Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2020-05-15
22:35:52 mnaser melwitt: :D
22:36:25 mnaser I tried to make it as backportable as possible
22:37:00 melwitt good call. this doesn't look too bad at all, so thank you for that
22:38:09 melwitt mnaser: I assume you tried out this patch? what was the perf result, just curious
22:39:50 ikla I'm trying to get pci_passthrough to work and followed the documents. I keep seeing Pci claimed failed. Anyway to drill this down. Config looks perfect.
22:40:34 mnaser melwitt: I have not yet actually! It’s Friday and the load dropped so it wasn’t hitting as hard
22:41:07 mnaser I can try on Monday though but I imagine it’ll drop scanning instance mappings and I only have 2 cells really (0 and 1)
22:41:10 melwitt mnaser: I guess, this seems like the opposite of what you wrote in the bug report. it's always limiting to project_id and you're adding some that would look at all cells?
22:42:04 mnaser melwitt: sorry I’ve been pretty slow today. But yes before it was limited to project ID and this just doesn’t limit and checks all cells anywyas
22:42:27 melwitt like before your patch it's always only looking at a particular project_id and quota check is always scoped to a project, so there's no need to look at all cells
22:43:23 melwitt mnaser: if that's the case, that would mean there's a bug in objects.CellMappingList.get_by_project_id if it's not actually scoping down to a project
22:44:28 mnaser melwitt: that function does actually scope by project. The only problem is that to do that, it needs to scan all instance mappings to find all cells for that project which is where I took the performance hit
22:44:55 mnaser After my patch, nova checks all cells, regardless if a user has instances in it or not
22:45:00 melwitt and no worry, I've been slow today too. spent all day/night yesterday dealing with fallout from a nova gate breakage and trying to get it fixed. so exhausted
22:45:18 mnaser Sorry if my commit message or bug aren’t clear, I’m just feeling bleeeh and dazed all day lol
22:47:49 melwitt mnaser: sorry this just doesn't seem to make sense. before your patch, it's always limiting to one project cells. after you patch, it's capable of not limiting. I don't see how that would help with any instance mapping scanning?
22:48:48 melwitt oh hang on
22:49:08 mnaser melwitt: I think we may be talking about two scans. One for the quota calculation and one to find which cells the user has instances in
22:49:19 melwitt ok so you're saying when you can look at all cells, you actually go _faster_ because you don't do the instance mapping thing? wow ok
22:49:39 melwitt yeah ok it's a tradeoff, slowly remembering now
22:49:49 mnaser melwitt: https://github.com/openstack/nova/blob/master/nova/objects/cell_mapping.py#L275-L285
22:49:57 melwitt the project limiting behavior is actually worse for single cell-ers
22:50:00 melwitt yeah, I see now
22:50:01 melwitt sorry
22:50:11 mnaser All good! Yeah, it hurt me more in this case
22:50:24 melwitt yeah, I see
22:50:28 melwitt nice catch
22:50:47 mnaser not the catch I wanted on a Friday but it is what it is :P
22:50:54 melwitt haha right
22:51:13 melwitt ok, I'll review this properly a bit later. thanks for opening the bug and doing the patch
22:55:27 mnaser No worries
23:05:21 ikla Timed out waiting for nova-conductor. Is it running? Or did this service start before nova-conductor?
23:05:29 ikla any ideas?
23:06:57 ikla or normal
23:07:33 openstackgerrit Merged openstack/nova master: docs: Add evacuation pre-conditions around the src host https://review.opendev.org/723853
23:10:00 melwitt ikla: nova-conductor is a required piece of nova, it needs to be running so if it isn't, that would be a problem
23:31:59 openstackgerrit Merged openstack/nova master: Reproduce bug 1869050 https://review.opendev.org/714997
23:31:59 openstack bug 1869050 in OpenStack Compute (nova) "migration of anti-affinity server fails due to stale scheduler instance info" [Low,In progress] https://launchpad.net/bugs/1869050 - Assigned to Balazs Gibizer (balazs-gibizer)
23:32:20 ikla Insufficient compute resources: Claim pci failed. - is there anyway to see why it failed?
#openstack-nova - 2020-05-16
00:19:47 openstackgerrit Merged openstack/nova master: Update scheduler instance info at confirm resize https://review.opendev.org/714998
01:36:46 openstackgerrit Brin Zhang proposed openstack/nova master: Bump hacking min version to 3.1.0 https://review.opendev.org/727589
01:58:46 openstackgerrit Merged openstack/python-novaclient master: Switch to newer openstackdocstheme and reno versions https://review.opendev.org/728270
02:01:13 ikla any idea why instances would get stuck in deleting and building state? anything I can do?
02:31:07 openstackgerrit Merged openstack/nova master: Switch to newer openstackdocstheme and reno versions https://review.opendev.org/728263
04:25:31 openstackgerrit Takashi Natsume proposed openstack/nova master: Remove six.PY2 and six.PY3 https://review.opendev.org/726512
04:25:51 openstackgerrit Takashi Natsume proposed openstack/nova master: Remove six.add_metaclass https://review.opendev.org/726565
04:25:59 openstackgerrit Takashi Natsume proposed openstack/nova master: Remove six.reraise https://review.opendev.org/726898
04:26:12 openstackgerrit Takashi Natsume proposed openstack/nova master: Remove six.moves https://review.opendev.org/727224
04:26:22 openstackgerrit Takashi Natsume proposed openstack/nova master: Remove six.iteritems/itervalues/iterkeys https://review.opendev.org/727757
04:26:33 openstackgerrit Takashi Natsume proposed openstack/nova master: Remove six.byte2int/int2byte https://review.opendev.org/727777
04:26:43 openstackgerrit Takashi Natsume proposed openstack/nova master: Remove six.binary_type/integer_types/string_types https://review.opendev.org/728094
04:26:53 openstackgerrit Takashi Natsume proposed openstack/nova master: Remove six.text_type (1/2) https://review.opendev.org/728109
04:27:20 openstackgerrit Takashi Natsume proposed openstack/nova master: Remove six.text_type (2/2) https://review.opendev.org/728117
07:30:56 openstackgerrit Andreas Jaeger proposed openstack/nova-specs master: Switch to newer openstackdocstheme and reno versions https://review.opendev.org/728611
07:33:44 openstackgerrit Andreas Jaeger proposed openstack/nova-specs master: Switch to newer openstackdocstheme and reno versions https://review.opendev.org/728611
07:34:03 openstackgerrit Andreas Jaeger proposed openstack/os-vif master: Switch to newer openstackdocstheme and reno versions https://review.opendev.org/728612
22:40:29 openstackgerrit Merged openstack/nova stable/queens: Add config option for neutron client retries https://review.opendev.org/722822
#openstack-nova - 2020-05-17
05:39:47 openstackgerrit Nalini Varshney proposed openstack/nova master: Add migration to make key field type VARBINARY in aggregate_metadata table, https://review.opendev.org/725522
06:29:23 openstackgerrit Nalini Varshney proposed openstack/nova master: When unshelving an instance that was booted from a volume, "shelved_image_id" is None. Therefore, when performing unshelve, the image metadata could not be retrieved and ImagePropertiesFilter scheduling will fail. https://review.opendev.org/728675
#openstack-nova - 2020-05-18
04:36:49 openstackgerrit Wenping Song proposed openstack/nova master: delete sub resource provider when delete resource provider https://review.opendev.org/719163
06:52:14 openstackgerrit Harshavardhan Metla proposed openstack/nova master: [Nova] Add reference to Placement installation guide https://review.opendev.org/726936
07:19:51 openstackgerrit Wenping Song proposed openstack/nova master: delete sub resource provider when delete resource provider https://review.opendev.org/719163
07:34:01 bauzas good morning Nova
07:43:15 openstackgerrit vinay harsha mitta proposed openstack/nova stable/queens: For Installing and Configure Controller Node for openSUSE and SUSE Linux Enterprise. Modified the Placement API Service Endpoint port from "8780" to "8778". https://review.opendev.org/728766
08:09:24 gibi good morning
08:19:32 openstackgerrit Balazs Gibizer proposed openstack/nova master: Reserve DISK_GB resource for the image cache https://review.opendev.org/728439
08:28:51 openstackgerrit Liang Fang proposed openstack/nova-specs master: Support volume local cache https://review.opendev.org/728778
08:36:38 openstackgerrit Balazs Gibizer proposed openstack/nova stable/ussuri: Reproduce bug 1869050 https://review.opendev.org/728781
08:36:38 openstack bug 1869050 in OpenStack Compute (nova) train "migration of anti-affinity server fails due to stale scheduler instance info" [Low,Triaged] https://launchpad.net/bugs/1869050
08:36:38 openstackgerrit Balazs Gibizer proposed openstack/nova stable/ussuri: Update scheduler instance info at confirm resize https://review.opendev.org/728782
08:43:58 openstackgerrit Jiri Suchomel proposed openstack/nova master: Bring back allowed_direct_url_schemes in support of RBD image download https://review.opendev.org/728095
08:51:58 openstackgerrit Huaqiang Wang proposed openstack/nova master: test_hardware: code formating for multiline construct https://review.opendev.org/728793
09:09:42 openstackgerrit Balazs Gibizer proposed openstack/nova stable/rocky: Clean up allocation if unshelve fails due to neutron https://review.opendev.org/713196
09:48:57 openstackgerrit Balazs Gibizer proposed openstack/nova stable/ussuri: Bump hacking min version to 3.0.1 https://review.opendev.org/728803
09:49:40 gibi pep8 is broken on stable/ussuri I'm trying ^^ to fix it
09:49:52 gibi elod, lyarwood, bauzas: ^^^
09:50:04 bauzas oh shit
09:50:10 bauzas ok thanks for working on it
09:50:25 bauzas the flake8 story I guess ?
09:50:35 openstackgerrit vinay harsha mitta proposed openstack/nova stable/queens: For Installing and Configure Controller Node for openSUSE and SUSE Linux Enterprise. Modified the Placement API Service Endpoint port from "8780" to "8778". https://review.opendev.org/728766
09:50:53 gibi bauzas: yes, hacking version
09:51:10 elod gibi: thx, looking
09:51:32 bauzas gibi: I see
09:52:46 lyarwood gibi: ack thanks was waiting to see some CI results this morning for that, I need to drop for a few hours, will review later today once I'm back if it's still open.
09:53:03 gibi thanks
09:54:53 bauzas I just said +2, let's see whether it fixes
09:56:07 elod I think 'hacking>=2.0,<3.1.0 would be better.
09:56:31 elod so just keep the lower constraint and just raise the upper
09:57:51 gibi elod: 3.0.0 is broken
09:57:54 gibi for us
09:58:31 gibi but I guess some of the 2.x as well borken
09:58:41 gibi checking ...
09:59:49 gibi hm
09:59:51 gibi elod: https://zuul.opendev.org/t/openstack/build/70853c9681d448f890da99bfe6ca310b/log/job-output.txt#541
09:59:59 gibi this is the run of https://review.opendev.org/#/c/728781/
10:00:10 gibi for some reason it picked up 3.1.0 from hacking
10:00:50 elod weird
10:02:53 elod actually I don't even know how it was working before, because as far as I know this should not work: 'hacking>=2.0<2.1' because of the missing ','
10:02:55 gibi based on that run I expect that the change https://review.opendev.org/728803 will fail too
10:03:27 bauzas that's becoming a saga

Earlier   Later