Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2018-12-06
21:51:44 gaudenz I'll look at jose's patch and contact him. Thanks again for helping out.
21:54:37 mriedem dansmith: unit test failures in https://review.openstack.org/#/c/623283/1
21:54:45 mriedem http://logs.openstack.org/83/623283/1/check/openstack-tox-py27/d6a560e/testr_results.html.gz
21:55:19 melwitt gaudenz: sounds good, thanks
21:57:34 mriedem efried_cya_jan: before i dig into the fewer placement calls from _ensure patch, you've addressed jaypipes' ironic concerns from https://review.openstack.org/#/c/615677/9/nova/compute/resource_tracker.py@812 ?
21:57:48 mriedem i assume so, but want to make sure i'm not going to waste time
22:11:32 dansmith mriedem: hmm, okay, I switched the order at the last minute and then realized that was probably not the best idea an hour later, so that's probably why
22:11:44 dansmith they were all passing before I did that
22:39:27 mriedem coreycb: your wish is my command https://review.openstack.org/#/c/579004/
23:13:01 openstackgerrit Gaudenz Steinlin proposed openstack/nova master: Extend volume for libvirt network volumes (RBD) https://review.openstack.org/613039
23:23:21 openstackgerrit Matt Riedemann proposed openstack/nova stable/rocky: Fix InstanceNotFound during _destroy_evacuated_instances https://review.openstack.org/623349
23:23:21 openstack bug 1794996 in OpenStack Compute (nova) rocky "_destroy_evacuated_instances fails and kills n-cpu startup if lazy-loading flavor on a deleted instance" [High,Triaged] https://launchpad.net/bugs/1794996
23:23:21 openstackgerrit Matt Riedemann proposed openstack/nova stable/rocky: Add functional regression test for bug 1794996 https://review.openstack.org/623348
23:44:34 openstackgerrit Merged openstack/nova master: Remove ironic/pike note from *_allocation_ratio help https://review.openstack.org/620154
23:44:44 openstackgerrit Merged openstack/nova master: Change the default values of XXX_allocation_ratio https://review.openstack.org/602803
23:59:17 mriedem omg
23:59:17 openstackgerrit Matt Riedemann proposed openstack/nova stable/queens: Fix InstanceNotFound during _destroy_evacuated_instances https://review.openstack.org/623355
23:59:17 openstack bug 1794996 in OpenStack Compute (nova) rocky "_destroy_evacuated_instances fails and kills n-cpu startup if lazy-loading flavor on a deleted instance" [High,In progress] https://launchpad.net/bugs/1794996 - Assigned to Matt Riedemann (mriedem)
23:59:17 openstackgerrit Matt Riedemann proposed openstack/nova stable/queens: Add functional regression test for bug 1794996 https://review.openstack.org/623354
#openstack-nova - 2018-12-07
00:11:02 openstackgerrit melanie witt proposed openstack/nova master: Fixed concurrent access to direct io test file https://review.openstack.org/515091
00:40:39 openstackgerrit Matt Riedemann proposed openstack/nova stable/pike: Fix InstanceNotFound during _destroy_evacuated_instances https://review.openstack.org/623359
00:40:39 openstack bug 1794996 in OpenStack Compute (nova) rocky "_destroy_evacuated_instances fails and kills n-cpu startup if lazy-loading flavor on a deleted instance" [High,In progress] https://launchpad.net/bugs/1794996 - Assigned to Matt Riedemann (mriedem)
00:40:39 openstackgerrit Matt Riedemann proposed openstack/nova stable/pike: Add functional regression test for bug 1794996 https://review.openstack.org/623358
01:11:34 openstackgerrit Zhenyu Zheng proposed openstack/nova master: Handle tags in _bury_in_cell0 https://review.openstack.org/621856
02:26:54 openstackgerrit melanie witt proposed openstack/nova-specs master: Propose counting quota usage from placement and API database https://review.openstack.org/509042
05:20:11 openstackgerrit Merged openstack/nova master: libvirt: Refactor handling of PCIe root ports https://review.openstack.org/620327
05:20:18 openstackgerrit Merged openstack/nova stable/queens: Make supports_direct_io work on 4096b sector size https://review.openstack.org/619220
05:20:27 openstack bug 1764883 in OpenStack Compute (nova) queens "Evacuation fails if the source host returns while the migration is still in progress" [Medium,In progress] https://launchpad.net/bugs/1764883 - Assigned to Lee Yarwood (lyarwood)
05:20:27 openstackgerrit Merged openstack/nova stable/queens: Add regression test for bug #1764883 https://review.openstack.org/621199
06:37:56 openstackgerrit Merged openstack/nova stable/queens: compute: Ensure pre-migrating instances are destroyed during init_host https://review.openstack.org/621200
06:39:33 openstackgerrit Merged openstack/nova stable/pike: Update docs for _destroy_evacuated_instances https://review.openstack.org/621203
09:30:50 openstackgerrit Balazs Gibizer proposed openstack/nova master: Ignore MoxStubout deprecation warnings https://review.openstack.org/623309
09:43:39 cdent stephenfin: I'm confused and frustrated and stuck on https://review.openstack.org/#/c/622972/ If you have any ideas, or can just fix it, that would be most helpful.
10:20:03 mdbooth lyarwood: https://review.openstack.org/#/c/618478/ I agree with melwitt's comment that instance.host checking is a risk. Could you audit it? We're hopefully ok.
10:21:08 lyarwood mdbooth: I've had a look already and I couldn't see a way for us to call cleanup on the source during a failure
10:21:40 lyarwood mdbooth: we would call it on the dest but the whole point of the workaround is to ensure the directory is cleaned when it _isn't_ shared
10:21:54 lyarwood mdbooth: so if it's enabled in that situation that's on the operator
10:22:06 mdbooth lyarwood: I would personally start by looking at everywhere in ComputeManager which sets instance.host, and work back from there. Don't assume it necessarily makes sense.
10:22:47 lyarwood mdbooth: there's zero point in looking at that if we don't call cleanup on the source during a failure
10:23:24 lyarwood mdbooth: which we wouldn't do for LM, either post-copy or pre-copy AFAIK
10:23:57 lyarwood I'll look again at where we are calling cleanup to confirm but I'm sure this isn't an issue
10:24:16 mdbooth lyarwood: I strongly suspect it isn't an issue.
10:25:53 lyarwood DNM 0 [-------x--] 1 LGTM
10:26:01 mdbooth lyarwood: +1
10:27:48 mdbooth lyarwood: Are there any circumstances where we could end up calling cleanup on the source (e.g. init_host, periodic) where instance.host is erroneously set to dest?
10:28:22 openstackgerrit Stephen Finucane proposed openstack/nova master: Correct lower-constraints.txt and the related tox job https://review.openstack.org/622972
10:31:28 lyarwood mdbooth: cleaning up evacuations that have failed in a weird way but again with this workaround enabled you'd only see console.log, kernel etc removed
10:31:57 lyarwood mdbooth: I can't say I've ever seen an evacuation fail in that way however
10:32:08 lyarwood mdbooth: instance.host updated but the instance is still on the source
10:32:22 mdbooth s/still on the source/isn't on the dest/
10:32:56 mdbooth although we still wouldn't have deleted the actual instance data as you say
10:33:02 mdbooth So probably not the worst.
10:49:16 openstackgerrit Surya Seetharaman proposed openstack/nova master: API microversion 2.68: Handles Down Cells https://review.openstack.org/591657
10:49:53 gibi alex_xu: have a nice vacation
11:36:59 openstackgerrit Dongcan Ye proposed openstack/nova master: Remove other volume snapshot type https://review.openstack.org/623456
12:22:46 openstackgerrit sean mooney proposed openstack/os-vif master: add isolate_vif config option https://review.openstack.org/612534
12:39:00 openstackgerrit Gaudenz Steinlin proposed openstack/nova master: Extend volume for libvirt network volumes (RBD) https://review.openstack.org/613039
12:47:13 cdent stephenfin: it looks like your change on the constraints job gets things to pass, which suggests that skipsdist is not working in a per-env setting?
12:47:37 stephenfin cdent: Yup, I think I called that out in a comment/the commit message?
12:47:39 cdent but it's also confusing that different behaviors are happening between my box and the gate.
12:47:46 stephenfin cdent: It's a global option
12:48:03 stephenfin I'd imagine it's the version of tox
12:48:14 cdent they are the same
12:48:19 cdent that's why it is confusing
12:48:54 stephenfin Hmm. Have you wiped your .tox directory?
12:49:19 cdent not the whole thing, but the specific env
12:49:43 cdent the vm I was using is at home, so I cant' look now, but will check when I'm back home
12:50:38 stephenfin I've no idea what's going on, in that case. Might be worth asking the tox devs, in case they've any ideas. The funkiness that pbr introduces could throw even them though
12:50:58 cdent I'll also check to make sure that the job is sitll working as designed
12:51:05 cdent yeah, pbr...
12:51:23 cdent if it's happy as is, I'm happy to just move ...
12:51:26 cdent on
12:51:49 cdent wasted way more time on this than I wanted
12:55:20 cdent "wasted way more time on this than I wanted" <- nova, defined
13:04:34 openstackgerrit Silvan Kaiser proposed openstack/nova master: Exec systemd-run without --user flag in Quobyte driver https://review.openstack.org/554195
13:33:36 openstackgerrit Merged openstack/os-vif master: always create ovs port during plug https://review.openstack.org/602384
13:34:29 openstackgerrit Lee Yarwood proposed openstack/nova master: WIP compute: reject migration request when source compute is disabled https://review.openstack.org/623489
13:39:38 openstackgerrit Merged openstack/nova master: Use tempest [compute]/build_timeout in evacuate tests https://review.openstack.org/623011
13:50:14 sc I need some help w/ configuring PCI passthrough in a newton based installation, I followed the newton admin guide but the scheduler returns 0 of 6 hosts.
13:51:19 sc what do you suggest to look for? In the compute host dmesg reports the iommu as enabled and I see a nice list of PCI cards, including the GPU I would like to pass to my VM
14:31:08 openstackgerrit Silvan Kaiser proposed openstack/nova master: Exec systemd-run without --user flag in Quobyte driver https://review.openstack.org/554195
14:49:26 openstackgerrit Merged openstack/nova master: Update compute API.get() stubs for test_*security_groups https://review.openstack.org/615344
14:49:39 openstackgerrit Merged openstack/nova master: Update compute API.get() stubs for test_disk_config https://review.openstack.org/615345
14:49:49 openstackgerrit Merged openstack/nova master: Update compute API.get() stubs in test_access_ips https://review.openstack.org/615346
14:50:25 openstackgerrit Lee Yarwood proposed openstack/nova stable/rocky: Add secret=true to fixed_key configuration parameter https://review.openstack.org/623507
14:51:00 openstackgerrit Lee Yarwood proposed openstack/nova stable/queens: Add secret=true to fixed_key configuration parameter https://review.openstack.org/623509
14:52:28 openstackgerrit Lee Yarwood proposed openstack/nova stable/pike: Add secret=true to fixed_key configuration parameter https://review.openstack.org/623510
15:03:53 openstackgerrit Dan Smith proposed openstack/nova master: Make compute rpcapi version calculation check all cells https://review.openstack.org/623284
15:03:53 openstackgerrit Dan Smith proposed openstack/nova master: Only warn about not having computes nodes once in rpcapi https://review.openstack.org/623282
15:03:54 openstackgerrit Dan Smith proposed openstack/nova master: Make service.get_minimum_version_all_cells() cache the results https://review.openstack.org/623283
15:04:23 dansmith mriedem: melwitt: ^ this should be good now I think. also fixed something in the base patch that was causing a non-deterministic run in the later patches
15:05:24 melwitt cool, thanks
15:07:21 mriedem on an unrelated note, i propose that we disable snapshot tests in the cells v1 job http://logs.openstack.org/47/623247/2/check/nova-cells-v1/18338f0/job-output.txt.gz
15:07:34 mriedem snapshot with cells v1 in tempest is historically racy https://bugs.launchpad.net/nova/+bug/1620761
15:07:35 openstack Launchpad bug 1620761 in OpenStack Compute (nova) "test_create_second_image_when_first_image_is_being_saved intermittently times out in teardown in cells v1 job" [Undecided,Invalid]
15:07:48 mriedem given the deprecated nature of cells v1 and the state of the gate, the less we can fail on the better
15:08:33 dansmith mriedem: fine with me
15:08:47 mriedem alright then
15:09:26 dansmith I also wouldn't be opposed to making a statement in stein that cellsv1 is deprecated fully, we're not going to test it anymore, and we'll be ripping the code out at whatever pace we feel is achievable
15:09:59 dansmith we know that the removal of v1 and n-net aren't trivial, so the actual removal will take time, but I think we could go ahead and drive the stake through the heart

Earlier   Later