| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2023-10-20 | |||
| 09:39:01 | kashyap | Or is the answer, "we'll figure out"? :) | |
| 09:41:07 | bauzas | I dunno, see the above discussion I had with frickler | |
| 09:41:20 | bauzas | but I hope I can trust OIF :) | |
| 10:12:52 | frickler | a longer answer is "we'll figure it out and hope to be able to fix issues if they arise. we also think the social benefits will outweight the possible technical issues" | |
| 10:16:52 | kashyap | What does it mean by "social benefits"? I can read that phrase in a few ways. Either way, I'm always happy w/ open source (and working) tooling! :) | |
| 10:17:12 | kashyap | At Linux Plumbers last year in Dublin, they used Jitsi ... and it was mixed results. There was a write-up of it on LWN.net | |
| 10:19:55 | kashyap | (Sorry, not Jitsi, but "BigBlueButton") | |
| 10:22:00 | sean-k-mooney | kashyap: we have meetpad is an integration of etherpad with jitsi meet | |
| 10:22:24 | sean-k-mooney | it has been aviable to use to use since the first vPTG | |
| 10:23:42 | kashyap | I see | |
| 10:23:57 | sean-k-mooney | bauzas: no one will be angrey of we dont use zoom we talked about using meetpad for the last 2 vPTGs | |
| 10:26:33 | gibi | let's see, in the past there were random issue with jitsi (meetpad) when the number of participants increased. Like some people did not hear each other. Sometimes it is browser dependent. Also there is an android jitis app as a viable alternative if the browser based does not work well | |
| 10:30:39 | sean-k-mooney | gibi: ya we also had issue with zoom in the past with the meeting canceling | |
| 10:30:48 | gibi | true | |
| 10:31:13 | sean-k-mooney | if we have issues with either i can also just use my personal paid google account to host a meeting | |
| 10:31:30 | gibi | yeah, my main point is also keep an alternative ready :) | |
| 10:31:41 | sean-k-mooney | the issue we ahve hadd with google meet or jisti meet in the past was more around connecting form china | |
| 10:31:58 | sean-k-mooney | zoom was allowed for some reason. | |
| 10:33:03 | sean-k-mooney | my paid account supprot 150 participant video meetings + recording | |
| 10:33:13 | sean-k-mooney | thats likely enough if we need a fallback | |
| 11:29:18 | frickler | the usual zoom rooms are still in place if they are needed | |
| 11:59:35 | sean-k-mooney | gibi: bauzas would ye mind looking at https://review.opendev.org/c/openstack/nova/+/884439 it would be nice if we could get that merged so we can start backporting it | |
| 11:59:54 | sean-k-mooney | assuming we end up removing some of the stable branches after the ptg | |
| 12:00:16 | sean-k-mooney | i woudl like to include that in the final releases of those branches if we can | |
| 12:19:29 | gibi | sean-k-mooney: +A | |
| 12:20:05 | sean-k-mooney | thanks :) | |
| 13:49:21 | dansmith | bauzas: I much prefer zoom, fwiw | |
| 13:49:51 | bauzas | well, the ship sailed apparently | |
| 13:50:49 | dansmith | ah I see, ugh | |
| 14:04:27 | bauzas | hmmm, first time I'm seeing OSC not installed on my devstack | |
| 14:04:29 | bauzas | weird | |
| 14:05:09 | bauzas | actually, pip freeze shows it to me | |
| 14:05:11 | bauzas | python-openstackclient==6.3.0 | |
| 14:05:17 | dansmith | bauzas: in the venv? | |
| 14:05:34 | bauzas | I'm not sourcing any venv | |
| 14:05:42 | sean-k-mooney | what os | |
| 14:05:42 | bauzas | and I did set GLOBAL_VENV=False | |
| 14:05:46 | bauzas | RHEL9.2 | |
| 14:05:54 | sean-k-mooney | its proably not in your path | |
| 14:05:57 | bauzas | yeah | |
| 14:06:06 | bauzas | osc is installed | |
| 14:06:27 | sean-k-mooney | the default path on rehl does not ahve /usr/local i belvie | |
| 14:07:00 | bauzas | [stack@lenovo-sr655-01 ~]$ echo $PATH | |
| 14:07:00 | bauzas | /opt/stack/.local/bin:/opt/stack/bin:/sbin:/bin:/usr/sbin:/usr/bin:/usr/local/sbin | |
| 14:07:14 | bauzas | you're correct | |
| 14:07:25 | dansmith | but has /usr/local/sbin, that's .. odd :) | |
| 14:08:08 | bauzas | anyway, problem solved | |
| 14:08:26 | bauzas | if I just awaited 2 mins before pinging here, I wouldn't feel stupid | |
| 14:09:07 | sean-k-mooney | meh if it saves you 2-n mins of bashing yoru head agaist it why not ask | |
| 14:09:18 | sean-k-mooney | i only know this becasue it has bit me before | |
| 14:10:07 | sean-k-mooney | we fix up the patch in devstack iself but we dont persit that anywhere | |
| 14:10:37 | sean-k-mooney | https://github.com/openstack/devstack/blob/5613db3caa9193dcf84edabe04400341644ecab0/stack.sh#L70-L73 | |
| 14:11:55 | sean-k-mooney | i fixed that last yer https://github.com/openstack/devstack/commit/eca9783a0ad9c7d7e2e8267457be8d4dd8b55502 | |
| 14:12:56 | sean-k-mooney | when i was fixing centos 9 supprot https://review.opendev.org/q/topic:fix-install-on-centos9-cloud-image | |
| 14:36:52 | opendevreview | Merged openstack/nova master: Translate VF network capabilities to port binding https://review.opendev.org/c/openstack/nova/+/884439 | |
| 14:49:11 | bauzas | sean-k-mooney: interesting, I have your patch in my devstack but for some reason the PATH isn't good | |
| 14:49:13 | bauzas | anyway | |
| 14:50:31 | bauzas | oh, that's because this is just in the scropt | |
| 14:50:35 | bauzas | anyway, all good | |
| 14:56:43 | sean-k-mooney | ya if you sourced local.sh it would proably end up in your shell | |
| 14:56:51 | sean-k-mooney | but if you just run it it wont | |
| 14:57:04 | sean-k-mooney | *stack.sh not local.sh | |
| 14:57:54 | bauzas | I just added it in /etc/env | |
| 14:59:11 | opendevreview | Alexey Stupnikov proposed openstack/nova stable/2023.1: Translate VF network capabilities to port binding https://review.opendev.org/c/openstack/nova/+/898945 | |
| 14:59:16 | sean-k-mooney | i have this in my .bashrc | |
| 14:59:19 | sean-k-mooney | if [[ ! "${PATH}" =~ "/usr/sbin:" ]]; then | |
| 14:59:21 | sean-k-mooney | export PATH="/usr/sbin:$PATH" | |
| 14:59:23 | sean-k-mooney | fi | |
| 14:59:25 | sean-k-mooney | if [[ ! "${PATH}" =~ "$HOME/.local/bin" ]]; then | |
| 14:59:27 | sean-k-mooney | export PATH="$HOME/.local/bin:$PATH" | |
| 14:59:29 | sean-k-mooney | fi | |
| 14:59:54 | sean-k-mooney | actuly that is not fixing that specifically but that the pattern in normaly follow | |
| 15:01:49 | bauzas | I guess my devstack shell was just in transient state, because when I sshed it with another window, I got the PATH right | |
| 15:57:03 | dougszu | Any hints on debugging libvirt /qemu live migrate failures? I have one where even with debug logs, there is no obvious failure. | |
| 21:45:01 | opendevreview | Merged openstack/nova master: Revert "Add upgrade check for compute-object-ids linkage" https://review.opendev.org/c/openstack/nova/+/898741 | |
| 23:04:19 | opendevreview | Dmitriy Rabotyagov proposed openstack/nova stable/2023.2: Revert "Add upgrade check for compute-object-ids linkage" https://review.opendev.org/c/openstack/nova/+/898953 | |
| 23:13:07 | opendevreview | melanie witt proposed openstack/nova stable/2023.1: Fix exception catch when volume mount fails https://review.opendev.org/c/openstack/nova/+/898954 | |
| #openstack-nova - 2023-10-21 | |||
| 01:58:20 | opendevreview | efineshi proposed openstack/nova master: Don't allow to evacuate when it is in evacuate state https://review.opendev.org/c/openstack/nova/+/898200 | |
| 02:01:27 | opendevreview | efineshi proposed openstack/nova master: Don't allow to evacuate when it is in evacuate state https://review.opendev.org/c/openstack/nova/+/898200 | |
| #openstack-nova - 2023-10-23 | |||
| 02:24:57 | hamburgler | Heyo :), was wondering is there a way when creating an instance - such as with 'openstack server create --flavor 04eefe73-e27c-4981-9d28-171dac2a8c34 --availability-zone az2 --block-device source_type=image,uuid=d739b3eb-29cb-445f-97e3-7df07fd2555b,destination_type=volume,volume_size=20,volume_type=premium,delete_on_termination=false,boot_index=0 --network=7ee05a21-629f-490e-bdaf-fa64e291012b vm_in_az2' for | |
| 02:24:57 | hamburgler | example, to pass the nova availability zone to cinder when it's creating the volume? when using a volume as the boot device, nova does not seem to pass the az name to cinder. Yes I can create a volume in the az off an image, and follow up by attaching it to a new server, but this is an additional step, and seems like broken or missing functionality? each az for example has it's own ceph backend and works fine, and | |
| 02:24:58 | hamburgler | works with ephemeral provisioning which points to each respective ceph cluster, but when it comes to volumes with server create it always defaults to the default availability zone. Do not want to use cross az attachment. TIA :) | |
| 07:31:07 | bauzas | hamburgler: do you know the existence of https://docs.openstack.org/nova/latest/configuration/config.html#cinder.cross_az_attach ? | |
| 07:31:18 | bauzas | by default, the value is True, so we don't verify the Cinder ZE | |
| 07:31:21 | bauzas | AZ* | |
| 10:03:32 | bauzas | eeek, someone emptied the whole ptg etherpad https://etherpad.opendev.org/p/nova-caracal-ptg | |
| 10:08:11 | johnthetubaguy | bauzas: I was just going to ask about that, I found this looks better, but I forget how you restore it: https://etherpad.opendev.org/p/nova-caracal-ptg/timeslider#13560 | |
| 10:08:30 | bauzas | we need an infra admin in order to do it | |
| 10:08:39 | johnthetubaguy | ah, OK | |
| 10:08:50 | bauzas | I just asked in the -infra chan | |
| 10:10:23 | bauzas | johnthetubaguy: I could copy/paste but I think we'll loose the colors | |
| 10:10:36 | johnthetubaguy | yeah, I was thinking the same | |
| 10:11:24 | bauzas | I gonna send an email | |
| 10:11:43 | bauzas | every single cycle, our etherpad got trashed | |
| 10:11:48 | bauzas | I'm just done with that | |
| 10:12:02 | bauzas | this time, this is the first time it's trashed *before* | |
| 11:27:13 | gibi | bauzas: maybe somebody hates nova :) | |
| 11:34:00 | bauzas | gibi: well if someone hates nova, I'm happy to discuss with them to know why | |
| 11:48:32 | gibi | sure, but they are passive agressives :D | |
| 11:49:36 | bauzas | emptying an etherpad isn't really a passive argument | |