18:31:15 <SumitNaiksatam> #startmeeting Networking FWaaS
18:31:16 <openstack> Meeting started Wed May 27 18:31:15 2015 UTC and is due to finish in 60 minutes.  The chair is SumitNaiksatam. Information about MeetBot at http://wiki.debian.org/MeetBot.
18:31:17 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
18:31:19 <openstack> The meeting name has been set to 'networking_fwaas'
18:31:40 <SumitNaiksatam> i believe people are still recovering from Vancouver ;-)
18:31:57 <SumitNaiksatam> and some might not be able to attend
18:32:26 <SridarK> SumitNaiksatam: +1
18:32:27 <SumitNaiksatam> i am proposing we cover two topics today - Bugs, and Vancouver Retrospective
18:32:29 <SridarK> :-)
18:32:35 <SumitNaiksatam> everyone okay with that?
18:32:36 <qwebirc32260> Hi Sumit
18:32:41 <pc_m> hi
18:32:53 <SridarK> SumitNaiksatam: yes - i have a hard stop at noon for another mtg
18:32:53 <SumitNaiksatam> qwebirc32260: pc_m: hi
18:32:55 * pc_m in another meeting, but will try to juggle
18:33:09 <SumitNaiksatam> qwebirc32260: is that vikram?
18:33:24 <SumitNaiksatam> #topic Bugs
18:33:38 <qwebirc32260> HI I am Yanping. I forgot change my name :-)
18:33:38 <badveli> fine sumit, i need some  info as per pc_m email exchange
18:33:56 <SumitNaiksatam> qwebirc32260: ah, hi yanping, good to see you :-)
18:34:06 <SumitNaiksatam> #link https://bugs.launchpad.net/neutron/+bug/1455863
18:34:06 <openstack> Launchpad bug 1455863 in neutron "FWAAS- FW Rule editing puts FW to error state " [Undecided,New] - Assigned to vikram.choudhary (vikschw)
18:34:29 <SumitNaiksatam> its not clear how to reproduce the above, so i have put a comment in the bug report
18:34:38 <SumitNaiksatam> #link https://bugs.launchpad.net/horizon/+bug/1454974
18:34:38 <openstack> Launchpad bug 1454974 in OpenStack Dashboard (Horizon) "FWAAS " [Undecided,New]
18:34:59 <SumitNaiksatam> that seems like a bug, but i have not noticed it before
18:35:15 <SumitNaiksatam> perhaps our Horizon guru vishwanathj can chime in on that ;-)
18:35:56 <SumitNaiksatam> other than that no new bugs report
18:35:59 <vishwanathj> Let me try later this week....am on new born baby duty since yesterday
18:36:13 <SumitNaiksatam> vishwanathj: oh wow, sorry, i forgot that
18:36:23 <SumitNaiksatam> vishwanathj: so the baby is here?
18:36:27 <badveli> congrats viswantahj
18:36:31 <SridarK> vishwanathj: congrats i was surprised to see u here
18:36:34 <vishwanathj> arrived yesterday afternoon
18:36:45 <vishwanathj> FWaaS is also like a baby to me :)
18:36:50 <SumitNaiksatam> vishwanathj: awesome, hearty congrats on behalf of the entire team!
18:36:55 <vishwanathj> thanks all
18:37:03 <SridarK> vishwanathj: ur getting bad vibes from ur spouse now or ur avoiding diaper duty ;-)
18:37:05 <SumitNaiksatam> vishwanathj: absolutely appreciate your committment for the cause
18:37:06 <yushiro> vishwanathj,  awesome!
18:37:14 <SridarK> SumitNaiksatam: +1
18:37:23 <SumitNaiksatam> on that happy note lets transition to the next topic
18:37:31 <SumitNaiksatam> #topic Vancouver Retrospective
18:37:49 <SumitNaiksatam> personally, it was great for me to see a huge section of the team
18:37:56 <vishwanathj> +1
18:37:58 <SumitNaiksatam> our team has grown quite a bit
18:38:17 <SumitNaiksatam> i believe the only people missing were badveli and trinath
18:38:35 <vishwanathj> has trinath ever come to a summit?
18:38:38 <SumitNaiksatam> regardless, i think we had some very helpful offline discussions
18:38:50 <SumitNaiksatam> vishwanathj: good question, perhaps not :-(
18:38:54 <badveli> next time
18:39:07 <SridarK> absolutely was great to meet all
18:39:10 <SumitNaiksatam> unfortunately its a pretty expensive trip if you are coming from India
18:39:34 <SumitNaiksatam> SridarK: special thanks to you for your relentless coordination
18:39:39 <vishwanathj> +1
18:39:59 <SridarK> SumitNaiksatam: no worries - was really great to meet all and have good discussions on FWaaS
18:40:26 <SumitNaiksatam> of the many things we discussed, we touched on the direction attribute in the FWaaS rules
18:40:34 <SumitNaiksatam> is slaweq here?
18:40:48 <SumitNaiksatam> i believe SridarK you had a follow up with slaweq and vikram today?
18:40:58 <SridarK> SumitNaiksatam: yes we had a long discussion
18:41:20 <SumitNaiksatam> anyone would like to summarize for the benefit of the team?
18:41:32 <SridarK> mainly to make sure that we have a model that works well and provides a consistent interface for good user experience
18:41:39 <SridarK> SumitNaiksatam: i will try
18:42:05 <SridarK> Mainly we wanted to apply the direction attribute at the Policy or (Firewall, Policy) level
18:42:18 <SridarK> so this will be consistent for zones
18:42:42 <SridarK> zones would be direction ++
18:42:56 <SridarK> so it would an evolution rather than make a rework
18:43:01 <SumitNaiksatam> SridarK: yes, thanks for bringing up that angle
18:43:06 <SumitNaiksatam> (the zones that is)
18:43:19 <SridarK> so we would like these attributes to be at the same level and avoids any complex validation logic
18:43:31 <SridarK> Slawek and Vikram are okay with this
18:43:56 <SumitNaiksatam> slaweq: there?
18:43:57 <SridarK> SumitNaiksatam: the other thing on related note was to explore multiple policies on a FW
18:44:42 <SumitNaiksatam> SridarK: okay
18:44:43 <SridarK> so we can support multiple values on a single FW (an ingress policy and an egress policy) for example
18:45:00 <SridarK> SumitNaiksatam: we discussed this and few others also would like to see this
18:45:17 <SridarK> SumitNaiksatam: let us put some thought into this as well
18:45:26 <SumitNaiksatam> SridarK: the only issue with multiple policies is backward compatibility
18:45:33 <SumitNaiksatam> SridarK: but we can think a little more
18:46:07 <SumitNaiksatam> the other big item was obviously was the design summit discussion regarding FWaaS and Security Groups
18:46:11 <SridarK> SumitNaiksatam: ok yes that is a valid point but may be this can be done with optional
18:46:20 <SumitNaiksatam> SridarK: true
18:46:28 <SridarK> SumitNaiksatam: i am done with the update we can discuss more later
18:46:32 <SumitNaiksatam> an entire session was dedicated to this
18:46:38 <SumitNaiksatam> SridarK: thanks!
18:46:44 <SridarK> np :-)
18:47:02 <SumitNaiksatam> during the session we discussed the requirements for getting FWaaS out of experimental
18:47:24 <SumitNaiksatam> which led to discussion on what are the differences between FWaaS and Security Groups
18:47:48 <SumitNaiksatam> a hand poll was taken to see if anyone objected to having two separate APIs
18:48:28 <SumitNaiksatam> i believe there were more people in the room in support of separate APIs versus one consolidated API
18:49:07 <SridarK> SumitNaiksatam: i felt that way too
18:49:08 <SumitNaiksatam> however towards the end of the meeting the point was made that it needs to be clearly articulated as to what the points of diffirence between the two APIs and feature sets were
18:49:13 <SumitNaiksatam> and how they could be consumed
18:49:16 <SumitNaiksatam> SridarK: okay
18:49:38 <SumitNaiksatam> that was my short summary of the 40 minute session
18:49:51 <SumitNaiksatam> others present in the session please feel free to add color
18:50:11 <SridarK> SumitNaiksatam: yes that was a very good summary
18:50:44 <blogan> RED!
18:51:08 <vishwanathj> My takeaways on getting FWaaS out of experimental, 1) Complete the documentation tasks 2) Ensure functional tests are adequate and completed
18:51:08 <SumitNaiksatam> also my understanding is that we as a FWaaS team, will continute to target features for Liberty (unless it was explicitly mentioned in some forum that we should not be doing so, and I missed it)
18:51:46 <SumitNaiksatam> blogan: hi, was that an acronym?
18:51:54 <blogan> I believe there was also a request to define the overlapping pieces with security groups and decide on whether those should be consolidated into the same API or not
18:52:05 <blogan> SumitNaiksatam: you asked to add color, i like red :)
18:52:28 <SumitNaiksatam> blogan: i know, but red was in agreement or disagreement?
18:52:29 <vishwanathj> I remember talking to mestery and doug, they said FWaaS being experimental should not prevent us from filing any blue prints for new FWaaS features
18:53:18 <blogan> SumitNaiksatam: is me being dumb a good answer?
18:53:35 <vishwanathj> :)
18:54:21 <SumitNaiksatam> blogan: is your comment “I believe there was also a request to define the overlapping pieces with security groups and decide on whether those should be consolidated into the same API or not”, different from what I mentioned earlier “however towards the end of the meeting the point was made that it needs to be clearly articulated as to what the points of diffirence between the two APIs and feature sets were"
18:55:00 <blogan> SumitNaiksatam: yeah those are the same, sorry I didn't see you say that
18:55:34 <SumitNaiksatam> blogan: no worries, i just want to make sure that for those who didnt attend the session they got the right summary out of this
18:55:41 <blogan> SumitNaiksatam: though I will add that I remember something about not having both of the same features in the both APIs
18:55:55 <SumitNaiksatam> such that people can accordingly decide how to orient their efforts
18:56:01 <blogan> agreed
18:56:22 <SumitNaiksatam> blogan: right
18:56:34 <SridarK> SumitNaiksatam: there was also mention of SG alignment with AWS for portability
18:56:50 <SumitNaiksatam> SridarK: right, that was good point from sc68cal
18:58:02 <SumitNaiksatam> and it was one of the main reasons we started with the FWaaS API and abstraction in the first place, to be able to deal with features which are not represented or apply to security groups
18:58:29 <SumitNaiksatam> any other thoughts, comments, concerns on this discussion?
18:59:13 <SumitNaiksatam> #topic Open Discussion
18:59:18 <SridarK> SumitNaiksatam: i guess we will need to evolve this discussion to drive next steps.
18:59:48 <SumitNaiksatam> SridarK: yes absolutely, the team is  still absorbing this I believe :-)
18:59:54 <SridarK> SumitNaiksatam: i will go to juggle mode - i have to get on another internal meeting
18:59:59 <SridarK> yes
19:00:15 <SumitNaiksatam> also, i think several people expresed the interest to work on independent features
19:00:26 <SumitNaiksatam> we will discuss those as the specs are posted
19:00:46 <mestery> You folks are aware of the new specs process, no?
19:01:00 <mestery> Certainly since you're part of hte Neutron Stadium you must be paying attention to things right?
19:01:12 <vishwanathj> Specs can be submitted at any time
19:01:26 <mestery> That's a small part of it
19:01:26 <mestery> https://review.openstack.org/#/c/177342/
19:01:36 <mestery> Also: https://review.openstack.org/#/c/186095/ is under review
19:02:05 <SumitNaiksatam> mestery: thanks for those links
19:02:22 <mestery> I'll let you all digest that as you move forward, the tl;dr is we've moving to feature requests and documentaiton landing with the patches
19:02:28 <mestery> Separating the "what" from the "how"
19:02:47 <mestery> The intent is to make things easier to submit as well as review.
19:02:55 <SumitNaiksatam> mestery: +1
19:03:02 <SridarK> mestery:  thanks yes for sure
19:03:39 <slaweq> SumitNaiksatam: hello
19:03:48 <SumitNaiksatam> slaweq: ah, you made it
19:03:57 <slaweq> now I'm here, sorry for late but I had something to do :)
19:04:06 <SumitNaiksatam> slaweq: no worries
19:04:06 <slaweq> do You want something from me?
19:04:17 <pc_m> mestery: So do request as LP bug, and then when approved do a spec?
19:04:20 <SumitNaiksatam> SridarK: summarized the discussion he had with you and Vikram
19:04:27 <SumitNaiksatam> you can find it in the back scroll
19:04:30 <slaweq> great
19:04:40 <slaweq> thx SridarK :)
19:04:40 <mestery> pc_m: No, waterfall design is a thing of the past, it wasn't working.
19:04:48 <SumitNaiksatam> i believe the plan is that you are going to update the spec accordingly
19:05:07 <mestery> pc_m: Do the RFE, if it's something huge whose intent is to make changes in many places, maybe a spec, otherwise, move forward with devref and code.
19:05:07 <SridarK> slaweq: np at all we can discuss more as needed
19:05:09 <slaweq> yes
19:05:22 <pc_m> mestery: I have two BP that I want to create, could use some guidance on new process.
19:05:35 <slaweq> SridarK: probably we will :) I will be in touch with You
19:05:49 <dougwig> pc_m: file an RFE bug, a driver or LT will confirm it and put it in a milestone, the drivers may request a spec sometimes, otherwise once the bug is marked by the drivers, just go for it.
19:05:50 <SridarK> slaweq: np catch up later
19:05:52 <slaweq> now I will modify our specs as we discussed today
19:05:52 <mestery> pc_m: The second patch (under review) indicates up until Liberty-1 we'll review old-style specs, but if you haven't filed them, an RFE is the way to go.
19:06:21 <vishwanathj> mestery, the patch set https://review.openstack.org/#/c/182905/ is also part of the new spec process, right?
19:06:29 <pc_m> mestery: OK. thanks. so spec upon request then. cool.
19:06:41 <mestery> vishwanathj: Yes, exactly.
19:06:57 <mestery> pc_m: Even without a spec, devref documentation is the new thing.
19:07:28 <pc_m> mestery: OK. And that is described in one of these reviews?
19:07:38 <mestery> pc_m: Yes.
19:07:45 <pc_m> mestery: ok. will check
19:08:56 <SumitNaiksatam> alright folks, anything else for today?
19:09:32 <SumitNaiksatam> thanks for joining
19:09:35 <SumitNaiksatam> #endmeeting