| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2017-02-01 | |||
| 17:35:42 | mriedem | btw, just let me say, i'm a professional engineer | |
| 17:36:00 | bauzas | LOL | |
| 17:36:56 | bauzas | I even use Atom | |
| 17:36:58 | mriedem | so, are https://review.openstack.org/#/c/427782/ and https://review.openstack.org/#/c/427775/ coexisting? or does the first supersede the second? | |
| 17:37:22 | melwitt | I use pdb occasionally | |
| 17:37:36 | bauzas | mriedem: I don't really like https://review.openstack.org/#/c/427775/1/nova/conductor/manager.py | |
| 17:37:36 | dansmith | mriedem: they can co-exist.. my preference is this: | |
| 17:37:56 | dansmith | mriedem: poke mine a bit and see if we hit the bug with it. If not, I'd like to merge it and get data before we merge yours | |
| 17:38:17 | bauzas | yeah, I think dansmith's change could help us getting a better situation | |
| 17:38:18 | dansmith | mriedem: yours doesn't hurt anything other than performance and maybe making the race harder to catch if it's still there, so I just wanted to delay | |
| 17:38:26 | dansmith | mriedem: but, we're also up against the deadline, so.. | |
| 17:38:36 | mriedem | -W for now | |
| 17:41:17 | cdent | bbl | |
| 17:52:26 | openstackgerrit | Oliver Walsh proposed openstack/nova: DNM - TripleO CI test https://review.openstack.org/427536 | |
| 17:59:15 | dims_ | mriedem : melwitt : dansmith : ServersNegativeTestJSON failed in the gate about 8-10 at least for the past 4 hours. looks like we are waiting on https://review.openstack.org/#/c/427775 | |
| 17:59:39 | dims_ | ah which transitively is waiting on https://review.openstack.org/#/c/427782/ | |
| 17:59:40 | dansmith | dims_: we've been working on it all morning | |
| 17:59:50 | dims_ | ack dansmith | |
| 18:00:21 | dims_ | just reporting back from watching the CI that's all :) | |
| 18:01:28 | mriedem | you're gonna get stabbed dims | |
| 18:01:47 | mriedem | lots of trolls in this cave right now :) | |
| 18:01:48 | dims_ | LOL | |
| 18:02:03 | mriedem | ok got this return/continue test finally passing | |
| 18:02:04 | mriedem | woot | |
| 18:02:22 | melwitt | dims_: yeah, it turns out https://review.openstack.org/#/c/427782 might supersede the other patch | |
| 18:03:14 | dims_ | thanks melwitt | |
| 18:05:57 | openstackgerrit | Matt Riedemann proposed openstack/nova: Continue processing build requests even if one is gone already https://review.openstack.org/427839 | |
| 18:05:58 | mriedem | melwitt: dansmith: bauzas: dtp: ^ | |
| 18:06:34 | dtp | k | |
| 18:37:25 | openstackgerrit | Dan Smith proposed openstack/nova: Allow scheduler to run cell host discovery periodically https://review.openstack.org/426826 | |
| 18:37:51 | dansmith | sfinucan: mriedem: ^^ 0 means "default" which means "every minute" which means "meltdown" | |
| 18:38:12 | dansmith | also, I added run_immediately, which will mean the scheduler will do this on startup and make this more useful for things like devstack | |
| 18:38:49 | sfinucan | dansmith: Oh, damn. I should have remembered that 🙈 | |
| 18:39:05 | dansmith | me too :/ | |
| 18:39:30 | sfinucan | Good catch, nonetheless | |
| 18:39:35 | mriedem | yar https://github.com/openstack/oslo.service/blob/master/oslo_service/periodic_task.py#L108 | |
| 18:39:44 | mriedem | https://github.com/openstack/oslo.service/blob/master/oslo_service/periodic_task.py#L122 | |
| 18:40:41 | dansmith | yeah | |
| 18:41:06 | mriedem | i think the last time i had to deal with the innards of the periodic task code was in ATL at the juno summit | |
| 18:41:21 | dansmith | mriedem: but, setting it =3600 will mean it runs once an hour, but at scheduler startup which will be useful for ironic people | |
| 18:41:29 | mriedem | i specifically remember myself and dims whispering in the back of someone's session about it | |
| 18:41:57 | mriedem | whispering, maybe some giggling | |
| 18:42:17 | mriedem | yeah on startup seems ok | |
| 18:45:23 | mriedem | ha | |
| 18:45:24 | mriedem | "I think Russian already owns the bear." | |
| 18:46:44 | melwitt | oh yeah, the mascot angst thread | |
| 18:47:15 | mriedem | i didn't have the context, i thought he was just saying, generally, | |
| 18:47:22 | mriedem | russia owns all likenesses of bears | |
| 18:47:23 | mriedem | period | |
| 18:47:43 | mriedem | which is something i'd expect from a russian given my US stereotype of them from the 80s | |
| 18:47:44 | dansmith | I thought Bear == Trump | |
| 18:47:58 | mriedem | gonna need to call in rocky balboa to settle this one | |
| 18:48:08 | melwitt | I think it's explained by a later reply about a russian meme | |
| 18:48:20 | mriedem | melwitt: yeah i read further | |
| 18:48:25 | mriedem | dansmith: trump is more of a peacock | |
| 18:48:37 | mriedem | beautiful to behold, but functionally useless | |
| 18:48:44 | dansmith | um. | |
| 18:49:03 | mriedem | what about "donal trump is beautiful to behold" don't you get?! | |
| 18:50:01 | melwitt | mriedem: I get what you meant now. it sounded like that to me too before someone else explained it | |
| 18:55:55 | mriedem | random thought: no one has asked me lately if i can smell what the rock is cooking | |
| 19:06:11 | artom | Oh God, Bear/Bare metal | |
| 19:08:39 | openstackgerrit | John Garbutt proposed openstack/nova-specs: WIP: write up policy ideas for pike https://review.openstack.org/427872 | |
| 19:13:15 | mriedem | johnthetubaguy: have you taken a gander at this xen live migration patch? https://review.openstack.org/#/c/424428/ | |
| 19:13:30 | mriedem | it mods the rpc version on the migrate_data object so won't be able to backport it if we don't get it in | |
| 19:19:52 | mriedem | dansmith: looks like your change is failing some functional tests http://logs.openstack.org/82/427782/2/check/gate-nova-tox-db-functional-ubuntu-xenial/f53a7e9/console.html#_2017-02-01_17_41_39_148413 | |
| 19:20:08 | dansmith | mriedem: I know, already fixed, just wanted to get a devstack run before I push it up again | |
| 19:20:12 | dansmith | looks like I'm good | |
| 19:20:13 | mriedem | ok | |
| 19:20:27 | mriedem | fwiw i agree with bauzas on keeping the tests but changing them to assert that InstanceNotFound is raised | |
| 19:20:43 | dansmith | I know, and I hate you guys for that, but working on that too | |
| 19:20:49 | mriedem | \o/ | |
| 19:22:38 | mriedem | bauzas: https://review.openstack.org/#/c/427667/ passed after the recheck, are you happy? | |
| 19:24:01 | mriedem | nvm, i +W | |
| 19:24:10 | johnsom | I just did a fresh reclone devstack this morning and cirros is going to error on boot with: | |
| 19:24:12 | johnsom | VIFBridge(active=False,address=fa:16:3e:6f:0e:84,bridge_name='qbrd3377ad5-43',has_traffic_filtering=True,id=d3377ad5-4397-474c-b6dd-e29b9a52d277,network=Network(d2582e93-d6e5-43ac-be7b-fe3fc3c7e593),plugin='ovs',port_profile=VIFPortProfileBase,preserve_on_delete=False,vif_name='tapd3377ad5-43') | |
| 19:24:12 | johnsom | n-cpu.log:2017-02-01 11:13:32.876 ERROR os_vif [req-17c8b4e4-2197-4205-aed3-007d0f2837e4 admin admin] Failed to plug vif | |
| 19:24:25 | johnsom | Is this a known thing? | |
| 19:24:39 | mriedem | johnsom: no, is there an actual error? | |
| 19:24:52 | mriedem | either in nova or neutron? | |
| 19:24:59 | johnsom | \"/opt/stack/nova/nova/compute/manager.py\", line 1780, in _do_build_and_run_instance | | |
| 19:24:59 | johnsom | | fault | {"message": "Failure running os_vif plugin plug method: Failed to plug VIF VIFBridge(active=False,address=fa:16:3e:6f:0e:84,bridge_name='qbrd3377ad5-43',has_traffic_filtering=True,id=d3377ad5-4397-474c-b6dd-e29b9a52d277,network=Network(d2582e93-d6e5-43ac-be7b-fe3fc3c7", "code": 500, "details": " File | |
| 19:25:29 | openstackgerrit | Dan Smith proposed openstack/nova: Remove pre-cellsv2 short circuit in instance get https://review.openstack.org/427782 | |
| 19:25:30 | dansmith | mriedem: ^ | |
| 19:25:33 | mriedem | johnsom: gonna need a stacktrace or something from n-cpu logs | |
| 19:25:37 | mriedem | johnsom: in a paste | |
| 19:25:43 | mriedem | the error could be on the neutron side too | |
| 19:25:56 | johnsom | https://www.irccloud.com/pastebin/1IAgqzBE/ | |
| 19:25:59 | johnsom | Ah, found it | |
| 19:26:43 | mriedem | cool | |
| 19:27:21 | dansmith | lol | |
| 19:29:51 | manjeets | hi I just hit the issue in devstack saying placement is not configured | |
| 19:29:53 | manjeets | http://paste.openstack.org/show/597262/ | |
| 19:30:15 | manjeets | is there any doc that explains that configuration ?? | |
| 19:30:16 | mriedem | manjeets: http://lists.openstack.org/pipermail/openstack-dev/2017-January/111295.html | |
| 19:32:47 | openstackgerrit | Vasyl Saienko proposed openstack/nova: Revert "Fresh resource provider in RT must have generation 0" https://review.openstack.org/427877 | |
| 19:33:30 | mriedem | vsaienko: is that just for testing? | |
| 19:33:59 | vsaienko | mriedem yes | |
| 19:37:57 | johnsom | So, looking at that trace, I'm not sure if it's a nova, neutron, or oslo issue... | |
| 19:38:39 | mriedem | dansmith: want me to work https://review.openstack.org/#/c/426826/ into https://review.openstack.org/#/c/424737/ or should we just mention the periodic task in the cells v2 devref? | |
| 19:38:53 | manjeets | thanks mriedem | |