tweining | skraynev: FYI, we intend to discuss your change in our Octavia meeting at 1800 UTC, so might want to join. | 09:57 |
---|---|---|
tweining | *you might | 09:58 |
frickler | isn't that 1600 UTC? | 10:04 |
skraynev | @tweining thank you for review, @gthiemonge could you also please review PR: https://review.opendev.org/c/openstack/octavia/+/883063 and also who else could help with it? | 11:03 |
tweining | Frickler: you're right thanks | 11:08 |
tweining | skraynev: today 1600 UTC we'll discuss your Patch in our meeting | 11:09 |
skraynev | tweining: awesome! thank you | 11:10 |
tweining | Yeah so if you can you're welcome to join the discussion of course | 11:11 |
noonedeadpunk | hey folks! I was wondering - with removal of amphorav1 - how previously create loadbalancers are going to be managed? What's the way to convert them to amphorav2? | 15:45 |
gthiemonge | noonedeadpunk: hi, there's nothing to do, an amphora/LB created by amphorav1 can be controlled by v2 too | 15:47 |
noonedeadpunk | aha, I see | 15:48 |
gthiemonge | noonedeadpunk: the amphora doesn't even know what v1 or v2 are. the API between the controllers and the amphorae didn't change | 15:48 |
noonedeadpunk | Yeah, I wasn't sure if octavia will try to manage v1 when there's nothing that handles v1 | 15:48 |
johnsom | The only thing to keep an eye on is rolling upgrades. Do the DB, then API, then controllers. | 15:50 |
gthiemonge | noonedeadpunk: people who explicitly created LB with the "amphorav1" provider (and not "amphora" provider) may have issues when updating because this amphorav1 no longer exists | 15:50 |
johnsom | I think we just left it as an alias right? | 15:51 |
gthiemonge | no, it was removed | 15:51 |
gthiemonge | we only have the alias amphora -> amphorav2 | 15:51 |
noonedeadpunk | Looking at wha twe have - it;s only octavia and amphora | 15:52 |
johnsom | Yeah, so you are in good shape | 15:53 |
gthiemonge | #startmeeting Octavia | 16:00 |
opendevmeet | Meeting started Wed Sep 27 16:00:30 2023 UTC and is due to finish in 60 minutes. The chair is gthiemonge. Information about MeetBot at http://wiki.debian.org/MeetBot. | 16:00 |
opendevmeet | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 16:00 |
opendevmeet | The meeting name has been set to 'octavia' | 16:00 |
gthiemonge | o/ | 16:00 |
tweining | o/ | 16:00 |
johnsom | o/ | 16:00 |
gthiemonge | #topic Announcements | 16:01 |
gthiemonge | * Project Team Lead Election Conclusion and Results | 16:01 |
gthiemonge | in case you missed the news, I'll be your PTL for the Caracal release cycle | 16:02 |
tweining | yay. congrats! | 16:02 |
gthiemonge | thanks ;-) | 16:02 |
gthiemonge | * 2023.2 Bobcat Release Schedule: R-1 | 16:02 |
gthiemonge | this is the week of final RCs | 16:03 |
johnsom | Congratulations! | 16:03 |
gthiemonge | all good here, RC1 will be our Bobcat release (next week) | 16:03 |
gthiemonge | johnsom: thanks | 16:03 |
tweining | what? I thought we'll backport the mypy patch before the release?! | 16:04 |
tweining | just kidding | 16:04 |
gthiemonge | :D | 16:04 |
gthiemonge | * 2024.1 Caracal vPTG | 16:04 |
gthiemonge | I've scheduled the Octavia session | 16:05 |
gthiemonge | October 24th (Tuesday) - 14:00 UTC - 18:00 UTC | 16:05 |
gthiemonge | (note sure if we need 4 hours) | 16:05 |
gthiemonge | Reminder, the etherpad is | 16:08 |
gthiemonge | https://etherpad.opendev.org/p/caracal-ptg-octavia | 16:08 |
gthiemonge | you can add your topics there | 16:08 |
johnsom | gthiemonge FYI, that is not the etherpad linked from the PTG site: https://ptg.opendev.org/etherpads.html | 16:08 |
johnsom | You can update that link using the bot | 16:09 |
gthiemonge | oh, I have to fix that | 16:09 |
tweining | uh, active-active topology would be really nice I guess | 16:09 |
gthiemonge | yup | 16:09 |
johnsom | native BGP | 16:09 |
johnsom | SR-IOV | 16:09 |
tweining | also, don't forget to register (even though it's optional) | 16:11 |
gthiemonge | +1 | 16:11 |
gthiemonge | let's skip the CI status section, I have no update | 16:11 |
gthiemonge | #topic Brief progress reports / bugs needing review | 16:11 |
gthiemonge | 896255: Set default value for HSTS options to None | https://review.opendev.org/c/openstack/python-octaviaclient/+/896255 | 16:12 |
gthiemonge | could you review ^ guys | 16:12 |
gthiemonge | I've updated it because the release note was msising | 16:12 |
tweining | ok | 16:13 |
gthiemonge | thanks | 16:13 |
gthiemonge | FYI I'm working on a patch that retries (for a very long time) to set the provisioning_status of a LB if the DB is down | 16:14 |
gthiemonge | still wip | 16:14 |
gthiemonge | that should fix | 16:14 |
gthiemonge | https://bugs.launchpad.net/octavia/+bug/2036952 | 16:14 |
gthiemonge | #topic Add option to limit recursion in to_data_model | 16:20 |
gthiemonge | special topic for | 16:20 |
gthiemonge | https://review.opendev.org/c/openstack/octavia/+/883063 | 16:21 |
gthiemonge | https://bugs.launchpad.net/octavia/+bug/2019311 | 16:21 |
gthiemonge | it's about an issue with GET /members requests that take more that 2sec | 16:22 |
gthiemonge | note that I also opened a launchpad for a similar issue in the driver-agent | 16:22 |
gthiemonge | https://bugs.launchpad.net/octavia/+bug/2033280 | 16:22 |
gthiemonge | the idea is to limit the recursion of the to_data_model call | 16:23 |
gthiemonge | (for instance to get a pool, octavia may fetch all the children resources but also the parents resources - we observed 19 SQL queries in the driver-agent) | 16:24 |
johnsom | I had a look at this today. I think it looks ok to me. I have issues with the whole "to_data_model" thing in general, but in the members case, this is a reasonable fix IMO | 16:24 |
tweining | so how does it work currently? does that member GET respone only contain the direct child nodes information? | 16:24 |
gthiemonge | tweining: https://docs.openstack.org/api-ref/load-balancer/v2/index.html?expanded=show-member-details-detail,list-members-detail#list-members | 16:25 |
gthiemonge | AFAIK the only recursive call is for the tags, we need to check if the patch doesn't break it | 16:26 |
tweining | okay, so it should be fine then | 16:26 |
gthiemonge | I will test the possible corner cases in this patch | 16:26 |
gthiemonge | ok, anything else on this topic? | 16:28 |
tweining | I didn't test it but the unit test looks convincing, so I'll +2 it | 16:29 |
tweining | apart from that nothing | 16:29 |
gthiemonge | ack | 16:30 |
gthiemonge | #topic Open Discussion | 16:30 |
gthiemonge | any other topics folks? | 16:31 |
gthiemonge | I guess that's a no | 16:33 |
tweining | nothing that needs your attention now ;) | 16:33 |
tweining | I know you're busy with other stuff already | 16:34 |
gthiemonge | ok | 16:34 |
gthiemonge | then that's it for today! | 16:34 |
gthiemonge | thank you guys! | 16:34 |
gthiemonge | #endmeeting | 16:34 |
opendevmeet | Meeting ended Wed Sep 27 16:34:43 2023 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 16:34 |
opendevmeet | Minutes: https://meetings.opendev.org/meetings/octavia/2023/octavia.2023-09-27-16.00.html | 16:34 |
opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/octavia/2023/octavia.2023-09-27-16.00.txt | 16:34 |
opendevmeet | Log: https://meetings.opendev.org/meetings/octavia/2023/octavia.2023-09-27-16.00.log.html | 16:34 |
opendevreview | Merged openstack/python-octaviaclient master: Set default value for HSTS options to None https://review.opendev.org/c/openstack/python-octaviaclient/+/896255 | 21:18 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!