18:30:52 <SridarK> #startmeeting Networking FWaaS
18:30:53 <openstack> Meeting started Wed Sep  9 18:30:52 2015 UTC and is due to finish in 60 minutes.  The chair is SridarK. Information about MeetBot at http://wiki.debian.org/MeetBot.
18:30:54 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
18:30:57 <openstack> The meeting name has been set to 'networking_fwaas'
18:31:00 <badveli> hello all
18:31:03 <SridarK> #chair xgerman
18:31:04 <openstack> Current chairs: SridarK xgerman
18:31:22 <xgerman> o/
18:31:37 <bharath> SridarK: I have to run for a quick errand. You could discuss the FWaaS fip bug with xgerman and he would fill me in later.
18:31:47 <SridarK> probab a quick one today as there is not too much in terms of bp stuff
18:31:56 <xgerman> sure
18:31:59 <SridarK> bharath: no issues
18:32:04 <SridarK> #topic bugs
18:32:16 <SridarK> #link https://bugs.launchpad.net/bugs/1492142
18:32:17 <openstack> Launchpad bug 1492142 in neutron "FWaaS: FIP namespace created after/before Firewall creation doesn't contain FW rules" [Undecided,Confirmed] - Assigned to Sridar Kandaswamy (skandasw)
18:32:54 <SridarK> in terms of update, clearly an issue, have been running thru the code and with the L3Agent refactor things have changed
18:33:07 <SridarK> thanks to bharath: for nailing this down
18:33:16 <xgerman> bharath +1
18:33:22 <yushiro> Hi SridarK, sorry for late :(
18:33:47 <SridarK> i will continue investigation and update the bug, but we may need to cover this as part of the observer hierarchy so we can get updates on events
18:34:02 <SridarK> let me look thru the code more and then discuss
18:34:10 <badveli> fine sridark
18:34:22 <xgerman> ok, we are on some tight deadline so if we can help let us know
18:34:29 <badveli> please let me know if you need any help
18:34:49 <SridarK> xgerman: sure, do we want to call this a Must Fix for this release ?
18:35:00 <badveli> sridark did we miss this as part of our testing
18:35:20 <SridarK> badveli: the code is very different from when we worked on this in Juno
18:35:21 <badveli> i think we tested one scenario
18:35:22 <badveli> to ping
18:35:36 <badveli> ok, thanks
18:35:48 <xgerman> SridarK yep
18:35:52 <SridarK> essentially there are parts in the guts of L3 Agent that trigger a notification to FWaaS
18:36:18 <vishwanathj> pc_m had recommendation early April to move to an Observer hierarchy, i did this for vyatta firewall, had plans to have it available for neutron-fwaas but somehow it slipped
18:36:26 <SridarK> the intent of the observer model was to clean up the L3 Agent from other Service stuff
18:37:10 <badveli> should we go with the observer model
18:37:13 <SridarK> vishwanathj: yes, for FWaaS community too this was the plan but then with some of the uncertainty - i think we did not really prioritize this
18:37:25 <vishwanathj> SridarK, agree
18:37:31 <SridarK> badveli: yes that is a call we want to take now
18:37:53 <xgerman> I like clean code but RC-1 is around 9/21 — so real tight
18:37:56 <SridarK> if we want to do that now as part of RC
18:38:03 <SridarK> xgerman: yes exactly
18:38:31 <SridarK> xgerman: so if we can manage with the workaround for L then we can target observer hierarchy for M
18:38:41 <xgerman> agreed
18:38:58 <vishwanathj> SridarK, what's the workaround?
18:39:07 <SridarK> vishwanathj: restart the agent
18:39:32 <xgerman> well, you can just re-add the router
18:39:40 <xgerman> on the command line
18:39:47 <SridarK> xgerman: ok
18:40:02 <vishwanathj> The impact is only to the reference implementation, right?
18:40:21 <xgerman> I haven’t tested on other implementations
18:40:21 <SridarK> xgerman: perhaps we can discuss more offline and update the bug and decide on next steps
18:40:29 <xgerman> +1
18:40:45 <xgerman> but if somebody can test on 3rd party and update the bug that would be great
18:41:09 <SridarK> vishwanathj: can u do that for vyatta ?
18:41:34 <vishwanathj> I will certainly take it up with my team
18:41:45 <xgerman> thanks
18:41:52 <badveli> sridark since the update was not proper, should it be the same for others implementation
18:41:53 <SridarK> xgerman: +1
18:42:20 <SridarK> badveli: it would depend on each, for cisco for ex - we have our own agent so no issue
18:42:41 <SridarK> for vyatta - i think vishwanathj: has fixed it
18:43:02 <xgerman> ok, cool
18:43:08 <SridarK> ok good lets discuss this offline and publish the next step into the bug
18:43:19 <xgerman> +1
18:43:24 <vishwanathj> SridarK, yes we use the observer hierachy, hence was trying to confirm that this impacted only the reference implementation
18:43:40 <SridarK> vishwanathj: thx yes agreed
18:44:05 <SridarK> the other bug that got some discussion time last mtg was:
18:44:09 <SridarK> #link https://bugs.launchpad.net/horizon/+bug/1491637
18:44:11 <openstack> Launchpad bug 1491637 in OpenStack Dashboard (Horizon) "Error when adding a new Firewall Rule" [Undecided,New] - Assigned to Rob Cresswell (robcresswell)
18:44:22 <SridarK> this has been picked up by Rob
18:44:50 <SridarK> i did not get a chance to check in with him on it - i don't have any further updates - others pls add if u know more
18:45:30 <xgerman> can we decide for each bug on the importance?
18:45:57 <SridarK> xgerman: agreed
18:46:19 <SridarK> i think this was critical according to the submitter as he could not create rules
18:46:26 <SridarK> on horizon
18:47:29 <SridarK> i don't believe there are any other critical ones to discuss
18:47:38 <SridarK> unless other folks have things they want to bring up
18:48:40 <badveli> only on bug front or Blue prints
18:48:45 <SridarK> on bugs
18:48:53 <vishwanathj> SridarK, i verified that hte horizon bug indeed exists, more of FYI
18:49:38 <SridarK> vishwanathj: thx, yes and it is assigned - so should get addressed - i will ping Rob as well
18:50:13 <SridarK> Ok lets move on
18:50:29 <SridarK> #topic Roadmap
18:51:01 <badveli> sridark before this i have one update
18:51:12 <SridarK> yes pls
18:51:24 <badveli> i have my neutron patch with the unit tests working as mentioned in previous meeting
18:51:37 <badveli> except the reference implementation
18:51:44 <SridarK> ok lets cover this as part of this topic as well
18:51:53 <badveli> do we have a chance
18:52:02 <badveli> to put in rc?
18:52:15 <SridarK> u will need to get an FFE
18:52:25 <SridarK> seems like a hard sell now
18:52:28 <xgerman> yep, and they like to have Blueprints
18:53:27 <badveli> xgerman:does it mean we cannot get FFE
18:53:27 <SridarK> badveli: i know u have been working hard on this, but perhaps it is easiest to shoot for M and get it in M1
18:54:07 <xgerman> badveli we can try — if you give me the blueprint I can ask mestery
18:54:22 <xgerman> but it’s tight
18:54:26 <SridarK> badveli: my sense is that if the patch was in active review and just missed the deadline ur chances are higher
18:54:36 <badveli> thanks xerman
18:54:42 <SridarK> but worth as shot as xgerman says
18:55:09 <badveli> fine sridark, xgerman
18:56:03 <badveli> but with reference implementation missing, sorry this is what i could do with my time
18:56:38 <SridarK> badveli: no worries, lets see if something happens else we get this in for M1
18:56:42 <SridarK> #link https://etherpad.openstack.org/p/neutron-fwaas-roadmap-mitaka-summit
18:56:52 <badveli> ok thanks
18:57:13 <SridarK> could we start capturing things that we can bring up for discussion at the summit on this etherpad
18:57:41 <SridarK> i tried to put together some prelim stuff and reflect on our recent discussions
18:57:42 <xgerman> also we need to put some topics into the design session ether pad: https://etherpad.openstack.org/p/neutron-mitaka-designsummit
18:58:07 <xgerman> SridarK that looks good
18:58:15 <SridarK> xgerman: +1 perhaps we can put a summary and point to the details in our etherpad
18:58:27 <xgerman> yep, that’s what I am thinking
18:58:38 <SridarK> xgerman: ok perfect
18:58:54 <SridarK> i have linked mickeys etherpad here as well
18:58:58 <badveli> ok i will outline the bp link for DVR east west
18:59:04 <xgerman> cool
18:59:12 <SridarK> badveli: i added service groups and u can expand that out
18:59:38 <badveli> yes and the DVR east west blue print
19:00:26 <SridarK> Also do u think we can get some early inputs from deployers for helping drive requirements and priorities
19:00:40 <SridarK> i just threw some names out there, but pls add
19:01:08 <xgerman> well, I think we need to do some serious API design
19:01:24 <xgerman> to map use cases to API...
19:01:34 <SridarK> xgerman: yes for sure and hoping that use cases will help
19:01:54 <SridarK> xgerman: but yes this can be a 2 way relationship
19:02:05 <xgerman> +1
19:02:40 <SridarK> xgerman: do u think we can get some inputs for HP
19:02:45 <SridarK> *from HP
19:02:48 <xgerman> yes, we can
19:02:52 <SridarK> perfect
19:02:52 <jwarendt> yes
19:03:10 <xgerman> was just saying jwarendt is in touch with our FWaaS guys
19:03:17 <xgerman> or FW
19:03:38 <SridarK> ok great
19:04:03 <SridarK> I know RAX had some interest and Walmart Labs as well
19:04:18 <xgerman> RAX should have more bandwidth after the summit
19:04:29 <SridarK> will be nice to kick start the summit with a few common use cases
19:04:42 <xgerman> yeah
19:04:43 <SridarK> xgerman: ok
19:04:56 <SridarK> will help our prioritization
19:05:10 <SridarK> ok let me stop rambling :-)
19:05:31 <xgerman> sballe is now with Intel so we might get them as well ;-)
19:05:45 <SridarK> so we can try to populate areas in the etherpad and also reach out to deployers
19:05:52 <SridarK> xgerman: hmm did not know that
19:06:01 <SridarK> but intel is very much interested too
19:06:07 <xgerman> yep
19:06:17 <SridarK> so good we can get some use cases going
19:06:46 <xgerman> we have all the use cases in trello so we can prioritize and work on the API mappingnin Tokyo
19:06:59 <SridarK> +1
19:07:00 <xgerman> and then start implementing for M
19:07:49 <SridarK> Anything else anyone wants to bring up
19:08:32 <badveli> will there be a recording for the sessions
19:08:43 <badveli> or watch
19:08:49 <SridarK> badveli: yes usually put on youtube
19:08:51 <xgerman> mmh, usually things get noted on the etherpad
19:08:56 <xgerman> for design summit
19:09:08 <badveli> thanks xgerman
19:09:12 <xgerman> summit talks are on youtube as SridarK mentioned
19:09:41 <SridarK> #topic Open Discussion
19:09:55 <xgerman> I am hoping we can have a mid cycle which might be easier to reach for the people not going to Tokyo
19:10:25 <xgerman> but that’s more for the December-February timeframe
19:10:53 <SridarK> xgerman: i think timing wise that will be good - time for some push on things we can get in and some retrospective
19:11:27 <yushiro> xgerman: sounds great. I'd like to attend that.
19:11:38 <SridarK> xgerman: are u planning to propose this in Hawaii ? ;-)
19:11:59 <xgerman> well, we need to look at travel budgets
19:12:24 <xgerman> also often they are where some compnay can host with conference rooms and wifi
19:12:39 <SridarK> xgerman: just kidding
19:12:57 <xgerman> I am all in for Hawaii :-)
19:14:17 <xgerman> sc68cal probably will root for Philly ;-)
19:14:35 <SridarK> in Dec hmm :-(
19:14:51 <SridarK> yushiro: i hope u also saw the alternate time next week so it is more friendly for APAC
19:15:33 <yushiro> SridarK: yeah. Thanks. However, I don't feel sleepy :)
19:15:42 <SridarK> yushiro: :-)
19:16:06 <SridarK> so we do alternate on times week by week
19:16:21 <SridarK> #link http://eavesdrop.openstack.org/#Firewall_as_a_Service_(FWaaS)_Team_Meeting
19:16:41 <SridarK> also note different channel on the other week
19:16:54 <SridarK> ok if nothing else, we can close things out
19:17:05 <xgerman> +1
19:17:17 <hoangcx> SridarK: SridarK: Sorry for last week, i could not attend the meeting because of Vietnam national holiday
19:17:35 <SridarK> hoangcx: no worries - we were just not sure if things got missed
19:18:12 <yushiro> SridarK: I couldn't attend the previous meeting because I was attending the wrong channel ... :(  I do take care of the channel.
19:18:29 <SridarK> yushiro: oops ok
19:18:45 <SridarK> ok then bye all until next week
19:18:56 <SridarK> #endmeeting