17:16:04 #startmeeting ovn_community_development_discussion 17:16:05 Meeting started Thu Oct 29 17:16:04 2020 UTC and is due to finish in 60 minutes. The chair is mmichelson. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:16:06 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 17:16:08 The meeting name has been set to 'ovn_community_development_discussion' 17:16:34 Hello 17:16:44 hi! 17:16:48 hi blp 17:16:53 blp, hi 17:17:39 I can get things started, I guess. 17:17:53 I got a new revision of unit tests put up for review. 17:18:02 aaand, that's about it. 17:18:39 I've got a patch brewing that is going to help OpenShift out, but I gave them a rough-cut version of the patch to test to start with. The actual patch I'll submit to the mailing list will likely differ. 17:18:55 But it is a patch that will allow for user-selection of the CT zone to use for SNAT. 17:19:14 And that's all I wanted to bring up for the moment. 17:19:16 OK... sounds good. 17:19:22 May I go next? 17:19:58 sure 17:20:28 OK, I've submitted a number of series of patches building up toward adding ddlog support for ovn-northd. 17:20:42 Thanks to everyone for reviewing them. The reviews have gone so quickly and smoothly! I am grateful for that. 17:20:59 I have more series to go. I will send out another one today. 17:21:16 This next one will focus on refactoring the test infrastructure to make it easier to understand 17:21:25 and better for interpretation. 17:21:56 I had to spend a lot of time wrestling with the tests in various ways to make the ddlog implementation pass, but these improvements should help everyone. 17:22:08 That's my report. 17:22:09 thanks for the patches so far. 17:22:13 blp, some of us Red Hat OVN people were talking today, and if you wanted to submit more patches at once it would be OK. We're in a holding pattern right now for new submissions to ovn-northd, and so we'll definitely get on reviews quickly so that we can get back to submitting new northd content. 17:22:50 I think there are probably 2-3 series left. Today's should be the biggest set of changes to existing code. 17:22:56 blp, ack 17:22:57 The final series will essentially be just new code. 17:23:16 blp, were you ever able to do any scale testing of the ddlog code? 17:23:26 no, not that I found convincing 17:25:30 Who's next? 17:25:45 blp, I've one question. 17:26:06 blp, I was trying to compile the ovn code from your ddlog6 branch 17:26:13 and I'm seeing this error - northd/ovn-northd-ddlog.c:45:10: fatal error: northd/ovn_northd_ddlog/ddlog.h: No such file or directory 17:26:22 blp, any pointers what could be wrong ? 17:26:48 I wanted to try it out and have a scale test run 17:27:04 That is a generated file. It is generated by the ddlog program at build time. If it is not being generated, then something is odd. 17:27:24 someone ddlog generation is skipped. 17:27:42 I'll debug later and see. 17:28:12 s/someone/somehow 17:28:24 OK. 17:28:40 I can go real quick. I addressed review comments from mmichelson and submitted v2 of hairpin improvment patches. 17:29:04 thanks mmichelson for the reviews. 17:29:23 I still need to address dceara's comments. 17:29:51 And I did blp's and few other patch reviews. 17:29:57 That's it from me. 17:31:14 If someone wants to go next. 17:33:56 I'm not sure who else is here. I know dceara said he was going to be out today. 17:35:06 OK, I'm going to get back to polishing up my patches then. 17:35:34 OK, we'll be watching the mailing list 17:35:40 All right, I guess that'sit 17:35:44 #endmeeting