Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2019-08-06
11:04:10 openstackgerrit Stephen Finucane proposed openstack/nova master: tests: Split NUMA object tests https://review.opendev.org/672336
11:04:10 openstackgerrit Stephen Finucane proposed openstack/nova master: Rework 'hardware.numa_usage_from_instances' https://review.opendev.org/672565
11:04:11 openstackgerrit Stephen Finucane proposed openstack/nova master: trivial: Rewrap definitions of 'NUMACell' https://review.opendev.org/674395
11:04:11 openstackgerrit Stephen Finucane proposed openstack/nova master: libvirt: '_get_(v|p)cpu_total' to '_get_(v|p)cpu_available' https://review.opendev.org/672693
11:04:12 openstackgerrit Stephen Finucane proposed openstack/nova master: hardware: Differentiate between shared and dedicated CPUs https://review.opendev.org/671800
11:04:12 openstackgerrit Stephen Finucane proposed openstack/nova master: libvirt: Remove unnecessary argument https://review.opendev.org/674396
11:04:13 openstackgerrit Stephen Finucane proposed openstack/nova master: Add support for translating CPU policy extra specs, image meta https://review.opendev.org/671801
11:04:13 openstackgerrit Stephen Finucane proposed openstack/nova master: objects: Rename 'fields' import to 'obj_fields' https://review.opendev.org/674103
11:22:57 openstackgerrit Merged openstack/nova master: Add nova-osprofiler-redis job to experimental queue https://review.opendev.org/674326
11:23:06 openstackgerrit Merged openstack/nova master: doc: fix physets typo https://review.opendev.org/674667
12:01:28 openstackgerrit Alex Xu proposed openstack/nova master: Moves the allocation retrieving early https://review.opendev.org/670783
12:01:28 openstackgerrit Alex Xu proposed openstack/nova master: Add the virt driver interface for claim and unclaim the devices https://review.opendev.org/670782
12:01:29 openstackgerrit Alex Xu proposed openstack/nova master: Add DeviceManager to the libvirt virt driver https://review.opendev.org/671388
12:01:29 openstackgerrit Alex Xu proposed openstack/nova master: Calling the virt driver's claim/unclaim_for_instance in resource tracker https://review.opendev.org/670784
12:01:30 openstackgerrit Alex Xu proposed openstack/nova master: Using the claim/unclaim_for_instance for mdevs https://review.opendev.org/671222
12:01:30 openstackgerrit Alex Xu proposed openstack/nova master: Populates the existing mediated devices in the libvirt device manager https://review.opendev.org/670787
12:01:31 openstackgerrit Alex Xu proposed openstack/nova master: Adds functional test for creating the instance with vgpus https://review.opendev.org/671398
12:05:02 openstackgerrit Artom Lifshitz proposed openstack/nova stable/stein: Add glanceclient method call debug log https://review.opendev.org/674791
12:11:52 sean-k-mooney efried: have you seen https://review.opendev.org/#/c/674726/1 that would be a hard blocker on cyborg integration proceeding in this cylce if the only instance action that work are basically spawn and delete instance
12:14:26 sean-k-mooney not supporting live migration in v1 is resonable but i think everything else shoudl be supported IMO or its not ready to be integrated.
12:17:41 cdent tssurya: are you still working on https://review.opendev.org/623558 ? I've been doing some profiling in the scheduler and it's definitely clear that multiple-queries to the db is problematic and that stacks approach would be better
12:19:54 tssurya cdent: feel free to update the patch and work on it
12:20:21 tssurya I am not working on it atm
12:20:36 cdent okay. I'm not sure if I will have time or not, but I'll see if I can use it as a starting point
12:21:01 tssurya I guess that patch had that typo https://review.opendev.org/#/c/623558/2/nova/scheduler/host_manager.py@892 which made it a bit wobly
12:21:11 tssurya cdent: thanks :)
12:21:36 cdent :)
12:21:59 sean-k-mooney jangutter: stephenfin thanks for reviewing https://review.opendev.org/#/c/672834/2 ill adress the nits and respin it after lunch
12:22:58 jangutter sean-k-mooney: I was thinking about it afterwards: have you seen bridges pop up with ageing_time != 300 ?
12:25:25 openstackgerrit Alex Xu proposed openstack/nova master: Add DeviceManager to the libvirt virt driver https://review.opendev.org/671388
12:25:26 openstackgerrit Alex Xu proposed openstack/nova master: Using the claim/unclaim_for_instance for mdevs https://review.opendev.org/671222
12:25:26 openstackgerrit Alex Xu proposed openstack/nova master: Populates the existing mediated devices in the libvirt device manager https://review.opendev.org/670787
12:25:27 openstackgerrit Alex Xu proposed openstack/nova master: Adds functional test for creating the instance with vgpus https://review.opendev.org/671398
12:36:14 sean-k-mooney jangutter: without the patch?
12:36:21 sean-k-mooney jangutter: or with it
12:36:44 jangutter sean-k-mooney: with the patch, leaving out _setting_ ageing_time.
12:37:29 sean-k-mooney well it should default to your kernel default in that case
12:37:48 sean-k-mooney so it will be 300 unless you have changed it with something like sysctl.conf
12:38:38 jangutter sean-k-mooney: yep, or something helpful like NetworkManager changes it for you, etc. Was just wondering if it wouldn't make sense to have the functional test check for 300.
12:38:40 sean-k-mooney for ovs with hybrid plug we woudl still be hardcoding it to 0
12:38:54 sean-k-mooney ah
12:39:13 sean-k-mooney well we could but i dont know if those helpful service could break it
12:39:40 sean-k-mooney we could maybe assert its no 0 but im not sure if that would always be the case
12:41:19 sean-k-mooney do you want me to add one for that usecase? as i said im not sure if it would always be correct but i suspect it would be
12:41:48 jangutter sean-k-mooney: I think, whatever choice is made, there's going to be some edge case system that might break it. Both checking != 0 and == 300 could fail so I'd just say pick one and add a NOTE like stephenfin said.
12:42:52 sean-k-mooney ok ill do !=0 and add a note and if we get a bug we can either nuke the test or close it as wont fix
12:43:25 sean-k-mooney or actully maybe i can check sysfs to see if there is a default
12:43:38 sean-k-mooney and skip the test if the default ageing time is 0
12:44:26 jangutter sean-k-mooney: that might be useful, but I'm almost tempted to say that might point to some bad defaults ppl should investigate.
12:45:17 sean-k-mooney it might but how many people are going to realisticaly run the os-vif functional test beside us
12:45:53 jangutter sean-k-mooney: yep, realistically, the only people interested would be people packaging (i.e. distros).
12:50:17 sean-k-mooney looks like there is no way to set teh bridge againg time globally so i think we are safe with !=0
12:50:59 sean-k-mooney e.g. via the kernel networkmanager could always change thinks but that is out of scope for this patch
12:52:13 stephenfin Does boot from volume imply we will not have any swap or ephemeral disk also, or does it just apply to the root disk?
12:52:34 sean-k-mooney it just applies to the root disk
12:52:41 sean-k-mooney if the flaovr has swap those should be local
12:52:45 stephenfin Does that mean this is wrong? https://github.com/openstack/nova/blob/master/nova/scheduler/utils.py#L400-L401
12:52:50 sean-k-mooney same for ephemeral
12:53:23 stephenfin it should read something like
12:53:35 stephenfin if is_bfv:
12:53:43 stephenfin disk = (flavor.swap / 1024) + flavor.ephemeral_gb
12:53:44 stephenfin else:
12:54:00 sean-k-mooney 1 that is only used for nova-manage
12:54:09 stephenfin disk = (flavor.swap / 1024) + flavor.ephemeral_gb + flavor.root_gb
12:54:32 sean-k-mooney 2 its a sligtly differet impleatnion of what we do here https://github.com/openstack/nova/blob/master/nova/scheduler/utils.py#L436-L442
12:54:48 stephenfin Yeah, and an incorrect one?
12:55:43 sean-k-mooney yep i think so
12:55:55 sean-k-mooney but that is not the only reason that that is slightly broken
12:56:04 sean-k-mooney that being resources_from_flavor
12:56:20 stephenfin yeah, it also shouldn't be setting things to 0. placement doesn't allow that
12:56:46 sean-k-mooney well that is not what i ment but its another bug
12:58:03 sean-k-mooney its ignoring the image which means when you use the nova manage command that executes that function it can ignore the traits on the image
12:58:30 sean-k-mooney actully its not broken
12:58:32 sean-k-mooney its correct
12:58:45 sean-k-mooney rather ((0 if is_bfv else flavor.root_gb) +
12:58:47 sean-k-mooney swap_in_gb + flavor.ephemeral_gb)
12:58:49 sean-k-mooney is correct
12:58:55 stephenfin how?
12:59:10 sean-k-mooney the bracket after root_gb
12:59:16 sean-k-mooney the if only applies to the root gb
12:59:50 sean-k-mooney i guess if you had not swap or ephemeral_gb it woudl still be wrong
12:59:58 stephenfin ah, indeed
13:00:04 stephenfin I missed that. Good spot!
13:00:28 stephenfin Yeah, we should be checking for that but it's not as serious
13:00:33 sean-k-mooney that is handel in the other implemataion here https://github.com/openstack/nova/blob/master/nova/scheduler/utils.py#L456-L459
13:01:19 sean-k-mooney if you want to fix it just condtionally append https://github.com/openstack/nova/blob/master/nova/scheduler/utils.py#L406
13:18:42 efried dustinc: left a comment. Better as a @property so you don't have to do icky things like passing arbitrary *args, **kwargs or method names as strings
13:19:26 openstackgerrit Chris Dent proposed openstack/nova master: single pass instance info fetch in host manager https://review.opendev.org/623558
13:19:26 openstackgerrit Chris Dent proposed openstack/nova master: add InstanceList.get_all_uuids_by_hosts() method https://review.opendev.org/623557
13:19:27 openstackgerrit Chris Dent proposed openstack/nova master: DNM: Play with cache of host mappings in nova scheduler https://review.opendev.org/674811
13:19:27 openstackgerrit Chris Dent proposed openstack/nova master: DNM: add profiling to scheduler https://review.opendev.org/674810
13:19:48 cdent mriedem: then of that ^ stack has some interesting observations about host states
13:19:56 cdent s/then/the end/
13:21:17 sean-k-mooney cdent: sylvain is on vaction for a few weeks so i or someone else on the team will pick up his placement audit patch
13:21:25 sean-k-mooney cdent: quick question on https://review.opendev.org/#/c/670112/2/nova/cmd/manage.py@2764
13:21:39 sean-k-mooney cdent: are you asserting that pagination should not be required
13:21:45 cdent hi sean-k-mooney , reading
13:21:58 sean-k-mooney e.g. i can just just remove that not/question
13:22:19 cdent sean-k-mooney: yes, just remove that note, I think. as there's no plan to do pagination (server-side) in placement
13:22:41 sean-k-mooney cool i think the other comment does need to be address

Earlier   Later