17:59:45 #startmeeting trove 17:59:46 Meeting started Wed Nov 2 17:59:45 2016 UTC and is due to finish in 60 minutes. The chair is amrith. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:59:47 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 17:59:49 The meeting name has been set to 'trove' 17:59:57 anyone here for the trove meeting? 18:00:02 o/ 18:00:03 o/ 18:00:13 o/ 18:00:30 hello ... let's give folks a couple of minutes to come on in 18:01:15 o/ 18:01:27 hi folks 18:01:37 o/ 18:02:20 o/ 18:02:27 ./ 18:02:33 let's get going 18:02:37 we have a longish agenda 18:02:45 #agenda https://wiki.openstack.org/wiki/Meetings/TroveMeeting 18:02:57 #topic Update CONTRIBUTING.rst 18:03:06 (yes, I skipped the review status) 18:03:21 I'm updating the numbers in parallel and will get back to that. 18:04:22 o/ 18:04:28 hi SlickNik 18:04:37 so, one of the things from barcelona 18:04:39 hi folks — sorry running a bit late this morning 18:04:47 We had a discussion in Barcelona and I've summarized that as a change to CONTRIBUTING.rst. See https://review.openstack.org/#/c/392194/ 18:05:04 no worries SlickNik; i'm still running on barcelona time :) and I'm wondering why there's no wine. 18:05:14 please take a minute to review https://review.openstack.org/#/c/392194/ 18:05:47 we've for a long time now been operating loosely on these principles 18:06:03 but several projects are (like us) feeling the need to be more explicit about this. 18:06:16 and I've proposed this language to contributing.rst as we discussed. 18:06:29 I think the text is rather straightforward. 18:07:13 but just in case anyone has concerns about it, let's discuss now. 18:08:15 WF+1'd, so I think we're good ;) 18:08:52 thx peterstac 18:09:01 Yes, it looks good to me — makes perfect sense. 18:09:09 still, if there are concerns, we can always revisit. 18:09:27 so without further ado, ... let's move to the previous item on the agenda 18:09:35 #topic Review status 18:09:42 #link https://docs.google.com/spreadsheets/d/1vJxNaoR3VVNS1Cpiz7U--1zyJRZ6ybMxzJoayrjdduo/edit?usp=sharing 18:10:19 it is clear that we've had a steep decline in the #reviews 18:10:32 and an increase in the number of outstanding change sets. 18:11:06 we will have to do a couple of things. first, I've got to make sure I update these numbers weekly (working on automating that). the second is that we've all got to pick up the pace on reviews. 18:11:27 nothing new or interesting here but I think the next two topics on the agenda may help. 18:12:20 anyone have anything to add ... 18:12:51 hearing none ... 18:12:53 #topic Merge these changes as a priority 18:13:20 OK, so this is one of two things that I'd like to start doing to help move things through in the areas that we have prioritized. 18:13:32 each week, let's have a short list of things that we want to focus reviews on. 18:14:43 in keeping with the projects that we thought we should get in for the ocata release, these are some of the reviews that we need to merge. some of them may have merged already, others may be on the way to merging (have already been verified). but this here is a starting point ... does anyone have questions about any of these, and how do we want to g about merging these 18:15:24 1 and 2 are already merged 18:15:41 7 is merged 18:15:58 8 is in merge conflict ... vgnbkr see https://review.openstack.org/#/c/382493 18:16:16 the rest are in need of reviews 18:16:26 Yeah, F'ing jenkins. 18:17:25 I updated the agenda 18:17:30 please refresh your webpages 18:17:56 the list now consists of the things that are in need of review (modulo the one that vgnbkr has merge conflicts on) 18:18:02 The numbers correspond to the reviews here: 18:18:06 #link https://wiki.openstack.org/wiki/Meetings/TroveMeeting 18:18:13 yes 18:18:17 now the reviews outstanding are: 18:18:25 https://review.openstack.org/#/c/387528 18:18:29 https://review.openstack.org/#/c/381241 18:18:34 https://review.openstack.org/#/c/370971 18:18:39 https://review.openstack.org/#/c/296667 18:18:44 https://review.openstack.org/#/c/382493 18:18:49 https://review.openstack.org/#/c/362338 18:18:54 https://review.openstack.org/#/c/362341 18:18:58 https://review.openstack.org/#/c/388100 18:19:10 and here is what each of those is ... 18:19:11 387528 Change the way redstack figures the default network 18:19:11 381241 Fix RPC Versioning 18:19:11 370971 make trove-dashboard use upper-constraints 18:19:11 296667 Replace obsolete oslo-incubator apiclient 18:19:12 382493 Multi-Region Support (client side, server side has -1's not listed here) 18:19:13 362338 Cluster Upgrade (server side) 18:19:15 362341 Cluster Upgrade (client side) 18:19:17 388100 Guest-agent metadata (spec, code already proposed) 18:19:27 We are required to finish 296667 for Ocata (goal across projects) 18:20:02 the cluster upgrade and multi-region changes are things that we decided to do for Ocata (among other things) 18:20:30 RPC versioning is going to be more important moving forward as we have upgrade now supported and may soon want to start actually versioning some of the API's 18:20:44 387528 is needed when we get to neutron (in the gate) 18:20:57 though I just noticed that the description says redstack which is wrong 18:21:42 for 296667, have we addressed the issues Ronald referred to in changeset 1? 18:22:26 johnma, yes. I think they have to the best that we could 18:22:48 ok sounds good. 18:22:53 the common class approach that he refers to is something that must be done across projects and I've not seen anyone leap forward to do that. 18:23:04 most people have taken the approach (an approach) similar to the one here. 18:23:18 ok 18:23:29 I think this is a stop-gap solution as we all think openstack client should be coming soon 18:25:39 any other thoughts on these ... 18:27:01 I'll see if I can take a look at it today ... 18:27:13 thx peterstac 18:27:41 I should be able to review quite a few of them today as well. 18:27:55 Thanks for coming up with the list — this is helpful. 18:28:04 np SlickNik 18:28:08 I'm going to star them 18:28:14 so it will pop up on the dashboard 18:29:06 ++ 18:29:19 done 18:29:47 I've edited the page to use ordered lists — makes them a bit easier to talk about. 18:30:22 cool thanks SlickNik 18:30:26 yes, much better to read 18:30:46 and for those who use the trove dashboard 18:30:55 https://review.openstack.org/#/dashboard/?foreach=status%253Aopen++%2528project%253Aopenstack%252Ftrove+OR+project%253Aopenstack%252Ftrove%252Dintegration+OR+project%253Aopenstack%252Ftrove%252Dspecs+OR+project%253Aopenstack%252Fpython%252Dtroveclient+OR+project%253Aopenstack%252Ftrove%252Ddashboard%2529&title=Trove+Review+Dashboard&My+Patches+Requiring+Attention=owner%253Aself+%2528label%253AVerified%253C%253D%25 18:30:56 2D1%252Cjenkins+OR+label%253ACode%252DReview%253C%253D%252D1+OR+label%253AWorkflow%253C%253D%252D1+OR+NOT+is%253Amergeable%2529+%2528project%253Aopenstack%252Ftrove+OR+project%253Aopenstack%252Ftrove%252Dintegration+OR+project%253Aopenstack%252Ftrove%252Dspecs+OR+project%253Aopenstack%252Fpython%252Dtroveclient+OR+project%253Aopenstack%252Ftrove%252Ddashboard%2529&Patches+waiting+longer+than+14+days=label%253AVeri 18:30:57 fied%253E%253D1%252Cjenkins+NOT+label%253AWorkflow%253C%253D%252D1+NOT+label%253ACode%252DReview%253C%253D%252D1+is%253Amergeable+age%253A14d&Patches+waiting+longer+than+7+days=label%253AVerified%253E%253D1%252Cjenkins+NOT+label%253AWorkflow%253C%253D%252D1+NOT+label%253ACode%252DReview%253C%253D%252D1+is%253Amergeable+age%253A7d+NOT+age%253A14d&Specs+requiring+review=project%253Aopenstack%252Ftrove%252Dspecs+NOT+ 18:31:02 label%253AWorkflow%253C%253D%252D1++NOT+label%253AVerified%253C%253D%252D1%252Cjenkins+NOT+label%253ACode%252DReview%253C%253D%252D2&Trove%253A+Priority+code+reviews=%2528starredby%253Aamrith+OR+starredby%253Aself%2529+%2528project%253Aopenstack%252Ftrove+OR+project%253Aopenstack%252Ftrove%252Dintegration%2529+NOT+label%253AWorkflow%253C%253D%252D1+is%253Amergeable+NOT+label%253AVerified%253C%253D%252D1%252Cjenkin 18:31:07 s+NOT+label%253ACode%252DReview%253C%253D%252D2&Trove+Client+and+Dashboard%253A+Priority+code+reviews=%2528starredby%253Aamrith+OR+starredby%253Aself%2529+%2528project%253Aopenstack%252Fpython%252Dtroveclient+OR+project%253Aopenstack%252Ftrove%252Ddashboard%2529+NOT+label%253AWorkflow%253C%253D%252D1+is%253Amergeable++NOT+label%253AVerified%253C%253D%252D1%252Cjenkins+NOT+label%253ACode%252DReview%253C%253D%252D2& 18:31:12 Changes+needing+Final+Approval=NOT+label%253AWorkflow%253E%253D1+NOT+label%253AWorkflow%253C%253D%252D1+label%253AVerified%253E%253D1%252Cjenkins+NOT+owner%253Aself+label%253ACode%252DReview%253E%253D2+NOT+label%253ACode%252DReview%252D1+is%253Amergeable&Changes+on+branches+other+than+master=NOT+branch%253Amaster&Needing+attention=%2528%2528%2528label%253AVerified%253C%253D%252D1%252Cjenkins%2529+NOT+label%253AWor 18:31:17 kflow%253C%253D%252D1+NOT+label%253ACode%252DReview%253C%253D%252D2%2529+OR+NOT+is%253Amergeable%2529 18:31:20 what a nice link :) 18:31:22 there are two sections there 18:31:31 trove priority code reviews and client and dashboard priority code reviews 18:31:44 I'll work through some of them as well today 18:31:49 hopefully we can get it done 18:31:58 so unless there are other comments about this, ... 18:32:13 #topic Abandon old reviews 18:32:20 As we discussed at the work session, one of the things I want to do is abandon a number of old reviews that have not seen any action in weeks. If you would like any of these to be retained, please either be at the meeting, or update the agenda below. We will likely NOT go through each one at the meeting so updating the review will not suffice. 18:32:31 Here's the list of reviews that I'm going to abandon 18:32:48 384298 Update .gitignore to ignore .idea of PyCharm 18:32:48 356270 trove-integration should use zuul user 18:32:48 347800 Reduce code duplication 18:32:48 367626 Handle Keystone V3 in module-list 18:32:48 373670 Updated the path of config files 18:32:49 353906 remove tenant_id from the injected guest_info file 18:32:50 347398 Reduce code duplication 18:32:52 370971 make trove-dashboard use upper-constraints 18:32:54 373236 Fix py34 jenkins gate errors 18:32:58 364608 Show flavor name instead of ID in show_instance 18:33:00 365412 Fix is_root_enabled to return True/False 18:33:02 369124 [install-guide]add Nova inject_partition configuration 18:33:04 356701 Create net/subnet for alt_demo tenant 18:33:06 303671 Make timeouts datastore specific 18:33:08 362416 [WIP] Cluster Upgrade for PXC 18:33:10 358693 Trove user page doesn't show allowed hosts AND databases 18:33:12 322826 Convert DIB elements from Fedora to CentOS7 18:33:14 288297 fix delete backup tenant_id error 18:33:16 315619 Superconductor Spec 18:33:18 348743 Add support for module ordering on apply 18:33:20 323590 Add support for Cassandra 2.2 and 3.x 18:33:22 note that this list does NOT include #22 which trevormc clarified 18:33:24 #337914 is pending waiting for a change to merge in oslo. 18:34:10 so, if there are no concerns about these reviews, I intend to make them go away after this meeting with a message that they've been abandoned per discussion at the Trove meeting 11/2/2016 or something short like that 18:35:27 wait 18:35:33 you sue this is a good idea? 18:35:39 Shortened the link: https://goo.gl/ytwZYL :) 18:35:42 I see a number of them that aren't dead... 18:35:45 just waiting for activity 18:36:27 362416 is [WIP] because nobody is reviewing my other cluster upgrade changes. 18:38:32 so how do we tell which is which? 18:38:53 if you have reviews on the list that you'd like to not have abandoned, please update the list in the agenda with a note 18:38:58 and I won't abandon those. 18:39:17 or do you have a better solution to how we can get rid of the ones which are in fact dead, for all intents and purposes. 18:39:58 Maybe we can put a note on the review this week. If folks reply to the note that they're still working on it, we keep the review. If there are no replies, we abandon the reviews next week. 18:40:07 Well as a first step, don't abandon things that are WIP without talking to the person. Afterall, they're WIP. 18:40:37 SlickNik, we talked about that some months ago and realized that putting that note in the review basically updated the review and it "appeared" active in the list :) 18:40:57 hence the idea was to not update the reviews and instead discuss them at the meeting. not saying that's the best solution, just what we talked about. 18:41:47 heh, yeah — we'd have to keep track of the reviews separately (like links off of that Meeting wiki) so we know which ones we're tracking (since we won't be able to use updated time as a metric for stale reviews anymore). 18:42:24 vgnbkr, you are refering to 362416? 18:42:32 Yes 18:42:38 But in general, too. 18:43:26 I don't know about the "in general" part. see https://review.openstack.org/373236 18:43:33 what do you suggest? 18:43:42 the thing has been comatose for a while 18:43:48 Somebody review the other changes that it is waiting on? 18:44:24 yes, that's been the request all along. that people should review changes. 18:44:33 but I'm asking about https://review.openstack.org/#/c/373236/ 18:44:50 what do you suggest? that we contact the submitter for this and each such thing with a WF-1? 18:47:22 anyone have other ideas? 18:50:53 . 18:51:03 I like the idea of having the submitter respond 18:51:29 But am trying to think of a way without updating the review that'll get their attention 18:51:58 The only other way I can think of is emailing them, but that has a different set of disadvantages. 18:52:05 mvandijk, which ones were you thinking were not really dead? 18:52:16 the WIPs 18:52:19 let's just take those off the list 18:52:25 just take them off the list in the wiki 18:52:31 I took vgnbkr's off 18:52:43 I'm sure that there are plenty that aren't WIP 18:53:14 ok, none of mine are on there so I can't really speak to the rest 18:53:28 Could we flag the the with like [TBA] (to ba abandoned) to the review title and if no response to the review for 7 days they are abandoned 18:54:26 ++ amrith: Let's do that — I'll make a separate list for the WIP ones on that page. And I'll track them and put a note in them. If I don't hear back in a week, we can abandon them. 18:54:47 sounds good to me, let's just say we carry this list forward to the next meeting (1 week) 18:54:57 do we want to try and review them in the next week or so and then do the following: 18:54:58 1) if they are good, merge 18:54:59 and if they are still there we abandon them. 18:55:00 2) if it needs more work, update it accordingly with comments and then wait for activity/inactivity and then abandon at that point 18:55:02 would that work? 18:55:11 I think thats fair amrith 18:55:22 amrith: ++ I think that's fair 18:55:33 i'm fine with either johnma's proposal or SlickNik's proposal 18:55:47 to johnma's proposal, we have a challenge getting review bandwidth 18:56:28 do I *really* want to waste review bandwidth on https://review.openstack.org/#/c/384298? no ... 18:56:50 nor 2, 3, 7, 9, ... 18:56:59 but I'm fine with the approach 18:57:06 let's pick one and move forward 18:57:09 what do people feel? 18:57:27 of course, the ones in the ocata review list definitely gets priority. the rest they either sit there for another week or as we get time we review 18:58:02 like, I have this review https://review.openstack.org/#/c/370971; it is on the list to abandon. do I want it abandoned? no. but it needs to be reviewed and that hasn't happened. so I don't want it to just sit there forever. 18:58:07 it should be a quick review 18:58:11 but it hasn't happened ... 18:58:32 and my concern is that the important ones are being lost in the noise 18:58:39 anyway, we're 1m from #end of meeting 18:58:43 so let's wrap up 18:58:49 SlickNik, you'll make a list? 18:59:58 * SlickNik SlickNik to make a candidate list of reviews to abandon and put a note in them. We can track this list and abandon all of the reviews that have no movement in them in a weeks time. 19:00:07 sounds good 19:00:12 #action SlickNik to make a candidate list of reviews to abandon and put a note in them. We can track this list and abandon all of the reviews that have no movement in them in a weeks time. 19:00:19 that's all we hav etime for today so let's call it a wrap 19:00:28 and if there's anything more to talk about move to #openstack-trove 19:00:32 #endmeeting