21:01:29 <markmcclain> #startmeeting Networking
21:01:29 <openstack> Meeting started Mon Mar 17 21:01:29 2014 UTC and is due to finish in 60 minutes.  The chair is markmcclain. Information about MeetBot at http://wiki.debian.org/MeetBot.
21:01:30 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
21:01:32 <openstack> The meeting name has been set to 'networking'
21:01:34 <markmcclain> #link https://wiki.openstack.org/wiki/Network/Meetings agenda
21:01:42 <sc68cal> hello
21:02:00 <markmcclain> #topic Announcements
21:02:21 <markmcclain> We're in the release stability phase of the project
21:02:27 <gongysh> hi
21:02:46 <markmcclain> so while it is good to start talking about features for Juno there is still work to be done to make sure Icehouse is ready for release
21:02:58 <markmcclain> #topic Bugs
21:03:09 <markmcclain> #link https://bugs.launchpad.net/neutron/+bugs?search=Search&field.importance=Critical&field.status=New&field.status=Confirmed&field.status=Triaged&field.status=In+Progress
21:03:10 <arosen> Hiya
21:03:39 <markmcclain> wanted to highlight a few critical bugs
21:03:57 <markmcclain> #link https://bugs.launchpad.net/neutron/+bug/1253896
21:04:35 <markmcclain> I think after many months and hard work by many folks we're nearly ready to close this one out
21:04:59 <markmcclain> unfortunately, that top bug has been replaced by a new one
21:05:11 <markmcclain> #link https://bugs.launchpad.net/neutron/+bug/1283522
21:05:30 <salv-orlando> I think we should close 1253896 anyway - and file more detailed bugs for the remaining connectivity issues
21:05:35 <salv-orlando> 1253896 is just way too wide
21:05:47 <markmcclain> agreed
21:06:21 <marun> I wonder if the bug in question is related to holding locks on port deletion....
21:06:42 <marun> markmcclain: I'm working on resolving that - need to ensure sqlite still works without manual deletion.
21:06:59 <salv-orlando> marun: ok thanks… can I stop working on it then?
21:07:05 <markmcclain> marun we don't sqlite in production
21:07:12 <markmcclain> or is this a sqlite unit test issue?
21:07:14 <marun> markmcclain: we use it in unit tests, though
21:07:17 <marun> markmcclain: yeah.
21:07:31 <marun> i will look into that today.
21:07:55 <marun> salv-orlando: you can probably pause until a port deletion patch lands.
21:08:09 <salv-orlando> marun: which is a nice way to say abandon ;)
21:08:20 <marun> salv-orlando: well, looking at the pastes they might be unrelated
21:08:21 <salv-orlando> no problem, I just spent probably 20 minutes on that patch, so it's ok
21:08:39 <marun> salv-orlando: i'll give it a look and follow up with you offline.
21:08:49 <markmcclain> those are the main two bugs I wanted to highlight
21:09:12 <salv-orlando> all the errors generating gate failures occurred because of locks between port delete and update port status
21:09:14 <salv-orlando> 100% of them
21:09:32 <marun> salv-orlando: ok.
21:09:59 <marun> markmcclain: btw, has there been any change in occurrence of the first bug since the notification patches landed?
21:10:15 <marun> markmcclain: I would assume that would have an impact if racing were the cause.
21:10:36 <markmcclain> marun: the '96 bug or '522 bug?
21:10:41 <marun> 96
21:11:12 <marun> we still might be racing with dhcp, mind you.  but I'd expect a decrease in incidence of nova winning races at least.
21:11:15 <markmcclain> so the '96 bug was a combination of two things
21:11:54 <markmcclain> we also had a wiring race if a port unplugged/plugged in the polling interval that went undetected
21:12:32 <arosen> markmcclain: where's the bug report for that one?
21:12:50 <arosen> is that related to the nova event stuff?
21:12:50 <markmcclain> https://review.openstack.org/#/c/66375/
21:12:57 <arosen> thx
21:13:03 <arosen> ah i see.
21:14:15 <markmcclain> we always knew there were multiple causes for it, so happy to see that both fixes have worked to close it out
21:14:35 <markmcclain> Are there any other critical/high bugs the team should discuss?
21:14:58 <markmcclain> Any items that are potential release blockers that are not rated high or above?
21:15:37 <beagles> mm
21:16:01 <beagles> what's the story with that security group related issue
21:16:06 <beagles> there was a nova side and a neutron side
21:16:27 * beagles scrambles for numbers
21:17:20 <sweston> beagles: do you mean this one? https://bugs.launchpad.net/neutron/+bug/1163569
21:17:30 <markmcclain> beagles: https://review.openstack.org/#/c/44596/ ?
21:17:48 <markmcclain> sweston: that is a different bug
21:17:52 <beagles> markmcclain, yup that's the one
21:17:54 <Swami> ls
21:18:13 <sweston> markmcclain: ok
21:19:02 <markmcclain> looks like there are still items to be worked out 44596 and I was hoping Nachi would be here today
21:19:19 <beagles> markmcclain,nachi isn't online so we'll have to ping him soonest
21:19:26 <beagles> yeah, what you said
21:20:15 <markmcclain> Any other bugs to discuss?
21:20:20 <sweston> markmcclain:  I can take point on that one, as far as ping Nachi ... I've already reviewed it as it's a bit related to 1163569
21:20:40 <sweston> as far as *pinging* him
21:22:06 <markmcclain> sweston:	ok.. it is sort of related but different too as the LBaaS plugin does not plug the vip port in the same manner as the nova does
21:22:50 <sweston> markmcclain:  yup.  understood ;-)
21:23:00 <markmcclain> #topic Docs
21:23:02 <markmcclain> emagana: hi
21:23:05 <emagana> markmcclain: Hi There!
21:23:13 <markmcclain> I've noticed you've been contacting everyone to help with docs.
21:23:14 <emagana> wuick update
21:23:19 <emagana> quick*
21:23:36 <emagana> yes, trying to get all bugs assigned.. so good response from almost everybody
21:23:43 <emagana> so are covered
21:23:57 <markmcclain> awesome
21:24:16 <marun> emagana: rats, I just realized i missed that meeting on friday.
21:24:18 <emagana> we have some "old" bugs that are not valid anymore with the latest restructure of the cloud guide
21:24:21 <marun> emagana: I'll talk with you offline.
21:24:33 <emagana> marun: Thanks! We really need your input!
21:25:06 <emagana> so, everything is moving properly, wiki is updated with all High and Medium bugs
21:25:11 <emagana> will continue updating the wiki
21:25:20 <markmcclain> emagana: perfect thanks for keeping it updated
21:25:35 <emagana> markmcclain: n.p.  I am done!
21:25:43 <markmcclain> emagana: thanks for the update
21:25:53 <markmcclain> #topic Nova Parity
21:25:55 <markmcclain> beagles: hi
21:25:57 <beagles> Hi!
21:26:43 <beagles> So .. whilst reviewing I came across some "conflicting information" about some of the bzs that were judged irrelevant
21:26:58 <beagles> these were the ones about unimplemented methods in neutronv2/api.py
21:27:17 <markmcclain> ok
21:27:29 <beagles> turns out they are pretty relevant and need to be revisited... which I'm going to be doing over the next few days
21:27:37 <markmcclain> cool
21:28:46 <beagles> I've started on some planning for juno
21:28:49 <beagles> link in the minutes
21:29:25 <beagles> so not critical for now..
21:29:29 <beagles> that's about it.
21:29:32 <markmcclain> just saw that… I'll add it to my reading list
21:29:35 <markmcclain> thanks for updating
21:29:52 <markmcclain> #topic Tempest
21:29:54 <markmcclain> mlavalle: hi
21:29:57 <mlavalle> Hi
21:29:59 <beagles> I know folks are pretty busy, but if you are interested in contributing to parity pls have a look
21:30:02 <beagles> all done
21:30:09 <mlavalle> api tests patch sets have continued to merge. In total we have merged 12 patch sets. There are another 4 that require just one more +2 from tempest core reviewers to merge
21:30:41 <mlavalle> After that,we have 5 still being developed and 5 abandoned
21:31:07 <mlavalle> I am contacting the authors of the 5 abandoned and reassigning them
21:31:14 <markmcclain> ok
21:31:28 <mlavalle> all in all, good progress thanks to a committed group of code contrinutords
21:31:40 <mlavalle> contributors^^^^
21:32:13 <mlavalle> with all those patch sets we will be covering the gap analysis developed at the end of 2013
21:32:21 <mlavalle> 100%
21:32:25 <markmcclain> that's great
21:32:38 <mlavalle> that's all I have
21:32:50 <markmcclain> mlavalle: thanks for the update
21:33:01 <markmcclain> #topic L3
21:33:03 <markmcclain> carl_baldwin: hi
21:33:05 <carl_baldwin> hi
21:33:16 <carl_baldwin> Just want to highlight a couple of things.
21:33:19 <markmcclain> Any Icehouse items or is everything mostly Juno?
21:33:40 <carl_baldwin> The only thing is the discussion about wrapper overhead when running system commands
21:33:54 <markmcclain> ah.. ok
21:33:55 <carl_baldwin> There is a section on the ether pad with action items for Icehouse.
21:34:05 <carl_baldwin> https://etherpad.openstack.org/p/neutron-agent-exec-performance
21:34:15 <markmcclain> #link https://etherpad.openstack.org/p/neutron-agent-exec-performance
21:34:32 <carl_baldwin> The rest of for Juno.
21:34:38 <carl_baldwin> Good progress on DVR.  L3 agent proposed changes doc is up for review.
21:35:23 <carl_baldwin> If anyone has any other ideas for root wrap perf issue for Icehouse....
21:35:28 <carl_baldwin> That is all I have.
21:35:44 <markmcclain> #link https://docs.google.com/document/d/1jCmraZGirmXq5V1MtRqhjdZCbUfiwBhRkUjDXGt5QUQ/edit?pli=1 L3
21:35:55 <markmcclain> #link https://docs.google.com/document/d/1depasJSnGZPOnRLxEC_PYsVLcGVFXZLqP52RFTe21BE/edit?pli=1#heading=h.5w7clq272tji L2
21:36:01 <markmcclain> carl_baldwin: thanks for the update
21:36:07 <markmcclain> #topic IPv6
21:36:14 <sc68cal> hello
21:36:54 <sc68cal> working mostly on getting patches merged for the FFE
21:37:26 <sc68cal> not much else to report
21:37:55 * sc68cal is on a tablet
21:38:38 <markmcclain> sc68cal: ok.. looks like we've got 2 items in the merge gate now
21:38:44 <markmcclain> sc68cal: thanks for updating
21:38:49 <markmcclain> #topic ML2
21:38:53 <rkukura> hi
21:38:55 <markmcclain> mestery, rkukura: hi
21:39:12 <rkukura> Our only FFE is marun's plugin migration tool
21:39:30 <rkukura> its in review at https://review.openstack.org/#/c/76533/
21:39:40 <markmcclain> armax has given a +2
21:39:46 <mestery> That tool is on my list of things to test out rkukura. I hope to run it tomorrow.
21:40:08 <rkukura> I'm reviewing it too, but am not a sqlalchemy wizard
21:40:12 <marun> I'll submit a patch removing the explicit sys.exit (as per markmcclain's comment), but otherwise I've addressed reviewer comments.
21:40:27 <rkukura> marun: Updating the current review?
21:40:29 <marun> rkukura: testing > reviewing in this case :)
21:40:31 <marun> rkukura: yes
21:40:45 <rkukura> OK - I'll review at least regarding the table content, etc.
21:40:57 <markmcclain> ok
21:41:12 <rkukura> We've got a couple of bug fixes in the works for RC1
21:41:19 <marun> rkukura: yes, if something has changed wrt ml2 tables in the past week, please make sure I've kept in sync.
21:41:49 <rkukura> marun: I'll check to make sure you've got the latest stuff - but schema shouldn't be changing at this point
21:42:00 <markmcclain> rkukura: ok.. make sure the priority levels are set correctly
21:42:24 <rkukura> need one more core on https://review.openstack.org/#/c/79511
21:42:39 <rkukura> That closes one bug and partially resolves another.
21:42:41 <mestery> Yes, that one would be great to get in!
21:43:08 <rkukura> I expect the have the remainder of the other bug in review tomorrow.
21:43:16 <markmcclain> ok
21:43:26 <markmcclain> wanted to return back to maru's patch
21:43:31 <markmcclain> I'm ok with it merging now
21:43:58 <markmcclain> and then if anything else pops up we'll address with a small bug fix
21:44:08 <rkukura> markmcclain: I'd be happy to do the +A after my review
21:44:10 <salv-orlando> agreed
21:44:12 <markmcclain> any objections?
21:44:37 <mestery> None here, that sounds good to me.
21:44:43 <salv-orlando> no objection. If two cores have tested it (unfortunately this is not gated) I'm fine
21:45:04 <salv-orlando> btw, do we need to add a gate test? is that even possible?
21:45:15 <mestery> I'm on the hook to test it out ASAP if we need another core.
21:45:38 <markmcclain> salv-orlando: grenade would be our only avenue, but that work is still ongoing
21:46:12 <rkukura> mestery: would be great if you could test, but I'd still like to finish looking over the table translation logic.
21:46:26 <mestery> rkukura: Lets do both in parallel.
21:46:31 <rkukura> +1
21:46:32 <markmcclain> rkukura: will you be able to look at it post meeting?
21:46:50 <rkukura> If not immediately, I will tonight
21:46:59 <markmcclain> ok that works
21:47:04 <markmcclain> Any other ML2 items?
21:47:18 <mestery> Nothing else from me
21:47:19 <rkukura> not from me - anyone else?
21:47:47 <markmcclain> Thanks for the update and the reviews
21:47:53 <markmcclain> #topic Open Discussion
21:48:43 <markmcclain> I don't think the remaining teams have any major open items for Icehouse.  If I've missed something feel free to raise it now.
21:49:03 <rkukura> markmcclain: Is there a target date for spinning RC1?
21:49:29 <markmcclain> rkukura: so the first cut of the RC is a function of our bug list being closed out
21:49:41 <SumitNaiksatam> there is a high priority bug for FWaaS: https://review.openstack.org/#/c/80715
21:50:03 <SumitNaiksatam> looking for another core
21:50:11 <markmcclain> SumitNaiksatam: I'll take another loko at that one
21:50:23 <SumitNaiksatam> we also some test clean up: https://review.openstack.org/#/c/78457/
21:50:30 <SumitNaiksatam> markmcclain: thanks
21:50:37 <rkukura> markmcclain: No more critical or high bugs, or just critical?
21:50:59 <markmcclain> Ideally only critical and high
21:51:09 <markmcclain> anything below medium will likely be deferred
21:51:19 <rkukura> markmcclain: thanks
21:51:38 <markmcclain> so feel free to PM me or raise any bugs with too low of priority in the IRC channel
21:52:56 <enikanorov_> can we discuss admin_state vs status?
21:53:20 <markmcclain> enikanorov_: for Icehouse or Juno?
21:53:30 <enikanorov_> Juno
21:54:29 <enikanorov_> I see recent proposal from salv-orlando to remove admin_state_up attribute
21:54:36 <enikanorov_> did i get the idea correctly?
21:55:45 <markmcclain> salv-orlando: ^
21:56:16 <enikanorov_> i guess we may have lost salv-orlando sleeping :)
21:56:25 <enikanorov_> anyway, we can continue on ML
21:56:48 <enikanorov_> there's an ongoing thread. please share your opinion everyone
21:57:17 <markmcclain> enikanorov_: yeah
21:57:24 <markmcclain> #topic Design Summit
21:57:36 <markmcclain> #link http://summit.openstack.org
21:57:59 <markmcclain> Some folks might have missed it, but design summit sessions are open the Atlanta summit.
21:58:16 <salv-orlando> enikanorov_: we can keep the discussion on the ML
21:58:28 <salv-orlando> It's just an idea I am gaming, not proposing anything
21:58:28 <enikanorov_> sure
21:58:33 <markmcclain> Like last summit we will be expect that the sessions have a blueprint associate with them
21:58:37 <salv-orlando> s/gaming/gauging/
21:59:43 <markmcclain> The deadline for the design summit sessions is April 20th.
22:00:04 <enikanorov_> markmcclain: what about general sessions like 'update on **aaS' ?
22:00:30 <markmcclain> Remember it is not first to file, so if you have an overlapping topic feel free to file again as we likely to combine as in the past
22:00:44 <marun> i think we need to have a broader session that raises the possibility of more granular status
22:00:56 <marun> yes/no is not sufficient to communicate problems
22:01:08 <marun> yes/no/maybe is also not enough
22:01:19 <markmcclain> enikanorov_: Update sessions typically do not make good design summit sessions as the discussion should actionable items
22:01:31 <enikanorov_> ok, understood
22:01:45 <markmcclain> marun: yes I expect there will be 1-2 filed on better status/workflows
22:01:55 <marun> +1
22:02:28 <markmcclain> Ok we've hit out time limit for the week… thanks to everyone who's been testing the Icehouse-3 milestone and reporting/addressing bugs
22:02:39 <markmcclain> Talk to everyone on IRC and mailing list
22:02:42 <markmcclain> Have a good week
22:02:44 <markmcclain> #endmeeting