14:00:44 <dulek> #startmeeting kuryr
14:00:45 <openstack> Meeting started Mon Aug  6 14:00:44 2018 UTC and is due to finish in 60 minutes.  The chair is dulek. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:00:46 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
14:00:48 <openstack> The meeting name has been set to 'kuryr'
14:00:57 <dulek> Anyone here for the Kuryr IRC meeting?
14:01:43 <ltomasbo> o/
14:02:00 <dmellado> the me with fever is half around too
14:03:18 <dulek> dmellado: Isn't fever nice in 40°C? Like you sweat anyway, but at least you feel it's colder.
14:03:53 <dmellado> dulek: well, you get the worst of the two worlds xD
14:04:25 <dulek> #topic announcements
14:04:36 <dulek> Okay folks, looks like we're in RC1 target week!
14:04:42 <dmellado> hoorray
14:04:44 <dmellado> xD
14:05:14 <dulek> Even though officially we're past feature freeze, I'm fine with merging some more new stuff.
14:05:22 <dmellado> dulek: me too
14:05:23 <ltomasbo> xD
14:05:32 <ltomasbo> good, I still need to land my last namespace patch set!
14:05:33 <dmellado> I'm looking forward to get the outcome from Wed meeting
14:05:42 <dmellado> as gary told me he might be asking for some FFE
14:05:43 <dulek> Without any particular exception process - we're quite small community and we know what others are working on.
14:05:46 <dmellado> so let's wait 'til that
14:06:04 <dmellado> but I'll have ltomasbo sing a song before merging any more of his patches
14:06:10 <dmellado> ltomasbo: which one would you present us with?
14:06:29 <dulek> dmellado: That's my point - I don't think we need a very complex FFE process, let's just focus on outstanding feature patchsets this week and if something's fine, let's merge it.
14:06:31 <ltomasbo> xD
14:07:07 <dmellado> dulek: exactly
14:07:14 <dmellado> I also have an announcement
14:07:18 <dmellado> and it's not that ltomasbo will sing
14:07:29 <dulek> ltomasbo: I've had enough of Pink Floyd yesterday, so Led Zeppelin might be a way to go. ;)
14:07:43 <dmellado> the TC decided to se tup liaisons that would follow the status of OpenStack projects so they have a better picture and can help
14:07:48 <ltomasbo> lol, I'll train myself and give it a try!
14:07:49 <ltomasbo> xD
14:07:50 * dulek wonders if "enough of Pink Floyd" is ever possible. :P
14:07:54 <dmellado> so in our case, we got dims and cdent
14:08:12 <dims> o/ :)
14:08:12 <dmellado> feel free to reach out to me or directly to them if you need something ;)
14:08:26 <dmellado> dims: I know I owe you a reply, writing atm ;)
14:08:46 <dims> thanks!
14:10:42 <dulek> Okay! So bottom line is - let's focus on outstanding feature patches, bonus points for those that aren't touching a lot of core code and mostly add plugins.
14:11:03 <dulek> We'll probably target the RC for Wednesday/Thursday, right?
14:11:07 <dmellado> dulek: yep
14:11:17 <dmellado> My plan is to do that on Thu after the Wed meeting
14:11:44 <ltomasbo> btw, at the moment the OpenShift gate is broken, right?
14:11:53 <dmellado> ltomasbo: didn't check today, tbh
14:11:56 <dmellado> what's the issue?
14:12:04 <dulek> ltomasbo: That's correct! Here's the fix: https://review.openstack.org/#/c/589084/
14:12:11 <ltomasbo> regsitry thing I suppose
14:12:20 <dulek> And no, I don't know why it failed Octavia gate.
14:12:25 <ltomasbo> dulek, I'm testing that on a local devstack...
14:12:57 <ltomasbo> dulek, timed_out, probably unrelated
14:13:20 <dulek> ltomasbo: Some "port already used" exceptions there, I'm rechecking that, so hopefully you're right.
14:13:39 <dmellado> ltomasbo: recheck and if you feel that it's a repeating issue
14:13:48 <dmellado> shout it out and we'll take a deeper look
14:14:22 <ltomasbo> dulek, my local devstack with your patch failed to deploy
14:14:38 <dulek> ltomasbo: In the same place?
14:15:11 <ltomasbo> dulek, I don't think so
14:15:15 <ltomasbo> seems to be related to the amphora
14:15:24 <ltomasbo> ++ /opt/stack/octavia/devstack/plugin.sh:build_octavia_worker_image:67 :   die_if_not_set 67 TOKEN 'Keystone failed to get token.'
14:15:41 <dmellado> ltomasbo: did you check keystone logs at that time?
14:15:41 <dulek> ltomasbo: Hm, so it shouldn't be related really…
14:16:19 <ltomasbo> ok, I'll dig a bit more later
14:17:11 <dulek> Okay, one more thing - I think that we still have a few bugs to get fixed in this release. My idea is to say loudly that we will plan to cut RC2, so we still have some time to include them.
14:17:28 <dulek> So feel free working on bugs. :)
14:17:50 <dmellado> dulek: let's put up an etherpad for this together
14:17:56 <dmellado> and start assigning things to ltomasbo
14:18:01 <dmellado> so he can sing even louder
14:18:05 <ltomasbo> xD
14:18:41 <dulek> Oh, this reminds me about on thing folks!
14:18:58 <dmellado> anyway folks, feeling like crap, so I guess I'll rest back to the sofa
14:18:59 <dulek> https://review.openstack.org/#/c/584377/, https://review.openstack.org/#/c/584421/ - those patches fix compatibility with older pod annotation format.
14:19:03 <dmellado> ping me if you need me for something
14:19:06 <dmellado> dulek: didn't I already approve that?
14:19:18 <dulek> It's quite important to include them to support upgrades of Kuryr.
14:19:44 <dulek> dmellado: Nope, no +2s there.
14:20:01 <dmellado> dulek: I'll check them out, thanks for bringing it
14:20:09 <ltomasbo> dulek, I took a look to the first one and it seems good to me, going to check the second one
14:20:15 <ltomasbo> and if possible, test both of them...
14:20:46 <ltomasbo> dulek, perhaps it will be nice to add some tempest test that checks the notation
14:21:46 <dulek> ltomasbo: I don't think tempest should depend on internal format of our annotations. We do depend because of some hacks that we needed to put into tempest plugin, but I don't think that's something we should test.
14:22:33 <dulek> ltomasbo: And I would love to have a test that does upgrading, but I don't think we'll be able to create a grenade gate in this release.
14:22:43 <dmellado> yeah, I was thinking about grenade as well
14:22:50 <dmellado> but that'd be for the S release
14:22:53 <ltomasbo> dulek, I was thinking more on a test that will ensure the format is not changed without noticing, as it seems we change the annotation format without thinking twice about them...
14:23:00 <dmellado> so dulek, for now that'd be ok
14:23:05 <dmellado> I'll take a deeper look in any case
14:23:19 <dmellado> ltomasbo: but those kind of tests would be in grenade
14:23:21 <dmellado> upgrades and so
14:25:19 <dulek> ltomasbo: Okay, I get that. So compatibility is not only annotations format and a complete solution would be grenade.
14:25:37 <ltomasbo> yep, I agree
14:26:52 <dulek> ltomasbo: Still it's not too convenient to test it in Tempest. It's possible, but I would need to either depend on o.vo serialization implementation and check it's dict field-by-field or depend from Kuryr o.vo's and that would not have any sense, because I would be changing o.vo when changing format.
14:27:54 <ltomasbo> true, and as you both mentioned, grenade will handle this anyway
14:28:28 <ltomasbo> I was just wondering if we had some (probably temploral) solution that even if it does not cover all the cases, at least it will trigger a red alarm when we do something like this
14:28:47 <dulek> I've created https://etherpad.openstack.org/p/kuryr-rocky-ffe-bugfixes - let's add outstanding patches there.
14:29:42 <dulek> #link https://etherpad.openstack.org/p/kuryr-rocky-ffe-bugfixes
14:30:00 <dulek> So I guess we're okay with announcements.
14:30:10 <dmellado> #action (dmellado) send an email to ML to announce the etherpad creation
14:30:26 <dulek> #topic kuryr-kubernetes or open discussion
14:32:50 <dulek> BTW seems like the bot doesn't answer us? ;)
14:32:53 <dmellado> I guess we can close it for today as we're also low on numbers
14:33:01 <pliu> For https://review.openstack.org/#/c/578009/, will it get merged in Rocky?
14:34:12 <dulek> pliu: It isn't touching core code at all, so I'm all for it to merge it.
14:34:27 <dmellado> pliu: yeah, I planned to discuss that on Wed meeting
14:34:33 <dmellado> but I'm also ok with that
14:34:42 <pliu> cool
14:35:09 <dulek> pliu: Can you add it to the etherpad?
14:35:15 <dmellado> I already spoke with gary that once we get to an agreement on that we'll make an exception
14:35:43 <pliu> dulek, sure
14:36:55 <dulek> Okay, so I'm closing this for today! Thanks everyone!
14:37:00 <ltomasbo> pliu, I will try to review it asap
14:37:02 <dulek> (unless celebdor has something to say :D)
14:37:03 <dmellado> so, anyone has anything else that you'd like to discuss? let's close this elsewhere
14:37:06 <pliu> thanks
14:37:17 <dmellado> dulek: celebdor will say he likes pablo casado
14:37:19 <dmellado> so let's close this
14:37:59 <dulek> #endmeeting