14:00:23 <SridarK> #startmeeting fwaas
14:00:25 <openstack> Meeting started Thu May 17 14:00:23 2018 UTC and is due to finish in 60 minutes.  The chair is SridarK. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:00:26 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
14:00:29 <openstack> The meeting name has been set to 'fwaas'
14:01:33 <SridarK> #chair xgerman_
14:01:34 <openstack> Current chairs: SridarK xgerman_
14:01:40 <mlavalle> o/
14:01:54 <SridarK> I see yushiro cannot make today's mtg
14:02:19 <SridarK> I think today was his turn - i can chair
14:02:59 <SridarK> We are getting close to summit
14:03:26 <xgerman_> o/
14:03:37 <xgerman_> SridarK: sounds good
14:03:40 <SridarK> Look fwd to meeting folks that are travelling to summit next week
14:03:48 <xgerman_> +1
14:04:09 <annp> +1
14:05:06 <SridarK> xgerman_: any announcements u want to bring up
14:05:21 <xgerman_> quiet week
14:05:50 <SridarK> ok lets move on
14:06:06 <SridarK> #topic Rocky Pluggable backend driver
14:06:13 <SridarK> doude: pls go ahead
14:06:20 <doude> ok
14:06:39 <doude> so I had two +1 reviews
14:06:44 <annp> doube, thanks for your great work on that.
14:06:51 <doude> one from annp , thanksq
14:07:08 <njohnston> o/
14:07:08 <annp> doube, you're welcome :)
14:07:21 <SridarK> annp: were u able to verify the multinode ?
14:07:26 <doude> and one from NSX developer after I explain how they have to use it
14:07:49 <annp> SridarK, yes. I already tested doube's patch with multinode.
14:08:09 <doude> and I just realized a +2 from Yushiro
14:08:10 <SridarK> annp thx
14:08:14 <SridarK> doude: yes
14:08:17 <annp> SridarK, It worked fined in my local environment.
14:08:30 <doude> with that comment
14:08:34 <doude> 'I tested devstack with AIO and multi nodes. There is no issue at present. Thanks for your great work, doube!!'
14:08:35 <SridarK> doude: let me run thru it once today and i can +A
14:08:49 <annp> SridarK, +1
14:08:52 <SridarK> doude: yes i just saw his coment too
14:09:22 <SridarK> doude: give me  a couple of hours  -
14:09:34 <SridarK> doude: BTW are u at the summit ?
14:10:05 <doude> cool
14:10:07 <doude> thank SridarK
14:10:32 <doude> no I could not (that was the first I missed since 5 years now)
14:11:07 <SridarK> doude: ok :-( yes u are usually there
14:11:28 <SridarK> lets move on
14:11:39 <SridarK> #topic Remote FWG
14:11:44 <SridarK> xgerman_: pls go ahead
14:11:45 <doude> yes but baby is coming :)
14:11:57 <SridarK> oh doude congrats
14:12:00 <mlavalle> doude: congrats!
14:12:04 <doude> thanks
14:12:07 <mlavalle> first one?
14:12:18 <annp> doude, congrats!
14:12:27 <doude> no second
14:12:32 <xgerman_> sqlalchemy has been the bane of my existsnce the last few weeks — I even got someone else to look at it with no resolution
14:12:42 <xgerman_> SridarK: +1
14:12:43 <SridarK> doude: cool keep us posted
14:12:49 <xgerman_> +1
14:12:55 <SridarK> xgerman_: ok
14:12:59 <doude> ok thanks guys
14:13:18 <annp> xgerman_, I think revise id not matching with down_revision
14:13:38 <xgerman_> ah, that could be
14:13:59 <xgerman_> also saw your comment — not sure if we want to cascade deletion of the ruke/fwg
14:14:19 <xgerman_> I rather have a user delete the rule/fwg before they can delete the rest
14:16:29 <SridarK> xgerman_: let me also look on the delete
14:16:51 <SridarK> we can talk more during the next week as well
14:17:04 <annp> SridarK, +1
14:17:12 <xgerman_> +1
14:17:20 <annp> xgerman_, let's discuss at the summit.
14:17:26 <SridarK> ok
14:17:45 <SridarK> #topic FWaaS Logging Spec
14:18:01 <annp> Hi SridarK,
14:18:10 <SridarK> I think we are almost ready to get this in
14:18:14 <SridarK> annp: pls go ahead
14:18:38 <annp> Currently, The spec is waiting +A from Akihiro or Miguel
14:18:43 <SridarK> mlavalle: if u can also take a look to see if u are good with it
14:18:49 <mlavalle> annp: I;ll review it today
14:18:55 <SridarK> mlavalle: thx
14:18:57 <mlavalle> sorry for the delay
14:19:02 <amotoki> annp: sorry for late.
14:19:05 <annp> mlavalle, thanks.
14:19:08 <mlavalle> I've been reviewing a lot of specs over the past few days
14:19:14 <SridarK> amotoki: thx
14:19:16 <annp> amotoki, no problem. :)
14:19:32 <xgerman_> SridarK: +1
14:19:33 <SridarK> mlavalle: yes indeed
14:19:44 <SridarK> amotoki: i think ur last set of comments were also addressed
14:19:57 <annp> mlavalle, I know you were busy, Everyone want to reach you :)
14:20:16 <amotoki> SridarK: yeah, at a glance, it has been addressed.
14:20:38 <amotoki> i just would like to look through it.
14:20:44 <SridarK> amotoki: +1
14:20:50 <annp> amotoki +1
14:21:31 <SridarK> ok lets move on
14:21:44 <SridarK> #topic Rocky Address Group spec
14:21:49 <SridarK> wkite: pls go ahead
14:21:57 <wkite> ok
14:22:07 <SridarK> If we can get a few more eyes on this for review
14:22:18 <SridarK> #link https://review.openstack.org/557137
14:23:07 <mlavalle> is this similar to what I proposed last week about address groups?
14:23:45 <SridarK> mlavalle: i think we have good alignment - i just want to look thru for any specific differences
14:23:59 <SridarK> I also see ZhaoBo has reached out to wkite
14:23:59 <mlavalle> I will take a look at this spec
14:24:12 <mlavalle> if we are making progress in this front, it is just great
14:24:35 <SridarK> mlavalle: thx - this was in the orig spec and we chose not to cover it in the initial implementation
14:24:51 <SridarK> wkite: has picked it up with a few changes
14:25:02 <SridarK> i think we can converge on this quickly
14:25:26 <wkite> this my github https://github.com/wkite/neutron_fwaas
14:25:35 <mlavalle> cool!
14:25:43 <SridarK> mlavalle: we can discuss this in detail next week when we meet to go over the Huawei proposal too
14:25:51 <mlavalle> yeap
14:26:08 <SridarK> wkite: did u want to discuss something specific here
14:26:13 <mlavalle> but it will be good if I look at this spec, so I'm aware of the possible overlaps
14:26:35 <wkite> SridarK: ok
14:27:00 <SridarK> mlavalle: sounds good
14:27:15 <xgerman_> +1
14:27:20 <annp> +1
14:27:36 <SridarK> ok lets move on
14:27:50 <SridarK> #topic Open Discussion
14:28:01 <njohnston> mlavalle asked me to look at bug https://bugs.launchpad.net/neutron/+bug/1770575
14:28:02 <openstack> Launchpad bug 1770575 in neutron "FWG status will be overided by mutilple l2 agent" [Undecided,New]
14:28:05 <xgerman_> PSA early bird for Denver ends today
14:28:18 <njohnston> it seems to me that this is a legit case but I am not sure how to rate it in terms of severity
14:28:26 <SridarK> I was swamped last few days so we could not connect on bug scrub
14:28:58 <SridarK> njohnston: ok
14:29:07 <njohnston> I have not been able to reproduce locally (issues getting my testing setup working again) but the theory seems sound
14:29:11 <amotoki> re: bug 1770575, I think we need to use provisioning block to maintain the status of FWG
14:29:11 <openstack> bug 1770575 in neutron "FWG status will be overided by mutilple l2 agent" [Undecided,New] https://launchpad.net/bugs/1770575
14:29:15 <SridarK> annp: perhaps u can take a look too on ur inputs
14:29:42 <annp> Sridark, I will look
14:30:31 <annp> amotoki, are you going to vancouver?
14:30:41 <amotoki> annp: yes, I will be there
14:30:54 <SridarK> In general the FWG status is something we need to look into
14:31:11 <annp> amotoki, I happy to hear that. :)
14:31:25 <amotoki> if multiple agents or drivers are involved, it is suggested to use provisioning block.
14:31:33 <SridarK> with multiple ports - and the fact that all ports will not always cooperatate in unison w.r.t to the FWG status
14:32:19 <annp> SridarK, I haved discussed with zhaobo in irc about this bug
14:32:20 <SridarK> amotoki: ok i will also look into that, but i think even with a single agent we can have some issues
14:32:27 <SridarK> annp: ok
14:32:32 <xgerman_> +1
14:32:46 <mlavalle> njohnston: thanks for looking at it ;-)
14:32:54 <annp> SridarK, Maybe we need to check FWG status before we update it.
14:32:56 <amotoki> SridarK: your point looks important too. we might need to revisit what is the status of FWG.
14:33:08 * mlavalle and njohnston are both multitasking in the same meetings.... LOL
14:33:10 <xgerman_> annp: that might be a possibility
14:33:50 <annp> SridarK, In case of if FWG status=ERROR, we shouldn't update status of FWG from rpc callback
14:34:03 <SridarK> amotoki: yes this is something that we need to get specific on this
14:34:37 <xgerman_> also we need to lock the status properly - so both don’t update at the same time
14:34:45 <SridarK> annp: yes but we will need to indicate that some specific port associate is in ERROR
14:34:46 <amotoki> annp: it sounds like compare-and-swap. perhaps we need to discuss what it should be.
14:34:46 <njohnston> +1
14:35:35 <SridarK> At the extreme - we can have a port specific status for every port that has a FWG associated
14:35:52 <annp> amotoki, yes. compare-and-swap or revision_number
14:35:54 <SridarK> that may be an overkill - we can possibly come to some midground
14:36:23 <xgerman_> yeah, I thibk proper locking would be enough — no need to keep history
14:36:49 <SridarK> xgerman_: yes no need for history
14:38:02 <chandanc> if you want to support port as part of multiple FWG in future, we need to think anout that case too
14:38:15 <SridarK> on a different topic, annp thx for the email - so we can sync up some time early in the week regarding the L7 topics for forum
14:38:24 <SridarK> chandanc: yes good point
14:38:32 <xgerman_> chandanc: +1
14:39:18 <annp> SridarK, xgerman_, amotoki, I'm asking Yushiro, Maybe he can help us arrange a meeting room at the summit
14:39:28 <SridarK> annp: amotoki: xgerman_ : we can coordinate on a time to meet to review
14:39:40 <SridarK> annp: yes sounds good
14:39:53 <xgerman_> that would be sweet — otherwise there is usually some developer “lounge”
14:39:54 <SridarK> we can coordinate on that email thread
14:39:55 <amotoki> yeah. I just started to check my session schedule today :(
14:40:10 <annp> Tomorrow, I will ask him again. I will let you know later
14:40:12 <xgerman_> still need to sogn up for booth duty…
14:40:16 <SridarK> amotoki: ok but it is good u are able to make it
14:40:22 <amotoki> :)
14:40:26 <xgerman_> +1
14:40:53 <annp> +1
14:40:55 <amotoki> i thin yushiro can join every time :)
14:40:56 <chandanc> agree, lucky to make it
14:40:56 <xgerman_> again if you missed it: prices for Denver PTG later this year are going up tomorrow — so book today to save $200
14:41:18 <SridarK> xgerman_: thx
14:41:29 <amotoki> early bird deadline is approaching too fast. good reminder
14:41:29 <xgerman_> yeah, they approved me today to save that $$
14:41:43 <mlavalle> that's cool
14:41:53 <annp> amotoki, yes, yushiro is always available :)
14:42:01 <SridarK> xgerman_: i am asking my boss today - i forgot earlier - thx for the reminder
14:42:33 <annp> xgerman_, congrats!
14:43:08 <xgerman_> thanks — not my doing the others prodeed
14:43:20 <amotoki> even if you cannot make it, you won't lose $$. there is a good cancelation policy (though we don't want it to happe)
14:43:20 <SridarK> :-)
14:43:43 <xgerman_> :-)
14:44:20 <annp> :)
14:44:59 <amotoki> btw, do we plan to cover some new features in rocky in the dashboard?
14:45:30 <amotoki> I don't know the plan, but if any I can help a bit.
14:45:39 <amotoki> CLI too.
14:45:45 <SridarK> amotoki: yes we will need to address that
14:45:52 <xgerman_> +1
14:46:15 <SridarK> chandanc: can u pls check in with SarathMekala - i think he has been busy on other things
14:46:24 <chandanc> I will SridarK
14:46:28 <SridarK> chandanc: thx
14:46:48 <mlavalle> SridarK: I have one request for you
14:46:54 <amotoki> perhaps we can list priority things first and then choose some of them
14:47:06 <SridarK> mlavalle: pls go ahead
14:47:09 <SridarK> amotoki: +1
14:47:23 <mlavalle> I am giving a Neutron project update on Monday
14:47:37 <mlavalle> and I want to cover FWaaS as well
14:47:57 <mlavalle> there is a set of slides with the features released in Queens
14:48:14 <mlavalle> for this, I looked at the release notes and already added my underdtanding
14:48:22 <mlavalle> however, I would like you to review it
14:48:29 <SridarK> mlavalle: ok surely thx
14:48:30 <mlavalle> is 2 or 3 highlights
14:48:45 <mlavalle> also there is another set of slides on what is planned for Rocky
14:49:00 <mlavalle> I would like you to list highlights in that section
14:49:09 <SridarK> mlavalle: ok will do
14:49:16 <amotoki> i think we need to clarify what is the initial version of FWaaS v2
14:49:45 <amotoki> the initial blueprint of fwaas v2 covered a lot of things, but we selected more important things
14:49:49 <mlavalle> amotoki: you mean in the update next week?
14:50:19 <amotoki> mlavalle: hopefully so (but it is not a requirement)
14:50:44 <amotoki> I heard some confusions on fwaas v2 from folks who see the fwaas-v2 spec.
14:51:06 <mlavalle> I'll be glad to talk about that
14:51:12 <amotoki> it is a comment from operators perspective.
14:51:17 <mlavalle> if we can do it in a few minutes
14:51:33 <mlavalle> and if SridarK gives me a hand
14:51:55 <SridarK> amotoki: ok we can clarify
14:52:02 <SridarK> mlavalle: surely
14:52:10 <mlavalle> cool
14:52:11 <amotoki> one good example is that address group hasn't landed.
14:52:21 <mlavalle> that's a good point
14:53:19 <amotoki> anyway we made a good progress in queens and hopefully rocky. it would be nice to cover this :)
14:54:04 <SridarK> amotoki: +1
14:54:57 <SridarK> Ok sounds good - if nothing else we can call it a meeting
14:56:21 <SridarK> ok folks have a great week and thx for joining
14:56:34 <mlavalle> are we meeting next week?
14:56:40 <SridarK> we can possibly cancel the meeting next week
14:56:48 <mlavalle> yeah, that makes sense
14:56:59 <SridarK> i will send a notification
14:57:17 <SridarK> Thx and bye all
14:57:23 <SridarK> #endmeeting