| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2020-06-30 | |||
| 15:24:26 | stephenfin | dansmith: I tried to do this https://review.opendev.org/#/c/738583/ and it exploded. Guess we're relying on fields staying dirty or something :( | |
| 15:25:44 | stephenfin | so I've gone with the 'keypairs' field-only fix in https://review.opendev.org/#/c/683043/ though I suspect the issues I highlighted in ^ might still be true | |
| 15:26:47 | dansmith | stephenfin: one failure in unit tests that is asserting something equal? | |
| 15:27:26 | dansmith | meaning, something that compared two objects and failed because of the changed list, which is likely just someone encoding what caused the test to pass before | |
| 15:28:16 | dansmith | what explosion specifically? | |
| 15:34:17 | stephenfin | dansmith: No, the functional test | |
| 15:34:20 | stephenfin | *s | |
| 15:35:08 | dansmith | the numa one seems like it *has* to be a bug | |
| 15:35:14 | dansmith | I dunno about the port ones | |
| 15:35:56 | stephenfin | If I filter out the numa_topology and resources fields before doing the rest, the few failing tests I tried passed again | |
| 15:36:24 | stephenfin | but idk where to even start figuring out what's going on | |
| 15:42:11 | openstackgerrit | Stephen Finucane proposed openstack/nova stable/ussuri: libvirt: Don't allow "reserving" file-backed memory https://review.opendev.org/736219 | |
| 15:55:54 | mandre | hi, when setting glance annotation to enable a specific driver, for instance virtio-scsi, that restricts this image to a given hypervisor, correct? For example with virtio-scsi, this limits it to libvirt/kvm, right? BM nodes won't be able to boot from the same image? | |
| 17:38:46 | sean-k-mooney | well for the equal case we shoudl never compure object with out serialising them as the dirty state is actully checked | |
| 17:39:34 | sean-k-mooney | so just resetting all fields recurcivly woudl break any chang that was looking at raw equals comparison | |
| 17:40:18 | sean-k-mooney | its one of the reason we call to_primative and compure the dits normally then jsut doing ovo == other_ovo | |
| 17:40:47 | sean-k-mooney | or self.assertEquals(ovo, other_ovo) i guess | |
| 17:44:01 | sean-k-mooney | mandre: if the dirver does not support it it will ignore it | |
| 17:44:10 | sean-k-mooney | mandre: so it should work on an ironic node | |
| 17:44:25 | sean-k-mooney | i have that set on my images and they can be used with ironic | |
| 17:50:35 | mandre | great, thanks sean-k-mooney! | |
| 17:51:46 | sean-k-mooney | the only thing that would prevent you from sharing the image beyond driver in the image for the hardware would be if you enabled a filter or prefilter that was makeing some assertion regrating support for a feature in the schduler | |
| 17:51:59 | sean-k-mooney | non of the defualt ones shoudl be an issue | |
| 17:52:52 | sean-k-mooney | well things like have a numa toplogy in the image might cause the node to be rejected if you have the numa toplogy filter enabled because we dont report a numa toplogy for ironic nodes | |
| 17:53:24 | sean-k-mooney | but hw_disk_bus wont break anything unless you enable the prefilter that checks that | |
| 17:53:44 | mandre | ack, that shouldn't be an issue in my case I think but I'll keep that in mind | |
| 17:57:26 | sean-k-mooney | mandre: strangly enough i was just deploy ironic nodes with vm image and i just hit what i assume was a dirver issue | |
| 17:57:56 | sean-k-mooney | cetos8.1 booted fine form me by ubuntu 18.04 and 20.04 was not getting an ip correctly | |
| 17:58:17 | sean-k-mooney | when i used my vm image on my ironic hosts | |
| 17:59:30 | sean-k-mooney | all of those iamge have hw_disk_bus=scsi and hw_scsi_model=virtio-scsi and booted fine other then that | |
| 19:41:35 | ajitha | Hi all, We are running a third party cinder driver, we are testing multi attach functionality. and i am getting "Multiattach volumes are only supported starting with compute API version 2.60" error. | |
| 19:42:39 | ajitha | i found as using --os-compute-api-version is one option | |
| 19:43:13 | ajitha | how to set the compute-api-version permanently | |
| 20:41:16 | lseki | ajitha: https://docs.openstack.org/tempest/latest/sampleconf.html | |
| 20:41:26 | lseki | https://www.irccloud.com/pastebin/wA3xj98c/ | |
| 20:44:17 | lseki | I believe you'll need to set `min_microversion` and/or `max_microversion`, so that their range will contain a version >= 2.60 | |
| 20:52:23 | openstackgerrit | Gonéri Le Bouder proposed openstack/nova master: QEMU/KVM: accept vmxnet3 NIC https://review.opendev.org/738695 | |
| 20:52:54 | openstackgerrit | Gonéri Le Bouder proposed openstack/nova master: QEMU/KVM: accept vmxnet3 NIC https://review.opendev.org/738695 | |
| 21:05:30 | mriedem | stephenfin: fyi on docs things, https://docs.openstack.org/nova/latest/admin/networking-nova.html#metadata-service must have moved at some point but the cloud-init docs reference it https://cloudinit.readthedocs.io/en/latest/topics/datasources/openstack.html | |
| 21:06:06 | mriedem | guessing that is https://docs.openstack.org/nova/latest/admin/metadata-service.html now | |
| #openstack-nova - 2020-07-01 | |||
| 01:04:11 | openstackgerrit | Gonéri Le Bouder proposed openstack/nova master: QEMU/KVM: accept vmxnet3 NIC https://review.opendev.org/738695 | |
| 01:04:11 | openstackgerrit | Gonéri Le Bouder proposed openstack/nova master: QEMU/KVM: accept vmxnet3 NIC https://review.opendev.org/738695 | |
| 01:05:14 | openstackgerrit | Ghanshyam Mann proposed openstack/nova master: DNM: testing nova gate on ubuntu focal https://review.opendev.org/738126 | |
| 01:37:41 | openstackgerrit | Kevin Zhao proposed openstack/nova master: [WIP] CI: add tempest-integrated-compute-aarch64 job https://review.opendev.org/714439 | |
| 01:49:18 | openstackgerrit | Ghanshyam Mann proposed openstack/nova master: DNM: testing nova gate on ubuntu focal https://review.opendev.org/738126 | |
| 02:10:27 | openstackgerrit | Tony Su proposed openstack/nova master: Provider Config File: YAML file loading and schema validation https://review.opendev.org/673341 | |
| 02:10:28 | openstackgerrit | Tony Su proposed openstack/nova master: Provider Config File: Enable loading and merging of provider configs https://review.opendev.org/693460 | |
| 02:10:28 | openstackgerrit | Tony Su proposed openstack/nova master: Provider Config File: Functions to merge provider configs to provider tree https://review.opendev.org/676522 | |
| 02:10:29 | openstackgerrit | Tony Su proposed openstack/nova master: Provider Config File: Function to further validate and retrieve configs https://review.opendev.org/676029 | |
| 02:22:38 | openstackgerrit | Ghanshyam Mann proposed openstack/nova master: DNM: testing nova gate on ubuntu focal https://review.opendev.org/738126 | |
| 02:27:42 | openstackgerrit | Ghanshyam Mann proposed openstack/nova master: Fix user creation with GRANT in MySQL 8.0(Ubuntu Focal) https://review.opendev.org/738723 | |
| 02:28:06 | openstackgerrit | Ghanshyam Mann proposed openstack/nova master: DNM: testing nova gate on ubuntu focal https://review.opendev.org/738126 | |
| 02:30:14 | gmann | dansmith: melwitt gibi Ubuntu Focal fix for tool/test-setup.sh - https://review.opendev.org/#/c/738723/ Testing it in https://review.opendev.org/#/c/738126/ . | |
| 02:31:11 | gmann | Result should be ready by morning. I will be on holiday tomorrow, so pinged before gate result. | |
| 03:04:29 | openstackgerrit | Wenping Song proposed openstack/nova master: Add instance project_id for cyborg arq https://review.opendev.org/738428 | |
| 06:55:14 | openstackgerrit | Jiri Suchomel proposed openstack/nova master: Move image verification and writing out of download method https://review.opendev.org/738738 | |
| 07:43:55 | huaqiang | a | |
| 07:58:50 | openstackgerrit | Jiri Suchomel proposed openstack/nova master: Move image verification and writing out of download method https://review.opendev.org/738738 | |
| 10:57:46 | stephenfin | bauzas: Can you take a punt at https://review.opendev.org/#/c/733827/ ? | |
| 10:58:05 | bauzas | stephenfin: will look this afternoon | |
| 10:58:34 | stephenfin | Thanks | |
| 13:08:24 | lseki | Hello folks, could you give some suggestions on how to address this bug? | |
| 13:08:35 | openstack | Launchpad bug 1870367 in Cinder "Fail to extend attached volume using generic NFS driver" [High,In progress] - Assigned to Silvan Kaiser (2-silvan) | |
| 13:08:35 | lseki | https://bugs.launchpad.net/cinder/+bug/1870367 "Fail to extend attached volume using generic NFS driver" | |
| 13:08:45 | lseki | in summary, when trying to resize a volume in-use, `qemu-img resize` command fails to get the write lock, because `qemu` process has locked the volume file upon VM start. | |
| 13:08:51 | lseki | This was discussed on Cinder PTG, but we didn't come up with a solution yet | |
| 13:16:23 | lyarwood | lseki: you can use --force-share to workaround the locks but I'm not sure that it would be the best approach here tbh | |
| 13:17:04 | lyarwood | lseki: but then again we were essentially doing that prior to QEMU taking the locks | |
| 13:18:11 | lyarwood | lseki: but without quessing the disk on the compute side first that might be problematic | |
| 13:21:03 | gibi | lyarwood: welcome back! | |
| 13:21:15 | lyarwood | gibi: \o hey thanks, glad to be back finally :) | |
| 13:24:32 | gibi | :) | |
| 13:33:39 | sean-k-mooney | lyarwood: we do quiese the disk for vm snapshots right | |
| 13:34:02 | sean-k-mooney | if we are doing a live snapshot via the qemu-guest agent? | |
| 13:34:20 | sean-k-mooney | we use the monitor to invoke freeze | |
| 13:34:38 | sean-k-mooney | could we do the same for volume snapshots? | |
| 13:36:22 | lyarwood | sean-k-mooney: we do for snapshots, we could for extend as well I guess but I'm not entirely sure if we need to tbh | |
| 13:36:26 | lseki | lyarwood: in my machine (without openstack, just qemu-kvm), I managed to resize an attached disk with the command | |
| 13:36:31 | lseki | `sudo virsh qemu-monitor-command ubuntu18.04 --hmp "block_resize drive-virtio-disk1 2G"` | |
| 13:37:29 | sean-k-mooney | lseki: well that is having qemu do the resize | |
| 13:37:41 | sean-k-mooney | lseki: which s not normally how a volume resize would work | |
| 13:37:59 | sean-k-mooney | lseki: a volume resize for other driver would have the backend to the resize | |
| 13:38:13 | sean-k-mooney | lseki: i think that only works because qemu is the thing with the lock | |
| 13:38:37 | sean-k-mooney | we could special case but that would fail for iscsi backend and proably ceph | |
| 13:38:37 | lyarwood | well, c-vol resizes the underlying device, n-cpu then resizes the host device on the compute and then we resize the device within QEMU | |
| 13:39:03 | lseki | sean-k-mooney: hmm would it be possible to add a condition to do that only when the backend driver is qemu? | |
| 13:39:26 | lyarwood | the issue here is that we are asking qemu-img to resize the host device, in this case a file. | |
| 13:39:26 | sean-k-mooney | lyarwood: right but presumably the nfs driver cant resize the file because qemu has locked the file right | |
| 13:39:31 | lyarwood | yeah | |
| 13:40:04 | lyarwood | but I guess c-vol is actually calling this | |
| 13:40:06 | sean-k-mooney | lseki: you mean the backedn driver is nfs | |
| 13:40:26 | lseki | sean-k-mooney: oops, yes | |
| 13:40:33 | sean-k-mooney | lyarwood: yes becasue if it was not attached to a vm then it would have to do the resize | |
| 13:40:37 | lyarwood | I think we should be safe to add --force-share to the qemu-img resize call in c-vol tbh | |
| 13:41:09 | sean-k-mooney | lyarwood: will that also resize the filesystem | |
| 13:41:16 | lyarwood | sean-k-mooney: no | |
| 13:41:28 | sean-k-mooney | jsut the qcow/raw file | |
| 13:41:36 | lyarwood | right, virt-resize does both | |
| 13:41:40 | lyarwood | or it's up to the user | |
| 13:41:54 | lyarwood | once n-cpu has called QEMU to resize the device within the domain | |