Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2018-03-27
15:20:49 jaypipes bauzas: right.
15:20:58 bauzas so each PGPU will be a leaf
15:21:11 sean-k-mooney[m] stephenfin: this wont work in general however as this would only work with vhost-user. the approch you need to enable need to work for any switch backend
15:21:22 bauzas with one inventory about the total number of vGPUs it can create *for the type defined by the opt*
15:21:25 jaypipes bauzas: each pGPU group, but yes.
15:21:33 bauzas plus the trait telling which type it is
15:21:42 bauzas jaypipes: libvirt doesn't have the notion of groups
15:21:43 jaypipes bauzas: right.
15:21:49 jaypipes bauzas: I know, but xen does.
15:22:21 bauzas sure, but from a placement perspective, a "PGPU" RP is, from a libvirt perspective, a PCI device, and from a xen perspective, a PGPU group
15:22:25 stephenfin sean-k-mooney[m]: Sure, but vhost-user would be a start. Once we have a way to expose this information, we can extend other backends
15:22:30 bauzas but both are reconciled
15:22:37 jaypipes ack
15:22:41 bauzas it's just a driver-only thing
15:23:00 bauzas okay, now question
15:23:07 bauzas does that need to be spec'd up ?
15:23:11 bauzas jaypipes:^
15:23:57 sean-k-mooney[m] stephenfin: if we wanted to do anything regardign the interface rx queue it likely should be an os-vif thing where we calulate teh best pmd our selves and set that not the other way around
15:24:13 jaypipes bauzas: yes. new CONF option, new way of behaving for the virt drivers. I would say yes.
15:24:35 openstackgerrit Merged openstack/nova master: tox: Remove unnecessary configuration https://review.openstack.org/556544
15:26:09 sean-k-mooney[m] stephenfin: most backend wont that this info so vhost-user is not something we should build on
15:26:36 stephenfin sean-k-mooney[m]: What other backends would there be?
15:26:54 openstackgerrit Kashyap Chamarthy proposed openstack/nova master: libvirt: Allow to specify granular CPU feature flags https://review.openstack.org/534384
15:26:57 sean-k-mooney[m] stephenfin: for example kernel ovs with kernel vhost has kerne vhost-treads it spwans per interface, it dose not have pmds, so while we can taskset those threads we cant affinites them via looking a queues
15:27:18 bauzas jaypipes: fine by me
15:27:51 bauzas artom: had some concerns about the virt driver "calling" the conductor in https://review.openstack.org/#/c/552722/6
15:28:24 artom bauzas, replied on the patch, but we can continue here if you want. Basically, no new calls are added, it's just existing methods/returns with new data in them
15:28:30 bauzas artom: maybe I misunderstood but you mean that the compute service would call back the conductor ?
15:28:30 artom Should I clarify the spec?
15:28:31 alex_xu_ kashyap: np
15:28:45 bauzas artom: that would help
15:29:03 edmondsw mriedem I don't remember ever discussing a service-protected-server case in our RBAC discussions, but I don't think you need to use service tokens there... just have a role that's given permission to manage these
15:29:12 sean-k-mooney[m] stephenfin: in any case queue mapping or tasksetting of vhost threads i think is out of scope of nova
15:29:26 edmondsw that spec needs a lot of work if anything's to happen there
15:30:10 bauzas artom: also, on the fact we would check the compute version
15:30:32 bauzas artom: does a live migration work with two different compute versions already ?
15:30:34 openstackgerrit Merged openstack/nova master: Move placement test cases from db to placement https://review.openstack.org/553149
15:30:40 mriedem bauzas: yes
15:30:48 mriedem we support mixed version computes for live migration
15:30:49 artom bauzas, I didn't see anything in the code that would suggest it wouldn't
15:30:58 bauzas mriedem: context is https://review.openstack.org/#/c/552722/6/specs/rocky/approved/numa-aware-live-migration.rst@244
15:30:59 sean-k-mooney[m] linux bridge, sriov, hardware offloaded ovs, vpp, iovision(ebpf), mini net, calico and macvtap are the main ones used beyond ovs/ovs-dpdk
15:30:59 mriedem we have a grenade + live migratoin job that also does this back and forth
15:31:25 bauzas mriedem: then, I was claiming we could microversion the behavioural change artom is going to introduce
15:31:26 sean-k-mooney[m] oh there is also snabb switch
15:31:27 mriedem that's a bit drastic
15:31:32 artom bauzas, also, we'd only fail it if the instance has "NUMA" characteristics
15:31:36 mriedem sec
15:31:50 artom Which is currently broken anyways, so not as drastic as it sounds
15:31:55 mriedem see https://review.openstack.org/#/c/522537/13/nova/conductor/tasks/live_migrate.py
15:31:55 bauzas artom: I agree, and we don't test that AFAIK
15:32:06 mriedem you only need to know if the source and dest computes can do the new hotness
15:32:35 mriedem otherwise it's a novalidhost if you can't find a pair
15:32:42 bauzas mriedem: yeah I remember that change
15:33:29 sean-k-mooney[m] stephenfin: on and NTT have a dpdk soft patch panel thing , the point is nova cant have special case code for all of these
15:35:21 bauzas mriedem: if the implementation of artom's spec would go into that direction, then yes we wouldn't need a microversion
15:35:35 mriedem i think we can agree that we don't want to add subnet to the requested network turducken in the compute api https://review.openstack.org/#/c/518227/
15:37:11 bauzas mriedem: oh please, yes
15:37:33 bauzas just create the port in neutron and pass it to nova
15:37:50 bauzas aha, jinxed by mriedem's comment
15:37:58 artom mriedem, that link you posted in my NUMA migration spec, that's basically just checking min compute version, right?
15:38:04 artom So the same thing I was suggesting?
15:38:35 artom Just for clarity in my head, because I feel like we're talking about the same thing as though they were different
15:39:00 gibi jaypipes: I'm open to debate to simulate or not the neutron backend ordering preference via a scheduler weigher but for that debate we need some neutron heavy guys, like mlavalle
15:39:45 openstackgerrit Jay Pipes proposed openstack/nova-specs master: Handle nested providers for allocation candidates https://review.openstack.org/556873
15:39:47 bauzas artom: yeah basically
15:39:55 bauzas artom: it's for a different usage tho
15:39:56 jaypipes mriedem, bauzas: ^^ added not about microversion.
15:40:05 artom bauzas, right, but the mechanism is the same
15:40:08 stephenfin sean-k-mooney[m]: Well damn. That's unfortunate
15:40:08 kashyap Darn, lost edits to wiki.openstack.org, as it logged me out mid-way. And didn't give me a way to get the changes back.
15:40:27 bauzas artom: yup, the pattern should be the same
15:40:29 artom (We might eventually standardise that in a utils somewhere, since we seem to be doing a lot of it, btw)
15:40:39 mriedem artom: kind of, but it depends on where you're using it,
15:41:01 mriedem if it's the API and you're checking min compute service version across the entire cell, that's different than just comparing the source and candidate dest host
15:41:16 bauzas at least, we can assume we run a pre-flight check on the conductor that checks both compute versions
15:41:27 bauzas that could be generalized
15:41:38 artom mriedem, ah, so you're saying we should be more granular and just check the (source, dest) pair
15:41:41 artom Hrmm
15:41:46 artom Could we do that in the scheduler?
15:41:48 mriedem artom: yes, see that patch i linked
15:41:51 bauzas but that's an implementation detail, IHMO
15:41:58 artom So that straight away we have a dest that supports it?
15:42:07 mriedem artom: we could if we exposed the capability as a trait on the compute and added a pre-request filter
15:42:19 bauzas artom: the scheduler doesn't know whether it's a migration or a boot
15:42:26 bauzas but what mriedem wrote
15:42:34 mriedem artom: that would be an optimization imo
15:42:40 mriedem worth noting in the spec though
15:42:41 bauzas zactly
15:42:45 openstackgerrit Chris Dent proposed openstack/nova master: [placement] Filter allocation candidates by forbidden traits in db https://review.openstack.org/556660
15:42:45 openstackgerrit Chris Dent proposed openstack/nova master: [placement] Filter resource providers by forbidden traits in db https://review.openstack.org/556472
15:42:46 openstackgerrit Chris Dent proposed openstack/nova master: [placement] Support forbidden traits in API https://review.openstack.org/556820
15:42:46 openstackgerrit Chris Dent proposed openstack/nova master: [placement] Parse forbidden traits in query strings https://review.openstack.org/556819
15:42:54 artom Mmmm, forbidden traits
15:42:58 artom /homer
15:42:59 gibi mriedem: I will be available during the notification subteam meeting timeslot or before that but I have nothing to really talk about so we can simply skip the meeting if you agree
15:43:11 SamYaple so i just ran into an annoying time consuming issue. i was going an upgrade and my nova-osapi_compute service changed to reporting its hostname to the database. this created new nova-osapi_compute services in the services table. the old service entires had version 9 and this was causing all GETs to instances to fail with instance not found
15:43:20 SamYaple code in question https://github.com/openstack/nova/blob/ed55dcad83d5db2fa7e43fc3d5465df1550b554c/nova/compute/api.py#L2268
15:43:31 SamYaple no logs anywhere describing the issue :/
15:43:52 mriedem gibi: yes let's skip

Earlier   Later