| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2021-01-04 | |||
| 15:06:36 | sean-k-mooney | "One key problem that would need to be addressed by this selection is a scenario | |
| 15:06:38 | sean-k-mooney | where compute service goes down, while the VMs continue operating (and their | |
| 15:06:40 | sean-k-mooney | volumes remain attached) - we don't want to lose this agent in this case. | |
| 15:06:42 | sean-k-mooney | " | |
| 15:06:56 | sean-k-mooney | that is the main motivator for a seperate deamon process right | |
| 15:07:48 | sean-k-mooney | https://review.opendev.org/c/openstack/cinder-specs/+/766732/12/specs/wallaby/nvme-agent.rst#71 | |
| 15:10:49 | zoharm | right | |
| 15:11:18 | zoharm | open to all suggestions here | |
| 15:14:40 | zoharm | we want it to be a periodic task that is launched only when nvme volumes are connected by the nvme connector | |
| 15:28:35 | kashyap | stephenfin: One more, sorry: vTPM 2.0 and TPM passthrough -- both are supported in upstream Train, yeah? | |
| 15:29:27 | stephenfin | kashyap: no, vTPM is supported since Victoria. TPM passthrough is not supported afaik | |
| 15:29:59 | sean-k-mooney | TPM passtough can only be done via pci passtough | |
| 15:30:12 | sean-k-mooney | but that is for stateless devices only | |
| 15:30:23 | sean-k-mooney | so using it wiwth a tpm which has state would be invalid | |
| 15:30:36 | kashyap | stephenfin: sean-k-mooney: Nod; thx. Context on these random questions, I got a ping elsewhere about these, so trying to fill in details | |
| 15:31:08 | sean-k-mooney | if cyborg had a TPM dirver it could provide a tpm to us and do the required cleaning but they dont as far as i know | |
| 15:31:37 | kashyap | stephenfin: I have one more: I take it, your recent virtio device addition work means, we now support all these devices, corecct? — | |
| 15:32:03 | kashyap | virtio-serial, virtio-vga, virtio-balloon, virtio-input, virtio-keyboard, virtio-mouse, virtio-tablet | |
| 15:32:06 | kashyap | Perhaps not the -mouse yet | |
| 15:32:21 | sean-k-mooney | you conflating different things | |
| 15:32:35 | kashyap | sean-k-mooney: What am I conflating? | |
| 15:32:39 | sean-k-mooney | virtio-serial has been supproted for year with the serial-console | |
| 15:33:08 | sean-k-mooney | virtio-ballon cant be turned off yet so you always get it because libvirt adds one we agreeed to change that | |
| 15:33:09 | kashyap | sean-k-mooney: Yeah; I know some of them have existed forever; all I care about is - does the latest upstream release support all the above devices or not | |
| 15:33:35 | kashyap | Right. (I know -balloon is supported; and we recommend it to turn off for especially the real-time cases) | |
| 15:33:37 | sean-k-mooney | virti-input i think is not but im not sure | |
| 15:33:51 | sean-k-mooney | kashyap: yes but right now you cant turn it off | |
| 15:33:56 | kashyap | (And I recall reviewing the -keyboard and -input patches from Stephen, recently) | |
| 15:34:02 | sean-k-mooney | you uses to be able to by disbleing mermoy stat monitoring | |
| 15:34:11 | sean-k-mooney | but at some point libvirt started always adding it | |
| 15:34:38 | sean-k-mooney | kashyap: im just sure what -input is | |
| 15:35:00 | sean-k-mooney | -tablet is definetly supported | |
| 15:35:03 | kashyap | sean-k-mooney: You mean not sure? | |
| 15:35:09 | kashyap | https://www.kraxel.org/blog/2016/09/using-virtio-input-with-libvirt/ | |
| 15:35:09 | sean-k-mooney | sorry yes | |
| 15:35:52 | sean-k-mooney | ya ok so stephenfin was changing that but i havent revied it in a while | |
| 15:36:04 | sean-k-mooney | but he wrote a nice table in the commit message | |
| 15:36:54 | sean-k-mooney | https://review.opendev.org/c/openstack/nova/+/756552 | |
| 15:37:26 | sean-k-mooney | oh the table was in https://review.opendev.org/c/openstack/nova/+/756551/3 | |
| 15:40:37 | kashyap | sean-k-mooney: Right; inded, I was looking at that table a minute ago | |
| 15:41:04 | kashyap | Okay; thanks. I may come back with more "does upstream support X?" questions, if I can't figure it out myself :-) | |
| 17:06:41 | qqmber | Hi! How can I make a auth token to last longer? Because when I try to create a snapshot of a big instance takes around 120 minutes and fails because a 401 error, due a timeout ( auth expired). I tried to configure "expiration_time = 7200" in [cache] section, but no luck.... | |
| 17:06:46 | qqmber | in keystone.conf | |
| 17:08:37 | sean-k-mooney | you can use service tokens | |
| 17:08:45 | sean-k-mooney | but it should not take 120mins to spawn | |
| 17:09:15 | qqmber | hmmm... sorry, I don't get it.. what do you mean? | |
| 17:10:15 | qqmber | How can I use service tokens? Never did before.. | |
| 17:14:13 | sean-k-mooney | you have to configure them in the nova config | |
| 17:14:16 | sean-k-mooney | one sec | |
| 17:14:59 | sean-k-mooney | https://specs.openstack.org/openstack/nova-specs/specs/pike/implemented/use-service-tokens.html | |
| 17:27:54 | lyarwood | qqmber: https://docs.openstack.org/cinder/latest/configuration/block-storage/service-token.html might be useful assuming the instance is boot from volume? | |
| 17:28:15 | qqmber | sean-k-mooney: thanks | |
| 17:28:28 | qqmber | lyarwood: it does not boot from volume | |
| 17:38:03 | qqmber | sean-k-mooney: actually I'm in Pike.. and as says there.. is a change proposal... so.. I guess I can't use it in Pike :( | |
| 17:48:21 | sean-k-mooney | qqmber: it was added in pike | |
| 17:48:32 | sean-k-mooney | so that the first release you cna use it in | |
| 17:49:07 | qqmber | oh, cool. thx | |
| 17:55:45 | sean-k-mooney | qqmber: you should still look in to why it took 2 hours to boot | |
| 17:56:22 | qqmber | because it's a snapshot of 160GB and the host is pretty busy | |
| 17:56:24 | qqmber | that's all | |
| 17:56:28 | sean-k-mooney | unless you have reallly slow HDD and are copying TBs of root disk over 1G links it should not be that slow | |
| 17:56:50 | sean-k-mooney | i see | |
| 17:56:58 | sean-k-mooney | well if you know why i guess that is ok | |
| 17:57:05 | qqmber | sean-k-mooney: that's why I'm doing the snapshot, to take off those instances to a new cluster | |
| 17:57:07 | qqmber | a faster one | |
| #openstack-nova - 2021-01-05 | |||
| 01:35:56 | brinzhang0 | gmann: hi | |
| 01:36:40 | brinzhang0 | gmann: https://review.opendev.org/c/openstack/tempest/+/769120 this tempest depends-on the nova patch, but it seems didnot active, could you give some guide? | |
| 04:45:44 | openstackgerrit | zhulipeng proposed openstack/nova master: Fixed multipath residue when evacuating the VM https://review.opendev.org/c/openstack/nova/+/767468 | |
| 07:17:17 | openstackgerrit | Mamduh proposed openstack/os-vif stable/train: Refactor code of linux_net to more cleaner and increase performace https://review.opendev.org/c/openstack/os-vif/+/765912 | |
| 07:25:15 | openstackgerrit | Mamduh proposed openstack/os-vif stable/train: Fix - os-vif fails to get the correct UpLink Representor https://review.opendev.org/c/openstack/os-vif/+/765970 | |
| 07:36:54 | brinzhang0 | gibi: pls review cyborg shelve/unshelve patch while you are free https://review.opendev.org/c/openstack/nova/+/729563, thanks | |
| 07:42:43 | openstackgerrit | Mamduh proposed openstack/os-vif stable/stein: Refactor code of linux_net to more cleaner and increase performace https://review.opendev.org/c/openstack/os-vif/+/765914 | |
| 07:46:02 | openstackgerrit | Mamduh proposed openstack/os-vif stable/stein: Fix - os-vif fails to get the correct UpLink Representor https://review.opendev.org/c/openstack/os-vif/+/765974 | |
| 08:28:42 | gibi | brinzhang0: done, I have one more issue to fix then it is good to go from my perspective | |
| 08:29:24 | brinzhang0 | gibi: good news, thanks | |
| 08:32:37 | brinzhang0 | gibi: could you please review the PS21, comment by your suggestion https://review.opendev.org/c/openstack/nova/+/729563/21 | |
| 08:33:12 | brinzhang0 | or I missed your mean | |
| 08:38:38 | openstackgerrit | zhangtongjian proposed openstack/nova-specs master: remove unicode from code https://review.opendev.org/c/openstack/nova-specs/+/769260 | |
| 08:41:37 | bauzas | good morning Nova | |
| 08:55:22 | openstackgerrit | Mamduh proposed openstack/os-vif stable/rocky: Refactor code of linux_net to more cleaner and increase performace https://review.opendev.org/c/openstack/os-vif/+/765962 | |
| 08:55:23 | openstackgerrit | Mamduh proposed openstack/os-vif stable/rocky: Fix - os-vif fails to get the correct UpLink Representor https://review.opendev.org/c/openstack/os-vif/+/765977 | |
| 09:01:30 | openstackgerrit | Brin Zhang proposed openstack/nova master: Cyborg shelve/unshelve support https://review.opendev.org/c/openstack/nova/+/729563 | |
| 09:01:34 | brinzhang0 | gibi: done | |
| 09:58:38 | openstackgerrit | Mamduh proposed openstack/os-vif stable/rocky: Refactor code of linux_net to more cleaner and increase performace https://review.opendev.org/c/openstack/os-vif/+/765962 | |
| 09:58:39 | openstackgerrit | Mamduh proposed openstack/os-vif stable/rocky: Fix - os-vif fails to get the correct UpLink Representor https://review.opendev.org/c/openstack/os-vif/+/765977 | |
| 10:09:55 | stephenfin | gibi: I'm looking at a bug report where cold migration fails for an SR-IOV instance with "Port update failed for port XXX: Unable to correlate PCI slot 0000:xx:yy.z". Does that ring a bell? | |
| 10:10:34 | stephenfin | It sounds familiar, but I can only find bugs related to unshelve, not cold migration | |
| 10:12:20 | stephenfin | Ah, got it https://bugs.launchpad.net/nova/+bug/1512880 | |
| 10:12:21 | openstack | Launchpad bug 1512880 in OpenStack Compute (nova) newton "Failed cold migration with SR-IOV" [Medium,Fix released] | |
| 10:55:56 | gibi | brinzhang0, stephenfin: sorry, I was away | |
| 11:45:34 | gibi | brinzhang0: one more thing https://review.opendev.org/c/openstack/nova/+/729563 | |
| 11:48:27 | brinzhang0 | gibi: ok, can be remove the excetpion block | |
| 11:48:33 | gibi | brinzhang0: yepp | |
| 11:48:37 | gibi | then I will be +2 | |
| 11:48:45 | brinzhang0 | cool | |
| 11:48:51 | openstackgerrit | Brin Zhang proposed openstack/nova master: WIP: Replace tenant* with project* in codes https://review.opendev.org/c/openstack/nova/+/769329 | |
| 11:49:27 | gibi | brinzhang0: I also started reviewing the tenant_id replacement series and left some comments. I haven't reached the top of that series but I will continue today | |
| 11:50:36 | brinzhang0 | gibi: thank you very much, I will update it after you completed review | |
| 11:50:45 | gibi | brinzhang0: OK, cool | |
| 11:51:29 | brinzhang0 | and there are some novaclient patch and sdk patach need to be submit, but it not urgent | |