| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2020-09-02 | |||
| 08:24:23 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Only unplug vif after the device is detached from libvirt https://review.opendev.org/744934 | |
| 08:24:23 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Make PCI claim NUMA aware during live migration https://review.opendev.org/748453 | |
| 08:24:24 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Use UUID as vif and network_id in vif tests https://review.opendev.org/748722 | |
| 08:28:07 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: [WIP] Support SRIOV interface attach and detach https://review.opendev.org/740995 | |
| 08:30:41 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: DNM: tracing the PCI allocation code path https://review.opendev.org/748454 | |
| 08:31:25 | stephenfin | jangutter: care to look at https://review.opendev.org/#/c/744816/ ? | |
| 08:38:00 | openstackgerrit | Luyao Zhong proposed openstack/nova master: Track error migrations in resource tracker https://review.opendev.org/745281 | |
| 08:38:00 | openstackgerrit | Luyao Zhong proposed openstack/nova master: Rollback when reset running live migration https://review.opendev.org/747899 | |
| 08:38:01 | openstackgerrit | Luyao Zhong proposed openstack/nova master: Improve orphans tracking in resource tracker https://review.opendev.org/745282 | |
| 09:01:49 | jangutter | stephenfin: ack, looking! | |
| 09:03:57 | jsuchome | gibi: lyarwood artom: hi, please take a look at https://review.opendev.org/#/c/574301/ (patch updated with stephenfin idea)+ https://review.opendev.org/#/c/749437/ (release notes) | |
| 09:04:56 | gibi | jsuchome: looking... | |
| 09:09:40 | gibi | jsuchome: thanks. I'm +2 | |
| 09:10:47 | jangutter | stephenfin, sean-k-mooney: I see https://review.opendev.org/#/c/744816/ is already in the gate. +2 from me, with a potential followup to clean up one of the nits (if it's actually a nit). | |
| 09:13:10 | stephenfin | jangutter: Yup, makes sense to me | |
| 09:17:35 | lyarwood | gibi: thanks! | |
| 09:18:22 | lyarwood | jsuchome: https://review.opendev.org/#/c/743220/ - could you update that to use the new configurables? | |
| 09:18:47 | lyarwood | jsuchome: I'm going to be pretty busy downstream this week so I'll not be able to get to it for a while | |
| 09:19:06 | lyarwood | https://review.opendev.org/#/c/748212/ is the actual job sorry | |
| 09:20:03 | jsuchome | lyarwood: yep; later today I hope | |
| 09:20:53 | lyarwood | jsuchome: cool thanks | |
| 09:25:34 | suryasingh | sean-k-mooney: melwitt dansmith lyarwood Sorry to interrupt you all. I have simple query. Does the current nova-compute service(Data-plane) support graceful shutdown ? Will the nova-compute service stop(systemctl stop openstack-nova-compute.service) wait for the ongoing operation(instance provisioning) to finish ? | |
| 09:34:36 | jkulik | suryasingh, there's https://review.opendev.org/#/c/666245/ for spawning VMs at least | |
| 09:35:59 | openstackgerrit | Brin Zhang proposed openstack/nova-specs master: Unified virtual persistent memory restrictions https://review.opendev.org/749472 | |
| 09:38:51 | brinzhang0 | alex_xu: if you are free, please review https://review.opendev.org/#/c/749472/, modify the depency of vPMEM limit in the spec, make it keep consistent with admin guide docs | |
| 09:39:25 | openstackgerrit | Lee Yarwood proposed openstack/nova stable/queens: Add regression tests for bug #1889108 https://review.opendev.org/749232 | |
| 09:39:25 | openstackgerrit | Lee Yarwood proposed openstack/nova stable/queens: compute: Don't delete the original attachment during pre LM rollback https://review.opendev.org/749233 | |
| 09:39:26 | openstack | bug 1889108 in OpenStack Compute (nova) train "failures during driver.pre_live_migration remove source attachments during rollback" [Undecided,In progress] https://launchpad.net/bugs/1889108 - Assigned to Lee Yarwood (lyarwood) | |
| 09:40:30 | brinzhang0 | alex_xu: please check the dependency package version, is it reasonable? | |
| 09:46:08 | lyarwood | bauzas / elod: would you mind taking a look at https://review.opendev.org/#/c/748496/ | |
| 09:47:23 | suryasingh | @jkulik thanks for response and patch link. Did you tested this patch ? And is there any priority from nova-community side for this patch, if you know about it, sorry for asking. | |
| 09:56:28 | jkulik | suryasingh, we're running some version of this patch in production on a queens code-base. I haven't seen any push from the nova-community for this. | |
| 09:57:37 | bauzas | lyarwood: ack | |
| 09:58:02 | suryasingh | @jkulik I see, thanks for info. | |
| 10:05:45 | openstackgerrit | Merged openstack/os-vif master: deprecate ovs-vsctl driver and make native the default https://review.opendev.org/744816 | |
| 10:27:30 | frickler | what do I need to do in order to be able to extend an attached RBD volume? https://review.opendev.org/613039 makes me think that this should work since stein, but if I try via cinder, it wants to see an unattached volume. is there some special nova API for this? | |
| 10:30:38 | lyarwood | frickler: isn't that a microversion thing on the c-api side? | |
| 10:30:40 | lyarwood | frickler: https://docs.openstack.org/api-ref/block-storage/v3/index.html?expanded=extend-a-volume-size-detail#volume-actions-volumes-action | |
| 10:30:48 | lyarwood | frickler: Prior to microversion 3.42 the volume status must be available. Starting with microversion 3.42, attached volumes with status in-use may be able to be extended depending on policy and backend volume and compute driver constraints in the cloud. Note that reserved is not a valid state for extend. | |
| 10:31:10 | frickler | lyarwood: ha, I'll try that, thx | |
| 10:34:06 | elod | lyarwood: sure, I'll review those patches today | |
| 10:37:04 | frickler | lyarwood: works like a charm, though seemingly only with the cinder CLI, not OSC, but that's a different topic. thx again | |
| 10:37:26 | lyarwood | frickler: cool np :) | |
| 10:37:32 | lyarwood | elod: awesome thanks! | |
| 10:57:03 | openstackgerrit | Merged openstack/nova master: Add ability to download Glance images into the libvirt image cache via RBD https://review.opendev.org/574301 | |
| 11:02:18 | stephenfin | gibi: So, interesting problem. I'm trying to remove those XenAPI-only APIs and need to optionally disable their tests in Tempest on Victoria+. The pattern we have for that is to check if our max API version > the max of the past release, but Victoria's max is the same as Ussuri (we didn't add new microversions this cycle) | |
| 11:02:38 | stephenfin | So i wonder if there are any last minute API changes we want to make/ :-D | |
| 11:02:39 | stephenfin | ? | |
| 11:09:47 | alex_xu | brinzhang: that is better to have luyao to take a look at, but I think that is the version luyao tested | |
| 11:40:55 | gibi | stephenfin: hm, that is an interesting catch | |
| 11:41:14 | gibi | I'm not aware of inflight microversion bumps | |
| 11:42:48 | gibi | yeah, looking through our bp list I don't think we have a version bump https://launchpad.net/nova/+milestone/victoria-3 | |
| 11:45:18 | sean-k-mooney | gibi: i think that might be a first for m3 since they were added :) | |
| 11:46:06 | sean-k-mooney | stephenfin: you could just add a microverion for this | |
| 11:46:14 | gibi | I guess it is a definit signal that nova development is slowing down | |
| 11:46:42 | sean-k-mooney | to singal the xenapi apis being removed | |
| 11:47:10 | sean-k-mooney | then use that for tempest | |
| 11:50:12 | sean-k-mooney | gibi: by the way i didnt find any thing broken in your code although there is that latent bug for keeping the vfs in teh db in sync with reality after a vf is detached | |
| 11:50:27 | sean-k-mooney | was there anything in particalar you wanted me to test that i did not | |
| 11:50:46 | gibi | sean-k-mooney: thanks for the tests. I think you covered everything. | |
| 11:51:36 | gibi | sean-k-mooney: regarding the latent bug. when a pf is attached to a guest then the host deletes the vfs from the pf | |
| 11:51:46 | gibi | and then when the pf is freed from the guest | |
| 11:51:52 | gibi | the vfs are not recreated by the host | |
| 11:52:06 | sean-k-mooney | kind of | |
| 11:52:19 | gibi | I did added a bit of configuration to re-create the vfs | |
| 11:52:25 | gibi | I mean host config | |
| 11:52:35 | sean-k-mooney | that is the effect but its not somuch the host deleting the vf as we are unbining the pf from the kernel and the vf are destoyed when that happens | |
| 11:53:06 | sean-k-mooney | basically i allocated the VFs using /sys and echo at run time | |
| 11:53:09 | gibi | how the vfs was created in the first place? I guess by writing under /sys | |
| 11:53:13 | gibi | yeah | |
| 11:53:19 | gibi | so that steps is not persistent | |
| 11:53:26 | sean-k-mooney | but that configurtion info is lost when you allocate teh pf to the vm | |
| 11:53:30 | gibi | yepp | |
| 11:53:43 | sean-k-mooney | then wehn qemu is done it does a functionl level pci reset | |
| 11:53:44 | gibi | I think this is a host configuration task to keep the number of vfs perisitent | |
| 11:53:52 | gibi | persistent | |
| 11:53:58 | sean-k-mooney | it is | |
| 11:54:07 | gibi | so I put | |
| 11:54:07 | gibi | pre-up echo '4' > /sys/class/net/enp129s0f0/device/sriov_numvfs | |
| 11:54:08 | sean-k-mooney | but we could also protect against it eaisly in nova | |
| 11:54:23 | gibi | hm, how? | |
| 11:54:39 | gibi | does nova know what is the number of vfs to be re-created? | |
| 11:54:54 | sean-k-mooney | gibi: when we set teh vfs from unavaiable to avaiable after freeing the PF in the db we chould check with libvirt to make sure the exist | |
| 11:55:02 | sean-k-mooney | if they dont we can delete them in the db | |
| 11:55:24 | sean-k-mooney | or i guess we could count them and recreate them with that echo but deleteing seams simpler | |
| 11:55:43 | gibi | ohh, yes, we could sync our db with reality | |
| 11:55:48 | sean-k-mooney | yep | |
| 11:55:55 | gibi | I don't want to recreate anything host level | |
| 11:56:04 | sean-k-mooney | so syncing with reality shoudl be simple | |
| 11:56:08 | gibi | I see | |
| 11:56:40 | sean-k-mooney | the other thning is that unless your udev rules are correct when you unbind the pf form the vm | |
| 11:56:52 | sean-k-mooney | the pf might not bind back to the netdev vendor driver | |
| 11:57:03 | sean-k-mooney | it can stay bound to vfio-pci | |
| 11:57:22 | songwenping_ | bauzas: hi, i'm test vGPU with VM and meet some problems. I have install nvidia-smi driver and my GPU's already set nvidia driver, but i cannot find mdev_bus at /sys/class path. | |
| 11:57:22 | sean-k-mooney | in that case the vf wont be recreated by a pre-up command | |
| 11:57:35 | sean-k-mooney | but that is definetly out of nova scope | |
| 11:57:59 | sean-k-mooney | songwenping_: what kernel are you using | |
| 11:58:10 | gibi | sean-k-mooney: ack | |
| 11:58:20 | sean-k-mooney | songwenping_: also i dont think that is where it is | |
| 11:58:32 | bauzas | songwenping_: have you installed the nvidia driver in the compute service ? | |