| *** marlinc is now known as Guest1294 | 00:09 | |
| *** marlinc is now known as Guest1303 | 01:25 | |
| *** marlinc is now known as Guest1308 | 02:42 | |
| *** marlinc is now known as Guest1315 | 03:57 | |
| *** marlinc is now known as Guest1323 | 05:26 | |
| *** marlinc is now known as Guest1326 | 06:29 | |
| *** marlinc is now known as Guest1331 | 08:26 | |
| opendevreview | Dr. Jens Harbott proposed openstack/neutron-dynamic-routing master: Fix tox.ini for tox4 https://review.opendev.org/c/openstack/neutron-dynamic-routing/+/866927 | 08:40 |
|---|---|---|
| opendevreview | Luca Czesla proposed openstack/neutron stable/yoga: Add address scope to the OVN LSP port registers https://review.opendev.org/c/openstack/neutron/+/866968 | 09:26 |
| *** marlinc is now known as Guest1336 | 09:28 | |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: Fix tox.ini for tox4 https://review.opendev.org/c/openstack/neutron/+/866969 | 09:50 |
| opendevreview | Rodolfo Alonso proposed openstack/networking-sfc master: Fix tox.ini for tox4 https://review.opendev.org/c/openstack/networking-sfc/+/866971 | 09:52 |
| *** marlinc is now known as Guest1341 | 10:45 | |
| opendevreview | Merged openstack/neutron master: Fix firewall_driver examples in docs https://review.opendev.org/c/openstack/neutron/+/866793 | 10:55 |
| opendevreview | Slawek Kaplonski proposed openstack/neutron-tempest-plugin master: DNM: just a test for dropping lib/neutron https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/866976 | 11:01 |
| *** marlinc is now known as Guest1348 | 12:27 | |
| lajoskatona | amorin, ykarel: Hi, I just tried fullstack in a fresh VM (jammy), but failed | 12:56 |
| lajoskatona | amorin, ykarel: cloned devstack, neutron, created /opt/stack/logs, cloned requirements repo under /opt/stack | 12:56 |
| lajoskatona | amorin, ykarel: $ export VENV=dsvm-fullstack; export IS_GATE=False; tools/configure_for_func_testing.sh ~/devstack/ -i | 12:57 |
| lajoskatona | and I have this: utilities/ovn-dbctl.c: In function ‘do_dbctl’: | 12:57 |
| lajoskatona | utilities/ovn-dbctl.c:724:9: error: too few arguments to function ‘ctl_context_init_command’ | 12:57 |
| lajoskatona | 724 | ctl_context_init_command(ctx, c); | 12:57 |
| lajoskatona | | ^~~~~~~~~~~~~~~~~~~~~~~~ | 12:57 |
| lajoskatona | amorin, ykarel: so something is really missing in the script or in the doc :-) | 12:58 |
| lajoskatona | (I mean other than creating folder and cloning requirements repo) | 12:58 |
| *** dasm|off is now known as dasm | 13:00 | |
| opendevreview | liuxie proposed openstack/neutron-fwaas master: Support standard_attrs for fwaas resources https://review.opendev.org/c/openstack/neutron-fwaas/+/866988 | 13:07 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: Amend documentation for fullstack tests https://review.opendev.org/c/openstack/neutron/+/866854 | 13:32 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: Allow restoration of tun_ofports on agent restart https://review.opendev.org/c/openstack/neutron/+/860270 | 13:32 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: Add tests for br-tun of restoration https://review.opendev.org/c/openstack/neutron/+/866589 | 13:32 |
| ykarel | lajoskatona, but that looks ovs/ovn from source, isn't by default ovs/ovn pulled from packages? | 13:33 |
| ykarel | or you have created custom local.conf to enable that? | 13:33 |
| lajoskatona | ykarel: this is some default I think as I haven't added local.conf | 13:35 |
| ykarel | lajoskatona, ack as per https://github.com/openstack/devstack/blob/5e7afb779c469f593a1628e8f63c66989b7e2c49/lib/neutron_plugins/ovn_agent#L71 source install is disabled by default | 13:37 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: Add tests for br-tun of restoration https://review.opendev.org/c/openstack/neutron/+/866589 | 13:37 |
| lajoskatona | ykarel: ack, so this OVN_BUILD_FROM_SOURCE is an env variable? | 13:38 |
| ykarel | lajoskatona, /me not sure if it parses env vars, but atleast that's can be overridden in local.conf | 13:39 |
| lajoskatona | ykarel: seems as env var it is useless, let me try with local.conf | 13:40 |
| ykarel | lajoskatona, can you check if .localrc.auto created in devstack dir? | 13:40 |
| ykarel | and what's in it | 13:40 |
| lajoskatona | ykarel: no, I haven't .localrc.auto | 13:42 |
| amorin | lajoskatona on my side it worked, I did no change in local.conf at all | 13:42 |
| amorin | I used ubuntu 22.04, created a stack user with correct sudo rights (like in devstack doc) | 13:43 |
| amorin | cloned the neutron, requirements, and devstack repos in /opt/stack | 13:43 |
| amorin | then export VENV=dsvm-fullstack | 13:43 |
| amorin | tools/configure_for_func_testing.sh ~/devstack/ -i | 13:43 |
| amorin | I havnt done thaT: export IS_GATE=False | 13:44 |
| amorin | would it be the root cause of the failure? | 13:44 |
| *** marlinc is now known as Guest1356 | 13:45 | |
| lajoskatona | amrin: thanks, and for you ovs & ovn was from pkg or from source? | 13:48 |
| amorin | let me check | 13:48 |
| amorin | from sources lajoskatona | 13:49 |
| ykarel | https://github.com/openvswitch/ovs/commit/b8bf410a5c94173da02279b369d75875c4035959 | 13:49 |
| ykarel | that landed recently and so places using ovs master should be impacted | 13:50 |
| ykarel_ | seems we only have experimental jobs with that case | 13:56 |
| lajoskatona | ykarel_, amorin: yeah seems like going back in time with ovs (git checkout a77c7796f23a76190b61e2109a009df980253b0f ) configure-for-funct-testing works, thanks | 13:58 |
| ykarel_ | lajoskatona, triggered experimental jobs in https://review.opendev.org/c/openstack/neutron/+/863780 | 14:02 |
| ykarel_ | we should see this issue there too | 14:02 |
| opendevreview | Lajos Katona proposed openstack/tap-as-a-service master: Fix tox.ini for tox4 https://review.opendev.org/c/openstack/tap-as-a-service/+/866998 | 14:22 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: Add tests for br-tun of restoration https://review.opendev.org/c/openstack/neutron/+/866589 | 14:40 |
| ykarel_ | lajoskatona, https://zuul.opendev.org/t/openstack/buildset/bb6c601452de45aea948880ebfd99a87, seen in all ovs-master experimental job | 15:08 |
| ykarel_ | reported https://bugs.launchpad.net/neutron/+bug/1999154 | 15:15 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: Allow restoration of tun_ofports on agent restart https://review.opendev.org/c/openstack/neutron/+/860270 | 15:21 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: Add tests for br-tun of restoration https://review.opendev.org/c/openstack/neutron/+/866589 | 15:21 |
| lajoskatona | ykarel_: thanks | 15:22 |
| opendevreview | yatin proposed openstack/neutron master: Pin OVS_BRANCH to working commit https://review.opendev.org/c/openstack/neutron/+/867004 | 15:24 |
| ykarel_ | lajoskatona, ^ | 15:24 |
| *** marlinc is now known as Guest1368 | 15:26 | |
| lajoskatona | ykarel_: accidentally I tried to execute with tox > 4.0.0 and it brought another problems, like missing scripts from allowlist_externals | 15:28 |
| lajoskatona | ykarel_: I will open a bug for that also to track it' | 15:29 |
| ykarel_ | lajoskatona, hmm i see there is patch from ralonsoh for it | 15:29 |
| ralonsoh | ykarel_, the problem is in tempest or devstack | 15:29 |
| ralonsoh | the converstaion is in the #qa channel now | 15:29 |
| ralonsoh | and the mailing list | 15:29 |
| ykarel_ | ralonsoh, ohkk, so https://review.opendev.org/q/topic:tox4 still needed? | 15:31 |
| ralonsoh | yes | 15:31 |
| ykarel_ | ok | 15:31 |
| ykarel_ | there are fixes in other topic too https://review.opendev.org/q/topic:fix-tox | 15:32 |
| lajoskatona | ralonsoh: ack | 15:32 |
| lajoskatona | ykarel_: another thing which I got with fullstack, perhaps you know: "testtools.runtest.MultipleExceptions: ((<class 'ModuleNotFoundError'>, ModuleNotFoundError("No module named 'neutron.privileged.agent'"), <traceback object at 0x7f6a4e7b6780>), (<class 'fixtures.fixture.SetupError'>, SetupError({}), <traceback object at 0x7f6a4e7b4dc0>))" | 15:35 |
| lajoskatona | ykarel_: starnge that for me these things come only with jammy..... but I would like to collect all to update the doc with them | 15:36 |
| ykarel_ | lajoskatona, hmm that happens when the parent dir for neutron are not have executable bits | 15:37 |
| ykarel_ | ls -l /opt, check stack dir would not have +x bit then it's the same issue | 15:37 |
| ykarel_ | sudo chmod +x /opt/stack to fix it | 15:37 |
| ykarel_ | https://review.opendev.org/c/openstack/devstack/+/838645 for context | 15:39 |
| lajoskatona | ykarel_: hmm, I check it | 15:42 |
| ralonsoh | good to know this | 15:42 |
| ralonsoh | ykarel_, should we enforce this in the configure_for_func_testing srcipt? | 15:43 |
| lajoskatona | I collect these and update the doc with the latest things for jammy (and perhaps not just for jammy) | 15:43 |
| ykarel_ | ralonsoh, yes would be good to do that in script or atleast in docs | 15:46 |
| ralonsoh | ykarel_, perfect, I'll propose a patch | 15:46 |
| ykarel_ | Thanks | 15:46 |
| *** ykarel_ is now known as ykarel|away | 15:47 | |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: Fix homedir permissions https://review.opendev.org/c/openstack/neutron/+/867008 | 15:57 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: Fix homedir permissions https://review.opendev.org/c/openstack/neutron/+/867008 | 15:58 |
| amorin | hey again, I am looking for the code in charge of setting the interfaces up and down (the qr-xxx qg-xxx) in qrouter namespace | 16:06 |
| amorin | I notice that when using DVR mode, the interfaces are never set to DOWN | 16:06 |
| amorin | while in centralized mode, the interfaces are set to down on the standby node | 16:07 |
| amorin | (I am using HA/keepalived) | 16:07 |
| amorin | any hint on which code is responsible for this? | 16:07 |
| *** ministry is now known as __ministry | 16:07 | |
| amorin | somehwere in agent/ha_router.py and agent/dvr_something.py I guess | 16:08 |
| opendevreview | Brian Haley proposed openstack/neutron master: Fix text in OVN RevisionConflict exception to match code https://review.opendev.org/c/openstack/neutron/+/867009 | 16:35 |
| haleyb | otherwiseguy: is that correct? ^^^ or anyone that knows the ovn code better than i :) | 16:36 |
| otherwiseguy | haleyb: not sure off the top of my head, lucasagomes would probably know (and I'm on PTO) :) | 16:37 |
| *** otherwiseguy is now known as otherwiseguy-pto | 16:37 | |
| otherwiseguy-pto | haleyb: but also, "hi!" :D | 16:38 |
| haleyb | otherwiseguy: enjoy PTO, don't look at code it will burn your eyes! | 16:38 |
| haleyb | otherwiseguy-pto: hi! enjoy the time off, heard KC was balmy these days :D | 16:39 |
| otherwiseguy-pto | haleyb: I spent the last few years building up an immunity to dangerous code. :p | 16:40 |
| otherwiseguy-pto | and iocane powder | 16:40 |
| otherwiseguy-pto | haleyb: it's been reasonably nice for the time of year. Went through a bunch of 20 degree days, then some 50-60. | 16:42 |
| haleyb | well that's inconveivable | 16:42 |
| otherwiseguy-pto | :) | 16:42 |
| lucasagomes | haleyb, checking | 16:42 |
| haleyb | it was mid 50's here yesterday we're like 1-2 days behind you, wasn't complaining | 16:42 |
| haleyb | lucasagomes: thanks, someone here had a log full of those, that's another story but noticed the text | 16:43 |
| lucasagomes | haleyb, I thinik it was fine before ? Because we use that conflict to skip updating the resource in OVN right. So if the revision numbers already match or is higher in OVN it will be skipped | 16:43 |
| haleyb | lucasagomes: well the code uses a '>' so something is wrong :-/ | 16:44 |
| lucasagomes | unless we only check for the > | 16:44 |
| lucasagomes | haleyb, ah | 16:44 |
| lucasagomes | so yeah hehe | 16:44 |
| haleyb | the code is "old" from networking-ovn days so figured it was correct? | 16:45 |
| lucasagomes | haleyb, yeah | 16:45 |
| lucasagomes | haleyb, so in that case I think ur change makes sense | 16:46 |
| lucasagomes | haleyb, cause I also don't see how we get equal revision numbers | 16:46 |
| lucasagomes | should never happen theorically | 16:46 |
| * haleyb shrugs too, and i didn't want to change the code since i know i'll break something | 16:47 | |
| lucasagomes | haleyb, all good :D | 16:47 |
| lucasagomes | u wouldn't break it at all | 16:47 |
| lucasagomes | voted on the patch | 16:48 |
| * otherwiseguy-pto needs a job just providing 3XX responses to questions | 16:49 | |
| haleyb | lucasagomes: thanks. now i just need to figure out what triggered, but that's a me problem | 16:49 |
| lucasagomes | haleyb, that's an erro per se... Could be two updates racing ? Where the first one took more time to run than the second, so the revision number in OVN was bumped to something higher | 16:53 |
| lucasagomes | haleyb, then the first update tried to commit and the revision number was lower than the one already registered | 16:54 |
| lucasagomes | so it was skipped | 16:54 |
| lucasagomes | cause the OVN resouce is already newer and contain the changes from the 1st update | 16:54 |
| lucasagomes | haleyb, im calling it a day now, but tomorrow if needed I can help u take a look too | 17:00 |
| haleyb | lucasagomes: thanks for the info. that could be it, was an outage with all sorts of failures like db connection failures and leader changes. | 17:01 |
| lucasagomes | haleyb, I see, yeah probably that error is a red herring | 17:01 |
| lucasagomes | it shouldn't be a problem | 17:01 |
| haleyb | i'll look closer since i think it showed as ERROR when WARNING is maybe better, but yeah it is a red herring | 17:01 |
| lucasagomes | yeah true, maybe the log level is not ideal | 17:02 |
| lucasagomes | but yeah | 17:02 |
| haleyb | i don't see how it showed ERROR, but not a big deal | 17:03 |
| lucasagomes | haha yeah bad form | 17:03 |
| *** marlinc is now known as Guest1378 | 17:30 | |
| opendevreview | Arnaud Morin proposed openstack/neutron master: Allow restoration of tun_ofports on agent restart https://review.opendev.org/c/openstack/neutron/+/860270 | 18:31 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: Add tests for br-tun of restoration https://review.opendev.org/c/openstack/neutron/+/866589 | 18:31 |
| opendevreview | Brian Haley proposed openstack/neutron master: Test tox4 https://review.opendev.org/c/openstack/neutron/+/867048 | 19:10 |
| opendevreview | Brian Haley proposed openstack/neutron master: Test tox4 https://review.opendev.org/c/openstack/neutron/+/867048 | 19:11 |
| *** marlinc is now known as Guest1387 | 19:19 | |
| opendevreview | Brian Haley proposed openstack/neutron master: Fix tox.ini for tox4 https://review.opendev.org/c/openstack/neutron/+/866969 | 19:57 |
| opendevreview | Brian Haley proposed openstack/neutron master: Test tox4 https://review.opendev.org/c/openstack/neutron/+/867048 | 19:58 |
| *** marlinc is now known as Guest1392 | 20:32 | |
| *** marlinc is now known as Guest1395 | 21:46 | |
| *** dasm is now known as dasm|off | 21:50 | |
| opendevreview | Brian Haley proposed openstack/neutron master: Fix tox.ini for tox4 https://review.opendev.org/c/openstack/neutron/+/866969 | 22:21 |
| opendevreview | Brian Haley proposed openstack/neutron master: Test tox4 https://review.opendev.org/c/openstack/neutron/+/867048 | 22:22 |
| *** marlinc is now known as Guest1402 | 23:07 | |
| opendevreview | Merged openstack/neutron master: Pin OVS_BRANCH to working commit https://review.opendev.org/c/openstack/neutron/+/867004 | 23:47 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!