| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2023-02-10 | |||
| 16:30:17 | spatel | I can't modify existing VM correct? | |
| 16:30:23 | bauzas | spatel: no, you can't | |
| 16:30:31 | spatel | Perfect! now i got it what to do. | |
| 16:30:32 | bauzas | flavor is embedded into the instance data | |
| 16:30:59 | bauzas | if you modify a flavor, the instances that booted from that flavor won't magically update | |
| 16:31:00 | spatel | I will add fresh compute nodes with no HugePage and create new flavor without Pages | |
| 16:31:14 | bauzas | you'll be required to resize with another flavor | |
| 16:31:31 | spatel | bauzas I totally understand.. you want just change flavor and it will work magically :) | |
| 16:32:17 | spatel | This is what happened last week, one of memory module die which crash my whole compute nodes because of HugePage requirement :( | |
| 16:34:06 | bauzas | hah, that's a common failure | |
| 16:34:22 | bauzas | and yeah, relying on RAM can be dangerous | |
| 16:35:03 | spatel | Yes.. because of that crash it created loop in my switch (I don't know how but it lock up my switch because of STP) | |
| 16:35:31 | spatel | Just trying to re-produce this issue with multiple variable to see if i can re-create | |
| 18:32:06 | sean-k-mooney[m] | spatel you might want to look at the numa blancing config option | |
| 18:32:55 | sean-k-mooney[m] | packing_host_numa_cells_allocation_strategy | |
| 18:34:10 | sean-k-mooney[m] | spatel by the way if you are using cpu pinning but not hugepages you should set hw:mem_page_size=small | |
| 18:34:35 | sean-k-mooney[m] | if you dont then the vms will randomly get kill due to OOM events | |
| 18:36:58 | spatel | hmm is that a new option packing_host_numa_cells_allocation_strategy ? | |
| 18:37:32 | spatel | sean-k-mooney[m] this is interesting - by the way if you are using cpu pinning but not hugepages you should set hw:mem_page_size=small | |
| 18:37:48 | spatel | does it going to work if i don't configure hugepage in grub? | |
| 18:38:43 | sean-k-mooney[m] | packing_host_numa_cells_allocation_strategy is new and we backported it | |
| 18:39:31 | sean-k-mooney[m] | i think it was added in zed or yoga we changed the default to spread this cycle or last | |
| 18:39:58 | sean-k-mooney[m] | packing_host_numa_cells_allocation_strategy goes in the compute section of the nova.conf i belvie | |
| 18:40:16 | sean-k-mooney[m] | https://docs.openstack.org/nova/latest/configuration/config.html#compute.packing_host_numa_cells_allocation_strategy | |
| 18:41:16 | sean-k-mooney[m] | spatel: if you are using cpu pinnign hw:mem_page_size need to be set to some valid value to turn on numa aware memory allocation | |
| 18:41:42 | sean-k-mooney[m] | if you dont we will scudle based on the gloabl not numa local memory | |
| 18:42:12 | sean-k-mooney[m] | the OOM reaper in the kernel operates at the numa level | |
| 18:42:43 | spatel | ohhhh | |
| 18:43:27 | spatel | I know what you saying.. to run workload in NUMA we need to set hw:mem_page_size | |
| 18:43:46 | sean-k-mooney[m] | so the end result of not setting it is we will overcommit the numa node since we are only schduling based on the cpu in that case | |
| 18:44:01 | sean-k-mooney[m] | ya basically | |
| 18:44:50 | sean-k-mooney[m] | i have wanted to enforce this for a while but there were concerns that operators are depending on the incorrect behavior | |
| 18:45:36 | sean-k-mooney[m] | i have wanted to make hw:mem_page_size=any the default if you have a numa toplogy in the guest and dont set anything | |
| 18:45:37 | sean-k-mooney[m] | any is the same as small excpet it allows you to override it in the image | |
| 18:47:48 | spatel | hmmm | |
| 18:51:34 | sean-k-mooney[m] | the simple way to think about it is if its a numa vm you should set a mem_page_size as well | |
| 18:52:43 | sean-k-mooney[m] | well or use file backed memory but that is not a configuration that many people use | |
| 18:56:37 | spatel | I will do it.. | |
| 19:14:55 | melwitt | bauzas: done, thanks for reminding | |
| 22:00:47 | gmann | bauzas: yes, I am planning to progress on 864594 but let's see if i can push it before FF | |
| #openstack-nova - 2023-02-11 | |||
| 10:13:09 | opendevreview | Takashi Natsume proposed openstack/placement master: Fix a wrong assertion method https://review.opendev.org/c/openstack/placement/+/861489 | |
| 11:59:19 | opendevreview | Alexey Stupnikov proposed openstack/nova master: Fix logging in MemEncryption-related checks https://review.opendev.org/c/openstack/nova/+/873388 | |
| #openstack-nova - 2023-02-12 | |||
| 04:57:13 | opendevreview | Ghanshyam proposed openstack/nova master: Add service role in nova policy https://review.opendev.org/c/openstack/nova/+/864594 | |
| 20:28:52 | opendevreview | Ghanshyam proposed openstack/nova master: Add service role in nova policy https://review.opendev.org/c/openstack/nova/+/864594 | |
| #openstack-nova - 2023-02-13 | |||
| 01:49:33 | opendevreview | Ghanshyam proposed openstack/nova master: Add service role in nova policy https://review.opendev.org/c/openstack/nova/+/864594 | |
| 02:33:26 | opendevreview | Nobuhiro MIKI proposed openstack/nova master: libvirt: Add 'COMPUTE_ADDRESS_SPACE_*' traits support https://review.opendev.org/c/openstack/nova/+/873221 | |
| 04:12:10 | opendevreview | Ghanshyam proposed openstack/nova master: Add service role in nova policy https://review.opendev.org/c/openstack/nova/+/864594 | |
| 07:58:14 | ihti[m] | Hi, question: Does anyone know why we are using warnings module instead of LOG.warning() e.g for deprecations in oslo_policy. It breaks logs parsing for us, as the warnings emitted deviates from the standard logging formatting. | |
| 08:09:58 | opendevreview | Amit Uniyal proposed openstack/nova master: Remove "see nova-manage.log" string from console https://review.opendev.org/c/openstack/nova/+/873496 | |
| 08:38:12 | bauzas | morning Nova | |
| 08:53:16 | gibi | bauzas: o/ | |
| 08:54:41 | bauzas | currently writing some PTL nomination email, but will be back in 20 mins | |
| 09:07:45 | gibi | ack | |
| 09:15:42 | opendevreview | Jorge San Emeterio proposed openstack/nova master: WIP: Look for cpu controller on cgroups v2 https://review.opendev.org/c/openstack/nova/+/873127 | |
| 10:33:07 | opendevreview | Amit Uniyal proposed openstack/nova master: Remove "see nova-manage.log" string from console https://review.opendev.org/c/openstack/nova/+/873496 | |
| 10:38:17 | opendevreview | Jorge San Emeterio proposed openstack/nova master: WIP: Look for cpu controller on cgroups v2 https://review.opendev.org/c/openstack/nova/+/873127 | |
| 10:58:14 | bauzas | gibi: can't imagine https://review.opendev.org/c/openstack/tempest/+/873300 is still not merged | |
| 11:05:09 | kashyap | There is an emoji for that in Slack! | |
| 11:06:05 | kashyap | 🤯 | |
| 11:06:05 | kashyap | 🤯 | |
| 11:17:46 | bauzas | gibi: https://zuul.openstack.org/status#873300 will fuckingly fuckering fucker fail again | |
| 11:24:28 | bauzas | gibi: we may need to skip the ceph-multistore job by today if we wanna just merge some stuff in nova | |
| 11:41:17 | sean-k-mooney | bauzas: push a patch to make it non voting and add those test to the exclude regex and you can ping me to help merge it if you like | |
| 11:43:12 | opendevreview | Jorge San Emeterio proposed openstack/nova master: WIP: Look for cpu controller on cgroups v2 https://review.opendev.org/c/openstack/nova/+/873127 | |
| 12:34:21 | opendevreview | Edward Hope-Morley proposed openstack/nova stable/yoga: ignore deleted server groups in validation https://review.opendev.org/c/openstack/nova/+/867989 | |
| 13:11:34 | opendevreview | Konrad Gube proposed openstack/nova master: Use Cinder's os-extend_volume_completion volume action. https://review.opendev.org/c/openstack/nova/+/873560 | |
| 13:34:32 | gibi | bauzas: we can make ceph non voting for now, but it does not solve the issue that it seems to be impossible to merge anything in tempest now | |
| 13:34:55 | bauzas | gibi: let's see what's the latest result then | |
| 13:35:35 | bauzas | yet again a f.f.f. failure https://review.opendev.org/c/openstack/tempest/+/873300 | |
| 13:36:27 | gibi | bauzas: it seems you commented recheck at 12:23 and zuul voted at 12:37 so I guess you rechecked before the job finished | |
| 13:42:19 | gibi | I queued up again | |
| 13:53:05 | opendevreview | Amit Uniyal proposed openstack/nova master: nova-manage volume_attachment issues https://review.opendev.org/c/openstack/nova/+/873561 | |
| 14:00:24 | opendevreview | Amit Uniyal proposed openstack/nova master: nova-manage volume_attachment issues https://review.opendev.org/c/openstack/nova/+/873561 | |
| 14:11:05 | dansmith | bauzas: I wonder if we're getting to the point where we should make the ceph job n-v | |
| 14:11:13 | dansmith | I really hate to do that, but this is kinda nuts | |
| 14:11:27 | bauzas | dansmith: yeah, I dunno what to say | |
| 14:11:50 | dansmith | there are other such fixes in the queue and we've been unable to land them over the weekend as well | |
| 14:12:02 | bauzas | yup | |
| 14:13:42 | bauzas | so, how could we merge the changes that were accepted ? that's only my concern :) | |
| 14:14:42 | dansmith | the nova changes you mean? making the ceph job n-v at least gives you a chance to pass the other jobs | |
| 14:14:48 | dansmith | it'll still be a struggle, but.. | |
| 14:43:23 | gibi | dansmith, bauzas: can we ask infra to promot the tempest fix by skiping the gate? Or is it too dangerous? | |
| 14:44:08 | dansmith | we could, but I think given that it's not landing due to all the other broken, it's not really the emergent situation that normally demands that sort of thing | |
| 14:44:34 | dansmith | I think we reserve that for situations where we've got two things wedged against each other and one has to merge without tests | |
| 14:47:40 | bauzas | https://zuul.openstack.org/status#873300 looks like tempest-full-py3 can try twice | |
| 14:49:27 | bauzas | gibi: dansmith: so, given the time we have before FF, what should we be doing ? | |
| 14:49:33 | bauzas | punting the FF ? | |
| 14:49:45 | bauzas | or awaiting for the change to be merged ? | |
| 14:49:59 | dansmith | the 100% fail is only in the ceph job, so you can just make that job n-v for the moment and watch it carefully | |
| 14:50:00 | bauzas | or non-voting the job ? | |
| 14:50:05 | dansmith | but you'll still suffer all the other issues | |
| 14:50:16 | dansmith | how much is still needing to land for FF? | |
| 14:50:55 | bauzas | https://etherpad.opendev.org/p/nova-antelope-blueprint-status at least two series | |
| 14:51:47 | bauzas | + https://review.opendev.org/c/openstack/nova/+/872413 | |
| 14:53:30 | gibi | I'm OK to make the ceph job non voting to land the pending nova series | |
| 14:54:04 | bauzas | ok, then I'll create the change | |
| 14:54:17 | bauzas | and I'll also create a revert one | |
| 14:54:40 | dansmith | just make the revert one depends-on the fix and we can go ahead and have it pre-approved | |
| 14:54:45 | gibi | yeah and you can make a depends on in the revert to point to the tempest fix | |
| 14:54:57 | bauzas | cool | |