| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2018-07-12 | |||
| 14:58:43 | mriedem | i have a whole buttload of comments in https://review.openstack.org/#/c/564452/ related to this | |
| 14:58:55 | mriedem | https://review.openstack.org/#/c/564452/17/nova/compute/api.py@889 | |
| 14:59:02 | stephenfin | mriedem: Sorry, you said RequestSpec and I read NUMACell https://review.openstack.org/#/c/564441/21/nova/virt/libvirt/driver.py@6338 | |
| 15:00:04 | stephenfin | mriedem: Yeah, I should just be setting that to NetworkMetadata(physnets=set(), tunneled=False) | |
| 15:01:08 | mriedem | well also see https://review.openstack.org/#/c/564444/14/nova/tests/unit/network/test_neutronv2.py@4963 | |
| 15:01:11 | mriedem | similar concern | |
| 15:01:44 | mriedem | anyway, you have options | |
| 15:02:34 | mriedem | either use an 'empty' NetworkMetadata object, or support None, or handle unset | |
| 15:02:59 | mriedem | at this point i kind of think nullable=True makes sense for when we know we don't have networking | |
| 15:03:19 | mriedem | but it's like a 5% "makes sense" | |
| 15:04:15 | stephenfin | heh | |
| 15:04:23 | stephenfin | OK, I'll explore that | |
| 15:04:36 | stephenfin | and try to push something to demonstrate this possible o.vo bug | |
| 15:09:44 | mriedem | anyone have any idea why we wouldn't cache the host az here? https://github.com/openstack/nova/blob/da16690f4db6172659ffa4f804296100b5ba24d6/nova/availability_zones.py#L93 | |
| 15:10:04 | mriedem | we do it when getting the az for an instance that has a host set https://github.com/openstack/nova/blob/da16690f4db6172659ffa4f804296100b5ba24d6/nova/availability_zones.py#L194 | |
| 15:12:41 | mriedem | dansmith: i think i have a fix for the cinder/cross_az_attach=False up-call issue | |
| 15:12:44 | mriedem | 1 line fix | |
| 15:13:03 | jmlowe | dansmith: do you happen to know off the top of your head where the ceph free space is queried? | |
| 15:13:20 | mriedem | jmlowe: should be down in libvirt driver update_available_resource | |
| 15:13:30 | mriedem | get_available_resource? | |
| 15:13:42 | jmlowe | those are grepable, thanks | |
| 15:13:45 | mriedem | yeah _get_local_gb_info | |
| 15:13:53 | mriedem | info = LibvirtDriver._get_rbd_driver().get_pool_info() | |
| 15:13:53 | mriedem | elif CONF.libvirt.images_type == 'rbd': | |
| 15:20:02 | mriedem | https://bugs.launchpad.net/nova/+bug/1781421 | |
| 15:20:03 | openstack | Launchpad bug 1781421 in OpenStack Compute (nova) "CantStartEngineError due to host aggregate up-call when boot from volume and [cinder]/cross_az_attach=False" [Medium,Triaged] | |
| 15:20:16 | mriedem | + if not host or (host and instance.availability_zone): | |
| 15:20:16 | mriedem | - if not host: | |
| 15:20:17 | mriedem | fixed! | |
| 15:25:59 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Mention osc-placement for managing traits in docs https://review.openstack.org/582173 | |
| 15:28:43 | melwitt | jmlowe: and here's a proposal for another way to query free space that I've not fixed up with tests yet https://review.openstack.org/#/c/556692/1/nova/virt/libvirt/storage/rbd_utils.py | |
| 15:31:51 | dansmith | note that in placement, we don't report free space, but infer it from total - (allocated + reserved) | |
| 15:32:42 | melwitt | oh, okay | |
| 15:32:58 | mriedem | melwitt: you were +2 on this before jaypipes pointed out that we couldn't actually update consumer project/user info in placement, but that bug is fixed now and his -2 is gone, and i've added a functional test so can you review this again? https://review.openstack.org/#/c/574488/ - it's the heal_allocations thing for healing incomplete consumers | |
| 15:33:02 | mriedem | and would close out that bp | |
| 15:33:29 | melwitt | mriedem: ah yup, thanks for the heads up | |
| 15:33:32 | melwitt | will revisit that | |
| 15:39:55 | stephenfin | Pro tip #1: 'tox -e py27 -- --failing' runs all the tests that failed on a previous iteration | |
| 15:40:29 | stephenfin | Pro tip #2: 'tox -e py27 -- -n nova/tests/unit/virt/test_hardware.py' let's you run tests without needing to convert to a Python path | |
| 15:42:20 | mriedem | efried: alex_xu: arvindn05: karimull: on its way to the gate of horrors https://review.openstack.org/#/c/569498/ - i'll do a follow up for my comments | |
| 15:42:35 | efried | mriedem: cool mon | |
| 15:42:54 | mriedem | pass the dutchie | |
| 15:43:15 | mriedem | melwitt: ^ means another runway slot will be open today | |
| 15:43:16 | melwitt | that means we get to close out a bp right? woo | |
| 15:43:26 | melwitt | noyce. good news | |
| 15:44:45 | melwitt | stephenfin: protip #2 I didn't know about, thanks | |
| 15:46:13 | efried | I tend to use stestr run directly from the venv so I don't have to wait an extra 45 seconds for tox to do... whatever it does in "preparation". | |
| 15:47:15 | melwitt | I do too when I'm iterating on running tests | |
| 15:47:49 | stephenfin | efried: I think we discussed this before, but doesn't 'touch nova.egg_info' fix that for you? | |
| 15:48:50 | efried | stephenfin: That does sound familiar. Not sure. But I know this works :) | |
| 15:49:05 | stephenfin | Fair :) | |
| 15:49:07 | efried | stephenfin: I've got it in a bash function so I can kick it off easily | |
| 15:50:08 | stephenfin | I really need to figure out what's going on with tox there. It seems pretty useless, given how often I need to recreate stuff when dependencies change | |
| 15:51:11 | efried | i.e. very rarely, yah? | |
| 15:52:06 | stephenfin | Ever :) If it's doing a dependency check based on the timestamp of .egg_info, that check should be failing because dependencies have changed | |
| 15:52:07 | efried | Assuming it's dependency checks that are slow, if there was a tox --no-dep-check flag that let you bypass all that stuff, that would be coolness. | |
| 15:52:23 | stephenfin | If it's not failing, you'd have to wonder what the "dependency check" _is_ doing | |
| 15:52:24 | efried | oh, or that, if it's actually a bug that should just be fixed :) | |
| 15:52:52 | efried | well, occasionally tox does rebuild my venv. I assume it does that based on figuring out that something is out of date. | |
| 15:53:23 | stephenfin | Yeah, I'd rather some way to just disable it, personally. The gate will catch the rare case where a local version mismatch causes issues | |
| 15:53:44 | stephenfin | The only time I see a full rebuild is when tox.ini changes | |
| 15:53:45 | melwitt | mine never auto rebuilds the venv for me. I do it manually if I hit "no module X found" errors when things change | |
| 15:53:54 | stephenfin | (assuming you mean recreate) | |
| 15:53:58 | stephenfin | melwitt: Yeah, same here | |
| 15:54:18 | openstackgerrit | Eric Fried proposed openstack/nova master: WIP: Compute: Handle reshaped provider trees https://review.openstack.org/576236 | |
| 15:54:21 | stephenfin | Hence, what _is_ tox doing :) | |
| 15:54:52 | melwitt | yeah... | |
| 15:54:55 | efried | If I know a dep is out of date, I'll enter the venv and pip install/upgrade the sucker manually. | |
| 15:55:02 | melwitt | I'd like to know too but I've no idea how to find out | |
| 15:55:04 | efried | Takes way less time than rebuilding the whole damn venv. | |
| 15:55:29 | melwitt | heh, I've done that in the past but nowadays I'm like meh | |
| 15:56:00 | stephenfin | efried: That. I personally let it rebuild but Real World Devs (TM) often fix this stuff themselves. I think a flag would be a good idea | |
| 15:56:08 | efried | Usually when I'm running tox it's because I actually want to run tests, not because I want to kick off a command and go eat lunch and come back to see if it has finished yet. | |
| 15:56:25 | stephenfin | ...as soon as we figure out what is taking that time when it's not rebuilding (the thing 'touch nova.*' solves) | |
| 15:56:58 | efried | Cool mon, let me know if you want me to try anything out. | |
| 15:57:43 | stephenfin | Ack. I'll try take a look tomorrow once I get somewhere on these functional tests for numa-aware-vswitch | |
| 15:57:59 | efried | That's not using nrp, is it? | |
| 15:58:12 | efried | (I haven't read the spec in yonks, have totally forgotten what it actually does) | |
| 15:58:58 | stephenfin | Nope, nothing to do with placement | |
| 15:59:33 | stephenfin | Not now, anyway. A fancier version of it might use it in the future but that's 2-3 cycles down the road, I'd imagine | |
| 16:03:01 | mriedem | TheJulia: do you know if there are any ironic jobs that can be used to test this rescue patch? https://review.openstack.org/#/c/416487/3 | |
| 16:03:05 | mriedem | https://review.openstack.org/#/c/416487/ | |
| 16:03:20 | mriedem | i know tempest tests rescue, we'd just have to flip the switch in some ironic job config | |
| 16:03:42 | TheJulia | mriedem: we've not merged them yet because of the nova patch not merging, looking it up | |
| 16:03:56 | mriedem | TheJulia: ok but we can see a depends-on patch passing tests right? | |
| 16:03:59 | TheJulia | that being said, the logs show it working in the proposed tempest patch | |
| 16:04:07 | mriedem | ah https://review.openstack.org/#/c/528699/ | |
| 16:04:09 | TheJulia | mriedem: already did if my memory is correct | |
| 16:04:20 | TheJulia | yup, thats the one | |
| 16:05:18 | TheJulia | https://review.openstack.org/#/c/528704/ | |
| 16:06:03 | mriedem | ok that depends on the plugin change which depends on the nova change, i see | |
| 16:06:48 | TheJulia | and I see you already rechecked it | |
| 16:06:49 | TheJulia | thanks! | |
| 16:07:00 | openstackgerrit | MultipleCrashes proposed openstack/nova master: Retry decorator fix for instances which go into ERROR state during bulk delete https://review.openstack.org/563418 | |
| 16:12:23 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Follow up for Ie49d605c66062d2548241d7e04f5a2a6b98c011e https://review.openstack.org/582292 | |
| 16:13:01 | kashyap | mriedem: The above commit message is really odd | |
| 16:13:40 | kashyap | (It's logically correct. But it can be turned into a sentence w/o a commit has :-)) | |
| 16:17:36 | openstackgerrit | Matt Riedemann proposed openstack/osc-placement master: Resource provider examples https://review.openstack.org/553461 | |
| 16:27:38 | openstackgerrit | Stephen Finucane proposed openstack/nova master: scheduler: Start utilizing RequestSpec.network_metadata https://review.openstack.org/564452 | |
| 16:27:39 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Add '[neutron] physnets' and related options https://review.openstack.org/564440 | |