20:00:02 <johnsom> #startmeeting Octavia 20:00:03 <openstack> Meeting started Wed Jan 9 20:00:02 2019 UTC and is due to finish in 60 minutes. The chair is johnsom. Information about MeetBot at http://wiki.debian.org/MeetBot. 20:00:04 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 20:00:06 <openstack> The meeting name has been set to 'octavia' 20:00:09 <cgoncalves> o/ 20:00:11 <johnsom> Hi folks! 20:00:26 <johnsom> It's that time of the week once again for the Octavia show. I will be your host. 20:00:28 <johnsom> grin 20:00:47 <xgerman> o/ 20:01:10 <johnsom> Let's see if I can poke rm_work too 20:01:22 <johnsom> #topic Announcements 20:01:43 <johnsom> The call for papers for the Summit is still open to the 23rd. 20:02:12 <johnsom> Also, the foundation is asking for estimates of how many people will attend the PTG octavia sessions. They do this for space planning. 20:02:23 <xgerman> yeah, cgoncalves might wanna do something on logging 20:02:33 <johnsom> If you plan to go (you don't have to be committed yet, just planning) please let me know this week. 20:02:46 <cgoncalves> xgerman, hum? logging? 20:02:58 <xgerman> yeah, logging is an exciting feature 20:03:02 <xgerman> (new) 20:03:15 <cgoncalves> johnsom, I haven't had that conversation yet internally but I'd say someone from our side should go 20:03:22 <xgerman> but not sure if I can go... 20:03:27 <johnsom> We could do a new feature deep dive and include flavors. 20:03:37 <xgerman> +1 20:03:40 <cgoncalves> yes! 20:04:05 <cgoncalves> I have some other ideas but need to check before 20:04:08 <johnsom> And the TLS work 20:04:47 <johnsom> Ok, well, It's just an estimate, so if there is a chance, let me know and we will make sure you don't have to stand in the corner..... 20:05:23 <johnsom> Or work in the executive lounge corner. Oh, wait, we already did that. 20:05:45 <xgerman> it’s the same hotel so we cna always go to famous dave 20:05:54 <xgerman> I might have a hat somewhere 20:06:11 <johnsom> Is it? I thought they were doing it all downtown..... hmmm 20:06:22 <cgoncalves> yeah, should be downtown 20:06:23 <johnsom> The Summit is downtown at the convention center 20:06:27 <cgoncalves> trains will be missed, though (not!) 20:06:48 <johnsom> I hope the re-combine breaks the PTG curse.... 20:07:03 <johnsom> Trains, snow, bad water, etc.... 20:07:41 <johnsom> Ok, that is all I have for announcements. Any thing I missed? 20:07:50 <xgerman> or the curse now affects the whole cnference — Berlin had it’s issues 20:08:02 <johnsom> hahaha 20:08:29 <johnsom> #topic Brief progress reports / bugs needing review 20:08:50 <johnsom> I have been working on gate stuff and reviews mostly. 20:09:07 <johnsom> We now have a TLS gate that integrates with barbican: 20:09:12 <johnsom> #link https://review.openstack.org/#/c/628075/ 20:09:39 <rm_work> oh 20:09:40 <rm_work> o/ 20:09:55 <johnsom> This stores a cert in barbican as a non-privileged user, then creates and LB with a terminated TLS listener and passes some traffic through it. 20:10:00 <cgoncalves> johnsom, why do we need a new job? why can't we extend octavia-v2-dsvm-scenario? 20:10:30 <johnsom> cgoncalves I decided to do it this way so we can make it non-voting when barbican is broken 20:11:00 <johnsom> Given that their gates are all broken right now, it seemed like a good idea. 20:11:01 <cgoncalves> pin it to a stable release? kidding :) 20:11:15 <johnsom> #link https://review.openstack.org/625072 20:11:51 <johnsom> Do you strongly feel it should not be a separate gate? 20:12:18 <cgoncalves> no because I haven't reviewed your patch yet 20:12:24 <johnsom> OpenStack usually puts integration gates separate. 20:12:39 <cgoncalves> we integrate with nova, and glance, and neutron, and... 20:12:59 <rm_work> just seems wasteful to spin a whole devstack (40 minutes) for one test 20:13:14 <rm_work> but this is probably correct 20:13:18 <johnsom> Those are the privileged projects though.... 20:13:34 <cgoncalves> yes, my point. since I haven't reviewed it yet, I can't say I'm against having a new job 20:13:52 <johnsom> Agreed. It does need to expand beyond the one test. This is just a start. Plus the client cert and backend re-encrypt will also need to run there. 20:13:57 <cgoncalves> ok, we can easily change it later if we have/want to 20:15:34 <johnsom> Other than that, today I am looking at why DIB can't build bionic ubuntu-minimal images on bionic hosts. 20:16:06 <johnsom> #link https://bugs.launchpad.net/diskimage-builder/+bug/1811113 20:16:06 <openstack> Launchpad bug 1811113 in diskimage-builder "DIB fails to build bionic using ubuntu-minimal (rsyslog)" [Undecided,New] - Assigned to Michael Johnson (johnsom) 20:16:26 <johnsom> When they switched the nodepool instances to default to bionic this broke DIB.... sigh 20:17:06 <johnsom> Currently I am down to running dpkg in the "stupidlyverbose Insane amounts of drivel" debug mode. 20:17:18 * johnsom kids you not, that is what it is called 20:18:11 <johnsom> After that I am back to fixing the driver agent code, then finishing up the last bits on flavors. (tempest, an example capability such as nova flavor, etc.) 20:18:32 <johnsom> If you do need a bionic image for something, it builds fine on a xenial host. 20:19:28 <johnsom> Any other updates today? 20:20:40 <cgoncalves> no much from my side. I'm traveling this week 20:21:40 <johnsom> Ok then 20:21:46 <johnsom> #topic Open Discussion 20:21:50 <johnsom> Other topics today? 20:22:13 <rm_work> is it just OUR images, or a generic DIB problem? 20:22:32 <johnsom> Generic DIB issue. It happens before any of our stuff runs 20:22:32 <xgerman> doubt anyone but us is using DIB that much 20:22:35 <rm_work> seems like someone who actually works on DIB should be able to help out there 20:22:50 <xgerman> you have names? 20:22:51 <johnsom> Yeah, looking at the patches that is Ian and I..... 20:22:52 <rm_work> oh, and yes, planning on being in Denver 20:23:00 <johnsom> Thanks 20:23:00 <rm_work> T_T 20:23:31 <johnsom> I think it is limited to the minimal method too, so if they aren't using ubuntu-minimal it won't fail 20:23:35 <rm_work> finally maybe seeing some stability in my schedule, i've been in and out for the past few months (more out than in, technically) but should be able to start picking up reviews again 20:23:58 <johnsom> That would be great! 20:24:03 <xgerman> +1 20:24:33 <johnsom> Assuming I don't pull my hair out on the DIB issue, I will try to take a pass at the review list today as well 20:24:43 <johnsom> It likely needs some updating 20:24:53 <xgerman> sweet — so I need to fix the rebases of my patches :-) 20:25:23 <johnsom> Any other topics today? 20:28:30 <johnsom> Ok then. Have another great week folks! 20:28:35 <xgerman> o/ 20:28:42 <johnsom> #endmeeting