21:02:07 <markmcclain> #startmeeting Networking
21:02:08 <openstack> Meeting started Mon Mar 10 21:02:07 2014 UTC and is due to finish in 60 minutes.  The chair is markmcclain. Information about MeetBot at http://wiki.debian.org/MeetBot.
21:02:09 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
21:02:11 <openstack> The meeting name has been set to 'networking'
21:02:21 <markmcclain> #link https://wiki.openstack.org/wiki/Network/Meetings Agenda
21:02:33 <markmcclain> #topic Announcements
21:02:50 <markmcclain> #info Icehouse-3 has been released
21:02:53 <markmcclain> #link http://tarballs.openstack.org/neutron/neutron-2014.1.b3.tar.gz
21:03:17 <markmcclain> #info We had more requests for FFEs than we could accept
21:05:06 <markmcclain> Making the cut was not easy, but working with the OpenStack release team we drew the line at items that were rated High or greater and/or were stuck in the merge queue when I-3 was tagged
21:05:24 <markmcclain> Had the gate not backed up those items would have merged on time
21:06:08 <markmcclain> #info Entering the stability, documentation, and testing phase
21:07:02 <markmcclain> Now that the Icehouse release is almost feature complete we're entering the time where we need to turn our attention to making the release stable and fixing any lingering bugs
21:07:37 <markmcclain> #info Only bugfixes and FFEs will be merged all new items will be deferred until Juno opens
21:08:31 <markmcclain> The RC will be cut when all blueprints/bugs targeted for it are closed
21:08:36 <markmcclain> #link https://launchpad.net/neutron/+milestone/icehouse-rc1
21:08:56 <markmcclain> We have 69 bugs targeted for RC1 which is bit high
21:09:24 <emagana> markmcclain: Do we have owners for all of them?
21:09:25 <marun> hmm, ml2 migration not on the list?
21:09:52 <markmcclain> marun: it's the first bp in RC1
21:09:53 <marun> this one appears to be missing an owner: https://bugs.launchpad.net/neutron/+bug/1244255
21:10:07 * marun is blind
21:10:55 <markmcclain> yeah… since the report was marked as incomplete, I was waiting for more info before ensuring we have an owner for 1244255
21:12:07 <emagana> markmcclain: any new bug will be consider for RC1 or not?
21:12:27 <markmcclain> emagana: for now please target new bugs to RC1
21:12:49 <markmcclain> or for those that can't target, let me know and I'll add it to RC1
21:13:08 <markmcclain> starting next week we'll be more strict on which bugs we target to RC1
21:13:10 <emagana> markmcclain: great!
21:13:25 <markmcclain> Let's move onto bugs
21:13:28 <markmcclain> #topic Bugs
21:13:53 <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 RC1 Critical Bugs
21:14:03 <markmcclain> https://bugs.launchpad.net/neutron/+bug/1112912
21:14:09 <markmcclain> nati_ueno: what is the status of this?
21:14:57 <nati_ueno> markmcclain: this is broked by enable_securiy_config option
21:15:06 <nati_ueno> https://review.openstack.org/#/c/67281/22
21:15:19 <nati_ueno> We are in the discussion of ini file design
21:15:25 <nati_ueno> for 67281
21:15:42 <armax> arosen: can you have a look at the review?
21:16:02 <markmcclain> nati_ueno: ok… when patchset 23 is posted you add related to the commit message?
21:16:56 <markmcclain> https://bugs.launchpad.net/neutron/+bug/1280464
21:17:03 <nati_ueno> patch set for 23 for which?
21:17:17 <nati_ueno> markmcclain: Ah I got it.
21:17:21 <nati_ueno> markmcclain: sure I will
21:17:26 <markmcclain> nati_ueno: thanks
21:17:35 <markmcclain> looks like jlibosva is now offline
21:17:51 <markmcclain> I'll follow up with tomorrow on the status of 1280464
21:18:01 <markmcclain> https://bugs.launchpad.net/neutron/+bug/1288427
21:18:23 <markmcclain> I've got a fix ready to push that will mitigate this issue
21:18:47 <markmcclain> our last 3 bugs are the ones we have been carrying for months
21:18:58 <markmcclain> https://bugs.launchpad.net/neutron/+bug/1265495
21:19:11 <markmcclain> https://bugs.launchpad.net/neutron/+bug/1253896
21:19:17 <markmcclain> https://bugs.launchpad.net/neutron/+bug/1230407
21:19:35 <markmcclain> Both me, salv-orlando, and others have been looking into them
21:19:55 <salv-orlando> 1230407 is even more vague than 1253896
21:20:02 <markmcclain> agreed
21:20:13 <markmcclain> all 3 can be triggered by any number of things
21:20:15 <salv-orlando> but I'd say it reming us of deadlocks and lock wait timeouts
21:20:43 <salv-orlando> We have that bp from arosen, which in theory should nullify 1253896
21:20:49 <markmcclain> right
21:21:13 <salv-orlando> basically because it removes any chance that a VM can boot without network
21:21:19 <markmcclain> #link https://review.openstack.org/#/q/topic:bp/nova-event-callback,n,z
21:21:26 <salv-orlando> still, we might end up having a lot of "boot timeout" issues ;)
21:21:31 <marun> boot can still fail, but failure will be detected early
21:21:46 <marun> and more importantly - tracked.  so we can prioritize our optimization efforts.
21:21:50 <markmcclain> ++
21:22:23 <marun> i'm hoping we can squeeze in dhcp validation too, via a dhcping-like mechanism.
21:22:40 <marun> but that maybe doesn't have to go into the rc.  we can always backport.
21:22:46 <salv-orlando> nova-event-callback is also telling that the dream of compute independent of networking was probably a chimera, but this is another story.
21:23:01 <marun> chimera == bs ? ;)
21:23:53 <markmcclain> I think we all learned that coordination is necessary
21:23:55 <salv-orlando> marun: something you hoped was true but in the end was just a dream in your head… pretty much. So yes maybe bs is a good definition
21:24:19 <markmcclain> marun: agreed… I'm thinking backport is probably what we're looking at for that fix
21:24:54 <markmcclain> Any other critical bugs the team should be tracking?
21:25:00 <marun> markmcclain: but the primary fix has to make it in, right?  since nova is planning on waiting for notifications this cycle?
21:25:15 <markmcclain> marun: arosen's fix yes
21:25:27 <markmcclain> sorry wasn't clear the backport was for dhcp-ping
21:25:38 <marun> ok
21:26:26 <markmcclain> #topic Docs
21:26:41 <emagana> hi there!
21:26:49 <markmcclain> hi
21:26:56 <emagana> just updated the wiki to make it short
21:27:46 <emagana> we may need some redhat experts because the folks helping on the installation guide are having problems with red hat, ubuntu went fine
21:28:04 <emagana> any volunteers?
21:28:11 <markmcclain> marun: can you help connect the right folks?
21:28:40 <marun> emagana: Let's coordinate after to make it happen.
21:28:50 <markmcclain> marun: thank you
21:29:01 <emagana> as I have been mentioning, there are a mayor refactor work on the installation guide ad the neutron part is one of the most important, we want to have it right!
21:29:01 <markmcclain> https://bugs.launchpad.net/openstack-manuals/+bugs?field.tag=neutron
21:29:19 <emagana> marun: thanks! I will give your name to the folks working on this.
21:29:20 <markmcclain> Looks like we have a lot of bugs open
21:29:32 <markmcclain> and not a lot in-progress
21:29:52 <emagana> indeed, some of them become obsolete
21:30:05 <emagana> we need to do a deep clean up work on the bugs
21:30:24 <markmcclain> Is that something you can tackle this week? or do you need assistance?
21:30:52 <emagana> I can complete it by Wednesday and have you or anyone else to double check before the end of week
21:31:21 <banix> emagana: For new plugins, can you point me to where I can find out what I need to do
21:31:41 <emagana> banix: sure, please ping me on the mailing list
21:31:51 <banix> emagana: will do; thanks.
21:31:59 <markmcclain> emagana: perfect ping me and/or channel when we're ready for a 2nd set of eyes
21:32:11 <emagana> marun: sounds good!
21:32:26 <emagana> upps.. I mean markmcclain
21:32:28 <emagana> :-)
21:32:39 <markmcclain> Anything else for docs?
21:33:16 <emagana> we are good!
21:33:31 <markmcclain> emagana: thanks for updating
21:33:50 <markmcclain> #topic Tempest
21:33:51 <markmcclain> mlavalle: hi
21:33:58 <mlavalle> hi
21:34:14 <markmcclain> now that we're done frantically get code merged into Icehouse
21:34:15 <mlavalle> We have been seeing continue merging
21:34:32 <mlavalle> of api tests
21:34:43 <mlavalle> over the past 2 weeks or so, we have merged 9
21:34:59 <markmcclain> seems like we could help by getting some more eyes from the neutron team on the tempest patches
21:35:18 <mlavalle> I just sent a message to the ML with a list of 6 that only need one more +2 to merge
21:35:57 <markmcclain> cool.. I was looking at the bigger set of patches where we there's not a +1 from someone on the Neutron side
21:36:07 <mlavalle> that's good…. I will send a list to neutron debs with code that needs review
21:36:17 <markmcclain> great
21:36:33 <mlavalle> I am also contacting 5 authors who have abandoned patches
21:37:00 <markmcclain> thanks for doing that
21:37:03 <mlavalle> I will thank them for their effort, ask if they still have bandwidth and if the answer is no, reassign the tests to someone else
21:37:05 <Sukhdev> I have sent several requests for https://review.openstack.org/#/c/66921/
21:37:40 <mlavalle> but all in all, good progress so far
21:37:50 <mlavalle> that's all I have
21:37:53 <markmcclain> great
21:38:31 <mlavalle> Sukhdev: I will give visibility to your patch set during next Tempest meeting
21:38:32 <markmcclain> do you know what our current pass rate is for the full job?
21:38:48 <Sukhdev> mlavalle: Thanks
21:38:55 <mlavalle> I haven't been paying attention to that, but I can do it this week
21:39:27 <markmcclain> great would nice to activate the full job permanently if we're within an acceptable passing range
21:39:39 <mlavalle> will do
21:39:50 <markmcclain> Any other items you wanted to discuss?
21:39:59 <mlavalle> that's it, thanks
21:40:12 <markmcclain> mlavalle: thanks for the update
21:40:17 <markmcclain> #topic L3
21:40:20 <carl_baldwin> hi
21:40:21 <markmcclain> carl_baldwin: hi
21:40:25 <carl_baldwin> #link https://wiki.openstack.org/wiki/Meetings/Neutron-L3-Subteam
21:40:34 <carl_baldwin> Discussed l3-high-availability, neutron-ovs-dvr, and dynamic-routing in the meeting
21:40:43 <carl_baldwin> Information about reviewing and testing l3-high-availability added to the sub team page.
21:40:56 <carl_baldwin> Bring discussion for new L3 work for Juno to the meeting Thursday at 1500 UTC.
21:41:11 <carl_baldwin> I don’t know if this fits strictly under this subteam but it affects L3 agent ...
21:41:25 <carl_baldwin> I started an etherpad about performance problems in L3 and other agents due to execution wrapper overhead.  I think we should make some progress on this before Icehouse releases.
21:41:34 <carl_baldwin> #link https://etherpad.openstack.org/p/neutron-agent-exec-performance
21:42:13 <carl_baldwin> That is all I wanted to highlight.
21:43:12 <markmcclain> yeah I think that making more efficient use of the calls executed within the wrappers is smart
21:43:26 <markmcclain> and is something that makes the final release more stable
21:43:52 <markmcclain> carl_baldwin: thanks for the update
21:43:59 <markmcclain> #topic IPv6
21:44:02 <carl_baldwin> markmcclain: Agreed
21:44:23 <markmcclain> sc68cal is traveling today
21:44:50 <markmcclain> I wanted to call attention to some of the FFE IPv6 reviews
21:44:56 <markmcclain> https://review.openstack.org/#/c/74156/
21:45:14 <markmcclain> https://review.openstack.org/70649
21:45:29 <markmcclain> #topic ML2
21:45:32 <markmcclain> rkukura: hi
21:45:36 <rkukura> hi
21:46:02 <markmcclain> mestery is on PTO this week
21:46:05 <rkukura> a few bug fixes are in the works, but no FFEs except the migration tool
21:46:08 <markmcclain> do we need another core to help review: https://review.openstack.org/#/c/76533/
21:46:31 <rkukura> that's the one - need core reviews on that
21:47:17 <marun> markmcclain: the script should work for both lb and ovs now.
21:47:34 <markmcclain> awesome
21:47:36 <marun> markmcclain: want to talk to you about how to check that db target is icehouse
21:47:46 <marun> markmcclain: we can do that offline
21:48:01 <markmcclain> ok
21:48:14 <markmcclain> armax: would you mind being the other core on the migration review?
21:48:53 <armax> I'll allocate a few cycles the next couple of days
21:48:59 <markmcclain> armax: thank you
21:49:30 <markmcclain> rkukura: any other bug fixes the team should pay attention to?
21:50:00 <rkukura> I'm working on the DB transaction fix, which will probably come in 2 parts for easy review
21:50:11 <rkukura> hope to have 1st today, 2nd tomorrow
21:50:28 <markmcclain> perfect
21:50:44 <rkukura> the combination, plus one that merged, fixes at least 2 bugs
21:51:02 <markmcclain> good to know
21:51:44 <rkukura> What's the status of https://bugs.launchpad.net/neutron/+bug/1227336
21:52:24 <banix> no recent reviews
21:52:41 <rkukura> I'll review that one, and we need another core
21:52:42 <markmcclain> I've looked at it a couple of times, so I'll take another look
21:53:31 <rkukura> I think that's about it
21:53:38 <markmcclain> rkukura: thanks for the update
21:53:56 <markmcclain> #topic General Discussion
21:54:35 <markmcclain> I know many on the services side were frustrated with the process merging code for Icehouse.
21:55:20 <markmcclain> One of the things I've been thinking about recently is how we can alter the process to make it work better
21:55:44 <markmcclain> If you've got ideas feel free to PM, email me.
21:55:54 <lukego> I'm aiming to have the CI up and running for the tail-f driver early this week, FWIW, based on Jay's fine walkthrough.
21:56:12 <markmcclain> lukego: great… thanks for letting me know
21:56:41 <mandeep> markmcclain: Is the process for API review clearly documented somewhere?
21:57:03 <markmcclain> mandeep: It has been an informal process up until now
21:57:10 <mandeep> markmcclain: (that came up in FWaaS review, and Salvatore was needed to be pulled in last minute)
21:57:49 <markmcclain> For Juno, we will clarify and publish how that process should work
21:58:09 <mandeep> markmcclain: It would be helpful if that was clearly communicated on the wiki, it would help reduce some of that last minute frustraction
21:58:30 <markmcclain> mandeep: I agree 100%
21:58:51 <SumitNaiksatam> we plan to have the "advance services common requirements" IRC meeting this week on Thursday 18.00 UTC (hope to discuss flavors among other things)
21:58:55 <mandeep> markmcclain: Thanks!
21:58:59 <markmcclain> there are several items that the team has done informally over time that we need to publish to a wiki
21:59:38 <pcm_> SumitNaiksatam: I had some discussion on ML about vendor plugins for L3 services
21:59:55 <SumitNaiksatam> pcm_: sure we can add that to the agenda
21:59:58 <pcm_> Didn't know if we want to discuss then, or if I should setup a separate IRC?
22:00:07 <pcm_> cool
22:00:11 <SumitNaiksatam> pcm_: it was originally on the agenda as well
22:00:21 <SumitNaiksatam> until we took a break from the meetings
22:00:41 <pcm_> SumitNaiksatam: Sound good.
22:00:54 <SumitNaiksatam> pcm_: but of course if its a longer discussion you can schedule separate meetings
22:01:18 <pcm_> SumitNaiksatam: we can see how it goes, and if more discussion is needed do a separate mtg.
22:01:19 <SumitNaiksatam> pcm_: in this meeting we want to get all the advanced services participants together at the least
22:01:43 <pcm_> roger that
22:02:26 <markmcclain> ok we're at time for this week. I really want to thank everyone for the long hours they put in last week to get code reviewed, revised, and reviewed again for Icehouse-3
22:02:43 <markmcclain> Talk to everyone on the ML and IRC. Have a good week.
22:02:45 <markmcclain> #endmeeting