16:00:16 <gthiemonge> #startmeeting Octavia
16:00:17 <opendevmeet> Meeting started Wed Jun  2 16:00:16 2021 UTC and is due to finish in 60 minutes.  The chair is gthiemonge. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:00:18 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:00:21 <opendevmeet> The meeting name has been set to 'octavia'
16:00:29 <gthiemonge> Hi everyone!
16:00:33 <johnsom> o/
16:00:43 <gthiemonge> is it the right place?
16:00:46 <haleyb> hi
16:00:55 <johnsom> First meeting in our shiny new home
16:01:16 <haleyb> i dropped freenode completely
16:01:45 <rm_work> o/
16:01:59 <gthiemonge> I'm only in openstack-lbaas in case someone asks a question
16:01:59 <rm_work> made it to the inaugural meeting :P
16:02:04 <gthiemonge> great!
16:02:12 <johnsom> I dropped most of my freenode channels, but I'm hanging out in a few to direct people to the new location.
16:03:01 <gthiemonge> we should probably update the topic there, I don't know if I have the permission
16:03:38 <gthiemonge> anyway...
16:03:45 <gthiemonge> #topic Announcements
16:03:52 <johnsom> We don't want to update the topic on freenode. They will lock the channel if we do
16:04:12 <gthiemonge> ML2/OVN is now the default network backend in devstack - Octavia gates are broken
16:04:35 <gthiemonge> You may have noticed that ML2/OVN is now the default backend in devstack
16:05:07 <gthiemonge> It breaks our gates because the route to the IPv6 subnets is not set, our IPv6 load balancers are not reachable.
16:05:19 <gthiemonge> We need a change in devstack to fix it:
16:05:30 <gthiemonge> #link https://review.opendev.org/c/openstack/devstack/+/780628
16:05:34 <haleyb> they didn't re-revert it?  think they were trying
16:06:02 <gthiemonge> haleyb: It seems it was re-enabled on May 25th
16:06:25 <gthiemonge> I'll ping the folks on #openstack-qa to review that change
16:06:48 <gthiemonge> or should we wait for a 2nd revert?
16:06:55 <gthiemonge> :D
16:07:44 <haleyb> the 2nd revert was abandoned
16:09:03 <gthiemonge> haha, and this is no joke
16:09:16 <gthiemonge> ok, next announcement
16:09:19 <gthiemonge> IRC channel name change
16:09:34 <gthiemonge> As we're moving to OFTC, we requested to rename the IRC channel to #openstack-octavia.
16:09:42 <gthiemonge> #link https://review.opendev.org/c/openstack/project-config/+/793999
16:10:08 <gthiemonge> It hasn't officially been validated yet, and I'll send an email to the openstack-discuss ML when it's done.
16:10:27 <haleyb> I would +1 that, and sent this review out to change our in-tree doc
16:10:31 <haleyb> #link https://review.opendev.org/c/openstack/octavia/+/793216
16:11:06 <rm_work> hmmm i'm not sure I fully agree? but can be convinced (or outvoted :D)
16:11:39 <gthiemonge> do we need to vote?
16:12:07 <haleyb> neutron didn't keep openstack-quantum :p
16:12:53 <johnsom> I think it's nice to have a project name agnostic channel, but on the flip side we have someone about once a year that can't find the "octavia" channel.
16:15:44 <gthiemonge> rm_work: convinced? :D
16:15:59 <rm_work> hmmm
16:16:23 <rm_work> I personally liked to be "the place for lbaas", if people want to know about lbaas they go here, and find out the project is octavia
16:16:53 <rm_work> I dunno, I guess I'd rather have #openstack-compute than #openstack-nova too but I give the "core" projects a pass I guess for people knowing their names
16:17:26 <rm_work> that said, octavia is now "the lbaas project"
16:17:31 <johnsom> How dare you not call us "core"
16:17:32 <rm_work> so *shrug*
16:17:34 <rm_work> lol
16:17:47 <johnsom> grin
16:18:12 <gthiemonge> I don't think people will reach the "lbaas" IRC channel without knowing what Octavia is
16:18:25 <rm_work> I also hate to move channels without a GREAT reason, it just ends up splitting stuff forever
16:18:48 <rm_work> there will be people in this channel until the end of the network, lol
16:18:59 <johnsom> I could go either way really
16:19:32 <rm_work> yeah someone do some convincing?
16:20:00 <rm_work> or PTL just say we're doing it and I will say "ok" :D
16:20:16 <gthiemonge> we can continue this discussion in the project-config review? if we need it
16:20:32 <rm_work> I just think "The team wanted to rename it for a long time" is an interesting assertion, I don't know that I ever heard that, nor felt that way :D
16:20:38 <rm_work> yeah sure
16:20:47 <rm_work> if we have lots of other pressing business ;)
16:21:18 <gthiemonge> lol
16:21:40 <rm_work> bike-shedding moved to gerrit
16:22:25 <gthiemonge> rm_work: it has a W-1, everything is still possible
16:22:45 <rm_work> indeed, not the end of the world either way
16:23:25 <gthiemonge> yeah
16:23:40 <gthiemonge> any other announcements?
16:25:52 <gthiemonge> #topic Brief progress reports / bugs needing review
16:26:38 <gthiemonge> I have one open review for octaviaclient, it fixes errors when the user has more than 1000 octavia resources (>=1k load balancers, ...)
16:26:45 <johnsom> I had a little vacation time, so....
16:27:06 <gthiemonge> API calls were not paginated, so some commands failed because octaviaclient may not find the requested LB among the first 1000 entries
16:27:08 <johnsom> I also have a related patch open. The tests need work on it.
16:27:17 <gthiemonge> #link https://review.opendev.org/c/openstack/python-octaviaclient/+/792122
16:27:39 <johnsom> #link https://review.opendev.org/c/openstack/python-octaviaclient/+/792688
16:27:44 <gthiemonge> johnsom: do you need help on this one?
16:27:57 <johnsom> If you have time, please feel free to work on it.
16:28:17 <gthiemonge> johnsom: ok np
16:28:18 <johnsom> Basically the current tests blanket mock out the DB calls, so the new filters don't apply
16:28:38 <gthiemonge> ok
16:30:21 <gthiemonge> #topic Open Discussion
16:30:26 <gthiemonge> Any other topics today?
16:31:52 <rm_work> johnsom: lol yeah ... mocks are hard :D
16:32:19 <rm_work> nothing for me -- trying to figure out what my directives are going to be for the next quarter+ so I know how much time I actually have to spend upstream
16:32:30 <rm_work> (in that meeting simultaneously, now)
16:32:36 <johnsom> Well, it's just the way the old tests were written. I ran out of time I could put into that patch last week.
16:33:22 <johnsom> Many hats, just one clock. or something like that
16:34:15 <johnsom> It should provide one heck of a speedup on large clouds though.
16:38:41 <gthiemonge> I guess that's it for today
16:38:45 <gthiemonge> Thanks everyone
16:38:50 <gthiemonge> #endmeeting