18:01:06 <SumitNaiksatam> #startmeeting networking_policy 18:01:06 <openstack> Meeting started Thu Apr 2 18:01:06 2015 UTC and is due to finish in 60 minutes. The chair is SumitNaiksatam. Information about MeetBot at http://wiki.debian.org/MeetBot. 18:01:07 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 18:01:09 <openstack> The meeting name has been set to 'networking_policy' 18:01:18 <igordcard_> SumitNaiksatam, hi 18:01:24 <SumitNaiksatam> #info agenda https://wiki.openstack.org/wiki/Meetings/GroupBasedPolicy#April_2nd.2C_March_26th.2C_2015 18:01:47 <SumitNaiksatam> we still have the pending critical bugs from last time 18:01:59 <ivar-lazzaro> hi 18:02:07 <SumitNaiksatam> magesh-gv has made progress, and his UTs are passing on: #link https://bugs.launchpad.net/group-based-policy/+bug/1433530 18:02:08 <openstack> Launchpad bug 1433530 in Group Based Policy "GBP Kilo release should be in sync with Neutron Kilo" [Critical,In progress] - Assigned to Magesh GV (magesh-gv) 18:02:31 <SumitNaiksatam> #link https://review.openstack.org/165377 18:02:37 <SumitNaiksatam> ivar-lazzaro: hi 18:02:48 <SumitNaiksatam> however we will need to test this with the gate job as well 18:03:15 <SumitNaiksatam> ivar-lazzaro: on #link https://bugs.launchpad.net/group-based-policy/+bug/1432779 18:03:16 <openstack> Launchpad bug 1432779 in Group Based Policy "redirect actions don't work with external policies" [Critical,In progress] - Assigned to Ivar Lazzaro (mmaleckk) 18:03:26 <SumitNaiksatam> #link https://review.openstack.org/#/c/164920/ 18:03:34 <SumitNaiksatam> ah, last time magesh said he wanted to test 18:03:39 <SumitNaiksatam> so i think we are blocked on that 18:03:41 <ivar-lazzaro> yes 18:04:03 <SumitNaiksatam> okay i will follow up with him 18:04:13 <SumitNaiksatam> any other bugs we want to discuss today? 18:04:29 <ivar-lazzaro> also, there was the matter about checking whether a service could be provided by an EP or not 18:04:44 <SumitNaiksatam> i forgot to change the topic! ;-) 18:04:47 <SumitNaiksatam> #topic Bugs 18:05:04 <SumitNaiksatam> ivar-lazzaro: ah true 18:05:18 <SumitNaiksatam> ivar-lazzaro: was there a conclusion to that discussion? 18:05:51 <ivar-lazzaro> I think we certainly have to keep in mind that some services (eg. LB) can't be instantiated when the EP is a provider 18:06:10 <ivar-lazzaro> however, we didn't reach a consensus on whether this should be validated or ignored 18:07:02 <SumitNaiksatam> ivar-lazzaro: i am in favor of validation, might not be needed in this patch (can be filed as a follow up bug) 18:07:07 <ivar-lazzaro> we should think of this also in the context of the service chain refactor (assuming there's any on Kilo) 18:07:19 <ivar-lazzaro> SumitNaiksatam: yeah 18:07:25 <SumitNaiksatam> ivar-lazzaro: yes that is on the cards for kilo 18:07:47 <SumitNaiksatam> ivar-lazzaro: the extent of refactor itself can be discussed :-) 18:07:58 <ivar-lazzaro> SumitNaiksatam: so we may want to wait for that before even fixing #link https://bugs.launchpad.net/group-based-policy/+bug/1432779 18:07:59 <openstack> Launchpad bug 1432779 in Group Based Policy "redirect actions don't work with external policies" [Critical,In progress] - Assigned to Ivar Lazzaro (mmaleckk) 18:08:36 <SumitNaiksatam> ivar-lazzaro: hmmm, since you already have patch that is tested, i am inclined go with this fix for now 18:08:56 <ivar-lazzaro> SumitNaiksatam: except it's not tested :) 18:09:19 <ivar-lazzaro> but assuming magesh takes the item, then we could go through with it 18:09:24 <SumitNaiksatam> ivar-lazzaro: okay, point taken, let me rephrase, it does not break existing functionality :-) 18:09:48 <SumitNaiksatam> ivar-lazzaro: do you recommend reducing the priority of this bug? 18:10:02 <SumitNaiksatam> we cannot sit on a critical for this long 18:10:19 <ivar-lazzaro> mmmh no it's actually super important at least as far as the "consumer" use case is cocerned 18:10:26 <SumitNaiksatam> especially since the patch is already posted 18:10:44 <ivar-lazzaro> probably I should have done two different patches 18:11:00 <SumitNaiksatam> ivar-lazzaro: yeah, i was going to say, can you remove the part which you were not able to test functionally? 18:11:36 <ivar-lazzaro> That would be patch set 1 18:11:41 <SumitNaiksatam> ivar-lazzaro: okay 18:11:55 <SumitNaiksatam> ivar-lazzaro: and after that you factored in igordcard_’s comment? 18:11:59 <ivar-lazzaro> I can load another patchset which is a mirror of patchset 1 18:12:06 <ivar-lazzaro> yes 18:12:48 <ivar-lazzaro> then the difference between 1 and 2 could be pushed in a separate branch for magesh to test 18:12:55 <ivar-lazzaro> how does it look? 18:13:01 <SumitNaiksatam> ivar-lazzaro: that works for me, and then between you, magesh and igordcard_ we can figure out how to best test the case (and what constraints we need to check for) 18:13:17 <SumitNaiksatam> ivar-lazzaro: yes what you said ;-) 18:13:21 <SumitNaiksatam> ok moving on 18:13:25 <ivar-lazzaro> ok 18:13:30 <SumitNaiksatam> any other bugs that we need to discuss today? 18:13:39 <igordcard_> alright :) 18:13:58 <SumitNaiksatam> igordcard_: thanks 18:14:14 <SumitNaiksatam> #topic Functional/Integration Tests 18:14:38 <SumitNaiksatam> so my merged to start running the test ingegration tests was merged over the weekend 18:14:49 <SumitNaiksatam> this is currently an experimental job 18:15:01 <SumitNaiksatam> so you have to go “check experimental" 18:15:10 <ivar-lazzaro> SumitNaiksatam: nice! 18:15:23 <SumitNaiksatam> ivar-lazzaro: thanks 18:15:31 <ivar-lazzaro> SumitNaiksatam: is it running actual tests? I've seen it failing a couple of times 18:15:32 <SumitNaiksatam> but it does run a complete end to end tst 18:15:46 <SumitNaiksatam> ivar-lazzaro: yeah, i was getting to the tests 18:16:05 <SumitNaiksatam> currently the test script is in the form of a devstack exercises script 18:16:15 <SumitNaiksatam> so this is not a functional test, its an integration test 18:16:29 <SumitNaiksatam> but it does exercise all the resources 18:16:41 <SumitNaiksatam> it brings up a VM, etx 18:16:44 <SumitNaiksatam> *etc 18:17:00 <SumitNaiksatam> going forward we need to add functional tests 18:17:08 <SumitNaiksatam> and also tempest API tests 18:17:42 <SumitNaiksatam> also we will merges going forward only if the experimental job succeeds 18:18:00 <SumitNaiksatam> (we will be turning the experimental job to a voting job soon) 18:18:07 <SumitNaiksatam> *merge 18:18:15 <rkukura> SumitNaiksatam: What needs to happen before this becomes a voting job? 18:18:19 <SumitNaiksatam> does that sound reasonable to everyone 18:18:34 <SumitNaiksatam> rkukura: i was hoping that we observed the behavior for a little more 18:18:35 <rkukura> SumitNaiksatam: +1 18:18:47 <SumitNaiksatam> rkukura: and my hope was to also have added the functional tests by now 18:18:48 <Yi> +1 18:19:01 <SumitNaiksatam> rkukura: i spent time on it, but i made only limited progress 18:19:10 <SumitNaiksatam> rkukura: Yi: thanks for the confirmation 18:19:23 <yapeng> SumitNaiksatam: for every patch we manually trigger this test, right? 18:19:34 <SumitNaiksatam> yapeng: correct, and i know its painful 18:19:43 <SumitNaiksatam> so apologies in advance 18:19:44 <rkukura> Can we make it automatically trigger, but not vote yet? 18:19:48 <yapeng> no problem. +1 18:19:56 <SumitNaiksatam> but we cant make the job voting until we are confident that the job is solid 18:20:04 <SumitNaiksatam> rkukura: yes 18:20:12 <SumitNaiksatam> rkukura: we can make it non-voting 18:20:40 <SumitNaiksatam> rkukura: then it will be another infra patch to make it voting, so i was trying to save my trouble (and the reviewers time) ;-) 18:21:12 <SumitNaiksatam> okay anything more on that? 18:21:30 <SumitNaiksatam> if you are interested in contributing to the functional/integration tests, please do let me know 18:21:33 <SumitNaiksatam> i can help you get started 18:21:52 <SumitNaiksatam> #topic Packaging Update 18:21:58 <SumitNaiksatam> rkukura: anything at your end? 18:22:33 <rkukura> I havern’t had a chance to update the juno-based fedora packages with the stable release yet, but hope to in the next week 18:22:41 <SumitNaiksatam> rkukura: ok good 18:23:03 <SumitNaiksatam> rkukura: did we miss the end of march deadline for fedora 22? 18:23:23 <SumitNaiksatam> i had checked with magesh, and he said it would not have been possible to get his patch ready over the weekend 18:23:32 <SumitNaiksatam> so i gave up on that 18:23:36 <rkukura> SumitNaiksatam: I have not seen anything furtther on whethe they went with juno or kilo for f22. I need to check. 18:23:43 <SumitNaiksatam> rkukura: okay 18:23:49 <SumitNaiksatam> rkukura: thanks for the update 18:23:54 <SumitNaiksatam> #topic Docs 18:24:03 <rkukura> SumitNaiksatam: quick question 18:24:07 <SumitNaiksatam> #undo 18:24:08 <openstack> Removing item from minutes: <ircmeeting.items.Topic object at 0x912f250> 18:24:11 <SumitNaiksatam> rkukura: go head 18:24:17 <SumitNaiksatam> *ahed 18:24:20 <SumitNaiksatam> *ahead 18:24:39 <rkukura> Are there any additional patches we should include in an additional stable release before it gets packaged? 18:26:11 <SumitNaiksatam> rkukura: not that i am aware of 18:26:21 <rkukura> ok, thanks 18:26:22 <SumitNaiksatam> i already tagged the new stable 18:26:28 <SumitNaiksatam> but good point 18:26:42 <rkukura> I saw that for gbp, and assume the other repos also have stable releases 18:26:42 <SumitNaiksatam> if anyone wants to backport anything more to the stable, please speak up at the earliest 18:26:48 <SumitNaiksatam> rkukura: yes 18:27:17 <SumitNaiksatam> #topic Docs 18:27:24 <ivar-lazzaro> SumitNaiksatam: do we need to backport EP providing chains? 18:27:42 <SumitNaiksatam> ivar-lazzaro: your call 18:28:04 <ivar-lazzaro> I think we should 18:28:06 <SumitNaiksatam> ivar-lazzaro: i am good with backporting it since, its a contained change 18:28:11 <SumitNaiksatam> ivar-lazzaro: ok 18:28:16 <ivar-lazzaro> ad least for the consuming (I said providing before) at least 18:28:39 <SumitNaiksatam> so in that case i think we will have to wait for Ivar to post a new patch, wait for it to merge, and then do the backport 18:28:52 <SumitNaiksatam> lets plan to wrap this up in a day, since we had already reviewed the first patchset 18:29:06 <SumitNaiksatam> on the docs 18:29:23 <SumitNaiksatam> we do have readthedocs now building our docs: 18:29:24 <SumitNaiksatam> http://group-based-policy.readthedocs.org/en/latest/ 18:29:37 <ivar-lazzaro> SumitNaiksatam: (The new patch is there btw) 18:29:45 <SumitNaiksatam> its a different matter that we need to update the docs themselves (a little bit) :-) 18:30:12 <SumitNaiksatam> ivar-lazzaro: thanks 18:30:36 <SumitNaiksatam> the docs are currently being built in readthedocs on demand 18:30:51 <SumitNaiksatam> there is an infra job to automate that too, so we can do that eventually 18:30:57 <SumitNaiksatam> moving on 18:31:03 <igordcard_> SumitNaiksatam, where is the source for them? 18:31:18 <SumitNaiksatam> igordcard_: its in the tree 18:31:42 <SumitNaiksatam> #link https://github.com/stackforge/group-based-policy/tree/master/doc/source 18:31:54 <igordcard_> SumitNaiksatam, right! thanks 18:32:02 <SumitNaiksatam> so please feel free to suggest updates and additions 18:32:23 <SumitNaiksatam> you will need to submit a new patch just like any other code patch 18:33:00 <SumitNaiksatam> #topic GBP Project Proposal 18:33:23 <SumitNaiksatam> we had removed the project proposal patch from WIP last week 18:33:31 <SumitNaiksatam> and then we put it back to WIP ;-) 18:33:41 <SumitNaiksatam> there was a discussion in the TC meeting on this proposal 18:34:58 <SumitNaiksatam> the question asked was whether GBP was circumventing neutron 18:35:47 <SumitNaiksatam> since in the current model, it seems that the TC wants to preserve the interactions to the core infrastructure pieces through the current established projects (like neutron, nova, etc) 18:36:17 <SumitNaiksatam> some of the team members engaged in the discussion and provided clarifying comments 18:36:32 <SumitNaiksatam> we do not know if the opinion has changed or now 18:36:36 <SumitNaiksatam> *not 18:36:50 <SumitNaiksatam> so currently the patch is still in WIP, and we will try to work through this 18:36:55 <SumitNaiksatam> questions/comments? 18:37:09 <rkukura> SumitNaiksatam: Is it still on the agenda for the next TC meeting? 18:37:34 <SumitNaiksatam> rkukura: i do see it on the agenda, but i think it made it there by default (i did not request it) 18:38:02 <SumitNaiksatam> rkukura: we have to decide if we need more time, and accordingly we can request ttx to remove it from the agenda 18:38:35 <ivar-lazzaro> SumitNaiksatam: do we plan on addressing their concerns in a separate session / code walkthrough? 18:38:40 <rkukura> I’m thinking we need more time to really understand the objections and what we need to do about them. 18:38:52 <SumitNaiksatam> rkukura: i am very much agreement 18:39:02 <ivar-lazzaro> rkukura: +1 18:39:22 <SumitNaiksatam> ivar-lazzaro: i do not believe there is a process for that, but i think its a great idea if they have the time and inclination 18:40:03 <ivar-lazzaro> I think mainly there are some still seeing GBP as it was 1 year ago. It changed quite a bit but obviously we need to make a greater effort sending the message through 18:40:36 <SumitNaiksatam> ivar-lazzaro: sure 18:40:48 <SumitNaiksatam> so everyone in this team is empowered to send that message 18:41:16 <hemanthravi> ivar +1, need to reiterate GBP is a app oriented API that uses neutron to realize the networking required 18:41:44 <SumitNaiksatam> we do need to coordinate a little bit, but beyond that, please feel free to suggest options and follow up 18:42:25 <SumitNaiksatam> hemanthravi: do you have a forum in mind that you think we need to send that message to (and we havent already)? 18:43:16 <hemanthravi> SumitNaiksatam:this was done at the tc meeting, not sure of any other forum 18:43:24 <SumitNaiksatam> hemanthravi: yes 18:43:44 <SumitNaiksatam> okay if there are no more comments on this, we need to get to the remaining items 18:43:51 <SumitNaiksatam> #topic Re-factor Group Based Policy with Neutron RESTful APIs 18:44:04 <SumitNaiksatam> so #link https://review.openstack.org/#/c/156776 merged 18:44:20 <SumitNaiksatam> and we are waiting on: #link https://review.openstack.org/#/c/156856 18:44:35 <SumitNaiksatam> Yi: yapeng over to you 18:44:49 <Yi> I was syncning with Ivar on a couple of his concerns 18:45:18 <Yi> but otherwise, should have new update very soon 18:45:28 <SumitNaiksatam> Yi: thats great! 18:46:11 <SumitNaiksatam> Yi: yapeng: was there anything more you wanted to discuss? 18:46:18 <ivar-lazzaro> Yi: it's been an awesome job! ping me if you have questions on my comments 18:46:41 <yapeng> SumitNaiksatam, no issue right now. 18:46:53 <Yi> SumitNaiksatam: no -- but if other members have concerns, please make your comments now 18:47:08 <Yi> ivar-lazzaro: I will ping you right after this meeting 18:47:19 <SumitNaiksatam> Yi: yapeng thanks 18:47:39 <SumitNaiksatam> rkukura: it will help if you can take a look at https://review.openstack.org/#/c/156856 as well 18:47:55 <rkukura> SumitNaiksatam: I will 18:48:03 <SumitNaiksatam> rkukura: great 18:48:22 <SumitNaiksatam> i was expecting a longer discussion on this, hence try to hurry up earlier :-) 18:48:43 <SumitNaiksatam> oh regarding the floating IP support 18:49:19 <SumitNaiksatam> i think magesh is doing that on the background but needs the Kilo-parity to work before he can get the floating IPs going 18:49:32 <SumitNaiksatam> klio-parity with Neutron that is 18:49:44 <SumitNaiksatam> #topic Open Discussion 18:50:25 <SumitNaiksatam> did we miss anything today, or anyone want to circle back to the earlier topics? 18:50:37 <SumitNaiksatam> or get 10 mins of our time back? ;-) 18:51:18 <SumitNaiksatam> ivar-lazzaro: anything you wanted to discuss on the hands on lab? 18:51:39 <SumitNaiksatam> for vancouver summit 18:51:42 <igordcard_> SumitNaiksatam, I dropped a comment at https://review.openstack.org/#/c/149798/ a few days ago 18:52:04 <ivar-lazzaro> SumitNaiksatam: I'd like to set up a discussion for when I'm back from Italy 18:52:28 <SumitNaiksatam> igordcard_: sorry did not respond, i will get back to you, the issue is reproducible 18:52:29 <ivar-lazzaro> just to make sure we are well organized and understand how we want to drive the session 18:52:34 <SumitNaiksatam> ivar-lazzaro: sure 18:53:09 <SumitNaiksatam> igordcard_: that bug is in the context of shared resources 18:53:10 <ivar-lazzaro> obviously we want to show the awesome power of rainbow intents, but I'd like to sink up with the core team on the actual set of demonstrations we want to show 18:53:23 <SumitNaiksatam> ivar-lazzaro: yes 18:53:37 <s3wong> ivar-lazzaro: great 18:54:22 <igordcard_> SumitNaiksatam, okay... will reproduce it with shared resources then 18:54:29 <SumitNaiksatam> igordcard_: sorry i stand corrected 18:54:40 <SumitNaiksatam> igordcard_: i need to go back and check 18:54:46 <SumitNaiksatam> igordcard_: i will follow up 18:55:06 <SumitNaiksatam> alright, thanks everyone 18:55:08 <igordcard_> SumitNaiksatam, thanks 18:55:17 <SumitNaiksatam> bye! 18:55:20 <igordcard_> bye all 18:55:22 <Yi> later 18:55:26 <ivar-lazzaro> bye 18:55:31 <SumitNaiksatam> #endmeeting