Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2018-02-26
10:15:20 dayou After "enable_service n-cell" and "VIF_PLUGGING_TIMEOUT=0 VIF_PLUGGING_IS_FATAL=False ", I was able to start an instance.
10:15:46 dayou Somehow, I tried other serivces such as octavia, it failed to plug vip there, a 500 is back from nova-api
10:16:58 dayou https://github.com/openstack/nova/blob/master/nova/api/openstack/compute/attach_interfaces.py#L156
10:17:04 dayou From the line above
10:17:40 melwitt dayou: you shouldn't be using the n-cell service, that's the legacy cells v1 that has an extremely crippled API functionality
10:17:52 dayou Here is my local.conf: https://review.openstack.org/#/c/520878/5/devstack/samples/l3a2/local.conf
10:19:29 dayou melwitt, Any idea why I get the Host 'xxx' is not mapped to any cell problem?
10:23:08 melwitt dayou: this is the step in devstack that should be mapping the compute host to the API database
10:23:10 melwitt https://github.com/openstack-dev/devstack/blob/83ecb97fec7efde09f543fb1f5ef65bd6d4f1011/stack.sh#L1421
10:23:37 melwitt that should take care of avoiding the error 'host not mapped to any cell'
10:29:43 dayou https://github.com/openstack-dev/devstack/blob/master/stack.sh#L1394
10:29:51 dayou Ok, maybe this is the problem
10:30:04 dayou I have a local.sh that is creating a server before that line
10:30:24 dayou Then it failed, so the cell mapping step running later is not executed
10:30:46 dayou melwitt, can we propose a fix to move loca.sh to the last?
10:31:02 openstackgerrit Tetiana Lashchova proposed openstack/nova-specs master: Allow modification of user-data via the server update https://review.openstack.org/547964
10:33:14 openstackgerrit yuanyue proposed openstack/nova master: Whether to use config drive depends on instance.configdrive https://review.openstack.org/485930
10:34:39 bauzas jianghuaw_: I could have like 30 mins available around 11am for discussing with you about vGPUs
10:36:32 jianghuaw_ bauzas, Great. Thanks. Where can we meet?
10:36:53 jianghuaw_ I'm now in Hogan Mezzanine L4.
10:39:01 melwitt dayou: what do you mean? run local.sh as the last step in devstack?
10:40:03 dayou https://review.openstack.org/#/c/547976/
10:40:13 dayou I mean after cells are configured
10:40:41 dayou If no cells is created and user tries to create server in local.sh, then it will fail
10:42:54 bauzas jianghuaw_: I'm in L4 too, so let's meet up in the lobby there
10:43:23 jianghuaw_ bauzas, great. See you then:-)
10:50:21 openstackgerrit Tetiana Lashchova proposed openstack/nova-specs master: Allow modification of user-data via the server update https://review.openstack.org/547964
10:53:27 melwitt dayou: yeah. so this is a matter of the devstack phases and ordering. to be honest, I'm not too familiar with that. trying to find the documentation. I thought there was a phase for plugin use that runs at the very end
10:56:07 openstackgerrit Matt Riedemann proposed openstack/nova master: Update noVNC deployment docs to mention non-US keymap fix in 1.0.0 https://review.openstack.org/547985
10:56:19 mriedem stephenfin: ^
10:56:26 mriedem that came up in the public WG room today
10:57:16 stephenfin mriedem: I assume that is not tracked by global requirements?
10:57:24 dayou melwitt, thanks for you input and help, it is very useful
10:58:49 mriedem stephenfin: since it's a native package, no it's not
10:59:45 stephenfin mriedem: Fair. +2d
10:59:57 mriedem thanks
11:00:25 mriedem stephenfin: the notes from the denver ptg said that once the fix was released in novnc, we could move forward with your patch on deprecating the keymap config option
11:00:30 mriedem https://review.openstack.org/#/c/483994/
11:00:35 mriedem so maybe you want to refresh that
11:00:58 stephenfin Will do
11:02:18 vivsoni Hi Matt
11:04:56 openstackgerrit Ghanshyam Mann proposed openstack/nova-specs master: Spec for Granular API policy https://review.openstack.org/547850
11:05:13 gmann_ mriedem: alex_xu johnthetubaguy added spec for granular API policy - https://review.openstack.org/#/c/547850/
11:06:17 stephenfin mriedem: If you haven't moved on too far, did OVH folks mention anything about the issue Tushar reported in that review?
11:06:27 stephenfin https://review.openstack.org/#/c/483994/6
11:06:40 stephenfin I can fix it, but if they have it fixed already...
11:09:44 mriedem stephenfin: nope
11:10:36 stephenfin Lovely. Time to stack
11:12:48 openstackgerrit Merged openstack/nova master: Drop compute RPC 4.x compatibility https://review.openstack.org/543580
11:18:48 openstackgerrit Dan Smith proposed openstack/nova master: Add aggregates list to Destination object https://review.openstack.org/544729
11:18:48 openstackgerrit Dan Smith proposed openstack/nova master: Add AggregateList.get_by_metadata() query method https://review.openstack.org/544728
11:18:49 openstackgerrit Dan Smith proposed openstack/nova master: Add require_tenant_aggregate request filter https://review.openstack.org/545002
11:18:49 openstackgerrit Dan Smith proposed openstack/nova master: Add request filter functionality to scheduler https://review.openstack.org/544730
11:18:50 openstackgerrit Dan Smith proposed openstack/nova master: WIP: Make get_allocation_candidates() honor aggregate restrictions https://review.openstack.org/547990
11:18:50 openstackgerrit Dan Smith proposed openstack/nova master: WIP: Honor availability_zone hint via placement https://review.openstack.org/546282
11:23:35 vivsoni mriedem: Hi
11:23:58 mriedem vivsoni: hi?
11:24:24 vivsoni one of my tempest Test case is failing
11:24:28 vivsoni http://54.201.44.218/45/539145/2/check/3par-iscsi-driver-master-client-pip-ssa02-dsvm/e43ca1d/logs/testr_results.html.gz
11:24:41 vivsoni mriedem: this is something to do with force dettach
11:25:07 vivsoni mriedem: connector is empty and driver is not implemented that case
11:25:13 vivsoni can you please guide me
11:25:29 dansmith edleafe: last week or something you said you might be interested in doing the member_of thing for allocation_candidates.. are you interested enough to commit there or did you want to look at what is required first?
11:25:30 mriedem vivsoni: no idea, i'm in sessions at the ptg right now
11:25:49 vivsoni mriedem: ok
11:26:29 melwitt dayou: based on the ordering of the phases in stack.sh, it looks like you could use the 'test-config' phase in your plugin.sh to run the steps in your local.sh. see this example: https://github.com/openstack/devstack-plugin-ceph/blob/master/devstack/plugin.sh#L95
11:26:49 mriedem Feb 26 09:30:53.005199 d-p-c-p77-1659969 devstack@c-api.service[30317]: DEBUG cinder.api.openstack.wsgi [None req-f882bcca-b050-4909-afa9-95698a62c1c7 tempest-VolumesActionsTest-1767615534 tempest-VolumesActionsTest-1767615534] Action body: {"os-force_detach": {"attachment_id": "d8020aad-e1a5-4a24-8a93-764e2086100d", "connector": null}} {{(pid=30322) get_method /opt/stack/new/cinder/cinder/api/openstack/wsgi.py:986}}
11:26:49 mriedem vivsoni: i guess you're right
11:26:54 mriedem vivsoni: but force detach is purely a cinder API
11:26:59 mriedem it has nothing to do with nova
11:27:32 vivsoni mriedem: ok.. it needs to be handled in cinder
11:28:14 smcginnis vivsoni: The driver needs to fix handling that.
11:28:47 vivsoni smcginnis: can you please guide me in fixing that
11:30:59 smcginnis vivsoni: Not really. If the driver is not passed the connector in the case of force detach, it needs to expect and handle that to detach the volume. if not connector: do_driver_stuff()?
11:32:37 vivsoni smcginnis: ok
11:33:56 melwitt mriedem: this is the queens final releases patch for us, and I was wondering, what is the diff-start entry mean? https://review.openstack.org/#/c/547519/2/deliverables/queens/nova.yaml
11:34:36 mriedem melwitt: that's an excellent question for the release team nerds, like scm
11:34:41 mriedem smcginnis:
11:34:46 melwitt ah, thanks. sorry
11:34:53 mriedem >:(
11:34:59 mriedem np
11:35:09 mriedem melwitt: it goes into the docs i believe
11:35:15 smcginnis e
11:35:15 smcginnis melwitt: That's a reno thing. It tells reno where to compile the release note from to have one "queens" release note page rather than breaking out each mileston.
11:35:39 smcginnis melwitt: No problem, reno can be a confusing thing. ;)
11:35:52 mriedem ah https://docs.openstack.org/releasenotes/nova/queens.html
11:37:54 edleafe dansmith: I don't have any other tasks lined up for placement, so yeah, I'll commit
12:02:31 melwitt smcginnis: on the diff-start thing, why is it the pike rc1 version then? how does that compile the release note for queens?
12:02:42 melwitt https://review.openstack.org/#/c/547519/2/deliverables/queens/nova.yaml
12:21:58 openstackgerrit yuanyue proposed openstack/nova master: Whether to use config drive depends on instance.configdrive https://review.openstack.org/485930
12:28:46 openstackgerrit yuanyue proposed openstack/nova master: Whether to use config drive depends on instance.configdrive https://review.openstack.org/485930
12:31:26 openstackgerrit Merged openstack/nova master: Update noVNC deployment docs to mention non-US keymap fix in 1.0.0 https://review.openstack.org/547985
13:24:16 openstackgerrit Merged openstack/nova master: rp: GET /resource_providers?required= https://review.openstack.org/546837
13:31:13 dansmith edleafe: okay this is the change I have queued up that will utilize it: https://review.openstack.org/#/c/544729/
13:47:43 openstackgerrit Dan Smith proposed openstack/nova master: Add AggregateList.get_by_metadata() query method https://review.openstack.org/544728
13:47:44 openstackgerrit Dan Smith proposed openstack/nova master: Add request filter functionality to scheduler https://review.openstack.org/544730
13:47:44 openstackgerrit Dan Smith proposed openstack/nova master: Add aggregates list to Destination object https://review.openstack.org/544729
13:47:45 openstackgerrit Dan Smith proposed openstack/nova master: Add require_tenant_aggregate request filter https://review.openstack.org/545002
13:47:45 openstackgerrit Dan Smith proposed openstack/nova master: WIP: Make get_allocation_candidates() honor aggregate restrictions https://review.openstack.org/547990
13:47:46 openstackgerrit Dan Smith proposed openstack/nova master: WIP: Honor availability_zone hint via placement https://review.openstack.org/546282

Earlier   Later