Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2017-01-17
14:45:21 tdurakov acked, let me check it again
14:45:28 markus_z thanks!
14:49:23 sean-k-mooney mriedem just looking now
14:50:45 sean-k-mooney mriedem: looking at the code the vhost-user socket name must = the name of the openvswitch port
14:51:27 sean-k-mooney mriedem: if we called _get_vif_name() we would have to modify neutron to generate the socket path differently
14:52:19 sean-k-mooney my intention was to modify neutron to pass the name of the device in the port binding
14:52:36 sean-k-mooney and then nave nova use that if present or fall back to generating it if not set
14:54:32 mriedem ok so in other words this is fine
14:55:23 sean-k-mooney mriedem: yes what is there is currently fine but i would like to decouple it in the future so that we dont have to ensure that neutron and nova generate the name with the same algoritim
14:56:06 tdurakov markus_z: done, thank you
14:56:24 markus_z tdurakov: yeah, thanks!
14:56:32 tdurakov mriedem, sdague hi, could you please take a look https://review.openstack.org/#/c/347471/43
14:57:27 jaypipes sdague: well said, on the PWG thing. I won't say that publicly, of course, since I'd get my head chewed off internally.
14:58:43 cdent jaypipes: you must have very flexible teeth or jaw
15:03:18 mriedem sean-k-mooney: ok i've also gone through https://review.openstack.org/#/c/385061/
15:10:29 mriedem alex_xu: your intel buddies might want to look at this https://review.openstack.org/#/c/420960/1/nova/objects/fields.py
15:10:36 mriedem pkoniszewski: ^
15:12:33 cdent _gryf: I think that endpoint stuff is simply that I hadn't revisited that update_inventory code in a long time (perhaps 4 months?)
15:13:54 _gryf cdent, ok, cool
15:22:21 alex_xu mriedem: that sounds like that overlap with traits. in the future we can request specific trait in the image metadata, the requested trait can be cpu feature. there is the spec about request traits https://review.openstack.org/#/c/351063/3/specs/ocata/approved/request-standardized-capabilities.rst@66
15:27:05 pkoniszewski thanks mriedem, i will take a look
15:31:50 cdent edleafe: I think you're on PTO, but if not I fixed that bug in gabbi and made a new one: https://pypi.python.org/pypi/gabbi
15:34:13 openstackgerrit Diana Clarke proposed openstack/nova: nova-manage docs: add cells commands prep https://review.openstack.org/421320
15:48:26 openstackgerrit Sylvain Bauza proposed openstack/nova: Add ComputeNodeList.get_all_by_uuids method https://review.openstack.org/418134
15:48:26 openstackgerrit Sylvain Bauza proposed openstack/nova: WIP: Scheduler calling the Placement API https://review.openstack.org/417961
15:48:58 bauzas mriedem: jaypipes: folks I need to bail out for a few, but you can bite ^
15:49:09 openstackgerrit Vladyslav Drok proposed openstack/nova: Generate necessary network metadata for ironic port groups https://review.openstack.org/388756
15:53:59 openstackgerrit Diana Clarke proposed openstack/nova: nova-manage docs: cell_v2 simple_cell_setup https://review.openstack.org/421333
15:54:34 openstackgerrit Zhenyu Zheng proposed openstack/nova: Add query parameters white list for server list/detail https://review.openstack.org/408571
15:56:39 openstackgerrit Zhenyu Zheng proposed openstack/nova: Add sort_key white list for server list/detail https://review.openstack.org/415142
15:57:50 openstackgerrit Zhenyu Zheng proposed openstack/nova: Add new policy for server list/detail with all_tenants https://review.openstack.org/415330
16:08:16 mriedem tdurakov: markus_z: for https://review.openstack.org/#/c/347471/ i see it depends on a tempest patch to enable the test, do we have a patch up anywhere that depends on the nova change to see it all working and running the serial console tests in the live migration job?
16:08:43 mriedem or is it self-testing?
16:09:02 markus_z mriedem: it's self testing
16:09:31 mriedem oh really http://logs.openstack.org/71/347471/43/check/gate-grenade-dsvm-neutron-multinode-live-migration-nv/c427a53/console.html#_2017-01-17_13_10_19_246017
16:09:59 markus_z http://logs.openstack.org/71/347471/43/check/gate-grenade-dsvm-neutron-multinode-live-migration-nv/c427a53/console.html#_2017-01-17_13_10_22_840459
16:10:45 markus_z We have to skip in in grenade. Let me search for the Ocata-Ocata case.
16:11:04 mriedem oh right
16:11:17 mriedem this is the o-o one http://logs.openstack.org/71/347471/43/check/gate-tempest-dsvm-multinode-live-migration-ubuntu-xenial/0b45362/console.html#_2017-01-17_12_34_13_290332
16:11:21 mriedem still skipped
16:11:52 mriedem melwitt: i'm struggling trying to get discover_hosts to work on the subnode in a multinode job http://logs.openstack.org/81/420981/5/check/gate-tempest-dsvm-neutron-multinode-full-ubuntu-xenial-nv/6767a3a/logs/subnode-2/devstacklog.txt.gz#_2017-01-17_14_49_15_138
16:11:59 mriedem melwitt: please tell me there is something you know about this that i don't
16:12:11 mriedem the subnode nova.conf has the api_database connection and we're getting the cell mappings from the api db ok,
16:12:23 mriedem it's the target_cell that's failing on the non-cell0 cell
16:12:37 mriedem for some reason it's trying to connect to 127.0.0.1
16:13:13 markus_z mriedem: That was the first run before the hook enables the serial console. After the hook we have http://logs.openstack.org/71/347471/43/check/gate-tempest-dsvm-multinode-live-migration-ubuntu-xenial/0b45362/console.html#_2017-01-17_12_35_19_716558
16:13:56 markus_z mriedem: The teardown of the hook sets everything back, so we have http://logs.openstack.org/71/347471/43/check/gate-tempest-dsvm-multinode-live-migration-ubuntu-xenial/0b45362/console.html#_2017-01-17_12_36_38_329078
16:20:19 shaner hi all - any idea why live-migrations won't show up when issuing 'migrations-list'?
16:20:30 shaner (in kilo)
16:21:07 shaner only those that were migrated using 'migrate' show up .. vs 'live-migration'
16:25:31 jaypipes shaner: pls see /topic. best to ask on ask.openstack.org or mailing list for usage questions.
16:25:53 shaner jaypipes, thx, will do
16:26:23 mriedem migrate is cold migration
16:27:27 openstackgerrit int32bit proposed openstack/python-novaclient: Allow multiple tag add/delete from cli https://review.openstack.org/414635
16:28:12 shaner mriedem, shouldn't 'migration-list' list all types of migrations though?
16:28:59 mriedem i don't think until 2.23 http://docs.openstack.org/developer/nova/api_microversion_history.html#id21
16:29:23 mriedem fwiw this is the api code in kilo https://github.com/openstack/nova/blob/kilo-eol/nova/api/openstack/compute/contrib/migrations.py#L53
16:30:00 openstackgerrit int32bit proposed openstack/nova: Choose VMware datastore by the least provisioned space https://review.openstack.org/402632
16:36:36 mriedem shaner: was the migration in progress or complete? if the migration was complete i wonder if the migration record was deleted
16:36:50 mriedem at this point kilo is pretty far eol so it would take some investigating
16:37:47 shaner mriedem, it seems either way, in-progress/complete, it doesn't show up
16:37:58 shaner I've tested it several times to no avail
16:38:13 shaner appreciate the info!
16:38:46 mriedem have you looked at the migrations table in the nova db?
16:39:07 mriedem are there any records where deleted = 0?
16:39:39 shaner mriedem, let me check
16:43:13 mriedem melwitt: hmm this is interesting
16:43:28 mriedem the db connection for the cell0 mapping in this failed run has localhost in it http://logs.openstack.org/81/420981/5/check/gate-tempest-dsvm-neutron-multinode-full-ubuntu-xenial-nv/6767a3a/logs/devstacklog.txt.gz#_2017-01-17_14_43_53_943
16:43:44 mriedem nova-manage cell_v2 map_cell0 --database_connection 'mysql+pymysql://root:secretmysql@127.0.0.1/nova_cell0?charset=utf8'
16:45:15 mriedem when we call simple_cell_setup the transport_url has a valid ip in it nova-manage cell_v2 simple_cell_setup --transport-url rabbit://stackrabbit:secretrabbit@10.20.124.3:5672/
16:45:40 andreykurilin mriedem: hi! I remember you spent a bit of time investigation an issue with novaclient's gate failures (which was fixed by blocking openstacksdk 0.9.12). If you have any details please share them here - https://review.openstack.org/#/c/418663
16:47:28 mriedem andreykurilin: done
16:47:34 andreykurilin mriedem: thanks
16:50:12 mriedem hmm wtf http://logs.openstack.org/81/420981/5/check/gate-tempest-dsvm-neutron-multinode-full-ubuntu-xenial-nv/6767a3a/logs/devstacklog.txt.gz#_2017-01-17_14_27_19_600
16:50:19 mriedem BASE_SQL_CONN=mysql+pymysql://root:secretmysql@127.0.0.1
16:53:18 alaski mriedem: yesterday was a holiday for me so I missed your message. Thanks for not letting a national figure distract from the important birthday
16:54:07 mriedem alaski: you're MY national figure
16:54:13 shaner mriedem, all items in migrations db are cold migrations and indeed have 'deleted = 0'
16:54:44 shaner I don't see any from the past live-migrations I've done, so it seems, a live-migration isn't updating this table?
16:54:45 alaski mriedem: :)
16:54:50 mriedem DATABASE_HOST=MYSQL_HOST=SERVICE_LOCAL_HOST=127.0.0.1 blarg!
16:54:56 alaski mriedem: I see there was a question of naming, did that get resolved?
16:55:17 mriedem shaner: would have to check if live migrations were tracked with migrations records in kilo
16:55:52 mriedem alaski: sort of https://review.openstack.org/#/c/420439/
16:55:58 mriedem there was much gnashing of teeth
16:56:12 mriedem alaski: https://review.openstack.org/#/c/420439/5/nova/cmd/manage.py@1122
16:56:37 mriedem basically we weren't sure about the original intent of using the api_database.connection for the cell0 netloc and name
16:56:53 mriedem but all agreed that nova_api_cell0 was a confusing db name for a db that is not using the api db schema
16:57:18 alaski yep
16:57:19 mriedem but also agreed that cell0 might not be on the same host as the api db, or main db,
16:57:38 mriedem and finally agreed that us trying to guess is a bad idea and people should just provide the value explicitly
16:57:45 alaski right, in a large deployment I expect all dbs to be separate
16:57:58 mriedem we probably need to make a note in our docs for ^
16:58:16 mriedem i.e. .. note:: we'll guess, but might be wrong, and then cell0 won't work, don't say we didn't tell you
16:59:02 mriedem sdague: so in devstack, if you're not on the subnode DATABASE_HOST isn't set,
16:59:13 mriedem sdague: and on the api (primary) node, it gets defaulted to localhost: DATABASE_HOST=MYSQL_HOST=SERVICE_LOCAL_HOST=127.0.0.1
16:59:23 alaski yeah. the simple_cell_setup command seems to sit in the same place as something like devstack, i.e. it'll get you setup and show you what needs doing but probably isn't what you want for production
16:59:42 mriedem sdague: but the devstack docs say that DATABASE_HOST=$SERVICE_HOST by default if not set, and that's also what's used for the MQ host URL
16:59:49 mriedem sdague: so anything you know of that would be bad for me trying to fix that?
17:00:33 alaski mriedem: so the original reason for colocating cell0 with the api db vs the main db is that there is only ever expected to be one api db and one cell0 db, but there could be multiple "main" dbs. So thinking of where cell0 is gets confusing when there's multiple

Earlier   Later