| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2017-07-26 | |||
| 14:26:00 | sdague | mriedem: can you check out whatever you were verifying yesterday in https://review.openstack.org/#/c/487246/ ( the wait for service patch ), or at least point me to what I'm looking for | |
| 14:26:43 | bauzas | I do remember you modified the REST API to use a PUT /os-services/ instead of an action and you made it cell-aware, but I don't remember we said to not accept enablement/disablement for internal services but computes | |
| 14:28:40 | bauzas | mriedem: nevermind, saw the commit msg in https://review.openstack.org/#/c/464280/ | |
| 14:30:26 | dansmith | vdrok: so you can set this variable early enough to make it work right? https://review.openstack.org/487443 | |
| 14:30:52 | mriedem | bauzas: https://bugs.launchpad.net/python-novaclient/+bug/1700359/comments/2 | |
| 14:30:53 | openstack | Launchpad bug 1700359 in OpenStack Compute (nova) "test_os_service_disable_enable fails with "ERROR (NotFound): Could not find binary nova-scheduler on host"" [Medium,Confirmed] | |
| 14:31:40 | mriedem | bauzas: trying to perform actions on non-compute services literally won't work, | |
| 14:31:52 | mriedem | because we look up the service in the cell via the host mapping, which only exists for compute services | |
| 14:31:55 | bauzas | mriedem: yeah it made sense to me | |
| 14:32:10 | mriedem | i just have the todo to put something in the release notes for it | |
| 14:32:25 | bauzas | mriedem: and because we only care about the enable/disable flag by the scheduler only for finding a target | |
| 14:32:32 | jaypipes | jangutter: where are you getting the vif_plug_ovs package here? https://github.com/Netronome/agilio-ovs-openstack-plugin/blob/d3f53fd3e6eaaa3d4b5f98978b2f3aa504a7a8cd/vif_plug_agilio_ovs/agilio_ovs.py#L29-L31 | |
| 14:32:40 | bauzas | mriedem: given other services and not targets for instances | |
| 14:33:00 | jaypipes | jangutter: I don't see os-vif in the list of requirements.txt Python packages. | |
| 14:33:23 | jangutter | jaypipes: yep, there's a sizeable set of cleanups still on that repo. | |
| 14:33:26 | openstackgerrit | Merged openstack/nova master: Using plain routes for the microversions test https://review.openstack.org/485061 | |
| 14:34:49 | dansmith | jaypipes: we're claiming in the scheduler now right? | |
| 14:35:02 | dansmith | oh no | |
| 14:35:06 | dansmith | nm that one hasn't merged yet | |
| 14:35:32 | dansmith | jaypipes: hopefully after it does, the ironic people won't have reschedules during their test runs :/ | |
| 14:35:33 | jaypipes | dansmith: looking at the -1s on that now. | |
| 14:35:58 | cdent | gibi: on https://review.openstack.org/#/c/487327/ I assume the virtenv’s still build correctly before runnin the actual tests? That error message is produced by wsgi-intercept, which gets upset about proxy variables when using the urllib3 interceptor | |
| 14:36:46 | vdrok | dansmith: yup, seems like it should help, thanks! we'll have to add it to our base builder in project-config | |
| 14:37:34 | dansmith | vdrok: okay, you'll work on that/ | |
| 14:37:42 | vdrok | yup | |
| 14:37:53 | vdrok | will propose a project-config patch now | |
| 14:38:03 | dansmith | vdrok: once we merge the scheduler claiming patch, hopefully you won't get any more reschedules during test runs like this, which will probably be faster anyway | |
| 14:38:22 | dansmith | should make your runs faster I mean | |
| 14:38:39 | dansmith | probably good to do this devstack variable in the meantime anyway | |
| 14:38:55 | vdrok | ++ | |
| 14:41:35 | mriedem | sdague: question in https://review.openstack.org/#/c/486700/2/nova/tests/fixtures.py | |
| 14:44:12 | mriedem | oh dan https://review.openstack.org/#/c/487443/ | |
| 14:44:43 | dansmith | mriedem: I was rushing | |
| 14:45:42 | sdague | mriedem: looking | |
| 14:45:45 | mriedem | why do we have: local nomulticellflag="$1" | |
| 14:45:50 | mriedem | in start_nova_compute | |
| 14:45:55 | mriedem | nothing passes a variable to that function | |
| 14:46:03 | dansmith | mriedem: yes, grenade does | |
| 14:46:07 | dansmith | mriedem: and ironic's grenade does | |
| 14:46:18 | sdague | mriedem: so what is your concern there? | |
| 14:46:19 | mriedem | oh grenade | |
| 14:46:41 | mriedem | sdague: was trying to sort out if that's coming from something within nova, since we do use pkg_resources.parse_version, but nowhere near that code | |
| 14:47:17 | dansmith | mriedem: jlvillal was struggling with that in o.vo recently too | |
| 14:47:18 | sdague | mriedem: stevedore probably | |
| 14:47:25 | dansmith | something changed in setuptools I think | |
| 14:50:08 | sdague | mriedem: actually, I think it's pbr | |
| 14:52:11 | sdague | let me see if I can trigger a deeper stack warn to figure out exactly where | |
| 14:58:50 | sdague | dansmith: for https://review.openstack.org/#/c/487443/2/lib/nova why isn't the answer to disable multicell? | |
| 14:59:40 | dansmith | sdague: not sure what you mean.. default=True=do setup multiple tiering.. set to false will not do that | |
| 15:00:07 | dansmith | maybe my logic needs inverting to make more sense? like disable=True? | |
| 15:00:31 | mriedem | note we aren't actually multi-cell yet | |
| 15:00:35 | mriedem | we're multi-tier | |
| 15:00:57 | dansmith | mriedem: well, you could multi-cell with this, we just don't configure it that way | |
| 15:01:14 | mriedem | dansmith: so i can post a WIP to set NOVA_NUM_CELLS=2 and it should work? | |
| 15:01:36 | dansmith | mriedem: it should configure multiple cells, but not sure it will work :) | |
| 15:01:47 | dansmith | mriedem: you need NUM_CELLS=2 and the CELL_INDEX=2 for the third node | |
| 15:01:53 | mriedem | could tease out some interesting bugs we don't know about yet | |
| 15:02:16 | dansmith | well, I didn't know we had three node jobs until you told me yesterday | |
| 15:02:16 | mriedem | where is CELL_INDEX? | |
| 15:02:49 | dansmith | I rememberd the name wrong, but it's this: https://github.com/openstack-dev/devstack/blob/master/lib/nova#L66-L68 | |
| 15:03:45 | mriedem | couldn't that just be: NOVA_CPU_CELL=${NOVA_CPU_CELL:-$NOVA_NUM_CELLS} ? | |
| 15:03:48 | dansmith | I just want to say that I wish we had taken my original suggestion to call cellsv2 something other than cells | |
| 15:03:49 | mriedem | are they always the same? | |
| 15:03:57 | sdague | ok, I'm trying to figure out a way to explain this in stackrc that isn't going to require poking people in irc. | |
| 15:04:03 | dansmith | no, | |
| 15:04:16 | dansmith | mriedem: you need it to be different on each node you want to be in a different cell | |
| 15:05:02 | mriedem | ah ok | |
| 15:05:04 | sdague | so, the fleetify patch creates a conductor per compute. | |
| 15:05:10 | sdague | plus the api conductor | |
| 15:05:11 | mriedem | conductor per node | |
| 15:05:16 | dansmith | condutor per cell | |
| 15:05:26 | mriedem | well, | |
| 15:05:31 | sdague | right, it's good we're all on the same page :) | |
| 15:05:32 | mriedem | not really | |
| 15:05:45 | mriedem | dansmith: today the multinode job with this change is 1 cell | |
| 15:05:46 | openstackgerrit | Eric Fried proposed openstack/nova master: Add requirement for os-service-types https://review.openstack.org/487471 | |
| 15:06:01 | mriedem | there is a cell conductor and super conductor | |
| 15:06:02 | mriedem | yeah, | |
| 15:06:02 | dansmith | mriedem: it's conductor per cell, plus one super conductor | |
| 15:06:03 | sdague | mriedem: right, so 1 compute conductor + 1 api conductor, right ? | |
| 15:06:06 | mriedem | and the subnodes talk to the cell conductor | |
| 15:06:09 | mriedem | api talks to super conductor | |
| 15:06:28 | mriedem | fwiw this is why i wrote a todo to doc this in devstack later | |
| 15:06:48 | mriedem | because we have 3 nova confs now and multiple conductors so debugging it is a challenge at first | |
| 15:06:55 | openstackgerrit | OpenStack Proposal Bot proposed openstack/nova master: Updated from global requirements https://review.openstack.org/487473 | |
| 15:06:57 | sdague | so, if we introduced this as CELLSV2_SETUP=allinone | |
| 15:06:59 | dansmith | mriedem: trying to help by starting that nova doc with diagrams of course | |
| 15:07:04 | mriedem | dansmith: yup | |
| 15:07:09 | mriedem | that's higher priority IMO anyway | |
| 15:07:09 | sdague | CELLSV2_SETUP=condpercell | |
| 15:07:38 | sdague | or something, then we could global switch there? | |
| 15:07:42 | dansmith | sdague: if it wasn't for ironic (and grenade) I'd say we don't need a flag | |
| 15:07:56 | sdague | dansmith: sure, ironic is always going to be broken this way? | |
| 15:08:01 | dansmith | sdague: and we really don't want people doing the allinone if possible | |
| 15:08:11 | dansmith | no | |
| 15:08:41 | dansmith | ironic will stop being broken when we get the claims patch in, | |
| 15:08:41 | sdague | what's the ironic path fix here? | |
| 15:08:56 | dansmith | and when we have proper reschedules (which we've punted at this point) then double fixed | |
| 15:09:28 | mriedem | 1. ironic sets the tiered=False flag temporarily in devstack-plugin-ironic, | |
| 15:09:30 | mriedem | 2. we land https://review.openstack.org/#/c/483566/ | |