Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2022-04-26
16:55:38 bauzas gibi: well, I chose to stay :)
16:55:45 artom He replaced himself with a very small shellscript
16:55:56 bauzas Sophie will stare at me when I go off the office room
16:56:00 artom Turns out we was getting bash'ed this whole time
16:56:09 gibi artom: lol
16:56:17 bauzas artom: I wish I could
16:57:11 bauzas artom: but I've been told any IA bot turns crazy after you let people discussing with it
16:58:02 gibi it is our, humans, fault to make AI bots crazy
16:58:23 artom So we're to AIs what demons are to us, The Corruptors?
16:58:38 sean-k-mooney lol
16:58:41 sean-k-mooney but yes
16:58:42 gibi at least the naming sounds badass
16:59:20 artom "I'm a Corruptor" "What does that mean?" "Oh I make chatbots says rude words and stuff"
17:02:14 bauzas or I could sleep on the couch
20:36:45 opendevreview Merged openstack/nova master: Move centos stream testing to centos-9-stream https://review.opendev.org/c/openstack/nova/+/839275
21:15:49 melwitt bauzas: could you also add https://review.opendev.org/c/openstack/nova/+/821348 to your stable review list please? the -2 from zuul is bc I accidentally voted on it earlier
#openstack-nova - 2022-04-27
03:13:16 mnaser sean-k-mooney, bauzas: ricolin on our side will help drive this and i'll also be helping in shaping it from our side. we don't have a whole *lot* of background knowledge about it but we can do a bunch of testing about the vIOMMU stuff
03:38:14 opendevreview Merged openstack/nova stable/yoga: Fix eventlet.tpool import https://review.opendev.org/c/openstack/nova/+/836788
05:03:24 opendevreview Merged openstack/nova stable/ussuri: Fix the vGPU dynamic options race https://review.opendev.org/c/openstack/nova/+/831524
07:14:07 gibi good morning
07:16:51 bauzas gibi: morning
07:16:59 bauzas mnaser: ack, very much appreciated
07:26:30 bauzas elodilles: I made my homework I promised
07:26:41 bauzas melwitt: sent to the gate
08:12:00 elodilles bauzas: cool, thanks \o/
09:25:46 opendevreview Andrew Bonney proposed openstack/nova master: Fix segment-aware scheduling permissions error https://review.opendev.org/c/openstack/nova/+/839361
10:00:03 opendevreview Rajat Dhasmana proposed openstack/nova master: Add support for volume backed server rebuild https://review.opendev.org/c/openstack/nova/+/820368
10:00:03 opendevreview Rajat Dhasmana proposed openstack/nova master: Add conductor RPC interface for rebuild https://review.opendev.org/c/openstack/nova/+/831219
10:00:04 opendevreview Rajat Dhasmana proposed openstack/nova master: Add API support for rebuilding BFV instances https://review.opendev.org/c/openstack/nova/+/830883
10:21:47 ricolin bauzas: for vIOMMU, I will work on BP asap for further discussion. Also I'm not that familar with nova future develop, are these also require to raise a meeting topic, or BP is enough for first step
10:33:41 sean-k-mooney ricolin: you should ideally draft a spec
10:33:49 sean-k-mooney rather then just a blueprint
10:39:15 ricolin sean-k-mooney: agree
10:43:49 sean-k-mooney are you familar enough with that process https://github.com/openstack/nova-specs/blob/master/specs/zed-template.rst
10:44:06 sean-k-mooney the poc patch is a good start
10:44:20 sean-k-mooney and i left most of my feed back in that patch already from a design point of view
10:45:06 sean-k-mooney so for the spec really waht i am looking for is to defien the scope of the chagne, the interfac (extra_specs and image proerties) and the usecasses we want to adress.
10:45:44 sean-k-mooney there is also a schduling question e.g. can we rely based on min libvirt version on all host suport this or do we need a compute capablity trait and prefileter for schduling
10:46:05 sean-k-mooney i havent checked thet min version to determin that
10:47:03 sean-k-mooney the vIOMMU has been aroudn a while so i think our min version support it but the adress space option is newer so not sure about that part
11:27:19 gibi bauzas: you probably interested in https://review.opendev.org/c/openstack/nova/+/839361 as it fixes a bug in the routed net pre-filter
11:39:28 sean-k-mooney gibi: i was looking at that
11:39:38 sean-k-mooney im not sure why we would need admin rights
11:39:57 sean-k-mooney i would have expect the segment id to be viabel to a normal user
11:40:09 gibi because neutron does not return the segment_id for non admins in the GET /subnets response
11:40:25 gibi I tried in a devstack, it only returns it for an admin
11:40:36 sean-k-mooney hum that soudn like a design issue to me on the neutron side
11:40:47 sean-k-mooney i wonder why that desision was made
11:40:47 gibi of course other solution is to change neutron to return that for non-admins
11:41:02 gibi I haven't tracked down the reason of this on the neutron side
11:41:16 gibi I had many similar bugs in the qos work
11:41:32 gibi there the resource_request field being admin-only caused similar problems
11:41:34 sean-k-mooney right but i dont think using admin is the right move in general
11:41:52 gibi in general no
11:41:53 sean-k-mooney yep i was thinking of that and there is no reason that shoudl be admin-only
11:42:37 sean-k-mooney it shoudl not be writable form the api but it shoudl be readable by the owner
11:43:13 sean-k-mooney i think the segment id shoudl be the same readable by project_reader
11:43:25 sean-k-mooney but not writabel by project_member
11:45:51 gibi I'm OK to pursue this with the neutron team but I think we can merge the above fix as changing API behavior like that might no be backportable
11:50:00 sean-k-mooney its currently project_admin by policy https://github.com/openstack/neutron/blob/master/neutron/conf/policies/subnet.py#L54-L67= its a correct fix i guess
11:51:02 gibi hm, that is subnet creation
11:51:22 gibi but yes the GET is project admin too segment_id
11:51:27 gibi https://github.com/openstack/neutron/blob/6db61d070483df60ffbf992d57cef2c1903fe355/neutron/conf/policies/subnet.py#L98-L108
11:52:22 sean-k-mooney ya its there in both
11:52:45 opendevreview Wenping Song proposed openstack/os-traits master: Add nova and cyborg owner traits for resources https://review.opendev.org/c/openstack/os-traits/+/839128
11:52:54 sean-k-mooney what troubling is actully https://github.com/openstack/neutron-lib/blob/master/neutron_lib/api/definitions/subnet_segmentid_writable.py
11:53:27 sean-k-mooney apparently you can update the segment id of a subnet
11:54:04 sean-k-mooney i dont know if the vm/schdilign implciations fo that were considered
11:54:12 gibi I don't want to know what happens with the IP pool during such update
11:54:32 gibi and yes, you are right that can go against the currently scheduled host of the VMs
11:54:34 sean-k-mooney or if it went form not a routed netowk to a routed one
11:54:52 gibi yeah that too
11:55:24 sean-k-mooney i hope they only allow this before ports are added to the subnet
11:58:43 sean-k-mooney gibi: added +2 but now +w to see if bauzas wants to review. we can kick it into the gate if bauzas is busy and does not review by tomorrow
11:59:04 gibi sure. thanks
11:59:16 gibi btw the author is a first time nova contributor :)
12:01:20 sean-k-mooney i guess https://bugzilla.redhat.com/show_bug.cgi?id=1761903 woudl fail qe without thsi so once that lands we should backport it for them i guess
12:06:22 gibi yeah this fix needs to be backport
12:08:59 gibi up until wallaby upstream
12:20:51 opendevreview Wenping Song proposed openstack/os-traits master: Add nova and cyborg owner traits for resources https://review.opendev.org/c/openstack/os-traits/+/839128
12:57:19 gibi sean-k-mooney: the OWNER os-traits patch now looks good to me https://review.opendev.org/c/openstack/os-traits/+/839128 let's land it and propose a os-traits release
12:58:44 sean-k-mooney yep i see the added a release not. i guess that does not hurt but im not sure how many people actully consume that
12:58:52 sean-k-mooney ill take a look again shortly
12:59:09 gibi yeah reno does not hurt
12:59:09 sean-k-mooney i need to finsih reviewing your pci patches again too
12:59:20 gibi thanks for that!
12:59:59 sean-k-mooney the code looks fine on the first patch but i have not found a block of uninterupted time in the last few days to trace through the unit tests and line them up
13:00:23 sean-k-mooney im sure they are fine but that all i have left to do on the first patch
13:07:07 gibi no worries
13:26:36 opendevreview Merged openstack/os-traits master: Add nova and cyborg owner traits for resources https://review.opendev.org/c/openstack/os-traits/+/839128
13:28:33 bauzas gibi: sean-k-mooney: eventually accepted https://review.opendev.org/c/openstack/nova/+/839361
13:29:09 bauzas gibi: sean-k-mooney: that said, I'm sad we need to be an admin for knowing the API extensiion
13:31:43 sean-k-mooney well we can know if the extension is aviabel without being an admin but we cant consume the data form the extention without being an admin
13:32:02 sean-k-mooney i guess the assumetion here is that routed networks are provider networks and requies an admin to create
13:32:16 sean-k-mooney but you then dont need to be an admin to create ports on that network
13:32:26 sean-k-mooney so i really think it shoudl be readonly
13:32:34 sean-k-mooney for project_memeber
13:32:57 sean-k-mooney and project_admin to create the subnet and affinitese it to a segment
13:33:29 sean-k-mooney so we likely shoudl add neutron to the bug and ask them to change the default policy for get but not create to project_reader
14:19:39 bauzas sean-k-mooney: yeah, we should ask the neutron team to change the policy

Earlier   Later