| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2018-03-01 | |||
| 12:19:02 | danpawlik | I see that since pike (https://docs.openstack.org/nova/pike/user/placement.html) there is a plan to add speciall DB for placement. So it looks like there is a bug in puppet-openstack-integration for Ocata branch | |
| 12:19:04 | danpawlik | or? | |
| 12:22:12 | cdent | danpawlik: can you point me to the puppet code? | |
| 12:22:37 | cdent | danpawlik: the hope is that eventually there would be a different database, but at the moment it is usually the nova api_db | |
| 12:27:13 | openstackgerrit | Chris Dent proposed openstack/nova-specs master: Express forbidden traits in placement API https://review.openstack.org/548915 | |
| 12:31:28 | openstackgerrit | Eric Fried proposed openstack/nova master: placement: Return new provider from POST /rps https://review.openstack.org/548934 | |
| 12:31:55 | efried | jaypipes: cdent ^ | |
| 12:55:33 | openstackgerrit | Merged openstack/nova master: Remove unused LOG variables https://review.openstack.org/546097 | |
| 12:55:42 | openstackgerrit | Merged openstack/nova master: Check for leaked server resource allocations in post_test_hook https://review.openstack.org/538510 | |
| 12:55:49 | openstackgerrit | Merged openstack/nova master: Add check for redundant import aliases https://review.openstack.org/543995 | |
| 12:56:00 | openstackgerrit | Merged openstack/nova master: Address nits in change I7e01f95d7173d9217f76e838b3ea71555151ef56 https://review.openstack.org/532898 | |
| 12:56:12 | openstackgerrit | Merged openstack/nova master: Remove warning in feature support matrix page https://review.openstack.org/544017 | |
| 12:56:23 | openstackgerrit | Merged openstack/nova master: [placement] api-ref: Fix a missing response code https://review.openstack.org/546494 | |
| 13:09:35 | openstackgerrit | Merged openstack/nova master: Enable native mode for ScaleIO volumes https://review.openstack.org/545304 | |
| 13:15:20 | danpawlik | cdent: sure, https://github.com/openstack/puppet-openstack-integration/blob/stable/ocata/manifests/nova.pp#L111 | |
| 13:43:14 | stephenfin | melwitt: I'm trying to organize accommodation for tomorrow night. Will be back in he room as soon as that's sorted | |
| 13:43:44 | stephenfin | melwitt: If you're starting on the neutron session, maybe do the other stuff first and I'll be back for the NUMA-aware vSwitch section, please :) | |
| 13:46:58 | melwitt | stephenfin: ack | |
| 13:48:14 | melwitt | stephenfin: we've been told the croke park hotel has rooms available that are not released to the public. did you ask there first? | |
| 14:02:28 | danpawlik | cdent: probably you didn't receive the url | |
| 14:02:45 | danpawlik | cdent: https://github.com/openstack/puppet-openstack-integration/blob/stable/ocata/manifests/nova.pp#L111 | |
| 14:02:51 | cdent | danpawlik: yeah, sorry, left the network for lunch | |
| 14:02:58 | cdent | and now I'm on some horrible network | |
| 14:03:19 | danpawlik | xD | |
| 14:05:05 | cdent | interesting, I'm going to have to do some grepping | |
| 14:05:15 | cdent | which will mean some waiting on the network | |
| 14:09:10 | danpawlik | cdent: I can send you url for nova module if you want | |
| 14:09:40 | cdent | danpawlik: grepping done. looks like EmilienM did that change around about the time when we were recommending that a separate database was the way to go, but then we didn't quite make it there | |
| 14:10:00 | cdent | which probably means that though the puppet variable is set it may not be actually used | |
| 14:10:03 | danpawlik | cdent: so I can revert that change for Octa? | |
| 14:10:23 | danpawlik | ack | |
| 14:10:26 | cdent | danpawlik: I can't remember what you said before: is it breaking something, or did you just notice that the puppet was weird? | |
| 14:11:02 | danpawlik | cdent: I was wondering what DB should I use for [placement_database] connection param | |
| 14:11:41 | cdent | in ocata there's no support in placement itself for using a database connection other than the nova db api | |
| 14:11:44 | danpawlik | cdent: so to summarry what you told me: I can use nova_api DB for https://github.com/openstack/puppet-openstack-integration/blob/stable/ocata/manifests/nova.pp#L111 but normally it will be ignored in Ocata ye? | |
| 14:11:56 | cdent | yes | |
| 14:11:57 | danpawlik | ack | |
| 14:12:24 | cdent | danpawlik: I'll see EmilienM a bit later in the day and we'll chat about it | |
| 14:12:33 | danpawlik | cdent: thats what I was looking for | |
| 14:12:52 | danpawlik | I try to catch him on irc later :) | |
| 14:13:00 | danpawlik | cdent: thanks! | |
| 14:26:49 | mriedem | jroll: fyi https://docs.openstack.org/neutron/latest/admin/config-routed-networks.html | |
| 14:28:38 | mriedem | linked some summit videos in the etherpad as well | |
| 14:31:43 | jroll | mriedem: thanks, I was getting conflicting info earlier this week. making a todo to make sure nova docs are up to date with that | |
| 14:31:59 | jroll | or maybe it was pike neutron docs I had | |
| 14:39:25 | openstackgerrit | Merged openstack/nova stable/pike: Clean up ports and volumes when deleting ERROR instance https://review.openstack.org/546221 | |
| 14:45:29 | openstackgerrit | Merged openstack/nova stable/pike: Add functional test for deleting BFV server with old attach flow https://review.openstack.org/546812 | |
| 14:50:38 | openstackgerrit | Eric Fried proposed openstack/nova master: placement: Return new provider from POST /rps https://review.openstack.org/548934 | |
| 15:31:06 | Spazmotic | Work computer what did you do | |
| 15:36:44 | mriedem | jroll: could have been tied up in the stuff where we were going to move port creation to conductor so we could do network-aware scheduling | |
| 15:37:11 | mriedem | i'm still not really sure in how all of the neutron stuff works today | |
| 15:39:02 | mriedem | dansmith: if i don't care about reading deleted records from the cell db, or filtering those records based on project_id/user_id, then i don't need to use the model_query() in the DB API right? | |
| 15:39:31 | openstackgerrit | Merged openstack/nova stable/pike: Ensure attachment_id always exists for block device mapping https://review.openstack.org/546275 | |
| 15:39:49 | dansmith | mriedem: you'll get deleted records if you don't use model_query().. if that's what you don't care about, then yeah | |
| 15:40:04 | mriedem | i don't want deleted records....damn | |
| 15:40:05 | mriedem | ok | |
| 15:40:11 | mriedem | well, i can just filter on deleted=0 | |
| 15:40:18 | melwitt | yes, that | |
| 15:45:09 | openstackgerrit | Chris Dent proposed openstack/nova master: WIP: Move resource provider objects into placement hierarchy https://review.openstack.org/540049 | |
| 15:45:10 | openstackgerrit | Chris Dent proposed openstack/nova master: Isolate placement database config https://review.openstack.org/541435 | |
| 15:45:10 | openstackgerrit | Chris Dent proposed openstack/nova master: Optional separate database for placement API https://review.openstack.org/362766 | |
| 15:45:14 | dansmith | mriedem: yep | |
| 15:46:03 | openstackgerrit | Eric Fried proposed openstack/nova master: Stop assuming initial provider generation is 0 https://review.openstack.org/548975 | |
| 15:47:23 | stephenfin | gibi: would tx/rx queue size be good to bring up now? Kind of neutron stuff? | |
| 15:47:35 | gibi | stephenfin: could be | |
| 15:47:46 | stephenfin | Go for it? (assuming we have time) | |
| 15:47:54 | gibi | stephenfin: OK, I will try | |
| 15:49:04 | openstackgerrit | Chris Dent proposed openstack/nova-specs master: Express forbidden traits in placement API https://review.openstack.org/548915 | |
| 16:07:09 | openstackgerrit | Eric Fried proposed openstack/nova master: WIP: report client: placement API version discovery https://review.openstack.org/548983 | |
| 16:32:37 | lyarwood | mriedem: https://review.openstack.org/#/c/269867/ - before I forget can you restore your connector o.vo change? | |
| 16:33:39 | mriedem | lyarwood: done | |
| 16:34:00 | lyarwood | mriedem: tanks | |
| 16:35:41 | mriedem | yar, wood | |
| 16:36:34 | openstackgerrit | Lee Yarwood proposed openstack/nova stable/queens: Handle spawning error on unshelving https://review.openstack.org/548415 | |
| 16:38:41 | lyarwood | argh not a fan of the gerrit UI using bug/1234-stable/foo for cherry-picks | |
| 16:56:18 | mriedem | melwitt: dansmith: TheJulia: jroll: i think we need to get out of this room | |
| 16:57:26 | lyarwood | argh not a fan of the gerrit UI using bug/1234-stable/foo for cherry-picks~. | |
| 17:04:57 | openstackgerrit | Eric Berglund proposed openstack/nova master: PowerVM Driver: Snapshot https://review.openstack.org/543023 | |
| 18:12:40 | openstackgerrit | Merged openstack/nova master: Use dict.get() when accessing capabilities dict https://review.openstack.org/539715 | |
| 18:16:20 | kukacz | hi, I'm facing an issue of unbootable nova instance in Ocata - it happens when both boot from volume and config drive is applied while virtio-scsi controller is used. I believe the cause is wrong order of scsi devices ("unit" number in scsi address, specifically) generated in libvirt config, where configdrive is placed prior the root volume | |
| 18:16:41 | kukacz | the instance can boot correctly after swapping the "unit" numbers using `virsh edit` so that the root volume is assigned unit=0 | |
| 18:17:35 | kukacz | I could not find a bug for this. before filing up one I wished to ask if that could be just my misconfiguration? | |
| 18:18:33 | hrw | lyarwood: 'git review -d CHANGENUMBER'? | |
| 19:22:35 | MikeW | Hey guys what var in nova.conf controls a hypervisors availability zone? | |
| 19:24:28 | MikeW | nova_default_schedule_zone seems to be the only variable | |
| 21:32:41 | openstackgerrit | Tetsuro Nakamura proposed openstack/nova-specs master: Support shared and dedicated VMs in one host https://review.openstack.org/543805 | |
| 21:43:33 | openstackgerrit | Merged openstack/nova master: Convert users of tune2fs to privsep. https://review.openstack.org/519484 | |
| 21:43:41 | openstackgerrit | Merged openstack/nova master: Move makefs to privsep https://review.openstack.org/527510 | |
| 21:55:44 | openstackgerrit | Eric Berglund proposed openstack/nova master: PowerVM Driver: DiskAdapter parent class https://review.openstack.org/549053 | |
| 21:56:16 | openstackgerrit | Eric Berglund proposed openstack/nova master: PowerVM Driver: DiskAdapter parent class https://review.openstack.org/549053 | |
| 22:17:09 | openstackgerrit | Tetsuro Nakamura proposed openstack/nova-specs master: Support shared/dedicated vCPUs in one instance https://review.openstack.org/545734 | |
| 22:38:30 | mnaser | CRITICAL nova [None req-91b430be-aff9-4149-9167-fab5f4f9d5d7 None None] Unhandled error: MessageDeliveryFailure: Unable to connect to AMQP server on 10.30.170.88:5672 after None tries: Connection.open: (530) NOT_ALLOWED - access to vhost 'nova_cell1' refused for user 'stackrabbit' | |
| 22:38:38 | mnaser | and then nova dies right away without retrying | |
| 22:38:50 | mnaser | do we consider that a bug or it's ok that it doesnt retry? | |
| 22:39:10 | mnaser | https://github.com/openstack/nova/blob/stable/pike/nova/conductor/api.py#L61-L78 | |
| 22:39:21 | mnaser | it looks like this doesn't handle the exception if it is unable to connect | |
| 22:45:52 | gibi | mnaser: if this is a problem with credentials as the message suggest then I think it is OK not to retry as it is not something that will succeed without operator intervention | |
| 22:47:16 | mnaser | gibi: sometimes race conditions in deployments happen where control plane might not be ready yet. | |
| 22:47:33 | mnaser | gibi: question is — would anyone be opposed to patching something like that | |
| 22:58:54 | openstackgerrit | Tetsuro Nakamura proposed openstack/nova-specs master: Support shared and dedicated VMs in one host https://review.openstack.org/543805 | |