17:02:46 #startmeeting Designate 17:02:47 Meeting started Wed Dec 16 17:02:46 2015 UTC and is due to finish in 60 minutes. The chair is mugsie. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:02:48 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 17:02:51 The meeting name has been set to 'designate' 17:02:54 #topic roll Call 17:03:02 o/ 17:03:02 o/ 17:03:06 sorry - got distracted 17:03:11 o? 17:03:29 #topic Announcements 17:03:31 o/ 17:03:33 o/ 17:03:59 #link http://lists.openstack.org/pipermail/openstack-dev/2015-December/082256.html 17:04:11 The Designate mid cycle has been set for the HPE Offices, in Galway, 17:04:14 Ireland [0] from the 08th to the 10th of February. 17:04:22 #link https://ti.to/designate/mitaka-mid-cycle 17:04:29 if you are joining us ^ 17:04:37 Woo to no travel for me :) 17:04:39 please fill in that form 17:04:52 yay 17:05:05 #topic Action Items 17:05:25 Kiall bp e6adbb9 17:05:31 Kiall investiagte 1505605 17:05:42 Kiall: ^ 17:06:03 Humm, not that I know of. Havent looked in ~2 week 17:06:13 #action Kiall investiagte 1505605 17:06:16 1505605 doine 17:06:19 done* 17:06:21 as won't fix 17:06:24 #action Kiall bp e6adbb9 17:06:32 #action mugie bp ed2ba15 Merge "Check TXT record length limit" 17:06:38 I forgot about ^ 17:06:48 #topic Bug Triage (timsim - recurring) 17:06:52 I also think the BP is done.. need to check to 17:06:56 tho* 17:07:07 https://bugs.launchpad.net/designate/+bug/1525019 17:07:07 Launchpad bug 1525019 in Designate "mdns.get_serial_number retries for no reason" [Undecided,New] 17:07:18 Yea, it was done. 17:07:32 I still need to investigate 1516355 17:08:02 1525019 - not sure 17:08:19 I am not 100% what that function is supposed to do 17:08:32 Kiall: you any idea? 17:08:45 My impression was that it was a sync version of the async poll_for_serial number. 17:08:48 I believe timsim's comments in the ticket are correct, it should be just getting the current number 17:09:03 Which, if that's the case, it shouldn't be retrying waiting for anything, just getting the current value. 17:09:17 That's getting called a lot during `update_status` in the pool manager. 17:09:28 in that case H - M2 17:09:30 retrying, sure, if it fails to get a response.. but retrying because it's an unexpected serial number seems wrong 17:09:33 Basically whenever the first poll_for_serial comes back from mdns, pool manager goes and calls get_serial for every other target. 17:09:42 That isn't in the cache 17:09:52 timsim: oo, that also sounds like a bug 17:10:19 since we'll get those numbers back from the initial calls to poll for serial, we don't want to ask for it twice 17:10:46 If it didn't do that, and the last of the poll for serials gets lost, then the zone stays PENDING forever. 17:10:53 The way it works now, anyway. 17:11:03 Yea, I get that.. Still feels wrong! 17:11:15 No arguments here :P 17:11:28 That's a much trickier fix than this one. 17:11:30 Okay, well.. let's leave that issue for now ;) 17:11:33 yeah 17:11:38 https://bugs.launchpad.net/designate/+bug/1525199 17:11:38 Launchpad bug 1525199 in Designate "designate-dashboard: clear "Create Record" form fields" [Undecided,New] 17:11:48 should be in progress 17:11:52 assigned to federico3 17:11:56 #action kiall to file bug re update_status unnecessarily calling out to get_serial (see IRC meeting logs) 17:12:16 1525199 should be done, it failed gate due to a tox upgrade, but it otherwise +A'd 17:12:30 Kiall: I go more in detph about that here: https://github.com/TimSimmons/poolmanageroops#update_status 17:12:43 timsim: I've seen that doc, and mostly like ;) 17:12:53 1525199 might be a duplicate 17:13:04 federico3: about? 17:13:10 Gerrit appears to be down so we can't check the review :P 17:13:27 Nope, it's not - checked our JIRA and he refs that # 17:13:35 Ah "Gerrit is offline for a software upgrade from 17:00 to 21:00 UTC. " 17:13:36 It's in-progress, high, m2 17:13:41 cool 17:13:53 That's it 17:14:00 sweet 17:14:11 #topic Stable Backport Triage (kiall - recurring) 17:14:14 * Kiall panics.. 2 sec 17:14:38 http://replygif.net/i/1534.gif 17:14:49 mugsie: can you get the list? I actually don't have anything cloned, wiped PC yesterday. 17:15:14 nvm - got it 17:15:26 http://paste.openstack.org/show/482107/ 17:15:49 (0f5fd24 Fix tox ValueError: No closing quotation <-- hasn't landed yet, but it needs a backport as otherwise the gate is wedged) 17:16:02 ee5efbf Object model: require "priority" field 17:16:15 937e6dd Set zone in ERROR status on periodic sync fail 17:16:17 ee5efbf Object model: require "priority" field 17:16:25 Yep, all of those. 17:16:31 any others? 17:16:45 actually - that's all bar the setup.cfg one 17:16:45 anyone want to sign up to do these? 17:17:10 I'll ask federico3 to do em in the morning 17:17:17 * mugsie is on holidays from tomrrow 17:17:19 OK, cool 17:17:28 #action federico3 to do backports 17:17:43 #topic Unhandled exception during periodic 17:17:47 #topic Unhandled exception during periodic sync 17:18:09 this was federico3 as well, so we may push this for a week or so 17:18:18 / past the holidays 17:18:32 #topic Designate/Neutron Integration Update 17:18:49 mlavalle seems to be off ... 17:18:55 we will get an update post holidays 17:19:06 #topic Open Discussion 17:19:15 anything of agenda? 17:19:19 off* 17:19:32 Next week cancelled? 17:19:47 yeah - everyone OK skipping the meeting next week? 17:19:55 and the week after? 17:19:56 Yep, everyone's off here i think. 17:19:58 Same. 17:20:17 Okay, next meeting is... Jan 6th|? 17:20:24 #action mugsie Cancel meetings for 2 weeks 17:20:32 we will resume on the 6th 17:20:35 of Jan 17:20:38 2016 17:20:50 anything else? 17:20:53 I put up a pretty crazy spec here: https://review.openstack.org/258621 (https://gist.github.com/TimSimmons/510ae418de6804f5e5da while gerrit is down). It'd be a big change, but I think it'd be pretty cool. Give it a look and shout at me (and pglass) if you think we're mental. 17:21:17 http://replygif.net/i/1528.gif 17:21:25 Started from some elarson ideas, although I'm not sure how on board he is with the current state ;) 17:21:42 timsim: yea, I had a quick read of the doc on GH yesterday.. ++ to most of it :) 17:21:46 timsim: nah, I'm just thinking about the extreme cases of the design. not important 17:22:12 taking a trip to mars on my architect astronaut space ship :P 17:22:16 Well, I'd love to hear people's thoughts on the bits that they don't ++ :) 17:22:18 elarson: alwayts going to the extreme 17:22:19 * Kiall would like to see any "worker" service be TaskFlow based :) 17:22:30 which means ZK, basically 17:22:36 sorry timsim 17:22:38 * timsim would not like to see that at all 17:22:43 lol - why? 17:22:59 not like ZK or TF? 17:23:02 Both lol 17:23:09 What's wrong with RPC casts? 17:23:53 I'm not psyched about using another Openstack "NIH" type project (although I'll admit I don't know a ton about TaskFlow, other than knowing one of the OG people who started it, and has long since left it) 17:23:55 * mugsie thinks this will be a hot topic on Galway 17:24:19 TaskFlow enforces a nice structuring, and gives us stuff for "free".. e.g. define a flow for checking the status.. Flow(WaitForSerial(server1, expected=123, retries=10),WaitForSerial(server2, expected=123, retries=10), onfail=RollBackChange(change)) 17:24:20 etc 17:24:33 anyone able to reach gerrit? 17:24:35 * timsim will look into it 17:24:38 kobis: its down 17:24:47 they are upgrading it today 17:24:49 kobis: down for maint from 17:00 to 21:00 UTC 17:24:49 [17:08:32] [openstackstatus] NOTICE: Gerrit is offline for a software upgrade from 17:00 to 21:00 UTC. See: http://lists.openstack.org/pipermail/openstack-dev/2015-December/081037.html 17:24:49 !blame rsyed 17:24:50 elarson: Error: "blame" is not a valid command. 17:24:56 nice try 17:24:58 :) 17:25:03 thanks 17:25:18 elarson: http://replygif.net/i/1501.gif 17:25:49 OK, everyoine look at the outstanging specs (when gerrit is back) 17:26:03 mugsie is on :fire: with the giffage today 17:26:04 timsim: anyway, taskflow is nice, even if it's an OS NIH project.. 17:26:19 until then, unless there is anything else - lets call it a day? 17:26:19 * timsim has PTSD from oslo_* 17:26:25 might try convince one of the Cue folks to talk about it at the mid-cycle (via video..) ;) 17:27:03 timsim looking at olso_* code - http://replygif.net/i/961.gif 17:27:11 and on that note 17:27:17 #endmeeting