| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2020-02-07 | |||
| 00:11:16 | mriedem | dansmith: a few comments in your reno | |
| 00:36:17 | openstackgerrit | Sundar Nadathur proposed openstack/nova master: Add Cyborg device profile groups to request spec. https://review.opendev.org/631243 | |
| 00:36:18 | openstackgerrit | Sundar Nadathur proposed openstack/nova master: Define Cyborg ARQ binding notification event. https://review.opendev.org/692707 | |
| 00:36:18 | openstackgerrit | Sundar Nadathur proposed openstack/nova master: Create and bind Cyborg ARQs. https://review.opendev.org/631244 | |
| 00:36:19 | openstackgerrit | Sundar Nadathur proposed openstack/nova master: Pass accelerator requests to each virt driver from compute manager. https://review.opendev.org/698581 | |
| 00:36:19 | openstackgerrit | Sundar Nadathur proposed openstack/nova master: Compose accelerator PCI devices into domain XML in libvirt driver. https://review.opendev.org/631245 | |
| 00:36:20 | openstackgerrit | Sundar Nadathur proposed openstack/nova master: Delete ARQs for an instance when the instance is deleted. https://review.opendev.org/673735 | |
| 00:36:20 | openstackgerrit | Sundar Nadathur proposed openstack/nova master: Enable hard/soft reboot with accelerators. https://review.opendev.org/697940 | |
| 00:36:21 | openstackgerrit | Sundar Nadathur proposed openstack/nova master: Enable start/stop of instances with accelerators. https://review.opendev.org/699553 | |
| 00:36:21 | openstackgerrit | Sundar Nadathur proposed openstack/nova master: Enable and use COMPUTE_ACCELERATORS trait. https://review.opendev.org/699554 | |
| 00:36:22 | openstackgerrit | Sundar Nadathur proposed openstack/nova master: Bump compute rpcapi version and reduce Cyborg calls. https://review.opendev.org/704227 | |
| 00:36:22 | openstackgerrit | Sundar Nadathur proposed openstack/nova master: Add cyborg tempest job. https://review.opendev.org/670999 | |
| 08:29:01 | gibi | efried: ack, I will try to spend some time on the cyborg series today | |
| 10:22:01 | stephenfin | gibi, lyarwood: real quick, do either of you see anything obviously wrong in https://zuul.opendev.org/t/openstack/build/5b196c5a7cf944df8857a08dc15aa79f/console ? | |
| 10:22:15 | stephenfin | It's saying "TypeError: create_port_binding() missing 2 required positional arguments: 'port_id' and 'data'" | |
| 10:22:31 | stephenfin | But line it's failing on is obviously passing those? | |
| 10:22:32 | stephenfin | "binding = client.create_port_binding(port_id, data)['binding']" | |
| 10:23:29 | stephenfin | gibi: Also, low priority but I updated a commit message on a patch you'd previously approved, if you have time to revisit https://review.opendev.org/705655 | |
| 10:24:00 | gibi | stephenfin: could you link the patch that produced such test result? | |
| 10:24:09 | stephenfin | gibi: https://review.opendev.org/#/c/706295/ | |
| 10:25:30 | gibi | interesting :) | |
| 10:25:38 | stephenfin | right? | |
| 10:25:40 | gibi | pulling it down | |
| 10:29:02 | gibi | trying to reproduce locally.. | |
| 10:34:06 | gibi | stephenfin: nova.tests.fixtures.NeutronFixture.get_port_binding has get_port_binding(self, context, client, port_id, host) signature so when the code pass two positional args it fills only client and context and not the port and host | |
| 10:34:18 | stephenfin | gdi | |
| 10:34:40 | stephenfin | gibi++ Thanks /o\ :) | |
| 10:34:44 | gibi | no problem | |
| 10:34:48 | stephenfin | I'd been staring at that for an hour | |
| 10:34:50 | gibi | the error message was deeply missleading | |
| 10:35:01 | stephenfin | Yuuup. Bad coincidence :) | |
| 10:35:27 | gibi | yepp | |
| 10:50:40 | openstackgerrit | Stephen Finucane proposed openstack/nova master: Remove universal wheel configuration https://review.opendev.org/706466 | |
| 11:02:26 | openstackgerrit | Brin Zhang proposed openstack/nova master: Add new default roles in os-instance-actions policies https://review.opendev.org/706470 | |
| 11:05:12 | lyarwood | stephenfin: sorry missed that, still need me to take a look? | |
| 11:05:25 | stephenfin | lyarwood: Nope, we sussed it | |
| 11:05:45 | lyarwood | wonderful | |
| 11:06:47 | openstackgerrit | Stephen Finucane proposed openstack/nova master: WIP: Use neutronclient's port binding APIs https://review.opendev.org/706295 | |
| 14:19:04 | openstackgerrit | Huachang Wang proposed openstack/nova-specs master: Use PCPU and VCPU in one instance https://review.opendev.org/668656 | |
| 14:30:35 | huaqiang | stephenfin:sean-k-mooney:alex_xu spec https://review.opendev.org/668656 is updated, hope you guys have time to review it before freezing day | |
| 14:49:09 | stephenfin | gibi: Comments left on https://review.opendev.org/#/c/704759/ | |
| 14:49:23 | stephenfin | huaqiang: Ack. I'll hit that next Tuesday (spec review day) | |
| 14:55:07 | huaqiang | stephenfin: appreciate! | |
| 14:56:24 | gibi | stephenfin: thanks. I will check | |
| 15:01:58 | openstackgerrit | Brin Zhang proposed openstack/nova master: Add new default roles in os-instance-actions policies https://review.opendev.org/706470 | |
| 15:05:53 | openstackgerrit | Martin Midolesov proposed openstack/nova master: Implementing graceful shutdown. https://review.opendev.org/666245 | |
| 15:22:18 | openstackgerrit | Brin Zhang proposed openstack/nova master: Add SYSTEM_READER role to servers actions API https://review.opendev.org/706179 | |
| 15:26:33 | kashyap | efried: Want to have a stab at this spec? Already approved previously; and has a +2: https://review.opendev.org/#/c/693844/ | |
| 15:29:18 | Sundar | dansmith, gibi, sean-k-mooney, efried: About https://review.opendev.org/#/c/706083/, this is a refactor that affects pci requests, numa topology etc. that are not related to the Cyborg patch. I see that some of you are noncommittal and, in any case, probably needs further discussion. The Cyborg series is already long in the tooth; if we add this | |
| 15:29:19 | Sundar | general refactor, keeping that entire context in our minds will only prolong our efforts and strain. Plus, some of us are hitting time constraints. What do you think of considering the series as it is now, and feel free to add refactors on top of that later? Both of them could even go concurrently, but the discussions on the refactor need not hold | |
| 15:29:19 | Sundar | up the main series. | |
| 15:31:52 | Sundar | Just to be clear, I am not objecting to the refactoring. I am fine either way and can contribute to the reviews. | |
| 15:33:25 | sean-k-mooney | Sundar: im currently planing to redeploy with the code you pushed yesterday. | |
| 15:33:37 | Sundar | sean-k-mooney: Thanks. | |
| 15:33:45 | sean-k-mooney | i found that with the previous version i was unable to delete the cyborg vm | |
| 15:34:01 | sean-k-mooney | the conductor was not able to delete the arp binding | |
| 15:34:33 | sean-k-mooney | i was able to manually do it but could still not delete the nova vm | |
| 15:34:46 | sean-k-mooney | ill see if i hit the same issue with the new version and let you know | |
| 15:35:41 | Sundar | I see. In the worst case, for development purposes only, once could delete the ARQs using curl. | |
| 15:36:00 | sean-k-mooney | yes but you still cannot delte the nova vm | |
| 15:36:01 | Sundar | Sure. You are also updating the Cyborg side, right? | |
| 15:36:07 | sean-k-mooney | yep | |
| 15:36:31 | sean-k-mooney | i delete all my repos and confirm it pulle the correct version if i also set CYBORG_REPO and CYBORG_BRANCH | |
| 15:37:04 | Sundar | Could you expand on what you mean by Nova VM? Different from a devstack all-in-1 setup? | |
| 15:37:43 | sean-k-mooney | i mean if i do openstack server delete <uuid of vm with fake cyborg device> then you cant delete it | |
| 15:37:58 | gibi | Sundar: I agree that such refactor could move the focus away from the goal to merge the cyborg integration code to nova. And I understand the time pressure here. I'm fine doing the refactor on top of the series | |
| 15:38:23 | sean-k-mooney | the conductor got an unexpected reponce code form cyborg and the delete failed | |
| 15:38:56 | gibi | Sundar: I'm reviewing the series as we speak | |
| 15:39:56 | sean-k-mooney | Sundar: this could be related to the error handeling you and dansmith were talking about. i did not dig into it as it was late when i tried to delete the vm | |
| 15:40:20 | sean-k-mooney | Sundar: im going to recreate teh env with the latest version and see if the issue is still there | |
| 15:54:36 | Sundar | gibi: Thanks | |
| 15:55:18 | Sundar | sean-k-mooney: Please let me know how it goes. I may get on phone calls once in a while but I'll be around all morning PST. | |
| 15:58:39 | efried | Sundar: The refactor is already "on top" of the series, not folded in the middle. It shouldn't block anything else. | |
| 15:58:54 | efried | (It's on top of a patch that's in the middle, but it's its own tip) | |
| 15:59:22 | efried | I just put it out there to get agreement with dansmith about what the refactor should look like, and to satisfy myself that it could be done without too much trouble, so that I could be okay with the preceding patch as it is. | |
| 15:59:37 | efried | kashyap: ack, in queue | |
| 16:00:09 | kashyap | Thank you. | |
| 16:03:44 | sean-k-mooney | efried: which patch is the refactor https://review.opendev.org/#/c/706083/1 you dont mean https://review.opendev.org/#/c/704227/11 right | |
| 16:03:59 | sean-k-mooney | im planning to deploy https://review.opendev.org/#/c/704227/11 | |
| 16:04:00 | efried | the former | |
| 16:04:05 | sean-k-mooney | cool | |
| 16:05:52 | mriedem | dansmith: +2 on your hidden=null fix in case efried or melwitt want to approve | |
| 16:07:52 | spatel | sean-k-mooney: hey! i have few more question related NUMA, I did lots of testing with erlang and have pretty good result with numa tuning | |
| 16:08:09 | sean-k-mooney | thats good to hear | |
| 16:08:27 | sean-k-mooney | so memory seams to be your bottleneck then | |
| 16:08:35 | spatel | when i create vm with 16vcpu core just running on single numa node then result is freaking good | |
| 16:08:37 | dansmith | mriedem: thanks | |
| 16:09:00 | spatel | but if i create vm with 32vCPU and set numa_node=2 in flavor in that case result is 70% worst | |
| 16:09:49 | spatel | now question is how do i tell openstack to pin down vCPU with pCPU 1-0-1 ? | |
| 16:10:44 | spatel | I do have hw:cpu_policy=dedicated policy but its randomly mapping vCPU <--> pCPU | |
| 16:11:09 | spatel | sean-k-mooney: what do you suggest here? | |
| 16:11:43 | spatel | In short i want to make my VM fully NUMA map like bare metal has. | |
| 16:14:26 | sean-k-mooney | spatel: so it look like the erlang applciation is not numa aware | |
| 16:14:56 | spatel | sean-k-mooney: but if same erlang application i am running on bare metal it does work and result is good | |
| 16:14:59 | sean-k-mooney | meaning it is internally not numa affinitising its memory allocations | |
| 16:16:11 | sean-k-mooney | spatel: wehn you set hw:cpu_policy deicated it does a 1:1 mapping between cores on a virtual numa node to the a host numa node | |
| 16:16:24 | spatel | i thought if somehow openstack pin vcpu0 <--> pcpu0 , vcpu1 <--> pcpu1 so on... that would be great | |
| 16:16:30 | sean-k-mooney | so its not random it is numa aware | |
| 16:16:42 | sean-k-mooney | spatel: no it does not by design | |
| 16:17:05 | spatel | sean-k-mooney: look at this current mapping - http://paste.openstack.org/show/789225/ | |