18:30:44 <SumitNaiksatam> #startmeeting Networking FWaaS
18:30:45 <openstack> Meeting started Wed Feb  4 18:30:44 2015 UTC and is due to finish in 60 minutes.  The chair is SumitNaiksatam. Information about MeetBot at http://wiki.debian.org/MeetBot.
18:30:46 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
18:30:49 <vishwanathj> SridarK, pc_m, badveli, hi
18:30:50 <openstack> The meeting name has been set to 'networking_fwaas'
18:30:59 <badveli> hello viswantah
18:31:03 <pc_m> vishwanathj: hi
18:31:03 <SumitNaiksatam> #info metting agenda https://wiki.openstack.org/wiki/Meetings/FWaaS#Agenda_for_Next_Meeting
18:31:07 <SumitNaiksatam> #info Kilo-2 is Feb 5th
18:31:11 <SumitNaiksatam> that is tomorrow
18:31:12 <SridarK> hi all
18:31:31 <SumitNaiksatam> any other pieces of information worth sharing?
18:32:54 <SumitNaiksatam> #topic Bugs
18:33:23 <SridarK> SumitNaiksatam: nothing new that needs attention
18:33:41 <SumitNaiksatam> SridarK: yeah, my observation as well
18:33:52 <SumitNaiksatam> there were a few issues which cropped up during the week
18:34:06 <SumitNaiksatam> thankfully folks jumped in and patches are passing the gate
18:34:29 <SumitNaiksatam> * jumped in with a fix
18:34:42 <SridarK> SumitNaiksatam: :-)
18:34:49 <SumitNaiksatam> :-)
18:34:57 <SumitNaiksatam> badveli: anything to report on the bugs assigned to you?
18:35:09 <badveli> sumit looked like some other modules implemented the connection track
18:35:20 <badveli> i saw some review
18:35:28 <SumitNaiksatam> good, can you post the review?
18:35:42 <SumitNaiksatam> review link
18:36:14 <badveli> trying to check now, i saw it when i was checking all the open reviews
18:36:25 <badveli> i will send the link to the team
18:37:34 <SumitNaiksatam> badveli: thanks
18:37:39 <SumitNaiksatam> #topic Docs
18:38:03 <SumitNaiksatam> it seems that there are some doc bugs for the freescale fwaas plugin
18:38:32 <SumitNaiksatam> other than that i dont see anything new
18:38:41 <SumitNaiksatam> SridarK: did you notice anything?
18:38:48 <SridarK> SumitNaiksatam: nothing else
18:38:58 <SumitNaiksatam> SridarK: okay good
18:39:31 <SridarK> SumitNaiksatam: i will check with swami if there is anything pending on the dvr changes - i think it is done
18:39:39 <SumitNaiksatam> SridarK: thanks
18:39:48 <SridarK> SumitNaiksatam: np at all
18:40:14 <SumitNaiksatam> so just FYI for everyone - the freescale patch author added the DocImpact flag to his review
18:40:20 <SumitNaiksatam> that generated a bug like this:
18:40:26 <SumitNaiksatam> #link https://bugs.launchpad.net/openstack-manuals/+bug/1417237
18:40:55 <SumitNaiksatam> and that can be followed up witha documentation patch
18:41:00 <SridarK> SumitNaiksatam: thanks - good to know this
18:41:45 <SumitNaiksatam> yeah, since we have quite a few vendor patches in review
18:42:02 <SumitNaiksatam> and i would imagine that might need to be documented
18:42:17 <SumitNaiksatam> although things keep changing with regards to the vendor documentation
18:42:45 <SumitNaiksatam> and i am not sure that exactly what is the latest on that (in terms of incorporating vendor docs in the main docs)
18:42:53 <SumitNaiksatam> but just keep an eye on this
18:42:57 <SumitNaiksatam> okay moving one
18:43:00 <SumitNaiksatam> *on
18:43:14 <SumitNaiksatam> #topic Firewall Insertion
18:43:23 <SumitNaiksatam> SridarK: thanks for the WIP patch
18:43:31 <SridarK> Basic skeleton
18:43:41 <SridarK> #link https://review.openstack.org/#/c/134896/
18:43:58 <SridarK> sorry
18:44:01 <SridarK> wrong link
18:44:04 <SridarK> #link https://review.openstack.org/152697
18:44:26 <SumitNaiksatam> i suspect that this is now target for kilo-3
18:44:39 <SridarK> SumitNaiksatam: yes we should be good
18:44:47 <SridarK> basic extension and db changes
18:44:52 <SumitNaiksatam> so we will have to really push to ge this in at the front end of k-3
18:45:04 <SridarK> SumitNaiksatam: yes for sure
18:45:11 <SridarK> Trying to keep a separation between the extension & db code for existing fwaas and the insertion work now.
18:45:13 <SumitNaiksatam> since we might need to refactor the vendor plugins/drivers
18:45:40 <SridarK> Trying to keep that in mind for sure and see how we can minimize impact
18:45:43 <SumitNaiksatam> i guess the good thing is that the firewall extension have moved to the neutron-fwaas repo
18:45:48 <SumitNaiksatam> SridarK: thanks
18:45:53 <SridarK> yes that was good
18:46:04 <SumitNaiksatam> so we dont have to do the multi-repo acrobatics
18:46:15 <SridarK> SumitNaiksatam: surely is a blessing
18:46:19 <badveli> Sumit i did not really follow this
18:46:36 <badveli> could you please explain
18:46:40 <SumitNaiksatam> given that I think we should go full steam ahead and try to get this wrapped up
18:46:46 <SridarK> yes
18:46:48 <SumitNaiksatam> badveli: sorry, which part?
18:46:58 <SridarK> badveli: few mins pls
18:47:30 <SumitNaiksatam> SridarK: go ahead
18:47:40 <SridarK> one thought is that keeping the new extension and db as a separate entity will enable other plugins to pick up the the basic FW resource, policies, rules etc and if needed adopt the router insertion extension. If they need a different insertion strategy they can still use the fwaas extension (and db) and use a different extension / db implementation for their own insertion model. IMHO, keeping this separation will enabl
18:47:49 <SumitNaiksatam> anyone have a chance to look at SridarK’s patch?
18:48:07 <vishwanathj> nope
18:48:15 <SridarK> also this can help minimize impacts to other plugins
18:49:08 <SridarK> SumitNaiksatam: i think this will keep things clean in terms of the insertion have some separation
18:49:45 <SridarK> Anyways just trying to get some rationale out there - we can always discuss more offline or in the review
18:49:52 <SumitNaiksatam> SridarK: by keeping the extension and db separate, you mean using attribute extension (instead of, say, adding the router attribute to the firewall resource)?
18:50:01 <SridarK> SumitNaiksatam: yes
18:50:25 <SridarK> SumitNaiksatam: and also the db for these parts as  a separate table
18:50:55 <SridarK> SumitNaiksatam: i want to avoid polluting the existing firewall db code as much as possible
18:51:06 <SumitNaiksatam> SridarK: the db can still be a separate table (and I think it will need to be regardless of how you decided to extend the resource definition)
18:51:23 <SridarK> SumitNaiksatam: the plugin will include the existing fwaas extension and this new rtrinsertion extension
18:51:30 <SumitNaiksatam> SridarK: since its a 1:n relationship
18:51:36 <SridarK> SumitNaiksatam: we will keep the logic in the plugin
18:51:47 <SridarK> SumitNaiksatam: yes u are correct
18:52:14 <SridarK> SumitNaiksatam: i want other plugins to inherit from the existing firewall db class
18:52:28 <SridarK> SumitNaiksatam: irrespective of the insertion strategy
18:52:30 <SumitNaiksatam> SridarK: i think the choice will be between adding the “router_ids” as an optional attribute to the firewall resource, or using the attribute extension
18:52:51 <SridarK> SumitNaiksatam: yes the latter
18:53:13 <SumitNaiksatam> SridarK: sorry i forget, but did we state an explicit design decision in our spec?
18:53:15 <SridarK> so it is not in the firewall resource as such
18:53:35 <SumitNaiksatam> if we did then, we are just executing on that design
18:53:39 <SridarK> SumitNaiksatam: i think i did state both as possible options
18:54:12 <SumitNaiksatam> SridarK: ah ok
18:54:26 <SumitNaiksatam> other folks in the team please chime in
18:54:37 <SridarK> SumitNaiksatam: sorry my inability to communicate very effectively on irc :-)
18:55:42 <SridarK> SumitNaiksatam: if a plugin does not want the router insertion and just the fwaas extension for the fw resource - want it make it easy for them to do that
18:55:59 <SumitNaiksatam> SridarK: i think you are very articulate, certainly more than me! :-)
18:56:06 <SridarK> SumitNaiksatam: :-)
18:56:14 <SumitNaiksatam> SridarK: i very much agree with that intent
18:56:19 <vishwanathj> I might need to revisit the spec again to be an effective contributor
18:56:30 <SumitNaiksatam> vishwanathj: sure, please do :-)
18:56:33 <SridarK> SumitNaiksatam: perhaps i can also follow up with a email to the fwaas folks
18:56:40 <SumitNaiksatam> SridarK: sure
18:56:42 <SridarK> vishwanathj: no worries -
18:56:59 <SridarK> thats all from me
18:57:11 <SridarK> badveli: sorry - did not want to lose my train of thought
18:57:34 <badveli> no problem
18:57:58 <badveli> as things are changing i am a bit not sure which way we are going
18:58:00 <SumitNaiksatam> badveli: thanks for waiting
18:58:08 <SridarK> badveli: i think if u look at the email from SumitNaiksatam on patch from Doug u should get a good idea
18:58:28 <SumitNaiksatam> badveli: can you be more specific on what “things” and which “way” you are referring to?
18:58:52 <SumitNaiksatam> badveli: apologies upfront if i might have confused you
18:59:20 <badveli> about the extension part
18:59:26 <badveli> no problem since this is chaning
19:00:21 <badveli> my question is if i need to get the patch for service objects should i follow it/ should i wait for sridark patch
19:00:41 <SridarK> badveli: no u are not dependent on my patch
19:01:05 <SumitNaiksatam> right, they are not related, at least to the extent i know
19:01:51 <badveli> thanks sridar, sumit. So now we can move the extension part to the fwaas repo, correct?
19:02:38 <SumitNaiksatam> badveli: its already moved
19:02:52 <SumitNaiksatam> that was my email informing the team that it had moved
19:03:04 <SumitNaiksatam> the existing fwaas extension definition
19:03:31 <SumitNaiksatam> badveli: i am still pretty fuzzy on how you are planning to implement your extension
19:03:35 <badveli> thanks sumit, so i can start working my patch
19:04:14 <SumitNaiksatam> badveli: if its not fwaas-specific, then the extension definition should not be in the neutron-fwaas repo
19:04:32 <SumitNaiksatam> the extensions we are discussing here are not fwaas-specific
19:04:55 <SumitNaiksatam> sorry i meant - they *are* fwaas specific
19:05:50 <badveli> my problem would be if did not have the extension over fwaas, it would be tough to implement the reference impl
19:06:02 <SridarK> SumitNaiksatam: i guess badveli will need to push a patch to neutron for the extensions
19:06:04 <SumitNaiksatam> badveli: okay
19:06:33 <badveli> yes sridar, sumit. looks like i need to push the a patch for neutron
19:06:34 <SumitNaiksatam> SridarK: yeah, i meant to say, it depends on what is stated in the approved blueprint
19:06:46 <badveli> as the reviewers wanted it to be generic
19:07:15 <badveli> but would it make the reference implementation
19:07:46 <badveli> the fwaas usage not so easy?
19:08:27 <SumitNaiksatam> badveli: my thinking was that we had discussed these issues at design time (while proposing the spec)
19:08:45 <SumitNaiksatam> badveli: perhaps good idea to check with Yi
19:08:56 <SridarK> badveli: i think it should be ok - u can still implement the reference in fwaas - u will just need to manage 2 patches
19:09:32 <badveli> thanks sumit, sridark, yes we have discussed in getting two patches
19:10:11 <badveli> ultimately we want to get the code in
19:11:09 <SumitNaiksatam> badveli: yeah
19:11:23 <SumitNaiksatam> okay anyone have any questions for badveli on this?
19:11:48 <badveli> thanks sumit, as long as  we are in agree
19:12:16 <SridarK> I am good
19:12:25 <SumitNaiksatam> badveli: thanks for the update
19:12:38 <badveli> thanks sumit
19:12:57 <SumitNaiksatam> #topic Service Objects
19:13:01 <SumitNaiksatam> service objects was a separate agenda item, but we munged it with the earlier topic
19:13:26 <SumitNaiksatam> just added the topic for anyone going through the logs
19:13:34 <badveli> thanks sumit, let me start working on the patches
19:13:45 <SumitNaiksatam> badveli: great
19:13:46 <SumitNaiksatam> #topic FWaaS L3 agent refactoring/restructuring
19:13:54 <SumitNaiksatam> pc_m: hi
19:13:58 <pc_m> SumitNaiksatam: hi
19:14:01 <SumitNaiksatam> anything to discuss today on this?
19:14:06 <pc_m> nothing on this front.
19:14:12 <SumitNaiksatam> pc_m: ah okay
19:14:26 <pc_m> SumitNaiksatam: Waiting for FW insertion to complete first.
19:14:30 <SumitNaiksatam> pc_m: and again, many thanks for your continued patience
19:14:47 <pc_m> SumitNaiksatam: np. I've got enough other balls to juggle.
19:14:51 <pc_m> :)
19:14:55 <SumitNaiksatam> pc_m: _;_
19:14:58 <SumitNaiksatam> :-)
19:15:07 <SumitNaiksatam> #topic FWaaS gate jobs
19:15:11 <SumitNaiksatam> pc_m: since you are around
19:15:28 <SumitNaiksatam> pc_m: you had taken an AI for yourself to send us some pointers
19:15:32 <SumitNaiksatam> ;-)
19:15:40 <pc_m> SumitNaiksatam: Yes... here you go... https://wiki.openstack.org/wiki/Neutron/FunctionalGateSetup
19:16:05 <SumitNaiksatam> ah sweet!
19:16:10 <pc_m> SumitNaiksatam: Please let me know if you have any issues in following the steps (feel free to make corrections).
19:16:24 <SumitNaiksatam> pc_m: this is great info for everyone
19:16:35 <pc_m> For VPN, I have a commit out for review to make the check queue non-voting.
19:16:48 <SumitNaiksatam> pc_m: can you share the link to that review?
19:17:00 <pc_m> Once approved, and we use it for a bit, I'll do the last piece to make it voting.
19:17:05 <SridarK> pc_m: this is great
19:17:23 <pc_m> #link https://review.openstack.org/#/c/152602/
19:17:27 <SumitNaiksatam> pc_m: thanks
19:17:49 <SumitNaiksatam> i think the timing is good to this since the extension definition has moved
19:17:54 <pc_m> In the instructions, I put all the VPN review #s so you can see the actual diffs for VPN through the steps/
19:18:06 <pc_m> I'm going to send this out on the ML today.
19:18:38 <SumitNaiksatam> pc_m: sweet!
19:18:45 <SumitNaiksatam> i will go through it
19:19:08 <SridarK> SumitNaiksatam: Nikolay was working on the FWaaS Scenario tests - i have asked him if will be able to continue thru this and if so we can get some these tests in
19:20:11 <pc_m> SumitNaiksatam: You'll likely want to get the experimental queue going, and the gate hooks, and then people can do commits with functional tests.
19:20:31 <SumitNaiksatam> pc_m: definitely
19:20:35 <SumitNaiksatam> SridarK: that is awesome
19:20:47 <SumitNaiksatam> SridarK: can you add me to the conversations as well?
19:20:56 <SridarK> SumitNaiksatam: yes will do
19:21:04 <SumitNaiksatam> SridarK: thanks
19:21:12 <pc_m> For VPN, we had people hold off on committing functional test modules, until we got that in place.
19:21:27 <SumitNaiksatam> pc_m: that sounds logical
19:21:51 <SumitNaiksatam> pc_m: again thanks a bunch for documenting this
19:22:04 <pc_m> Otherwise, you get functional modules upstreamed, and then when the tests are working, there are failures to deal with.
19:22:06 <pc_m> np
19:22:08 <SumitNaiksatam> lots of people have done this in the past, you took the trouble of documenting it so others can use
19:22:16 <vishwanathj> +1
19:22:18 <SumitNaiksatam> in true spirit of community work!
19:22:53 <pc_m> thanks.
19:22:55 <SridarK> pc_m: this is like a book for "defense against the dark arts" :-)
19:23:09 <pc_m> :)
19:23:10 <SumitNaiksatam> SridarK: lol
19:23:23 <SumitNaiksatam> didnt i say you were very articulate! ;-P
19:23:25 <SridarK> there is some deep magic in those parts
19:23:30 <SridarK> :-)
19:23:31 <SumitNaiksatam> we have only few mins left and need to cover vendor drivers
19:23:38 <SumitNaiksatam> #topic Vendor Plugins/drivers
19:23:46 <SumitNaiksatam> the freescale plugin was merged, yay!
19:23:51 <vishwanathj> +1
19:23:54 <SridarK> yes finally
19:24:11 <SridarK> kudos to trinath for his persistence
19:24:14 <SumitNaiksatam> kudos to the author for his pereseverance
19:24:19 <SridarK> :-)
19:24:28 <SumitNaiksatam> vishwanathj: on to your patch
19:24:38 <SumitNaiksatam> the last i checked it had comments from SridarK
19:24:43 <SumitNaiksatam> are those addressed?
19:24:59 <vishwanathj> I have addressed comments from SridarK and pc_m
19:25:18 <SridarK> vishwanathj: yes thanks
19:25:30 <pc_m> +1
19:25:36 <SridarK> vishwanathj: i think we are good
19:25:48 <SumitNaiksatam> oh, trinath put a -1
19:25:49 <vishwanathj> Trinath also commented early this morning, he is asking to upload README that has link to WIKI page and CI contacts....is that a hard requirement
19:25:49 <SumitNaiksatam> hmmm
19:25:59 <SumitNaiksatam> vishwanathj: no
19:26:03 <SumitNaiksatam> vishwanathj: but do it
19:26:13 <SridarK> vishwanathj: yes he got pulled for that
19:26:16 <SumitNaiksatam> vishwanathj: i mean your call
19:26:20 <SumitNaiksatam> its not a hard requirement
19:26:24 <SridarK> vishwanathj: so good to do it i think
19:26:51 <vishwanathj> I dont have a WIKI page yet, what are the CI contact requirements...completely new to this
19:26:58 <SumitNaiksatam> vishwanathj: let us know when you do, and we can hopefully proceed quickly from there
19:27:17 <SumitNaiksatam> vishwanathj: wiki page is a 2 min job, so should not be an issue
19:27:28 <SumitNaiksatam> vishwanathj: you can update the wiki page independently
19:27:30 <vishwanathj> SumitNaiksatam, pc_m, SridarK, will let you guys know once I upload a new patch set to get your votes
19:27:37 <vishwanathj> ok
19:27:38 <pc_m> roger
19:27:43 <SridarK> vishwanathj: u put ur email id or better yet some other colleague in ur company who will get spammed if ur CI job goes down
19:27:47 <SridarK> :-)
19:27:51 <vishwanathj> :)
19:27:56 <SumitNaiksatam> for the CI contact requirements i believe you would already have a brocade contact, no?
19:28:21 <SumitNaiksatam> vishwanathj: see this - #link https://review.openstack.org/#/c/152229/
19:28:24 <SumitNaiksatam> fyi
19:28:28 <vishwanathj> I will check with natarajk, he was the contact earlier, maybe will keep as the contact :)
19:28:38 <SridarK> :-)
19:28:45 <SumitNaiksatam> check if that file has a brocade contact that can deal with your fwaas plugin
19:29:05 <vishwanathj> SumitNaiksatam, thanks for the link, ok
19:29:26 <SumitNaiksatam> there are two other vendor drivers:
19:29:28 <vishwanathj> SridarK, pc_m, SumitNaiksatam, appreciate your earlier reviews
19:29:37 <SridarK> SumitNaiksatam: yes first patch is up
19:29:38 <SumitNaiksatam> mcafee: #link https://review.openstack.org/152093
19:29:49 <SumitNaiksatam> cisco: #link https://review.openstack.org/152282
19:30:00 <SridarK> need to refactor for vendor repo and there will be one more on agent/driver
19:30:04 <vishwanathj> SumitNaiksatam, can brocade fwaas still make Kilo-2 by tomorrow
19:30:39 <SumitNaiksatam> vishwanathj: my earlier wish - “we can hopefully proceed quickly from there”
19:30:48 <vishwanathj> ok
19:30:54 <yamahata> regarding to mcafee, we're respining the patch and working on CI system.
19:31:13 <SumitNaiksatam> yamahata: great, thanks for joining, are you the point of contact for this?
19:31:28 <yamahata> Yes.
19:31:28 <vishwanathj> yamahata, hi
19:31:35 <yamahata> vishwanathj: hi.
19:31:36 <SumitNaiksatam> yamahata: i see a differenent author on the patch
19:31:42 <SumitNaiksatam> okay anyway
19:31:51 <yamahata> Yalei is also contact, but he's located in China.
19:31:57 <yamahata> It's midnight now in China.
19:32:00 <SumitNaiksatam> yamahata: and you are now in the bay area?
19:32:18 <yamahata> SumitNaiksatam: Yeah at last.
19:32:22 <SumitNaiksatam> ok good! :-)
19:32:34 <SumitNaiksatam> oh, we are two mins over
19:32:39 <SumitNaiksatam> #topic Open Discussion
19:32:43 <SridarK> i have updated the FWaaS wiki with links to most of the patches
19:32:43 <SumitNaiksatam> anything that we missed?
19:32:47 <SridarK> #link https://wiki.openstack.org/wiki/Neutron/FWaaS/KiloPlan
19:32:50 <SumitNaiksatam> SridarK: cool!
19:33:01 <SridarK> pls check if i have messed up someone's patch id etc
19:33:08 <SumitNaiksatam> SridarK: you will need to maintain it now :-)
19:33:09 <SridarK> yamahata: i will add intel as well
19:33:15 <SridarK> SumitNaiksatam: :-)
19:33:27 <SridarK> no worries
19:33:51 <SumitNaiksatam> and please note, that every feature has all the associated componentst that need to be completed
19:34:01 <SumitNaiksatam> like docs, tempest tests, CLI, horizon, etc
19:34:07 <SumitNaiksatam> as captured in the table on the wiki
19:34:28 <SumitNaiksatam> and, we as a team, need to plan to cover all those aspects
19:34:36 <SumitNaiksatam> anything else anyone wants to add?
19:34:36 <vishwanathj> +1
19:34:52 <SumitNaiksatam> not just this, but in general?
19:34:59 <vishwanathj> agreed
19:35:07 <SridarK> SumitNaiksatam: i am good
19:35:12 <SumitNaiksatam> vishwanathj: thanks for posting the sequence diagrams
19:35:27 <vishwanathj> will add more as I go along
19:35:33 <SumitNaiksatam> vishwanathj: that is a great community contribution!
19:35:40 <SumitNaiksatam> from a fwaas perspective
19:35:40 <SridarK> +1
19:35:46 <SumitNaiksatam> okay thanks all for joining, sorry for going over
19:35:51 <SumitNaiksatam> bye!
19:35:54 <vishwanathj> it was a win win situation
19:35:57 <SridarK> bye all thanks
19:35:57 <badveli> bye
19:36:00 <vishwanathj> bye
19:36:02 <pc_m> bye!
19:36:04 <SumitNaiksatam> #endmeeeting
19:36:16 <SumitNaiksatam> vishwanathj: absolutely, glad you see it that way
19:36:45 <yamahata> bye
19:36:51 <vishwanathj> thanks
22:00:21 <openstack> alaski: Error: Can't start another meeting, one is in progress.  Use #endmeeting first.
22:00:50 <dansmith> fail
22:00:58 <alaski> SumitNaiksatam: can you #endmeeting please?
22:01:33 <alaski> well, I guess we go off the record then
22:01:40 <alaski> Anyone here for the cells meeting?
22:01:48 <belmoreira> hi
22:02:18 <dansmith> o/
22:02:29 <alaski> small crowd today, but that's alright
22:02:42 <alaski> topic: Test failure
22:03:01 <alaski> There have been some intermittent test failures in the tempest job
22:03:16 <alaski> I spent some time trying to repro locally, but I'm not seeing them
22:03:20 <melwitt> did we start the meetbot
22:03:39 <alaski> melwitt: the last meeting didn't end properly :(
22:03:44 <dansmith> #endmeeting