| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2022-11-07 | |||
| 15:35:03 | opendevreview | Arnaud Morin proposed openstack/nova master: Add os_vif in nova-config-generator https://review.opendev.org/c/openstack/nova/+/863891 | |
| 15:35:30 | amorin | bauzas and sean-k-mooney ^ | |
| 15:35:36 | amorin | we talked about that two weeks ago | |
| 15:36:00 | amorin | I expect this change to also do the necessary change on docs.o.o (config reference_ | |
| 15:36:03 | amorin | ) | |
| 15:36:09 | amorin | am I correct? | |
| 15:43:36 | opendevreview | Dan Smith proposed openstack/nova master: Test ceph-multistore with a real image https://review.opendev.org/c/openstack/nova/+/860864 | |
| 15:50:49 | opendevreview | Rajat Dhasmana proposed openstack/nova-specs master: Clarify client changes in rebuild spec https://review.opendev.org/c/openstack/nova-specs/+/856164 | |
| 15:54:02 | bauzas | amorin: should work | |
| 15:55:05 | amorin | ack | |
| 16:53:07 | opendevreview | Amit Uniyal proposed openstack/nova stable/xena: Adds a repoducer for post live migration fail https://review.opendev.org/c/openstack/nova/+/863864 | |
| 16:53:08 | opendevreview | Amit Uniyal proposed openstack/nova stable/xena: [compute] always set instance.host in post_livemigration https://review.opendev.org/c/openstack/nova/+/863792 | |
| 16:57:02 | opendevreview | Amit Uniyal proposed openstack/nova stable/wallaby: Adds a repoducer for post live migration fail https://review.opendev.org/c/openstack/nova/+/863900 | |
| 16:57:03 | opendevreview | Amit Uniyal proposed openstack/nova stable/wallaby: [compute] always set instance.host in post_livemigration https://review.opendev.org/c/openstack/nova/+/863901 | |
| 17:01:16 | opendevreview | Amit Uniyal proposed openstack/nova stable/victoria: Adds a repoducer for post live migration fail https://review.opendev.org/c/openstack/nova/+/863902 | |
| 17:01:17 | opendevreview | Amit Uniyal proposed openstack/nova stable/victoria: [compute] always set instance.host in post_livemigration https://review.opendev.org/c/openstack/nova/+/863903 | |
| 17:38:51 | opendevreview | Dan Smith proposed openstack/nova master: Test ceph-multistore with a real image https://review.opendev.org/c/openstack/nova/+/860864 | |
| 17:41:17 | darkhorse | Hi team - is there a standard way or recommended way of debugging under-development nova component under devstack environment? | |
| 17:42:38 | darkhorse | I tried to kill nova service from devstack and start my under-development nova in debug mode so that requests can be forwarded to the under-development nova. | |
| 17:44:04 | opendevreview | Stephen Finucane proposed openstack/nova master: hyperv: Mark driver as experimental https://review.opendev.org/c/openstack/nova/+/863910 | |
| 17:44:05 | opendevreview | Stephen Finucane proposed openstack/nova master: vmwareapi: Mark driver as experimental https://review.opendev.org/c/openstack/nova/+/863911 | |
| 17:44:42 | stephenfin | melwitt: dansmith: sean-k-mooney: I think that's what was suggested at the PTG. We don't deprecate, we just mark it experimental and remove if it starts costing us? ^ | |
| 17:45:56 | dansmith | stephenfin: yes, although I thought we were using other words like "untested" (or at least I was) | |
| 17:46:11 | dansmith | experimental means "not yet ready for prime time, but soon" to me in most software features | |
| 17:46:36 | dansmith | so I'd rather use something that implies it has degraded to the current state, not that it's working towards a better one | |
| 17:47:03 | dansmith | (although it's not a huge deal of course) | |
| 17:51:04 | melwitt | stephenfin: pretty much. my understanding was/is no formal deprecation, no plan for removal, just a log message that warns the driver is untested. ppl will use the driver at their own risk. there was also mention of updating the driver matrix doc to mention something similar to the log message for the drivers in question | |
| 17:51:40 | melwitt | *suggestion of updating driver matrix | |
| 18:19:57 | opendevreview | Dan Smith proposed openstack/nova master: Test ceph-multistore with a real image https://review.opendev.org/c/openstack/nova/+/860864 | |
| 18:32:12 | opendevreview | Dan Smith proposed openstack/nova master: Add virt/node module for stable uuids https://review.opendev.org/c/openstack/nova/+/863915 | |
| 18:32:13 | opendevreview | Dan Smith proposed openstack/nova master: WIP Add get_available_node_uuids() to virt driver https://review.opendev.org/c/openstack/nova/+/863917 | |
| 18:32:13 | opendevreview | Dan Smith proposed openstack/nova master: WIP: Pass service ref to init_host(), if exists https://review.opendev.org/c/openstack/nova/+/863916 | |
| 18:32:14 | opendevreview | Dan Smith proposed openstack/nova master: WIP: Make resource tracker use UUIDs instead of names https://review.opendev.org/c/openstack/nova/+/863919 | |
| 18:32:14 | opendevreview | Dan Smith proposed openstack/nova master: WIP: Persist existing node uuids locally https://review.opendev.org/c/openstack/nova/+/863918 | |
| 18:32:15 | opendevreview | Dan Smith proposed openstack/nova master: WIP: Detect host renames and abort startup https://review.opendev.org/c/openstack/nova/+/863920 | |
| 18:32:50 | dansmith | sean-k-mooney: all of this is marked WIP ^ because I haven't added more tests to cover any of the changes that aren't already covered (if any) | |
| 18:33:13 | dansmith | and because I need to look at it in gerrit to break up my tunnel vision | |
| 18:33:40 | dansmith | there are two functionals still failing because they test things this set intends to prevent, which are re-creating a compute node on a service and getting a different uuid | |
| 18:34:12 | dansmith | so I think the fix there is to make the services un-delete the compute nodes, but I left them failing to highlight and for discussion | |
| 19:19:12 | opendevreview | Dan Smith proposed openstack/nova master: DNM: Testing cinder devstack change https://review.opendev.org/c/openstack/nova/+/863925 | |
| 19:57:36 | opendevreview | Dan Smith proposed openstack/nova master: Test ceph-multistore with a real image https://review.opendev.org/c/openstack/nova/+/860864 | |
| 22:56:13 | opendevreview | Dan Smith proposed openstack/nova master: Test ceph-multistore with a real image https://review.opendev.org/c/openstack/nova/+/860864 | |
| #openstack-nova - 2022-11-08 | |||
| 00:56:10 | opendevreview | Dan Smith proposed openstack/nova master: Test ceph-multistore with a real image https://review.opendev.org/c/openstack/nova/+/860864 | |
| 01:34:52 | opendevreview | Merged openstack/nova stable/yoga: Adds a repoducer for post live migration fail https://review.opendev.org/c/openstack/nova/+/861871 | |
| 05:58:04 | opendevreview | Amit Uniyal proposed openstack/nova stable/ussuri: add regression test case for bug 1978983 https://review.opendev.org/c/openstack/nova/+/862603 | |
| 05:58:05 | opendevreview | Amit Uniyal proposed openstack/nova stable/ussuri: For evacuation, ignore if task_state is not None https://review.opendev.org/c/openstack/nova/+/862604 | |
| 08:59:44 | ygk_12345 | hi all | |
| 09:01:01 | ygk_12345 | I am trying to use pdb trace in the nova-api service in OSA setup. but when I start it manually , it is not dropping into the pdb shell. can anyone help me here ? | |
| 09:01:45 | bauzas | ygk_12345: you need to run a WSGI server that can support pdb | |
| 09:04:59 | bauzas | ygk_12345: example with httpd mod_wsgi https://modwsgi.readthedocs.io/en/master/user-guides/debugging-techniques.html#python-interactive-debugger | |
| 09:06:06 | bauzas | simple way with uwsgi https://stackoverflow.com/questions/18427948/how-to-debug-python-application-under-uwsgi | |
| 09:14:42 | ygk_12345 | bauzas: doesn't OSA nova-api support pdb ? | |
| 09:15:35 | bauzas | ygk_12345: I suppose nova-api wsgi app runs on a wsgi server (which I guess is httpd) which runs on daemon mode | |
| 09:16:03 | bauzas | if it would be uwsgi, that would still be a daemon | |
| 09:16:51 | bauzas | if you wanna use pdb, you need to run the wsgi server on demand and not as a daemon thru systemd or whatever | |
| 09:17:35 | bauzas | or rather use a remote pdb | |
| 09:18:16 | bauzas | but then you would need to make sure that you run on a single nova-api worker | |
| 09:19:36 | ygk_12345 | bauzas: I am using this manual command "/openstack/venvs/uwsgi-20.2.1-python3/bin/uwsgi --ini /etc/uwsgi/nova-api-os-compute.ini " | |
| 09:20:51 | ygk_12345 | bauzas: I have inserted importpdb;pdb.set_trace() in the file "/openstack/venvs/nova-20.2.1/lib/python3.6/site-packages/nova/objects/instance.py . but I see no pdb prompt | |
| 09:23:34 | bauzas | ygk_12345: you should probably add --honour-stdin and --workers 1 to the uwsgi command | |
| 09:23:47 | ygk_12345 | bauzas: let me try that | |
| 09:24:22 | bauzas | ygk_12345: and then, add a breakpoint in some easy rest endpoint call, like in nova.openstack.compute.api.servers on the get_instances() method | |
| 09:25:24 | bauzas | oh wait, you're not getting a BdBQuit error ? | |
| 09:25:58 | bauzas | that means that you haven't walked yet to where you added the breakpoint | |
| 09:26:20 | ygk_12345 | bauzas: still no luck | |
| 09:26:33 | ygk_12345 | bauzas: i am running the command nova list | |
| 09:26:46 | bauzas | ygk_12345: yup, because as i said just above, you haven't went to that line of code | |
| 09:26:53 | bauzas | s/went/gone | |
| 09:27:21 | ygk_12345 | get_all functions is executed from /openstack/venvs/nova-20.2.1/lib/python3.6/site-packages/nova/objects/instance.py right ? | |
| 09:27:27 | bauzas | ygk_12345: oh, you're not looking at the nova-api logs ? | |
| 09:27:44 | ygk_12345 | bauzas: no, I am tracing the code | |
| 09:28:09 | bauzas | ygk_12345: just to clarify, you won't get a prompt on your client call | |
| 09:28:23 | ygk_12345 | bauzas: yes | |
| 09:28:43 | bauzas | the interactive debugging session will run on the nova-api process that executes the server-side code for listing the instances | |
| 09:29:18 | bauzas | you can trigger it thru a client call of course | |
| 09:31:11 | bauzas | ygk_12345: to answer your question, no, a nova list doesn't lookup the instance records | |
| 09:31:38 | bauzas | ahah, I see your problem | |
| 09:32:01 | bauzas | since Newton, we don't expose the Instance record when you nova list | |
| 09:32:30 | bauzas | we rather expose an InstanceMapping record, eventually a BuildRequest object | |
| 09:32:39 | bauzas | you should rather try a nova show | |
| 09:36:26 | ygk_12345 | can anyone help me with this ? | |
| 11:31:31 | opendevreview | Rajesh Tailor proposed openstack/os-vif stable/zed: Move mtu update request into ovsdb transaction https://review.opendev.org/c/openstack/os-vif/+/863993 | |
| 11:32:16 | opendevreview | Rajesh Tailor proposed openstack/os-vif stable/yoga: Move mtu update request into ovsdb transaction https://review.opendev.org/c/openstack/os-vif/+/863994 | |
| 11:34:21 | opendevreview | Rajesh Tailor proposed openstack/os-vif stable/xena: Move mtu update request into ovsdb transaction https://review.opendev.org/c/openstack/os-vif/+/863995 | |
| 12:15:05 | admin1 | hi all .. my nova-api logs is filled with ERROR oslo_service.periodic_task oslo_ db.exception.DBConnectionError: (pymysql.err.OperationalError) (2013, 'Lost connection to MySQL server during query' .. but it reconnects .. and this is happening every 1-2 mins .. other apis are fine | |
| 12:15:16 | admin1 | am I hitting some kind of nova limits or connection/cache timeouts | |
| 12:15:25 | admin1 | mysql connection limit is not reached | |
| 13:57:34 | opendevreview | Amit Uniyal proposed openstack/nova stable/ussuri: Adds a repoducer for post live migration fail https://review.opendev.org/c/openstack/nova/+/864006 | |
| 13:57:35 | opendevreview | Amit Uniyal proposed openstack/nova stable/ussuri: [compute] always set instance.host in post_livemigration https://review.opendev.org/c/openstack/nova/+/864007 | |
| 14:01:33 | ygk_12345 | bauzas: now I am able to get the pdb prompt in the uwsgi process, but whatever pdb commands I type are not executing. also the cursor is coming out of pdb prompt by one line and waiting there | |
| 14:36:24 | opendevreview | Dan Smith proposed openstack/nova master: Test ceph-multistore with a real image https://review.opendev.org/c/openstack/nova/+/860864 | |
| 14:47:31 | opendevreview | Amit Uniyal proposed openstack/nova stable/ussuri: Adds a repoducer for post live migration fail https://review.opendev.org/c/openstack/nova/+/864006 | |
| 14:47:32 | opendevreview | Amit Uniyal proposed openstack/nova stable/ussuri: [compute] always set instance.host in post_livemigration https://review.opendev.org/c/openstack/nova/+/864007 | |
| 14:54:14 | opendevreview | Dan Smith proposed openstack/nova master: Test ceph-multistore with a real image https://review.opendev.org/c/openstack/nova/+/860864 | |
| 15:01:44 | Kirill_ | Hi, do we have a meeting today? | |
| 15:03:26 | elodilles | Kirill_: if i'm not mistaken we will have @ 16:00 UTC | |
| 15:03:44 | Kirill_ | got it, thanks | |
| 15:16:20 | opendevreview | sean mooney proposed openstack/nova master: Support multiple config file with mod_wsgi https://review.opendev.org/c/openstack/nova/+/864014 | |
| 15:26:13 | bauzas | reminder : nova meeting in 34 min | |
| 15:42:32 | opendevreview | Amit Uniyal proposed openstack/nova stable/train: Adds a repoducer for post live migration fail https://review.opendev.org/c/openstack/nova/+/863806 | |