Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2018-09-18
09:21:56 openstackgerrit Chen proposed openstack/nova stable/queens: Fix soft deleting vm fails after "nova resize" vm https://review.openstack.org/603340
09:22:08 naichuans_ bauzas: about gpu rp name mangling, it make no difference on xenserver, because gpu with same type would be put into the same gpu group. So base on gpu or gpu type is same
09:22:39 bauzas naichuans_: I'll look at your change for xen
09:22:53 naichuans_ bauzas: thanks.
09:23:41 bauzas naichuans_: we don't need to reshape for multiple GPU types
09:23:54 bauzas again, see my upgrade impact section on the spec
09:24:07 naichuans_ bauzas: will check it
09:26:42 naichuans_ bauzas: you mean this patch: https://review.openstack.org/#/c/552924/?
09:27:07 bauzas nope, the reproposed 'multiple GPU types' one
09:28:08 naichuans_ bauzas: will check it
09:29:28 openstackgerrit Ghanshyam Mann proposed openstack/nova master: Merge extended_volumes extension response into server view builder https://review.openstack.org/596285
09:33:24 naichuans_ bauzas: We restrict the user scenario to avoid those troubles~ Fair enough :)
09:33:45 bauzas naichuans_: which troubles ? :)
09:34:19 openstackgerrit Wenran Xiao proposed openstack/nova-specs master: Move juno approved specs https://review.openstack.org/603344
09:35:13 naichuans_ bauzas: Directly upgrade the old vgpu allocation to multiple vgpu rp support is troublesome~
09:35:57 naichuans_ bauzas: As I mentioned in the email, we have difficluties to decide the old allocation belongs.
09:36:38 bauzas naichuans_: ah this
09:36:40 bauzas naichuans_: yeah
09:36:52 bauzas naichuans_: well, about finding which is which, I don't see it difficult
09:37:01 bauzas but I could be wrong
09:37:14 bauzas I mean, allocations are done with a consumer UUID
09:37:33 bauzas we know that the consumer UUID is either an instance or a migration
09:37:57 bauzas given we don't support migrations yet in libvirt, that simplifies
09:38:32 bauzas so, the idea is to lookup the VGPU allocations, find for each which instance it is, check the libvirt XML to know which card is attached to, and give it back
09:38:46 naichuans_ bauzas: Yes, we need to talk with hypervisor, looks not so natural for me~
09:39:03 bauzas naichuans_: I already do that for rebooting the host
09:39:33 bauzas naichuans_: given mdevs aren't persisted on reboots, every compute restart checks the existing allocations and recreates the mdevs
09:40:14 naichuans_ bauzas: awesome~
09:40:56 bauzas naichuans_: I need to go to the gym, but I can help you by showing you the libvirt codee
09:41:01 bauzas give me a sec
09:41:30 naichuans_ bauzas: Got it, thank you very much~ And for so much information :)
09:42:58 bauzas naichuans_: https://github.com/openstack/nova/blob/master/nova/virt/libvirt/driver.py#L566
09:43:12 naichuans_ bauzas: will check it
09:49:50 naichuans_ bauzas: According to your spec, multiple vgpu support would not be implemented in stein, right?
09:50:00 naichuans_ https://review.openstack.org/#/c/602474/2/specs/stein/approved/vgpu-stein.rst
09:52:11 kashyap tox -epep8 -- -HEAD
09:52:21 kashyap Oeps
09:54:10 openstackgerrit Kashyap Chamarthy proposed openstack/nova master: libvirt: Drop MIN_LIBVIRT_PARALLELS_SET_ADMIN_PASSWD https://review.openstack.org/562313
09:59:27 pvc hi bauzas
09:59:29 pvc it is okay now :)
09:59:42 pvc i have 13 supported types for my one p100
10:04:37 pvc do i need to do something more bauzas?
10:21:53 openstackgerrit Tetsuro Nakamura proposed openstack/nova-specs master: Spec: Support filtering by forbidden aggregate https://review.openstack.org/603352
10:32:07 openstackgerrit Vlad Gusev proposed openstack/nova stable/queens: Remove mox in libvirt/test_driver.py (3) https://review.openstack.org/603353
10:38:29 stephenfin gibi: Any chance I could get a copy of the commands you were using in the bandwidth-aware scheduling demo last week?
10:45:30 openstackgerrit Vlad Gusev proposed openstack/nova stable/queens: libvirt: Reduce calls to qemu-img during update_available_resource https://review.openstack.org/603358
10:58:11 Luzi Hello, I'm new to nova. How do you propose patches or ideas? Do you use Launchpad or Storyboard?
10:59:16 openstackgerrit Tetsuro Nakamura proposed openstack/nova-specs master: Spec: Support filtering by forbidden aggregate https://review.openstack.org/603352
11:05:53 s10 Luzi: Hello. We should use gerrit for patches proposal, see https://docs.openstack.org/infra/manual/developers.html . AFAIK, bugs and RFE are proposed in Nova by Launchpad.
11:07:01 Luzi yeah, i work in a team, which implements security enhancements and we want to contribute to openstack
11:07:32 Luzi for now we want to propose a signature check after an image is cached on a compute node
11:11:57 gibi stephenfin: hi
11:12:18 gibi stephenfin: let me talk to rubasov if he have an easy to digest version of the commands
11:13:08 gibi stephenfin: hi did promise to write up the whole thing as a blogpost but I don't know the timing of that
11:13:34 s10 Luzi: So basically you want to implement signature check that is described in https://docs.openstack.org/glance/latest/user/signature.html ?
11:14:33 Luzi s10, well nova already checks the signature when downloading an image from glance
11:14:56 Luzi but not when it was cached on the compute node
11:18:57 s10 Luzi: I think we should wait for nova cores to wake up (15:00-16:00 UTC+0) for their thoughts. I believe good starting point is to open a bug in https://bugs.launchpad.net/nova describing this issue
11:19:20 Luzi okay, thank you s10 :)
11:44:23 kashyap 'Citrix XenServer CI' is a 'voting' job, is it?
11:55:29 sean-k-mooney kashyap: i think so
11:56:01 gibi stephenfin: does it work for you if there will be a blogpost still this week?
11:56:09 kashyap sean-k-mooney: Trying to figure out if this change is failing because of it: https://review.openstack.org/#/c/602592/ ("Use 'virt' as the default machine type for ARMv7")
11:57:19 sean-k-mooney kashyap: well you are chaing the libvirt driver the cirtx xenserver ci should be using the native xen driver
11:57:21 kashyap sean-k-mooney: Also that 'neutron-grenade' failure is irrelevant
11:57:27 kashyap sean-k-mooney: Exactly
11:57:52 sean-k-mooney kashyap: the neutron-grenade job is voting hence the zuul -1
11:58:06 kashyap sean-k-mooney: That Grenade thing was failing due to the presence of untracked working tree files: http://logs.openstack.org/92/602592/3/check/neutron-grenade/bfeb3e6/logs/devstack-gate-setup-workspace-old.txt
11:58:16 kashyap Yesterday mriedem was saying he saw it, but no clue why we're hitting it
11:58:28 kashyap Who can I bother about "Grenade issues"?
11:58:46 sean-k-mooney kashyap: infra but i would do a recheck first
11:58:49 kashyap A 'recheck' today didn't fix it; and weirdly enough, it didn't result in fresh logs.
11:59:04 kashyap (As in, it's still posted yesterday's 'neutron-grenade' logs)
11:59:08 sean-k-mooney has it actully run again?
11:59:29 johnthetubaguy Luzi: curious about the image signature issue, Is this because you are worried about on disk changes happening after the image gets in the cache?
11:59:32 sean-k-mooney kashyap: zuul has not commented back since the recheck
11:59:49 kashyap sean-k-mooney: I ran at 11:33 AM ("Gerrit Time"). And it's showing '-1' isn't it?
12:00:08 sean-k-mooney that is from before the recheck
12:00:27 sean-k-mooney kashyap: your job is still running
12:00:47 Luzi johnthetubaguy, yes the image could be replaced, that's because a recheck would be fine
12:00:49 kashyap sean-k-mooney: Hmm, in that case, I thought it'd "reset" the -1.
12:01:01 johnthetubaguy Luzi: if you didn't bump across it yet, this was the original design spec: http://specs.openstack.org/openstack/nova-specs/specs/rocky/implemented/nova-validate-certificates.html
12:01:01 kashyap sean-k-mooney: You showed me a live console in the past, can you point me to that again?
12:01:19 johnthetubaguy Luzi: well it references the original bit at the bottom of there
12:01:27 sean-k-mooney kashyap: that is the last job that is running http://zuul.openstack.org/stream.html?uuid=e576a55aa1d1464fbbecbdea0cd574d8&logfile=console.log
12:02:04 sean-k-mooney it just finished devstack and is tarting the tempest and will be starting the tempest run soon
12:02:27 kashyap sean-k-mooney: Most excellent. I _guessed_ so, but thought it shouldn't take 3+ hours. Apparently it does :-)
12:02:50 sean-k-mooney kashyap: if you go to http://zuul.openstack.org/ and filter by 602592 you will see the rest
12:03:17 kashyap sean-k-mooney: You're clairvoyant. That was my next question :-) To remind how I can figure it out next time
12:03:20 kashyap Thanks!
12:07:58 johnthetubaguy Luzi: this is an email thread that might be worth reading through, around when we used to do image verification: http://lists.openstack.org/pipermail/openstack-operators/2016-May/010554.html
12:09:11 johnthetubaguy Luzi: largely some basic checks we removed because once you have access to the image cache, its basically game over for a heap of reasons. Could well be something wrong in that analysis, but its worth a read of the thread for context
12:09:40 Luzi johnthetubaguy, thank you, i will read through
12:24:38 openstackgerrit Merged openstack/os-traits master: import zuul job settings from project-config https://review.openstack.org/601377
12:33:18 openstackgerrit Merged openstack/os-traits master: switch documentation job to new PTI https://review.openstack.org/601378
12:33:20 openstackgerrit Merged openstack/os-traits master: add python 3.6 unit test job https://review.openstack.org/601379
12:44:06 bauzas naichuans_: nope, just the fact that we would only support multiple types once reshape is fully done
12:44:21 bauzas pvc: nope, you're all good, enjoy
12:47:19 pvc thank you so much bauzas :)
12:47:57 stephenfin gibi: Blogpost this week is a-ok. A dump of the commands, assuming there wasn't anything confidential in them, is also a-ok :)

Earlier   Later