18:03:54 #startmeeting Networking FWaaS 18:03:55 Meeting started Wed Dec 11 18:03:54 2013 UTC and is due to finish in 60 minutes. The chair is SumitNaiksatam. Information about MeetBot at http://wiki.debian.org/MeetBot. 18:03:57 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 18:04:00 The meeting name has been set to 'networking_fwaas' 18:04:07 thanks everyone for joining 18:04:56 the main agenda for today's meeting was to sync up with the PTL, my understanding was that he was going to join this meeting 18:05:24 however, i don't see mark on IRC 18:06:33 Hmm! should we wait for a bit to see if he will join 18:06:51 yeah 18:07:03 so i just pinged the -neutron channel 18:07:08 don't see him there either 18:07:13 bummer :-( 18:07:54 #topic service_type framework 18:08:08 https://review.openstack.org/#/c/60699/ 18:08:10 I submitted code review 18:08:16 garyduan thanks for patch 18:08:20 working on unit test 18:08:24 ok great 18:08:41 I see tempest test are failing 18:09:01 ok, tests are failing on what? 18:09:25 in fact, gate-neutron-python27 18:09:48 I guess it's unit test issue 18:09:53 I will fix it, no problem 18:09:58 ok 18:10:11 its breaking existing UT? 18:10:21 right 18:10:32 ok, that should be easier than tempest :-) 18:10:36 seems like, I will double check 18:10:43 ok thanks 18:10:51 i heard from one of my colleagues (pcm_) that he was seeing a lot issues with tox (if he upped the memory to 8G on his VM) things were okay 18:11:06 perhaps something to check - not sure if u have the same issue 18:11:10 SridarK: yes, i was following that 18:11:28 SridarK: however, gduan's tests are failing in the gate 18:11:34 ok 18:11:41 SridarK: and that infra should be set up properly 18:11:52 true, stand corrected 18:11:55 its a regression 18:12:12 we need tempest on this review too, right? 18:13:06 gduan: absolutely 18:13:13 ok 18:13:22 gduan: we need it for existing resources and api, and anything new that we add 18:13:47 gduan: however, we have to first get the feature merged before the tempos tests can be merged 18:13:57 its a kind of a chicken and egg 18:14:05 ok. I will add unit test first. 18:14:33 gduan: yeah, thats a good start (rather must :-P) 18:15:51 okay, i was still hoping that mark would join before we go to the next item in the agenda 18:16:16 #topic Service Objects 18:16:29 yisun: how is this coming along? 18:16:38 #link https://blueprints.launchpad.net/neutron/+spec/fwaas-customized-service 18:16:59 had most code, but stuck on some db access issues 18:17:05 yisun: ok 18:17:06 still trying to find out reason 18:17:19 yisun: do we still plan to target this for Icehouse? 18:17:32 I will try 18:17:34 ok 18:17:59 yisun: the reason i ask is, the blueprint's series goal is not set 18:18:15 really, I did it once 18:18:22 ok, I will set it again 18:18:27 yisun: this will not show up on the PTL's radar if the series goal and milestone target is not set 18:18:45 yisun: i had requested gduan to convey this to you last week 18:19:02 I will put I3 18:19:16 yisun: ok 18:19:21 yisun: that said i am not trying to push you for this 18:19:31 :-) 18:19:45 But I may need some help to fix the DB 18:19:49 yisun: since you wanted to get this in, i want to make sure that the procedural issues are taken care off correctly 18:20:02 yisun: and things don't slip through the cracks 18:20:03 no problem 18:20:25 My current issue is that if the zone can not get in ice houst 18:20:34 then we may have to push all to J 18:20:43 yisun: good segue 18:20:57 #topic zones 18:21:05 #link https://blueprints.launchpad.net/neutron/+spec/fwaas-zones-api 18:21:40 yisun: per last meeting, general feeling was to keep this in discussion for I and target patch for merge in J 18:22:12 ok 18:22:16 this is so that we don't introduce too much feature churn, and avoid disrupting the ongoing efforts on Neutron stabilization and parity for nova networking 18:22:59 we also need to do our bit in terms of tempest tests 18:23:01 Sridar/others: Can we meet f2f on this? 18:23:29 BrianTorres-Gil: you were not in the meeting last time, hope you are in the loop on this 18:23:30 Just want to split the work so that I can start working on it 18:23:52 SumitNaiksatam: RajeshMohan: Hoping we can do this tomorrow on the tempest tests 18:24:47 ok 18:24:59 Or we can chat after the meeting 18:25:11 RajeshMohan: lets do it after this meeting 18:25:26 Sridar: Any updates from last week? 18:25:41 okay, i guess BrianTorres-Gil is in silent agreement 18:25:45 You said that you were updating the document - any help on that? 18:26:07 RajeshMohan: i believe the plan was to discuss this week (but i will let SridarK respond) 18:26:09 RajeshMohan: Still need to figure out on the dependencies - will sync with u 18:26:20 ok, thanks. 18:26:26 hopefully tomorrow we can get to more clarity 18:27:25 Yea, works for me 18:28:24 Sumit, sorry, I was driving, if we push zone to J, I may push service object to J too 18:28:32 BrianTorres-Gil: ok thanks, i was just going to say that you were in silent agreement :-) 18:28:43 yisun: okay, we can discuss 18:29:28 summit, I'm not push it 18:29:46 We do need to have tempest done fist 18:30:03 I agree with everyone on it 18:30:23 making the system stable is more important 18:30:51 yisun: great, glad that we all agree on that 18:31:04 :-) 18:31:44 the other two items on the blueprint agenda were FWaaS service insertion, and apply/commit operation 18:32:09 there is nothing new here that we have not discussed within the team yet 18:32:33 so this was mostly discussion with the PTL 18:32:55 any new developments/thoughts you want to add on this? 18:33:54 any updates on tempest? 18:33:55 SumitNaiksatam: On Service insertion - have been in discussion with colleagues as well and we are on board 18:34:35 Sumit: last week you said you neeed some help 18:34:47 SridarK: thanks 18:35:06 RajeshMohan: yes, help always welcome :-) 18:35:14 let's discuss 18:35:22 Sumit: ok 18:35:24 btw, this patch was posted: 18:35:27 https://review.openstack.org/#/c/60190/ 18:35:38 is gongysh around? 18:36:50 we already have a blueprint for this activity, to incorporate service insertion (not just router binding) 18:37:02 #topic Service Insertion for Firewall 18:37:10 #link https://blueprints.launchpad.net/neutron/+spec/fwaas-service-insertion 18:37:33 so the patch https://review.openstack.org/#/c/60190/ will need to reconcile with this effort 18:37:48 i think gongysh is not around 18:37:55 ok moving on 18:38:10 #topic bugs 18:38:27 i just noticed this: 18:38:28 https://bugs.launchpad.net/neutron/+bug/1258438 18:38:29 Launchpad bug 1258438 in neutron "Can't create a firewall per tenant" [High,Confirmed] 18:39:13 we need to fix this asap 18:39:32 it seems that the bug reported has taken this up 18:39:57 akihiro has confirmed it 18:40:27 did anyone test with more than one tenant in their earlier tests? 18:40:45 i have to go back and check if there wasn't a UT for this case 18:41:07 SumitNaiksatam: I think Dan had done some of those tests 18:41:16 SridarK: okay 18:41:23 anyone has time to confirm this? 18:41:31 I have tested with more than one client 18:41:40 Initially 18:41:48 must be some new change 18:41:50 RajeshMohan: you mean, more than one tenant, right? 18:42:22 Yes - I have demos working with more than one client - close to havana release though 18:42:40 RajeshMohan: i think the check for limiting to one firewall per tenant went in later (but in Havana) 18:42:44 did we add any code to restrict number of firewall recently 18:42:56 RajeshMohan: it was towards the end of havana 18:43:11 I must have missed that change 18:43:11 RajeshMohan: if this bug is correct, it was that code which introduced the issue 18:43:40 at any, rate any one have time to verify this (you can follow up with the fix as well) 18:43:56 I will confirm today 18:44:07 RajeshMohan: great, thanks! 18:45:02 SridarK: any plans on this: https://bugs.launchpad.net/neutron/+bug/1250841 18:45:06 Launchpad bug 1250841 in neutron "Move FWaaS Noop driver to unit tests directory" [Low,Triaged] 18:45:14 we have this marked for I2 18:45:19 would be nice to get it out of the way 18:45:24 SumitNaiksatam: will push a patch out soon 18:45:44 needed a few clarifications b4 getting it out - will reach out to u 18:46:13 SridarK: thanks 18:46:19 SridarK: we also have: https://bugs.launchpad.net/neutron/+bug/1223472 18:46:20 Launchpad bug 1223472 in neutron "firewall status does not become ACTIVE when a router does not exist" [Undecided,Triaged] 18:46:29 i believe we are waiting on the insertion to do this? 18:47:45 SumitNaiksatam: I think some clarification needed to be provided 18:47:51 SridarK: ok 18:47:53 need to jog my memory on this 18:48:11 SridarK: alright, lets track it for discussion next time 18:48:21 will update - i recall that this was not an issue but let me clarify and take care of this 18:48:27 gduan: https://bugs.launchpad.net/neutron/+bug/1222968 18:48:28 SumitNaiksatam: thanks 18:48:28 Launchpad bug 1222968 in neutron "Namespace handling is missing in vArmour's FWaaS agent and driver" [Medium,In progress] 18:49:44 I saw there is a fix 18:49:56 but it has been pushed out from rc1 by mark 18:50:07 ok 18:50:12 Sumit, what should be do about it? 18:50:22 if this is fixed, then we should invalidate this bug 18:51:08 #topic vendor plugins/drivers 18:51:08 Gary submitted patch 18:51:14 #undo 18:51:15 Removing item from minutes: 18:51:16 but is was not merged to rc1 18:51:37 and it is marked as back port potential 18:51:40 SumitNaiksatam: we can discard this bug 18:51:50 gduan: ok, you can invalidate it 18:52:00 #topic vendor plugins/drivers 18:52:02 SumitNaiksatam: we will cover it with another BP 18:52:35 gduan yisun SridarK RajeshMohan BrianTorres-Gil: anything we need to discuss regarding your plans 18:52:48 We have a BP for rework our fw driver as a plugin driver 18:52:52 is there any dependency or blocker for you to move forward? 18:53:03 gduan: ok, bp link? 18:53:28 SumitNaiksatam: will need to push our BP out to J with a dependency on Zones 18:53:36 https://blueprints.launchpad.net/neutron/+spec/varmour-firewall-plugin 18:53:40 SridarK: okay, got it 18:53:46 gduan: thanks 18:54:05 It depends on the service type framework BP 18:54:07 We are spinning our wheels a lot trying to figure out some of the code and structure of Neutron. Still seeking a good developers guide. 18:54:37 Might also depend on Service Insertion 18:54:47 gduan: good, i think the PTL has also approved this for I3, so the loop is closed 18:54:59 BrianTorres-Gil: ok 18:55:24 BrianTorres-Gil: welcome to the world of open source :-P 18:55:44 BrianTorres-Gil: let the rest of the team know if you need help with anything particular 18:56:09 SridarK (and others in general): it will help to clearly state in your blueprint as to what the dependencies are 18:56:27 that will help us to prioritize features accordingly, and make their case 18:56:35 SumitNaiksatam: ok will update the dependency 18:56:49 SridarK: can you add a dependency to the zones bp if appropriate? 18:56:49 thanks 18:57:25 gduan: similar can you please put a dependency on the service type framework and service insertion blueprints 18:57:57 SumitNaiksatam: just added 18:58:42 gduan: thanks, don't see the dependency on service insertion 18:58:47 anyway 18:58:49 moving on 18:58:50 #topic Open Discussion 18:58:54 we have a couple of mins 18:59:02 trying to find the BP :-) 18:59:06 anything to dicuss? 18:59:22 gduan: https://blueprints.launchpad.net/neutron/+spec/neutron-services-insertion-chaining-steering 18:59:58 SumitNaiksatam: thanks 19:00:07 we didn't touch on tempest today, since that is on top of the mind and we are trying to collectively figure our way out 19:00:14 SumitNaiksatam: it might not be a dependency though 19:00:34 but that is being targeted with high priority 19:00:49 ok, thanks everyone for attending 19:00:59 have good one, bye 19:01:07 #endmeeting