Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2025-08-06
14:29:59 sean-k-mooney i dont know what the cause iss but thats a smoking gun
14:30:49 sean-k-mooney i.e. i dont knwo why it does not happen after restarting nova-comptue but apprently it does not always fail
14:33:06 stephenfin sean-k-mooney: dumped it into the bug https://bugs.launchpad.net/nova/+bug/2119114/comments/3
14:33:24 sean-k-mooney cool i didnt have the link to that
15:12:55 noonedeadpunk ++
15:24:08 cardoe sean-k-mooney: so not sure if you saw my follow up on the vlan spec. The issue you pointed out in chat about the id field and that dansmith brought up on gerrit itself isn't something that spec can really fix right now. It's just defining the behavior of nova code today (and the behavior that its been since June 2015).
15:25:04 cardoe So I'm really not sure what to fix cause that spec isn't about changing the behavior of existing "phy" interfaces. It's about including "vlan" interfaces. The "vlan" interface references to "phy" interface its running on.
15:26:22 sean-k-mooney the spec si free to extend or modify the schema and change the data that is stored in specific fileds
15:26:40 sean-k-mooney it is not possibel for nova ot know what the name of the Phy interface will be
15:26:50 sean-k-mooney so the value that are provdie are and alwasy were informational
15:27:03 sean-k-mooney you are ment ot match on things like the mac adress
15:27:17 sean-k-mooney when you are consumign this to identify the phy not the id or name
15:27:29 sean-k-mooney its just an identifyer for the parsing program
15:29:26 abhishekk o/ can I get a review on https://review.opendev.org/c/openstack/nova/+/950623, it has one +2, this will be important change to get rid of show_multiple_locations config option
15:38:16 noonedeadpunk fwiw, I've addressed comments in https://review.opendev.org/c/openstack/nova-specs/+/955783 and I think implementation is also available for preliminary review
15:45:56 melwitt thanks sean-k-mooney
15:47:29 sean-k-mooney abhishekk: does it work this time? :) if i recall we reverted it because there was soem issue on the glance side that needed to be adressed first
15:47:57 sean-k-mooney abhishekk: ah i see dan is already +2 that reverting the revert
15:48:03 sean-k-mooney ill take a look shortly
15:48:46 sean-k-mooney we can always have a revert^3 patch if needed i guess.
16:48:49 abhishekk sean-k-mooney: thank you!!! I have added rechecks 7-8 times on revert patch to ensure that failure does not occur again
17:08:23 sean-k-mooney noonedeadpunk: oh the parallel mgtions
17:08:36 sean-k-mooney we really need to create teh 2026.1 folder in the specs repo
17:08:45 sean-k-mooney so you can propose that to the right pplace
17:09:59 sean-k-mooney noonedeadpunk: ill try and loop back to that but it likely will be next week at this rate
17:15:39 opendevreview Stephen Finucane proposed openstack/nova master: api: Add response body schemas for servers APIs (2/6) https://review.opendev.org/c/openstack/nova/+/956236
17:15:39 opendevreview Stephen Finucane proposed openstack/nova master: api: Add response body schemas for servers APIs (1/6) https://review.opendev.org/c/openstack/nova/+/956234
17:15:40 opendevreview Stephen Finucane proposed openstack/nova master: api: Add response body schemas for servers APIs (4/6) https://review.opendev.org/c/openstack/nova/+/956238
17:15:40 opendevreview Stephen Finucane proposed openstack/nova master: api: Add response body schemas for servers APIs (3/6) https://review.opendev.org/c/openstack/nova/+/956237
17:15:41 opendevreview Stephen Finucane proposed openstack/nova master: api: Add response body schemas for servers APIs (6/6) https://review.opendev.org/c/openstack/nova/+/956240
17:15:41 opendevreview Stephen Finucane proposed openstack/nova master: api: Add response body schemas for servers APIs (5/6) https://review.opendev.org/c/openstack/nova/+/956239
17:15:42 opendevreview Stephen Finucane proposed openstack/nova master: tests: Invert validation check https://review.opendev.org/c/openstack/nova/+/956241
17:15:42 opendevreview Stephen Finucane proposed openstack/nova master: api: Add response body schemas for server shares APIs https://review.opendev.org/c/openstack/nova/+/956266
#openstack-nova - 2025-08-07
05:51:06 noonedeadpunk sean-k-mooney: but it's not yet feature freeze for 2025.2 if I read the schedule right? and this is is quite tiny, easy, so maybe it's the right place? ;)
09:58:33 sean-k-mooney noonedeadpunk: we are well pass the spec approval freeze so this is something for next cycle
09:58:51 sean-k-mooney noonedeadpunk: the spec freeze was milestone 2
10:02:10 noonedeadpunk ah, right, true
10:02:17 noonedeadpunk I was looking for feature freeze
10:03:22 sean-k-mooney that does not mean it cant merge early in 2026.1 as in late september/october
10:03:26 noonedeadpunk it's kinda sad though, as this is pretty much the only way so far to get live migrations working reliable until gnutls is patched and with memory being encrypted
10:03:59 sean-k-mooney well its not actully fixing the issue
10:03:59 noonedeadpunk or well... use tunnels...
10:04:10 noonedeadpunk yeah, true
10:04:24 sean-k-mooney it just make it less likely as each coonenction gets it own rekey timer
10:04:39 sean-k-mooney the workaroudn for now is to just disable tls3 with aes
10:04:44 sean-k-mooney via the crypto polices
10:04:56 noonedeadpunk which you can;'t do outside of EL
10:04:58 sean-k-mooney i.e. disable the specific policy that has the issue
10:05:05 sean-k-mooney EL?
10:05:10 noonedeadpunk *RHEL
10:05:22 sean-k-mooney oh well that not a rhel thing as far as im aware
10:05:34 noonedeadpunk as I tried to place a gnutls config on ubuntu and qemu just ignores it
10:05:43 noonedeadpunk so it's not a thing on debian/ubuntu
10:05:58 noonedeadpunk Maybe it does work on centos/rocky - dunno
10:06:00 sean-k-mooney i suspect it might be but uses a diffent location
10:06:32 noonedeadpunk Ubuntu just dropped policy package with comment it never-ever worked on ubuntu 24.04 to start with
10:06:44 sean-k-mooney https://wiki.debian.org/CryptoPolicy
10:06:51 noonedeadpunk https://bugs.launchpad.net/ubuntu/+source/crypto-policies/+bug/2069129
10:06:59 sean-k-mooney so apprently it s a fedora thing?
10:07:11 sean-k-mooney https://fedoraproject.org/wiki/Changes/CryptoPolicy
10:07:21 noonedeadpunk well, I think anyway it does not work on debian/ubuntu
10:07:27 sean-k-mooney form 2014 that they were loooking to upstream
10:07:42 noonedeadpunk "crypto-policies is not supported on Ubuntu: it requires deep integration with the packages and their configurations. This integration doesn't exist and is unlikely to ever exist outside of redhat-based distributions"
10:08:03 sean-k-mooney https://manpages.ubuntu.com/manpages/jammy/en/man8/update-crypto-policies.8.html
10:08:22 sean-k-mooney well the pakcage for doing it is in ubuntu
10:08:31 noonedeadpunk So 2069129 claims it never worked despoite being packaged
10:08:43 noonedeadpunk which is the reason it's dropped
10:08:56 sean-k-mooney ya so this is in rhel to make fips work
10:09:15 sean-k-mooney i know cannonicals fipes supprot is a paied extra
10:09:43 noonedeadpunk anyway, I didn't find the way so far to make ubuntu to use POLY1305
10:09:50 sean-k-mooney ack
10:10:29 noonedeadpunk even though I blacklisted AES in gnutls config and gnutls output was like that: https://paste.openstack.org/show/b2VjO9rv4976Y4prCtt9/
10:10:42 noonedeadpunk seems that qemu was still doing what it wanted
10:11:30 sean-k-mooney you using kolla right
10:11:34 noonedeadpunk osa
10:11:47 sean-k-mooney oh so not in a contianer then
10:11:56 sean-k-mooney i was going to ask if you updated it in the contianer
10:12:12 noonedeadpunk yeah, it's just on the node
10:12:13 sean-k-mooney or suggest you try the centos libvirt contianer to see if that woudl work around it
10:12:48 noonedeadpunk well... I probably could do that....
10:12:55 sean-k-mooney its possible that htey staic link it or something like that but debian is normlaly all in on dynmaic linking everything
10:14:26 sean-k-mooney in your case i think tunnled migration or using just tcp if you are allow too is the best approch unless you want to carry that patch downstream until its landed upstream
10:14:44 sean-k-mooney still to fix iti properly you will need the qemu fix when that is ready
10:15:14 noonedeadpunk we have local drives as nova drives, which we expect to live migrate as well, which does not work with the trunnel;
10:15:16 opendevreview Stephen Finucane proposed openstack/nova master: api: Add response body schemas for servers APIs (5/6) https://review.opendev.org/c/openstack/nova/+/956239
10:15:17 opendevreview Stephen Finucane proposed openstack/nova master: api: Add response body schemas for servers APIs (6/6) https://review.opendev.org/c/openstack/nova/+/956240
10:15:17 opendevreview Stephen Finucane proposed openstack/nova master: api: Add response body schemas for server shares APIs https://review.opendev.org/c/openstack/nova/+/956266
10:15:18 opendevreview Stephen Finucane proposed openstack/nova master: tests: Invert validation check https://review.opendev.org/c/openstack/nova/+/956241
10:15:21 noonedeadpunk anyway
10:18:04 opendevreview Dmitriy Rabotyagov proposed openstack/nova-specs master: Add 2026.1 specs folder https://review.opendev.org/c/openstack/nova-specs/+/956767
10:19:47 opendevreview Dmitriy Rabotyagov proposed openstack/nova-specs master: Propose enabling parallel live migrations for libvirt https://review.opendev.org/c/openstack/nova-specs/+/955783
10:35:15 opendevreview Merged openstack/nova master: Revert^2 "Support glance's new location API" https://review.opendev.org/c/openstack/nova/+/950623
10:45:03 opendevreview Takashi Kajinami proposed openstack/nova master: Migrate MEM_ENCRYPTION_CONTEXT from root provider https://review.opendev.org/c/openstack/nova/+/921814
10:45:04 opendevreview Takashi Kajinami proposed openstack/nova master: Detect AMD SEV-ES support https://review.opendev.org/c/openstack/nova/+/925685
10:45:04 opendevreview Takashi Kajinami proposed openstack/nova master: Add hw_mem_encryption_model image property https://review.opendev.org/c/openstack/nova/+/927706
10:45:05 opendevreview Takashi Kajinami proposed openstack/nova master: libvirt: Launch instances with SEV-ES memory encryption https://review.opendev.org/c/openstack/nova/+/926106
10:46:55 tkajinam Uggla, I wonder if I can ask for your help to move this series forward ^^^ I've been receiving emails regularly to ask progress about this work (from different parties, which is interesting) and I'm hoping that I can move it forward (getting it merged is the best but I at least need some feedback)
10:50:57 Uggla Hi tkajinam, I'll do my best to try to review it and warn cores for reviews, as this is an important feature.
10:51:48 tkajinam Uggla, thanks !
10:56:27 sean-k-mooney oh the sev work

Earlier   Later