18:32:30 #startmeeting networking_policy 18:32:31 Meeting started Thu Sep 21 18:32:30 2017 UTC and is due to finish in 60 minutes. The chair is SumitNaiksatam. Information about MeetBot at http://wiki.debian.org/MeetBot. 18:32:32 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 18:32:34 The meeting name has been set to 'networking_policy' 18:32:52 #info agenda https://wiki.openstack.org/wiki/Meetings/GroupBasedPolicy#Sept_21st.2C_14th.2C_7th_2017 18:33:12 #topic DB migration backport 18:33:32 gerrit seems to be down 18:33:43 :( 18:33:50 tbachman: you abandoned the patches i have linked in the agenda 18:34:01 SumitNaiksatam: those were replaced by the new migration script 18:34:08 but you have the new patch which we discussed late couple of weeks 18:34:12 tbachman: right 18:34:36 #link https://review.openstack.org/504976 <= new gerrit to address DB migrations (short-term solution) 18:34:43 i think your latest patch got over the migration script error, right? 18:34:45 (at least the link is right ;-) 18:34:48 SumitNaiksatam: ack 18:34:51 tbachman: :-) 18:35:16 the issue was a subsequent migration script (already committed) was missing the previous pointer 18:35:22 once I updated that, it worked 18:35:35 (even with the method that imports, rather than duplicates, the earlier migrations) 18:35:44 tbachman: cool 18:36:07 SumitNaiksatam: the plan is to backport this, dropping some of the migrations as needed, per-stable branch 18:36:10 so this is with the non __init__ workaround? 18:36:17 tbachman: that makes sense 18:36:23 SumitNaiksatam: it still needs the __init__ workaround 18:36:33 tbachman: thanks for getting this patch out! 18:36:37 SumitNaiksatam: np! 18:36:38 nice work 18:36:40 thx! 18:36:57 I know that rkukura won’t be able to review it until next week, due to the holiday 18:36:57 on the longer term strategy i dont think we have made that much head way 18:37:02 * tbachman nods 18:37:08 i am good with the patch 18:37:15 so if we get one more reviewer, we can merge it 18:37:24 k 18:37:30 on the longer term, i did talk one more colleague 18:37:36 tbachman: you know him :-) 18:37:40 :) 18:37:47 or you can guess 18:37:49 lol 18:38:04 the suggestion was to build the migration chain as a graph 18:38:23 where we would have multiple HEADs as leaf nodes 18:39:10 in theory this approach would work, but it will be pretty complicated to write the migration scripts since the author will have to spell out the dependency of a migration script on all other scripts (it depends on) 18:39:22 difficult to completely explain here 18:39:27 will revert to email 18:39:31 SumitNaiksatam: sounds good 18:39:50 you can think of a generalization of the multiple chain strategy i was thinking 18:40:17 tbachman: i assume you are not in hurry as far as the longer term strategy is concerned? 18:40:22 SumitNaiksatam: ack 18:40:29 ok thankds 18:40:47 #topic UI patches 18:40:52 I think we need to make sure we have a good strategy long-term, so spending the time to think and investigate is time well-spent, imho 18:40:58 tbachman: ack 18:41:19 #link https://review.openstack.org/#/q/status:open+project:openstack/group-based-policy-ui 18:41:32 so i am still seeing some issues with the patches posted 18:41:37 :( 18:41:40 these might very well be issues in my env 18:41:46 strange, b/c it sounds like they’re working for annak 18:41:50 annakk 18:42:05 * tbachman forgot the extra “K" 18:42:16 i have been testing some of the patches she has presumably not tested 18:42:36 presumed (based on the review comments) 18:43:07 we will eventually finish reviewing and merging these patches (hopefully soon) 18:43:27 the thing i am not sure about is whether we want to backport these to stable/ocata 18:43:38 i started of thinking that we were going to backport these 18:43:58 SumitNaiksatam: it might be good to get annakk’s insights here 18:44:03 but there is a huge amount of refactor in these patches and i am leaning towards not backporting these 18:44:05 to see if she’s tried them on stable/ocata as well 18:44:06 tbachman: rivht 18:44:11 *right 18:44:37 so unless its absolutely needed, my proposal will be to not backport it 18:44:52 will check with annak if she needs these to be backported 18:44:55 k 18:45:28 #topic py27 error messages and copious logs 18:45:37 SumitNaiksatam: thanks for all the patches on this! 18:45:41 there were a couple of patches posted to address this 18:45:44 tbachman: np 18:45:45 * tbachman still has to review and approve 18:45:53 one from annak and one which i posted 18:46:04 tbachman: yeah thanks, please take a look 18:46:12 SumitNaiksatam: will do 18:46:14 i think annak is planning another rev on her patch 18:46:23 k 18:46:50 there are other ones which are not addressed 18:46:59 #topic Open Discussion 18:47:09 tbachman: anything else you wanted to bring up? 18:47:16 SumitNaiksatam: nothing right now 18:47:25 tbachman: ok cool, thanks for joining 18:47:28 SumitNaiksatam: np! 18:47:32 thanks SumitNaiksatam! 18:47:32 bye! 18:47:35 bye! 18:47:37 #endmeeting