Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2017-02-24
18:53:16 EmilienM mriedem: but nothing says v2 is prohibited
18:54:34 EmilienM I don't see any release note in Nova that says "do not install nova with keystone v2 endpoints", or I missed it
18:55:01 mriedem https://review.openstack.org/#/c/435010/13/nova/api/openstack/identity.py@40 doesn't mean you can't have v2
18:55:03 mriedem it means you need v3
18:55:22 mriedem but i also don't know why whatever error is in ^ is not handled by that try/except
18:55:54 mriedem "# something is wrong, like there isn't a keystone v3 endpoint,"
18:56:04 mriedem so tripleo has a v3 endpoint but it's wrong?
18:56:11 mriedem i don't know the latest on what the actual break is
18:56:44 EmilienM mriedem: tripleo undercloud has keystone endpoints pointing to v2.0, I'm fixing it now
18:56:49 EmilienM mriedem: to be versionless
19:03:01 oomichi dansmith: can you take a look at https://review.openstack.org/#/c/425958 ? I could not +2 before you
19:04:48 openstackgerrit Sean Dague proposed openstack/nova master: Generate log error message if keystone looks funny https://review.openstack.org/438049
19:05:05 sdague EmilienM: I do agree that we should be nicer to the operator to explain things
19:05:12 sdague that's the patch above
19:05:23 EmilienM I have been testing tripleo with v3 endpoint and it seems to work
19:05:31 EmilienM our overcloud runs v3 by default
19:05:37 EmilienM so our case might be fixed by tonight
19:05:49 sdague But pointing identity endpoint to /v2.0 actually means that no other clients besides keystoneclient can work with v3
19:05:55 EmilienM but I'm wondering if other operators hit it
19:06:12 sdague EmilienM: right, I just added and upgrade release note about that
19:07:08 EmilienM sdague: ok. thx
19:07:29 openstackgerrit Balazs Gibizer proposed openstack/nova master: handle uninited fields in notification payload https://review.openstack.org/415857
19:07:33 satyar Hi everyone
19:07:41 satyar i am running the trunk
19:08:06 satyar it seems when VM goes to error the VCPU is not cleaning up after deletion of VM
19:08:26 satyar anyone aware of this issue?
19:09:01 gibi jaypipes: I updated the flavor.projects lazy load patch based on our today's discussion https://review.openstack.org/#/c/415857/
19:11:36 openstackgerrit Jay Pipes proposed openstack/nova master: placement: ironic inventory reporting https://review.openstack.org/437602
19:11:41 jaypipes gibi: awesome, thanks!
19:14:20 openstackgerrit Balazs Gibizer proposed openstack/nova master: handle uninited fields in notification payload https://review.openstack.org/415857
19:17:03 openstackgerrit Stephen Finucane proposed openstack/nova master: doc: Don't put comments inside toctree https://review.openstack.org/437924
19:17:03 openstackgerrit Stephen Finucane proposed openstack/nova master: setup.cfg: Add note to enable doc warnings https://review.openstack.org/437926
19:21:01 sfinucan johnthetubaguy, mriedem: typos fixed on the PCI weigher spec :) https://review.openstack.org/364468
19:27:32 jaypipes gibi: reviewed... perhaps we need to chat about it.
19:28:30 jaypipes gibi: I was under the impression that all instance notifications >= 1.3 object version would have the flavor.projects attribute set to null, not []...
19:28:58 jaypipes gibi: to indicate that the attribute is not set (versus that there are 0 projects the flavor is "shared with")
19:30:15 mriedem sfinucan: questions inline
19:30:29 sfinucan mriedem: (y)
19:36:33 dansmith jbernard: are you still around?
19:45:50 jaypipes gibi: melwitt's also added a good question in that patch
19:50:50 jbernard dansmith: i am
20:03:44 openstackgerrit Sean Dague proposed openstack/nova master: Generate log error message if keystone looks funny https://review.openstack.org/438049
20:07:51 gibi jaypipes, melwitt: thanks for the reviews. I'm working on to answer / fix your comments
20:42:22 melwitt sean-k-mooney: is this it? https://review.openstack.org/#/c/183876/
20:44:06 openstackgerrit Chris Friesen proposed openstack/nova master: show flavor info in server details https://review.openstack.org/434514
20:56:53 openstackgerrit Balazs Gibizer proposed openstack/nova master: handle uninited fields in notification payload https://review.openstack.org/415857
21:00:47 openstackgerrit Balazs Gibizer proposed openstack/nova master: handle uninited fields in notification payload https://review.openstack.org/415857
21:04:55 openstackgerrit Balazs Gibizer proposed openstack/nova master: handle uninited fields in notification payload https://review.openstack.org/415857
21:08:59 efried mriedem: Per PTG feedback, I'm working on splitting up the first PowerVM driver change set. Do y'all have a preference for whether the topmost one is still 391288 (thus preserving its history and 36 patch sets) or a fresh new one?
21:09:24 efried Just a matter of git paperwork, so no biggie either way to me.
21:09:27 mriedem efried: topmost as in the one that should be passing CI?
21:09:30 efried right
21:09:34 mriedem that works
21:09:37 efried which?
21:09:56 mriedem whatever you love
21:10:05 mriedem as long as it's in the series, either first or the one that's passing
21:10:08 efried :) roger wilco. Thanks.
21:10:23 mriedem just don't drop it b/c then we lose the history
21:10:33 efried Ah, okay, that answers my question.
21:11:04 efried (Oh, I see, you're saying I can slot it anywhere in the series, but it should stay in the series.)
21:30:08 openstackgerrit Matt Riedemann proposed openstack/nova master: Fix typo in config drive support matrix docs https://review.openstack.org/438084
21:32:05 diana_clarke LOL, recognize this code anyone? https://twitter.com/redhatway/status/834855421500354561 I wonder who picked this snippet.
21:32:45 diana_clarke someone should port that water bottle to mock
21:33:27 cdent diana_clarke++
21:41:48 mriedem ha
21:41:53 mriedem that's not py35 compliant
22:23:33 NostawRm Any reason there's no option for backups for differential backups for nova/ceph like there is with cinder/ceph other than no one has done it?
23:33:45 openstackgerrit Eric Fried proposed openstack/nova master: PowerVM Driver: spawn/delete #1: no-ops https://review.openstack.org/438119
23:40:20 openstackgerrit Eric Fried proposed openstack/nova master: PowerVM Driver: spawn/delete #1: no-ops https://review.openstack.org/438119
23:54:47 efried mikal ^^ 362LOC! And here's a thing: it _might_ actually pass jenkins/CI.
#openstack-nova - 2017-02-25
01:52:56 openstackgerrit John Garbutt proposed openstack/nova-specs master: Add spec to use Cinder's new attachment APIs https://review.openstack.org/373203
02:06:47 openstackgerrit Sean Dague proposed openstack/nova master: Generate log error message if keystone looks funny https://review.openstack.org/438049
02:45:31 openstackgerrit Sean Dague proposed openstack/nova master: Correct some spelling errors https://review.openstack.org/430646
02:51:21 openstackgerrit John Garbutt proposed openstack/nova-specs master: Add locked-by-service spec https://review.openstack.org/438134
03:08:14 openstackgerrit John Garbutt proposed openstack/nova-specs master: Add policy-docs spec https://review.openstack.org/433010
03:23:25 openstackgerrit John Garbutt proposed openstack/nova-specs master: Add policy-remove-scope-checks spec https://review.openstack.org/433037
10:56:43 openstackgerrit Sean Dague proposed openstack/nova master: Generate log error message if keystone looks funny https://review.openstack.org/438049
12:48:21 openstackgerrit Sean Dague proposed openstack/nova master: Generate log error message if keystone looks funny https://review.openstack.org/438049
12:58:30 openstackgerrit Sean Dague proposed openstack/nova master: api/metadata/vendordata_dynamic: don't import Requests for its constants https://review.openstack.org/434339
15:40:30 openstackgerrit John Garbutt proposed openstack/nova-specs master: Add additional-default-policy-roles spec https://review.openstack.org/427872
18:04:56 mriedem miss_hypothermia: i saw your remove check_attach patch failed some bfv tests in the grenade job, looks like some bdm is invalid error
18:09:52 miss_hypothermia mriedem: I saw that too
18:10:16 miss_hypothermia mriedem: I couldn't figure it out yet what exactly the problem is
18:10:47 miss_hypothermia mriedem: does that test run in other jobs as well BTW?
18:10:55 mriedem miss_hypothermia: i know what it is
18:11:03 mriedem http://logs.openstack.org/58/335358/41/check/gate-grenade-dsvm-neutron-multinode-ubuntu-xenial/d3194cc/logs/subnode-2/old/screen-n-cpu.txt.gz?level=TRACE#_2017-02-24_23_47_29_665
18:11:13 mriedem the pike api is reserving the bfv volume and putting it into 'attaching' status
18:11:21 mriedem the instance is running on the ocata n-cpu node
18:11:34 mriedem where check_attach is still a thing down in nova/virt/block_device.py
18:11:47 mriedem and check_attach fails because the volume is in 'attaching' status from the reserve call in the pike api
18:11:52 mriedem and then, kablammo!
18:12:01 mriedem so you're going to have to add that service version thing back in
18:12:50 miss_hypothermia Ohhh ok, forgot about the mixed env from this perspective
18:13:22 mriedem yeah
18:13:23 mriedem me too
18:13:33 mriedem i'm happy the test caught this though
18:14:00 miss_hypothermia But we're still fine from unreserve perspective so I will add that part back and we should be good to go
18:14:17 mriedem agree
18:14:19 miss_hypothermia mriedem: +1, me too!
18:14:36 mriedem btw, you're still hypothermic?

Earlier   Later