would fail
16:47:09 whoami-rajat will address the other ones
16:49:21 stephenfin so if I call 'openstack server rebuild --image $IMAGE $SERVER' and $IMAGE happens to be the same image that was originally used, it will pass?
16:49:46 stephenfin whoami-rajat: ^
16:50:53 whoami-rajat stephenfin, yes, it should, the nova side allows it
16:50:57 whoami-rajat at least
16:51:18 stephenfin that feels super janky :-D
16:52:08 stephenfin aren't you effectively preventing that on newer microversions with this change?
16:53:00 stephenfin i.e. if someone was relying this previously, they wouldn't be able to do so with OS_COMPUTE_API_VERSION=2.93 or later unless they also passed '--rebuild-volume' ?
16:53:39 stephenfin perhaps we could have a temporary stop-gap measure before preventing it entirely client side
16:53:47 stephenfin if microversion >= 2.93; block outright
16:54:41 stephenfin if microversion < 2.93; warn that this is unsupported, that it will no longer be allowed in the future, and that nova will reject the request if the image is different from the one originally used, but allow the request to continue (for now)
16:55:44 whoami-rajat yeah but we have implemented a generic case to rebuild any type of volume backed instance, people would prefer that instead of the hacky thing we have had before
16:56:02 whoami-rajat if microversion >= 2.93; block outright: in this case we also block image backed instances

Earlier   Later