14:01:11 <ihrachys> #startmeeting networking
14:01:12 <openstack> Meeting started Tue May 24 14:01:11 2016 UTC and is due to finish in 60 minutes.  The chair is ihrachys. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:01:13 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
14:01:15 <openstack> The meeting name has been set to 'networking'
14:01:18 <ihrachys> hello my friends!
14:01:20 <njohnston> o/
14:01:23 <HenryG> o/
14:01:25 <slaweq> hello
14:01:26 <yamahata> hello
14:01:27 <hoangcx> hi
14:01:34 <ihrachys> it's nice to see you all! :)
14:01:36 <amotoki> hi
14:01:39 * ihrachys waves
14:01:39 <boden> howdy  o/
14:01:57 <regXboi> o/ o\ o/
14:02:03 <ihrachys> ok, let's get it started
14:02:05 <twm2016> o/
14:02:06 <ajo_> o/
14:02:07 <ihrachys> #link https://wiki.openstack.org/wiki/Network/Meetings Agenda
14:02:31 <bcafarel> hi
14:02:33 <ihrachys> #topic Announcements
14:02:44 <HoloIRCUser> Hi
14:02:51 <ihrachys> as you should be aware of, newton-1 is already approaching
14:03:04 <ihrachys> it's actually happening in next week or two
14:03:11 <ihrachys> #link http://releases.openstack.org/newton/schedule.html Release schedule
14:03:58 <ihrachys> to keep us focused on delivering what's scheduled, if not in N-1 then at least in N-final, please keep in mind the following...
14:04:03 <yanyanhu> Qiming, xuhaiwei_ , host of node?
14:04:03 <yanyanhu> if node is a VM managed by nova, how enduser could know the information about nova-compute nodes
14:04:24 <xuhaiwei_> hi yanyanhu
14:04:24 <ihrachys> yanyanhu: please use another channel, we run a meeting. thanks.
14:04:43 <ihrachys> we have the review dashboard contributed by rossella_s
14:04:48 <ihrachys> #link http://status.openstack.org/reviews/
14:05:04 <ihrachys> note the link at the top in "Gerrit Dashboard links" section
14:05:08 <rossella_s> thanks for mentioning this ihrachys
14:05:22 <ihrachys> it requires to be logged in, and it's personal (f.e. it hides your own patches, or patches that you voted)
14:06:01 <dasm> rossella_s: thanks
14:06:08 <ihrachys> please make sure that 1) we give priority to patches on that dashboard; 2) if you see a critical patch that is not there, please make sure the gerrit topic is correct (belonging to a bug or blueprint that is targeted for Newton)
14:06:30 <ihrachys> that will hopefully help us not to loose focus.
14:06:48 <ihrachys> more general announcements to come...
14:06:58 <ihrachys> 1. we have a mentoring program running, and people ask for help:
14:07:00 <ihrachys> #link http://lists.openstack.org/pipermail/openstack-dev/2016-May/095667.html
14:07:23 <ihrachys> as per the email, it should not require too much time on your behalf, so please consider joining the good effort
14:07:45 * njohnston volunteered, heaven help his mentee
14:07:56 <rossella_s> njohnston, thanks for that!
14:07:56 <ihrachys> and another thing to mention, amuller was looking for more volunteers to get tempest tests for neutron to a better shape:
14:08:01 <ihrachys> #link http://lists.openstack.org/pipermail/openstack-dev/2016-May/094997.html
14:08:09 <dasm> ihrachys: i'm one of the mentors. can tell that it indeed doesn't require a lot of time.
14:08:10 <amuller> Thank you Ihar :)
14:08:23 <ihrachys> I know tests are boring, but please consider doing what's right ;)
14:08:24 <rossella_s> regarding mentoring people interested can ping me...I have been a mentor for a while for the outreachy program :)
14:08:36 <rossella_s> ihrachys ++
14:09:33 <ihrachys> also worth noting is the fact that the next Monday is a holiday in US.
14:09:52 <ihrachys> so I am wondering what you think about when we will have the next one
14:10:08 <ihrachys> we can keep it on Monday, and US folks will join on best effort, or we may skip it.
14:10:44 <dasm> ihrachys: second option seems to be good solution
14:11:23 <ihrachys> everyone in agreement with that ^?
14:11:28 <amotoki> it depends on our PTL
14:11:30 <ihrachys> or you are just sleepy :P
14:11:40 <ajo_> if there's something important we can always reach out via email, but let's leave last call to our PTL may be?
14:11:46 <ihrachys> amotoki: the PTL suggested to reach out to you guys hence I am asking
14:12:01 <hichihar_> N-1 closes limit
14:12:10 <amuller> Why not have it a week from now?
14:12:18 <hichihar_> N-1 is next week?
14:12:18 <amuller> So have it on this slot next week as a one time thing
14:12:22 <dasm> amuller: tuesday?
14:12:25 <amuller> yes
14:12:30 <ihrachys> ajo_: we will definitely render unto Caesar
14:12:39 <rossella_s> amuller, +1
14:12:50 <ihrachys> amuller: I suspect the slot will be used.
14:12:51 <dasm> amuller: makes sense...
14:12:54 <ajo_> amuller +1 : we can fall back to #openstack-neutron if the slot here is busy
14:12:56 <ihrachys> but we can run it in our channel
14:13:11 * dasm is wondering what kind of mess we will have with next meetings :D
14:13:19 <ajo_> lol
14:13:40 <amuller> ihrachys: indeed the slot is taken in this room next week
14:13:53 <yanyanhu> for heat stack, host's definition is more confusing...
14:14:00 <yanyanhu> #endmeeting
14:14:12 <ajo_> yanyanhu, I believe you're talking on the wrong channel
14:14:14 <amotoki> yanyanhu: wrong channel :(
14:14:19 <amuller> ihrachys: I never understood why IRC channels are a problem considering they're limitless, we can just have it on the main channel
14:14:26 <ihrachys> someone is eager to end our meeting! :(
14:14:32 <dasm> xD
14:14:37 <ajo_> :')
14:14:40 <hichihar_> ;)
14:14:48 <ihrachys> amuller: they say it's to avoid too much clash and allow people to join more of those held.
14:15:05 <dasm> amuller: i've seen couple of meetings in regular channel. maybe it is good solution for next week?
14:15:20 <dasm> with info on ML about temporary chnge
14:15:24 <dasm> *change
14:15:26 <amotoki> tuesday 22UTC looks open
14:15:52 <ihrachys> ok, let me update the PTL with ideas popping up here, and deliver it to armax. he will make a final call.
14:16:04 <dasm> ++
14:16:06 <hichihar_> +1
14:16:12 <njohnston> +1
14:16:28 <ihrachys> final thing... armax naively said he will go to vacation the first two weeks of June
14:16:53 <ihrachys> so keep in mind he may be not that responsive as usual
14:17:03 <ihrachys> meaning, time to wait goes as far as 30 mins!
14:17:28 <ihrachys> ok, that's it on announcement front
14:17:35 <ihrachys> #topic Gate
14:18:04 <ihrachys> as you probably know, a rare patch goes without a recheck these days due to functional test failures
14:18:09 <ihrachys> #link https://bugs.launchpad.net/neutron/+bug/1567668
14:18:11 <openstack> Launchpad bug 1567668 in neutron "Functional job sometimes hits global 2 hour limit and fails" [High,In progress] - Assigned to Jakub Libosvar (libosvar)
14:18:14 <ajo_> armax, you can get rid of us for a couple of weeks, but it won't be forever!... lol. Enjoy your vacation :)
14:18:24 <ihrachys> jlibosva: you were looking into it. is it still cloudy there?
14:18:58 <dasm> it seems to be one of the patches: https://review.openstack.org/#/c/317369/
14:18:59 <jlibosva> I tried to come up with a workaround - anyone feel free to comment: https://review.openstack.org/#/c/317369/
14:19:01 * ihrachys targeted the bug to n-1
14:19:19 <jlibosva> while running debugging patch with rechecks to gather yet more info of what's going on there
14:20:04 <ihrachys> ok, let's continue in gerrit on that one. I feel you are in good hands with kevinbenton and otherwiseguy on the patch.
14:20:10 <ihrachys> we have more gate failure bugs reported
14:20:11 <ihrachys> #link https://bugs.launchpad.net/neutron/+bugs?field.tag=gate-failure
14:20:12 <jlibosva> I consulted the workaround with otherwiseguy on impacts of that patch. Basically not blocking on select() passed more than 10 times in a row
14:20:33 <ihrachys> jlibosva: great. what's the usual failure rate?
14:20:56 <jlibosva> ihrachys: IIRC according graphite it's 40-50%
14:21:18 <ihrachys> ok, seems promising
14:21:33 <ihrachys> so, back to the list of gate failure bugs I mentioned above
14:21:42 <ihrachys> we have some bugs that currently have no assignees there
14:22:06 <ihrachys> f.e. https://bugs.launchpad.net/neutron/+bug/1584920 or https://bugs.launchpad.net/neutron/+bug/1584923
14:22:07 <openstack> Launchpad bug 1584920 in neutron "ExternalGatewayForFloatingIPNotFound exception raised in gate-tempest-dsvm-neutron-full" [High,New]
14:22:08 <openstack> Launchpad bug 1584923 in neutron "test_port_security_macspoofing_port fails with IndexError: list index out of range" [High,New]
14:22:19 <ihrachys> it would be cool to see someone taking care of those orphans
14:22:48 <ihrachys> any volunteers?
14:23:11 <dasm> will try to do this: https://bugs.launchpad.net/neutron/+bug/1584923
14:23:12 <openstack> Launchpad bug 1584923 in neutron "test_port_security_macspoofing_port fails with IndexError: list index out of range" [High,New] - Assigned to Darek Smigiel (smigiel-dariusz)
14:23:36 <ihrachys> dasm: ok, please assign it to yourself if you will work on it.
14:23:43 <dasm> ihrachys: done
14:23:46 <yamamoto> dasm: i think https://review.openstack.org/#/c/251423/ also fixes that bug
14:23:48 <ihrachys> oh you did. thanks.
14:24:02 <dasm> yamamoto: oh. thanks. i'll look into this
14:24:14 <jckasper> 
14:24:21 <ihrachys> yamamoto: hm. is it a dup?
14:24:28 <dasm> ihrachys: seems to be
14:24:42 <yamamoto> i haven't looked 1584923 but i guess so
14:24:51 <ihrachys> dasm: ok. let's check it and mark appropriately if that's the case.
14:25:02 <dasm> ihrachys: yamamoto. will check
14:25:18 <ihrachys> anyone to take care of  https://bugs.launchpad.net/neutron/+bug/1584920 ?
14:25:19 <openstack> Launchpad bug 1584920 in neutron "ExternalGatewayForFloatingIPNotFound exception raised in gate-tempest-dsvm-neutron-full" [High,New]
14:26:37 <ihrachys> ok, we'll find someone. or not.
14:26:46 <carl_baldwin> I'll work on finding someone.
14:26:54 <carl_baldwin> Or, take it later this week.
14:26:58 <ihrachys> thanks Carl!
14:27:05 <ihrachys> that's the spirit :)
14:27:21 <dasm> carl_baldwin: round robin? )
14:27:22 <dasm> :)
14:27:34 <ihrachys> ok, let's hope we have the most burning gate issues taken care of. let's move on.
14:27:36 <ihrachys> #topic Bugs
14:28:14 <ihrachys> hichihar_: you were the latest deputy. do you have anything significant to report?
14:28:31 <ihrachys> were docs sufficient for you to run with the role?
14:28:33 <hichihar_> ihrachys: Nothing. It was a quiet last week although it seems this week was exciting... :P
14:29:24 <carl_baldwin> I can confirm that, I've already go a backlog of new bugs I'm working through.
14:29:24 <ihrachys> ok, let's make sure no bugs don't fall thru the cracks and we pass the role to the next deputy with some overlap
14:29:25 <carl_baldwin> :)
14:29:42 <carl_baldwin> That's me!
14:30:03 * ihrachys bows to the mightly ubiquitous Carl
14:30:08 <ihrachys> *mighty
14:30:22 <ihrachys> #topic Docs
14:30:23 <ajo_> I've signed up for the next slot, after migthy  carl_baldwin and kevinbenton
14:30:36 <ihrachys> I don't see Sam-I-Am here
14:30:48 <ihrachys> anyone want to update on docs front?
14:31:48 <ihrachys> ok, I guess we skip it
14:32:01 <ihrachys> #topic Open Discussion
14:32:09 <ihrachys> the agenda on demand agenda is empty
14:32:12 <hichihar_> Client?
14:32:18 <ihrachys> so if anyone has a thing to raise, please do
14:32:22 <ihrachys> hichihar_: what's about it?
14:32:29 <ajo_> #link https://review.openstack.org/#/c/320439/
14:32:46 <hichihar_> I thought we have client section in meeting.
14:32:53 <ajo_> if anybody has time and interest, I have put for review an spec related to flow management in the l2-agent ovs extensions
14:33:22 <hichihar_> It may be my misunderstand.
14:33:26 <ihrachys> hichihar_: not in the official agenda but we can discuss it now. do you have anything on your mind worth attention of the whole team?
14:33:28 <dasm> ihrachys: hichihar_ is right. last time we had transition to osc and keystonev3
14:34:06 <hichihar_> I don't have idea about client but client folks?
14:34:07 <ihrachys> dasm: if we think it's worth a stable topic in the agenda, let's edit the wiki page to make sure it's covered: https://wiki.openstack.org/wiki/Network/Meetings
14:34:15 <dasm> ihrachys: will do.
14:34:27 <amotoki> we have a lot of time remaining. let's discuss ajo_'s and move other later.
14:34:42 <amotoki> i forgot to add the client topic to the agenda.
14:35:01 <ihrachys> ok, let's do the ajo_'s topic
14:35:24 <ihrachys> ajo_: do you plan for more details on the matter, or that's it, and we will see more when we come to implementation?
14:35:24 <ajo_> it doesn't need a lot of discussion here probably, but I just wanted to raise awareness
14:35:53 <ajo_> ihrachys, I didn't want to drill down into the implementation details if not necessary, we can discuss that over code if the high level is clear
14:36:28 <ajo_> may be we can define more clearly how the API interface itself will change to allow this, for example
14:36:36 <ihrachys> I see. one thing worth mentioning there is how we are going to test it performance wise. introducing lots of redirections and tables may not be easy on the switch (?)
14:37:22 <ajo_> ihrachys, well, you will have the same amount of redirections if you have the same extensions happening
14:37:40 <ajo_> ihrachys,  the only thing is that you will have the non interoperability
14:37:52 <ajo_> ihrachys, ovs firewall has already lots of table jumps, and it's still faster than iptables fw
14:38:18 <ihrachys> I see. anyway, I guess it does not required a discussion here and can be postponed to gerrit.
14:38:37 <ajo_> ihrachys, ack, btw, the fixed parts that I'm proposing in the pipeline, are mostly taken from ovs firewall itself
14:38:46 <ajo_> all of them, in fact :)
14:38:48 <ihrachys> anything more to cover? amotoki, wanna do the client update?
14:38:56 <amotoki> re client, we are reviewing the initial OSC plugin support: https://review.openstack.org/#/c/309587/ it seems close to merge.
14:39:19 <amotoki> once it is merged, we will enable OSC plugin gate in the openstack client side.
14:39:52 <amotoki> in addition, I am preparing a devref on command namespaces in neutron osc plugin.
14:40:26 <amotoki> we see dynamic-routing folks are active to implement osc plugin commands. we will support them.
14:41:13 <amotoki> that's the update. in the summit, we agree that we have the osc transition status in our meeting.
14:41:28 <ihrachys> amotoki: good progress
14:41:31 <amotoki> I will add a topic to the meeting agenda after the meeting.
14:41:38 <ihrachys> amotoki: yeah, please update the agenda, I missed that point.
14:41:41 <ihrachys> thanks
14:41:59 <dasm> so maybe i'll give status about keystone v3.
14:42:12 <ihrachys> dasm: please do
14:42:12 <dasm> first patch for neutronclient was sent yesterday: https://review.openstack.org/#/c/320090/
14:42:32 <dasm> i noticed errors on nova, because made wrong assumption. will update it today
14:42:40 <HoloIRCUser2> amotoki: post this, would lbaas etc. be enabled as plugin?
14:43:08 <dasm> that's all for now in keystone v3 topic
14:43:27 <amotoki> HoloIRCUser2: yes. all commands related to advanced services will go to the OSC plugin way.
14:43:33 <ihrachys> thanks dasm for the update.
14:43:49 <HoloIRCUser2> amotoki: okay, great....
14:44:19 * ihrachys notes we experience netsplit
14:44:55 <ihrachys> ok, I guess that's all we have for today.
14:45:02 <ihrachys> keep up the good work folks!
14:45:05 <ihrachys> #endmeeting