18:03:34 #startmeeting networking_policy 18:03:35 Meeting started Thu Mar 12 18:03:34 2015 UTC and is due to finish in 60 minutes. The chair is SumitNaiksatam. Information about MeetBot at http://wiki.debian.org/MeetBot. 18:03:36 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 18:03:38 The meeting name has been set to 'networking_policy' 18:03:57 #info agenda https://wiki.openstack.org/wiki/Meetings/GroupBasedPolicy#March_12th_2015 18:04:26 so i tweaked the meeting format a wee bit and added a “standing items” agenda at the top 18:04:38 #link https://wiki.openstack.org/wiki/Meetings/GroupBasedPolicy#Standing_items 18:04:46 hello 18:04:59 yapeng: hemanthravi rkukura: hi 18:05:10 hi 18:05:11 SumitNaiksatam: hi 18:05:19 we can add/delete from the standing items as necessary 18:05:19 the meeting time has changed? 18:05:31 yapeng: daylight savings 18:05:47 UTC time has not changed 18:05:54 yapeng: Its UTC time hasn’t changed 18:06:01 SumitNaiksatam: oh, i forgot this :) 18:06:25 yapeng: i sent out a reminder email, but i think i forgot to add you and Yi, apologies 18:06:35 #info Kilo-2 is March 16th 18:07:06 with that lets quickly run through the standing agenda items 18:07:13 #topic Bugs 18:08:01 I have not posted the new stable/juno yes because we still have some reviews pending which targeted for this 18:08:04 #link https://wiki.openstack.org/wiki/GroupBasedPolicy/kilo-bug-squash 18:08:43 rkukura: i believe ivar-lazzaro had comments on each of your two patches 18:09:41 yes, on https://review.openstack.org/#/c/161619/, he wanted me to change the UT approach 18:10:15 rkukura: okay, so once we wrap up these two bugs, and backport, i think we can cut the new stable 18:10:20 And just a nit on https://review.openstack.org/#/c/161968/ 18:10:39 rkukura: i will look out for the new patches (other cores please look out as well) 18:10:45 Changing the UTs would take me some time - probably next week 18:11:26 With neutron kilo-3 coming up, I need to prioritize that right now 18:12:09 rkukura: ivar-lazzaro: is there any middle ground on this? i would hate to delay the stable/juno even further 18:12:32 I’d think we could file a bug to improve the UT later 18:12:46 rkukura: okay, ivar-lazzaro does that work for you? 18:13:31 sure 18:13:31 I could also add a REVISIT comment in the current UT 18:13:42 rkukura: either one works for me 18:14:04 I’ll respin both patches today, adding the REVISIT and fixing the nit. 18:14:21 great, lets try to close this today, and the backport by tomorrow, so i can cut the new stable/juno over the weekend 18:14:24 rkukura: thanks! 18:14:33 SumitNaiksatam: sounds good 18:14:48 i dont think the bug which magesh-gv has is critical for the backport 18:15:04 okay, any other bugs that we need to discuss today? 18:15:23 i just noticed that the links i have posted for the bug categories are not working any more, will fix 18:15:38 i meant links in the meetings page 18:16:04 #topic Functional/Integration Tests 18:16:43 may be i will spend a couple of mins here explaining to the team what we need to do 18:17:33 first we need to configur an openstack-infra gate job such that it will trigger a run on every patch-set (just like we have pep8 and py27 UT jobs today) 18:17:39 for this i had posted the patch: 18:17:46 #link https://review.openstack.org/#/c/161511 18:18:04 still waiting for reviews on that 18:18:27 currently this is being proposed as an experimental job 18:18:51 which means it will not run automatically, you would have to do a “check experimental” to trigger it 18:19:20 once we are comfortable that the job works as desired, we will make it voting on all patches 18:19:48 for this job to run, we need to implement pre- and post-job hooks in GBP as well 18:20:05 i have posted a patch for this in GBP: #link https://review.openstack.org/#/c/161532 18:20:21 the above scripts are to help customize the job for GBP 18:20:51 the job can trigger all kinds of functional and tempest tests 18:21:06 nice 18:21:30 jishnub is working on adapting his test suite such that the job will run those tests 18:21:54 i would characterize those tests more in the functional category 18:22:03 we still need the tempest tests 18:22:37 good development here is that other projects have started moving their tempest tests from the tempest repo to the individual project repos 18:22:49 so there is some precedence that we can use 18:23:02 in terms of how this is done, and how the tempest lib is used 18:23:13 of course, we need to write the tempest tests as well 18:23:29 SumitNaiksatam: Are tempest-speciifc clients required, or can we use python-gbpclient? 18:24:17 rkukura: i am not clear on that level of detail, but i believe tempest used to have a client wrapper inside to make rest calls? 18:24:35 rkukura: i dont think that has changed 18:24:46 SumitNaiksatam: That’s what I was refering to 18:25:23 rkukura: so good point, we need someone to do this investigation, and hence... 18:25:33 any volunteers to work on this? ;-) 18:25:53 this is a critical area of the project, and will need ongoing effort and committment 18:26:08 we dont have to decide here, but please let me know if you would like to pitch in 18:26:35 as it stands, this is the responsibility of the entire team - so if you write a new feature, you sign up to write the tempest tests for that as well! 18:26:57 SumitNaiksatam: +1 on that 18:27:08 of course, we need to have the framework in place first, but assuming we have that 18:27:16 ivar-lazzaro: ack 18:27:51 any questions, on the above? 18:28:35 this aspect has been top of mind for a long time, but we really have to get this done now, and we will track the progress on a weekly basis 18:29:07 okay moving on :-) 18:29:20 #topic Packaging update 18:29:29 rkukura: anything significant to report? 18:29:53 we already discussed the stable branches 18:29:58 nothing recent - plan to update fedora and RDO once the stable/juno release happens 18:30:07 rkukura: ah so one more on that 18:30:19 i notice susaant is here too 18:30:33 we owe susaant at least three reviews on the heat/automation side 18:30:43 and couple of those are backport-worthy 18:31:01 susaant: apologies, i believe you have been patiently waiting for almost a month now 18:31:24 SumitNaiksatam: Not a problem. 18:31:27 rkukura: so we will need to include those in the stable/juno for gbp-automation 18:31:35 i will coordinate with susaant today and tomorrow 18:31:48 susaant: thanks! 18:31:54 #topic Docs 18:31:55 Will we be releasing for all 4 repos? 18:32:04 #undo 18:32:06 Removing item from minutes: 18:32:13 rkukura: yes 18:32:17 thanks 18:32:29 i have a patch that i need to post on UI side that i have been sitting on 18:32:46 so we need to get that reviewed and merged as well 18:33:00 phew! 18:33:38 #topic Docs 18:33:57 i dont have anything specific on this for today, but i wanted to note this as a standing agenda item 18:34:23 over course of time we need to track that each feature is documented (just like tempest test support) 18:34:53 btw, if you haven’t noticed already, we do have in-tree documentation: 18:34:56 #link https://github.com/stackforge/group-based-policy/tree/master/doc/source 18:35:24 would really appreciate more feedback on this and feel free to propose patches to edit/update this 18:36:22 now coming to the agenda items specifically for today 18:36:36 #topic GBP Project Proposal 18:36:44 #link https://review.openstack.org/#/c/161902/ 18:36:57 per discussion in last meeting, we went ahead and put forward the proposal 18:37:12 please read and feel free to comment 18:37:26 it should not surprise any one, since its mostly what we have been discussing all along 18:38:34 i changed this to WIP earlier today since there were some comments, and it was felt that we should address them before proposal gets discussed in the TC meeting 18:38:51 any thoughts on the above? 18:39:51 one of the things that we needed to do was to have the PTL elections 18:39:56 Are we going to wait for temest-based gate tests before removing WIP? 18:40:41 rkukura: at this point i think we would want the gate job to be running 18:41:05 SumitNaiksatam: Do you mean the current no-op test? 18:41:07 i do not believe this is an explicit requirement though 18:41:23 is that the only reqmt for removing wip 18:41:35 rkukura: perhaps more than just the no-op test :-) 18:41:48 hemanthravi: I think the PTL election is another one 18:41:51 hemanthravi: we have proactively put this into WIP 18:42:09 ivar-lazzaro: true 18:42:36 though even on that we did not get -1 18:42:37 Those seem reasonable requirements to me 18:43:08 since there is no prior precedence of PTL elections for stackforge projects 18:43:13 ivar-lazzaro: yeah 18:43:20 hence we proactively put into WIP 18:43:38 does everyone agree with the approach, or have other suggestions? 18:43:40 That's what we are here for right? setting precedents ;) 18:43:48 ivar-lazzaro: nice one! ;-) 18:44:15 hemanthravi: to your question, i dont know 18:44:47 hemanthravi: this is an evolving topic, so if you see the TC meeting last week, there were a number of things which came up during the discussion of the Magnum proposal 18:45:06 though to the best of my knowledge there are not new requirements, yet 18:45:39 i believe the TC will evaluate a few more proposals in the next meeting, and take if from there 18:45:59 we are running behind schedule today 18:46:22 #topic GBP PTL elections 18:46:24 #link http://lists.openstack.org/pipermail/openstack-dev/2015-March/058783.html 18:47:00 there is one candidate so far: #link http://lists.openstack.org/pipermail/openstack-dev/2015-March/058925.html 18:47:22 I expect the wiki page: #link https://wiki.openstack.org/wiki/GroupBasedPolicy/PTL_Elections_Kilo_Liberty 18:47:33 to be updated with the list of candidates and the results 18:48:10 if you have a commit in the GBP project, you get to vote 18:48:29 anye questions/comments on the above? 18:48:50 SumitNaiksatam: Do you know when the polling starts/ends? 18:49:10 --> March 17, 2015 - 1300 UTC March 24, 2015: PTL elections 18:49:12 rkukura: i think its mentioned on the wiki, i believe starts on march 18th, ends on march 24th 18:49:21 oh march 17th then 18:49:27 ivar-lazzaro: thanks 18:49:47 I see - thanks 18:50:05 moving on 18:50:08 #topic Re-factor Group Based Policy with Neutron RESTful APIs 18:50:18 i know yapeng and Yi are patiently waiting 18:50:30 so we spent significant amount of time on the reviews yesterday 18:50:38 ivar-lazzaro: and rkukura thanks for the reviews 18:50:52 Yi: and yapeng thanks for your time and indulging as well 18:51:08 Yi: yapeng do we need to discuss anything today, or are you good? 18:51:28 thanks for all the comments. I think we are all good for now 18:51:28 SumitNaiksatam: I am good here 18:51:29 https://review.openstack.org/#/c/159725, https://review.openstack.org/#/c/156776, https://review.openstack.org/#/c/156856 18:51:40 yapeng: Yi: sweet 18:51:53 #topic Floating IP support 18:52:12 magesh-gv has been sick for the past couple of days, so he is not able to make it 18:52:22 he still working on this though, and has sent an update 18:53:02 he says - “For floating IP support I am going ahead with Nat Pool and use admin context to do the floating IP create and attach in single step. “ 18:53:34 he also says - “The ability to specify floating IP in create operation is added recently only.” (he is referring to Neutron) 18:54:07 i was not sure about the later 18:54:15 so need to follow up with him 18:54:35 rkukura: if you are familiar with the changes there, please respond to the email thread 18:55:05 SumitNaiksatam: I’m not, so will need to investigate a bit 18:55:10 rkukura: okay 18:55:20 i will skip task flow investigation for today 18:55:29 #topic Cross Project Liaisons 18:56:20 as we are growing, and aspire to enhance our interaction with other OpenStack projects, we need to have people who will take lead on these 18:57:37 at the very least, we need to track the meeting logs of some the other projects we depend on, or have interlocks 18:58:16 so four projects come to mind upfront 18:58:24 oslo, neutron, keystone and nova 18:58:42 heat and horizon? 18:58:54 rkukura: you bet, sorry i missed those 18:59:02 SumitNaiksatam: I’d consider oslo and/or nova 18:59:18 rkukura: great, thanks 18:59:32 rkukura: i thought you would want to look at neutron since you are deeply involved :-) 18:59:48 :( 19:00:09 of course, anyone can look at any project, but we want one go-to person 19:00:18 so rkukura for oslo and nova 19:00:19 SumitNaiksatam: I’m lucky to keep up with just ML2 in neutron 19:00:25 rkukura: :-) 19:00:40 let me know if you have interest in other areas 19:00:45 #topic Open Discussion 19:01:01 ivar-lazzaro: had an excellent suggestion that we should plan the mid-cycle code sprint 19:01:08 I have one item 19:01:17 rkukura: please go ahead 19:01:50 Given that neutron has postponed the pecan BP post-kilo, should we consider having our master track neutron master, at least until we become a standalone server? 19:02:46 Couple of benefits: 19:02:57 rkukura: that requires some refactoring as well (although it's not the same amount of work) 19:03:02 rkukura: absolutely, we have to do that 19:03:08 1) We’ll be ready to take advantage of kilo features, especially the subnet pools 19:03:15 rkukura: that is a major item on to-do list 19:03:43 2) fedora rawhide, RDO, and probably other distros are packaging kilo, and we get left out for now 19:04:24 3) tracking only juno could come up during project proposal review 19:04:27 thats all 19:04:34 rkukura: agree 19:04:47 rkukura: and that is a kilo item we have to accomplish 19:05:01 rkukura: we just need to figure out the right sequence of refactor activities 19:05:12 rkukura: lets discuss on this, and get this rolling at the earliest 19:05:22 magesh-gv also brought this up in his email 19:05:46 regarding the code-sprint i am thinking end of March/beginning of April 19:05:48 I’m not saying we should postpone becoming standalone beyond our kilo, but if that happens, we are probably better of tracking kilo now than just at the end 19:06:02 SumitNaiksatam: I'll be in Italy at that time :) 19:06:05 rkukura: absolutely 19:06:08 ivar-lazzaro: oops 19:06:26 ivar-lazzaro: okay, lets coordinate in #openstack-gbp then 19:06:30 SumitNaiksatam: however I can attend remotely, no problems 19:06:47 rkukura: +1 19:06:55 we need to get all the specs sorted out before we get to code sprint 19:07:14 i noticed that nbouthors also joined earlier 19:07:22 not sure if he had an update 19:07:26 I’d be willing to travel for a sprint, but 1st week of April wouldn’t work for me I think 19:07:46 rkukura: ah okay, lets coordinate, i just wanted to throw in the idea 19:08:03 okay i will sync up with nbouthors offline 19:08:05 thanks all 19:08:09 bye 19:08:14 ciaoooo 19:08:18 bye 19:08:34 #endmeeting