Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2017-11-21
19:04:29 openstackgerrit Merged openstack/nova master: cleanup mapping/reqspec after archive instance https://review.openstack.org/515034
19:05:29 mriedem mgagne: a couple of small changes to make here and then i think we're good https://review.openstack.org/#/c/503088
19:06:26 mriedem artom: the scheduler is definitely aware of moves for placement allocations, even if it's not super obvious
19:06:35 mgagne mriedem: will work on it today. I think committed the Ironic release note by mistake, will reword as suggested
19:07:03 mriedem mgagne: yeah heh, that was my comment :)
19:07:14 mgagne didn't read yet :P, will do
19:07:59 artom mriedem, yeah, it decides which instances go to which hosts, to operations on instances that have to do with hosts aren't outside its domain :)
19:08:05 artom *so operations
19:08:28 artom If we started telling it about keypair updates I'd be more concerned ;)
19:08:37 mriedem artom: i mean this specifically https://github.com/openstack/nova/blob/master/nova/scheduler/client/report.py#L1035
19:16:14 efried cdent Still around?
19:16:23 cdent yessir
19:16:33 mgagne mriedem: so should I refactor _generate_configdrive or not?
19:16:38 efried cdent Is there a way in gabbi for me to say "This must *not* be in the response" ?
19:17:30 cdent in a body or header?
19:17:38 efried body
19:18:12 efried I know about checking the length of a chunk of the response; will fall back on that if nothing else is available.
19:18:39 cdent there are a couple of additional options:
19:19:22 cdent if the surrounding chunk of json is small and you know all of it, you can make the value of the jsonpath be the entire chunk, instead of just a scalar value
19:19:37 cdent or you might be able to do something with regular expressions
19:19:49 cdent (the value side of the jsonpath statement can be a regex)
19:20:09 cdent efried: if you got a pointer to an almost there example I can probably say something more concrete
19:21:27 efried cdent Like here: https://github.com/openstack/nova/blob/master/nova/tests/functional/api/openstack/placement/gabbits/resource-provider.yaml#L105
19:21:38 efried cdent I want to make sure there's no "traits" key
19:21:43 efried etc.
19:22:21 efried response_forbidden_json_paths:\n\t$.resource_providers[0].traits
19:22:34 cdent heh
19:23:02 cdent is your intent to validate the entire response? If so, you might consider:
19:23:38 cdent response_json_paths:\n$: @<my-big-response.json
19:24:18 mriedem mgagne: no, i think what you have is ok
19:24:21 efried cdent Right; though the docs kinda recommended against that.
19:24:27 mriedem the error handling being different makes a refactor less worthwhile
19:24:28 openstackgerrit Eric Young proposed openstack/nova master: Implement ScaleIO image backend https://review.openstack.org/407440
19:24:35 mgagne mriedem: got it
19:24:37 cdent efried: they also recommend not validating entire responses
19:24:56 cdent (in one test)
19:25:26 efried cdent Specifically in this case I'd like to ensure I don't return the extra fields when I don't request them. (Related to the traits/aggregates/inventories thing from this morning)
19:25:50 efried cdent (Yes, I realize I haven't abandoned this patch, but I'm taking it as an opportunity to learn gabbi.)
19:26:15 cdent :)
19:26:59 cdent another option would be to write your own CustomHandler that actually implements response_forbidden_json_paths
19:27:46 efried cdent That would live in nova, or that would be contributed back to gabbi?
19:28:30 efried oh, jsonpath_rw, that's a separate thing. /me looks too...
19:30:05 cdent I was thinking live in nova initially, and then merged to gabbi if it felt sufficiently useful/nice
19:31:29 openstackgerrit Eric Berglund proposed openstack/nova master: WIP(5): PowerVM driver: ovs vif https://review.openstack.org/422512
19:32:49 cdent efried: this might be a case where it would be better to have a unit test of your serializer
19:33:16 efried cdent Also.
19:34:08 dansmith mriedem: you have some changes against the set locally, right? will you make the hint change and the other things we discussed as part of that?
19:34:48 mriedem dansmith: i don't have anything local
19:35:11 dansmith oh okay I thought you were working on the reno and flipping some of the flags
19:35:22 efried cdent Thanks for the help. A CustomHandler might be more than I'm willing to bite off at the moment, but I'll pester you if I decide to have a crack at it.
19:35:26 dansmith I'll go ahead and do that other stuff then
19:35:44 mriedem dansmith: that got sidetracked by the review comments and waiting to hear back from some people
19:35:52 dansmith okay
19:37:48 openstackgerrit Vipin Balachandran proposed openstack/nova master: VMware: Support volumes backed by VStorageObject https://review.openstack.org/521690
19:44:32 openstackgerrit Chris Dent proposed openstack/nova master: [placement] Symmetric GET and PUT /allocations/{consumer_uuid} https://review.openstack.org/510626
19:44:33 openstackgerrit Chris Dent proposed openstack/nova master: [placement] POST /allocations to set allocations for >1 consumers https://review.openstack.org/500073
19:49:50 efried cdent Sanity check me here: POST /resource_providers/{uuid}/inventories is not documented?
19:50:39 cdent yeah, it was sort deprecated
19:51:04 cdent there’s a review somewhere adding it back in, that’s under discussion
19:51:27 efried cdent Oh, okay. So I won't hurry to document it.
19:51:33 openstackgerrit Merged openstack/python-novaclient master: Microversion 2.54 - Enable reset keypair while rebuild https://review.openstack.org/519572
19:53:52 cdent efried: here’s that review https://review.openstack.org/#/c/511342/
20:02:09 openstackgerrit Andreas Jaeger proposed openstack/nova master: Update bindep.txt for doc builds https://review.openstack.org/521755
20:04:26 openstackgerrit Merged openstack/nova master: Get auth from context for glance endpoint https://review.openstack.org/490057
20:09:13 openstackgerrit Andreas Jaeger proposed openstack/nova stable/pike: Update bindep.txt for doc builds https://review.openstack.org/521999
20:09:31 openstackgerrit Andreas Jaeger proposed openstack/nova stable/ocata: Update bindep.txt for doc builds https://review.openstack.org/522000
20:09:53 openstackgerrit Andreas Jaeger proposed openstack/nova stable/newton: Update bindep.txt for doc builds https://review.openstack.org/522001
20:12:26 openstackgerrit Hongbin Lu proposed openstack/nova master: Skip authentication on version discovery URI https://review.openstack.org/522002
20:16:51 openstackgerrit Hongbin Lu proposed openstack/nova master: placement: skip authentication on root URI https://review.openstack.org/522002
20:23:53 openstackgerrit Dan Smith proposed openstack/nova master: Alternate fix for validating image on rebuild https://review.openstack.org/521186
20:23:53 openstackgerrit Dan Smith proposed openstack/nova master: Add regression test for rebuild with new image doubling allocations https://review.openstack.org/521153
20:23:54 openstackgerrit Dan Smith proposed openstack/nova master: Fix doubling allocations on rebuild https://review.openstack.org/521662
20:24:13 dansmith mriedem: artom: ^
20:28:58 openstackgerrit Mathieu Gagné proposed openstack/nova master: Regenerate and pass configdrive when rebuild Ironic nodes https://review.openstack.org/503088
20:29:01 artom dansmith, looking
20:29:22 dansmith I think I got everything but we talked about lots of things
20:30:55 openstackgerrit Andreas Jaeger proposed openstack/nova master: DNM - test updated releasenotes and sphinx job https://review.openstack.org/520677
20:37:03 mriedem dansmith: comments inline
20:37:09 mriedem mostly just nits
20:37:25 mriedem i can update next week if you're tired of this
20:41:12 dansmith oh I'm tired of it for sure but:
20:41:13 openstackgerrit Dan Smith proposed openstack/nova master: Alternate fix for validating image on rebuild https://review.openstack.org/521186
20:41:13 openstackgerrit Dan Smith proposed openstack/nova master: Add regression test for rebuild with new image doubling allocations https://review.openstack.org/521153
20:41:14 openstackgerrit Dan Smith proposed openstack/nova master: Fix doubling allocations on rebuild https://review.openstack.org/521662
20:41:45 dansmith apparently I broke the top patch in the process here
20:43:57 mriedem did you drop the filter scheduler claim part?
20:45:34 mriedem oh nvm i'm looking at the wrong patch
20:45:52 mriedem dansmith: yeah this needs updating https://review.openstack.org/#/c/521662/5/nova/scheduler/filter_scheduler.py
20:47:36 mriedem +1 on the bottom change though
20:49:34 artom Hah, I think you fixed some of what I was about to post before I had a chance to post it
20:50:45 dansmith mriedem: yep, that's it
20:51:19 openstackgerrit Dan Smith proposed openstack/nova master: Fix doubling allocations on rebuild https://review.openstack.org/521662
20:52:47 openstackgerrit Merged openstack/nova master: Service token is not experimental https://review.openstack.org/521955
20:59:54 mriedem ok barring renos and such i think that series is good
21:02:42 artom dansmith, two nits from me. And I need to look at the tests eventually
21:04:24 dansmith artom: mriedem is going to write the reno, so I'll let him update those things if he agrees
21:04:31 dansmith I'm going to be fading out here in a bit
21:05:32 artom Sounds good
21:08:00 mriedem the numa one is probably a change, i forgot that was using image props
21:08:03 mriedem but yeah next week at this point

Earlier   Later