| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2017-02-03 | |||
| 14:01:15 | jroll | cdent: +1'd. now I'm wondering if we want to catch NodeNotFound and have a better error message, but it's probably fine | |
| 14:01:50 | mriedem | jroll: you could still probably get the NodeNotFound for other reasons | |
| 14:01:58 | mriedem | like some kind of race to delete the node and schedule to it at the same time | |
| 14:02:01 | mriedem | i'd just leave it | |
| 14:02:18 | jroll | mriedem: good point | |
| 14:02:38 | mriedem | bauzas: i don't really have that all in my head, but dansmith has a WIP up for some of this | |
| 14:02:55 | mriedem | https://review.openstack.org/#/c/428481/ | |
| 14:07:55 | figleaf | cdent: thanks for picking up https://review.openstack.org/428375 | |
| 14:08:39 | mriedem | +2 from me | |
| 14:08:53 | cdent | figleaf: you're welcome, seemed like that one had some rush on it. If you're busy today I could also pick up the allocation schema stuff too (on which I left some comments about some wrongness) | |
| 14:09:34 | figleaf | cdent: No, not busy. Just getting started, and the caffeine hasn't yet kicked in | |
| 14:09:47 | figleaf | Gonna look at those next | |
| 14:09:48 | cdent | roger that | |
| 14:09:54 | jroll | figleaf: it's 8am already, why so late | |
| 14:10:19 | figleaf | jroll: had to get the rest of the house out to work/school | |
| 14:10:24 | cdent | I guess that means I should think about writing a "placement update" which I've been avoiding because I have no clue, probably because caffeine | |
| 14:10:27 | figleaf | jroll: gotta be able to focus! | |
| 14:10:35 | jroll | figleaf: that was a joke, to be clear :P | |
| 14:10:54 | figleaf | jroll: so was my response :) | |
| 14:11:11 | jroll | :D | |
| 14:11:31 | cdent | s/before/because/ | |
| 14:16:26 | cdent | mriedem: imma fix that nit | |
| 14:17:12 | openstackgerrit | Chris Dent proposed openstack/nova master: Delete a compute node's resource provider when node is deleted https://review.openstack.org/428375 | |
| 14:20:08 | bauzas | cdent: figleaf: open questions in https://review.openstack.org/#/c/428375/9 | |
| 14:20:42 | figleaf | cdent: regarding https://review.openstack.org/#/c/416752/8/nova/api/openstack/placement/handlers/inventory.py, what should be the value for 'max_unit' in the INVENTORY_DEFAULTS (line 109) ? | |
| 14:22:06 | cdent | figleaf: I would think since we want to set a schema minimum of 1 (since less than that is meaningless or wrong) then the default should be 1 as well? | |
| 14:22:09 | cdent | bauzas: looking | |
| 14:23:18 | bauzas | mriedem: when are you removing -W for https://review.openstack.org/#/c/428531/ ? | |
| 14:24:39 | mriedem | bauzas: probably after https://review.openstack.org/#/c/428375/ is merged now | |
| 14:24:46 | mriedem | johnthetubaguy: +2 on ^ now | |
| 14:25:01 | johnthetubaguy | looking... | |
| 14:25:11 | figleaf | cdent: didn | |
| 14:25:13 | figleaf | doh | |
| 14:25:13 | mriedem | cdent: figleaf: a functional report client test as noted by bauzas for deleting RPs would be good, but i'd do that in a follow up | |
| 14:25:38 | mriedem | bauzas: i already talked to tonyb about blocking requirements yesterday, it's not a big deal right now | |
| 14:25:52 | mriedem | dims_: was telling everyone that yesterday was a soft date for rc1 and monday would be ok | |
| 14:25:52 | figleaf | cdent: didn't seem right to set the default 'max' to 1 | |
| 14:26:03 | cdent | figleaf: it is weird yeah, but is better than 0 | |
| 14:26:03 | mriedem | bauzas: but i'll get rc1 out today if we get this last fix in | |
| 14:26:29 | figleaf | cdent: sure, but even db.MAX_INT felt better (even though that seems wrong, too) | |
| 14:26:48 | bauzas | mriedem: okay, cool with me about RC1 | |
| 14:26:59 | mriedem | i've got to get my kid on the bus and head into the office | |
| 14:27:02 | bauzas | cdent: mriedem: about https://review.openstack.org/#/c/428375/8/nova/scheduler/client/report.py@772 my question is rather | |
| 14:27:49 | cdent | figleaf: hmmm. yeah. a good point. that is probably better than 1 as it results in things being left in a working state if a max is not provided | |
| 14:27:49 | bauzas | cdent: mriedem: if we're deleting a compute node, are we also deleting the nested instances elsewhere ? | |
| 14:28:22 | cdent | figleaf: which is what DEFAULTS ought to be going for | |
| 14:28:54 | cdent | figleaf: so yeah, step_size 1, min_unit 1, max_unit MAX_INT seems right for the defaults | |
| 14:29:00 | figleaf | cdent: "Hi, I'd like 1 MEMORY_MB for my VM, please" | |
| 14:29:29 | cdent | it's more that 640k so it's gotta be enough | |
| 14:31:08 | cdent | metoo@aol.com | |
| 14:35:51 | bauzas | johnthetubaguy: I had a concern on https://review.openstack.org/#/c/428375/ | |
| 14:36:07 | bauzas | I did +1 because I didn't wanted to pollute and say it was a blocker | |
| 14:36:19 | bauzas | but I'm still wondering if all goes the right way | |
| 14:37:21 | johnthetubaguy | bauzas: most of the time the list of instances will be empty, I think | |
| 14:37:45 | johnthetubaguy | bauzas: its hitting the 409 from placement I guess, that would cause an issue | |
| 14:38:16 | bauzas | johnthetubaguy: I need to look at code, because I wonder that | |
| 14:38:42 | bauzas | johnthetubaguy: say I'm booting inst1, then deleting computeA, do I get a nova list empty? | |
| 14:38:50 | bauzas | I suppose so | |
| 14:38:58 | bauzas | so I'm wondering where the thing is done | |
| 14:44:47 | openstackgerrit | Chris Dent proposed openstack/nova master: Add placement request id to log when GET or POST rps https://review.openstack.org/427740 | |
| 14:44:47 | openstackgerrit | Chris Dent proposed openstack/nova master: Add more debug logging on RP inventory update failures https://review.openstack.org/414230 | |
| 14:44:48 | openstackgerrit | Chris Dent proposed openstack/nova master: Add more debug logging on RP inventory delete failures https://review.openstack.org/426290 | |
| 14:44:48 | openstackgerrit | Chris Dent proposed openstack/nova master: Add placement request id to log when GET aggregates https://review.openstack.org/426874 | |
| 14:45:41 | cdent | johnthetubaguy: the bottom of that stack ^ was +W (+2 from you) but failing in the gate a few times, so I've rebased but now it is all bereft of its approval | |
| 14:46:27 | openstackgerrit | Sylvain Bauza proposed openstack/nova master: Cover again the ReqSpec record creation testing https://review.openstack.org/391059 | |
| 14:46:27 | openstackgerrit | Sylvain Bauza proposed openstack/nova master: Destroy the ReqSpec object when deleting the instance https://review.openstack.org/391060 | |
| 14:54:22 | openstackgerrit | Ed Leafe proposed openstack/nova master: Add check for invalid allocation amounts https://review.openstack.org/416752 | |
| 14:54:24 | figleaf | cdent: ^^ Addressed your concerns | |
| 14:57:38 | johnthetubaguy | cdent: ah, doh, will take a peak | |
| 15:06:30 | johnthetubaguy | cdent: I do wonder if we have too many new lines in some of the error messages in the logs, BTW | |
| 15:07:22 | johnthetubaguy | cdent: its a micro nit, it might be more to do with the error message omitted by the placement API and how that looks in the logs | |
| 15:07:31 | mriedem | that's the nature of the requests.response isn't it? | |
| 15:09:25 | johnthetubaguy | it could be | |
| 15:12:22 | cdent | johnthetubaguy, mriedem: that could be the way that webob is formatting error responses. the json_error_formatter stuff subclasses something which puts newlines in here and there for no apparently reason | |
| 15:12:57 | johnthetubaguy | cdent: I was looking at that with pumaranikar the other day, I think it might be a bit new line happy | |
| 15:13:06 | cdent | yes | |
| 15:13:06 | cdent | new line happy | |
| 15:13:11 | openstackgerrit | Sarafraj Singh proposed openstack/nova-specs master: Send service token along with user tokens https://review.openstack.org/426933 | |
| 15:13:32 | cdent | i'm not sure I'm keen on the decorator idea, but the current middleware idea won't work | |
| 15:13:44 | cdent | (but I've mostly been ignoring that) | |
| 15:16:10 | cdent | johnthetubaguy: in any case, the logging on the rt side shouldn't care. we should just fix it on the api side, yeah? | |
| 15:16:59 | johnthetubaguy | cdent: yeah, certainly if the extra new lines are from the API, which I think they are | |
| 15:17:22 | johnthetubaguy | cdent: it was more your patches reminded my about the niggle I spotted debugging that bug the other day | |
| 15:17:56 | cdent | let's make pumaranikar fix it :) | |
| 15:18:39 | johnthetubaguy | good idea | |
| 15:18:47 | johnthetubaguy | pumaranikar: we found you another placement bug | |
| 15:19:23 | EmilienM | mriedem: good morning, regarding the issue with Nova in WSGI in TripleO - does it make sense we stop doing that in Ocata (and revisit later)? | |
| 15:19:26 | mriedem | http://logs.openstack.org/60/428360/2/experimental/gate-tripleo-ci-centos-7-nonha-multinode/7a6e5bc/logs/subnode-2/var/log/nova/nova-conductor.txt.gz#_2017-02-03_05_17_35_826 | |
| 15:19:26 | mriedem | so on the tripleo failures from yesterday, with a run on my debug patch, i see the instance that heat is trying to create fails | |
| 15:19:28 | mriedem | EmilienM: good timing | |
| 15:19:32 | EmilienM | wow | |
| 15:19:43 | mriedem | scheduler couldn't find a host | |
| 15:19:56 | EmilienM | do you want me to stop deploing nova api in wsgi with apache and see if it helps? | |
| 15:20:15 | mriedem | given we haven't really flushed out testing with that, i think it's prudent at this point in ocata | |
| 15:20:26 | EmilienM | ok let me do it | |
| 15:21:03 | mriedem | the 404 isn't because the instance was created and we just can't find it though, it's not found because it fails to build and the instance isn't created (it looks like), but the dead instance should go into cell0 and we should pull it from cell0 | |
| 15:21:31 | mriedem | hrm, dansmith melwitt do we even look in cell0 for compute api get()? | |
| 15:21:39 | cdent | johnthetubaguy, pumaranikar It can probably be built right in with the json_formatter stuff because it's all of a piece | |
| 15:22:12 | melwitt | mriedem: we should, I think. do we not? checking | |
| 15:22:21 | dansmith | melwitt: yes | |