17:00:18 <dtantsur> #startmeeting ironic
17:00:19 <openstack> Meeting started Mon Dec 11 17:00:18 2017 UTC and is due to finish in 60 minutes.  The chair is dtantsur. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:00:20 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
17:00:20 <crushil> \o
17:00:22 <TheJulia> o/
17:00:23 <openstack> The meeting name has been set to 'ironic'
17:00:24 <jroll> \o
17:00:25 <jlvillal> o/
17:00:27 <rpioso> o/
17:00:32 <hshiina> o/
17:00:33 <dtantsur> hi all (again) :)
17:00:35 <fellypefca> o/
17:00:42 <mjturek> o/
17:00:44 <nicodemos> \o
17:00:49 <etingof> o/
17:00:51 <milan_> o/
17:00:57 <dtantsur> #link https://wiki.openstack.org/wiki/Meetings/Ironic
17:01:11 <rloo> o/
17:01:37 <dtantsur> okay, looks like enough people remember that our meeting is here now :D
17:01:37 <sambetts> o/
17:02:15 <dtantsur> #topic Announcements / Reminder
17:02:25 <dtantsur> #info Tempest plugin migration THIS WEEK Mon - Wed
17:03:00 <dtantsur> anything else?
17:03:00 <rloo> dtantsur: so the inspector patches are failing in stable/pike? do they need to work first before migration?
17:03:14 <rloo> ironic jobs seem to be working
17:03:24 <jlvillal> Good question. Does inspector use the ironic tempest code?
17:03:27 <dtantsur> I guess we should work it out in parallel..
17:03:30 <dtantsur> yes, it does
17:03:44 <dtantsur> my main concern is that if we delay the migration again, it will slip to mid-Jan
17:03:51 <dtantsur> which is a bit too late to destabilize the gate
17:04:07 <rloo> i guess we also have to get the multinode/nova patch fixed too, before migration?
17:04:11 <jlvillal> Okay. Not sure if we want to discuss now (annoucement section) or later...
17:04:12 <rloo> we can discuss after meeting
17:04:30 <dtantsur> rloo: totally, it has 1x +2 already
17:04:39 <dtantsur> maybe for open discussion..
17:04:46 <dtantsur> anything else to remind/announce/...?
17:05:50 <dtantsur> #topic Review action items from previous meeting
17:06:03 <dtantsur> #link http://eavesdrop.openstack.org/meetings/ironic/2017/ironic.2017-12-04-17.00.html
17:06:20 <dtantsur> we only have bug triaging effort by etingof and milan_, how did it go? :)
17:06:32 <milan_> dtantsur, 2 bus
17:06:38 <milan_> updated the ether pad
17:06:49 <milan_> I think both are good to implement
17:06:57 * milan_ hopes nothing missed
17:07:20 <dtantsur> cool
17:07:52 <dtantsur> #topic Review subteam status reports
17:08:06 <dtantsur> #link https://etherpad.openstack.org/p/IronicWhiteBoard line 148
17:09:10 <rloo> dtantsur: remind me, are we trying to get the nova changes for version negotiation done, in queens?
17:09:31 <dtantsur> rloo: ideally, because we bumped the hardcoded version again
17:10:13 <rloo> dtantsur: ok, cuz i think the nova feature freeze is late Jan. is it a nova bug or feature?
17:10:36 <TheJulia> rloo: it would be a bug realistically
17:10:40 <dtantsur> I see it more as a bug
17:11:19 <rloo> dtantsur: let us know when you might want reviews on whatever then. i don't remember how/when the deadline is for bugs
17:11:28 <dtantsur> never? :)
17:11:36 <rloo> might be good to verify with nova. that it is a bug...
17:11:53 <dtantsur> the only problem with this "bug" is that it'll require an ironicclient version bump
17:11:56 <rloo> dtantsur: well, for critical, high. but others, who knows..
17:11:58 <jroll> did we file a bug with nova? (or a blueprint?)
17:11:59 <dtantsur> so it cannot go to a stable branch
17:12:13 <dtantsur> jroll: not that I'm aware of. TheJulia?
17:12:31 <jroll> that's the first step, I seriously doubt nova takes that change without something in LP
17:12:45 <rloo> would be good to make sure ironic & nova are on the same something on this...
17:12:49 <TheJulia> jroll: it is more along the lines of "Hey ,you guys are shooting yourselves in the foot, please fix this"
17:13:09 <rloo> still, they will want a bug or something. i suspect.
17:13:12 <jroll> sure. I stand by what I said :)
17:13:16 <TheJulia> and really, the whole idea would have been to do it when we bumped a microversion in our virt driver code
17:13:22 <jroll> just needs to be written down / communicated
17:13:29 <jroll> this has been contentious before
17:13:34 <dtantsur> TheJulia: I guess we can fix the configdrive thing that merged this cycle
17:13:40 <jroll> (i.e. we tried to do this once already)
17:13:57 <TheJulia> jroll: yeah, things change though
17:14:04 <TheJulia> dtantsur: That might be a feature then
17:14:21 <jroll> TheJulia: agree. still 99% sure it needs something in LP, that's all I'm saying
17:14:30 <dtantsur> "Fixes compatibility with pre-Queens Ironic" ;) <-- see, a bug fix!
17:14:32 <rloo> is there an action item here? :)
17:14:37 <jroll> which will quickly answer the bug/feature question
17:14:57 <dtantsur> I can take an action of getting a bug and showing it to folks
17:15:09 <dtantsur> #action dtantsur to create a bug against nova for not hardcoding ironic API version
17:15:12 <dtantsur> something like ^^
17:15:18 <rloo> yay, candidate #1. dtantsur! thx!
17:15:21 <TheJulia> I feel like there is actually a bug someplace already, but I don't remember exactly
17:15:37 <dtantsur> TheJulia: please throw it my direction if you find it
17:16:40 * rajinir joined
17:16:55 <TheJulia> dtantsur: I'll try to take a look later today, but no promises
17:17:07 <dtantsur> thnx
17:17:08 <rloo> mgoddard, johnthetubaguy: any updated status related to traits work?
17:17:58 <dtantsur> jroll likes traits, let's make him finish it :D
17:17:59 <mgoddard> hi rloo, john is unlikely to be joining, as he has a new human to look after
17:18:06 <dtantsur> \o/
17:18:13 <jroll> >.>
17:18:21 <rloo> mgoddard: yay!
17:18:23 <TheJulia> jroll: fyi we almost were blocked last cycle on the microversion bump because they wanted to see it more graceful then. We ended up shooting ourselves in the foot with the microversion pin due to grenade testing where they made some major changes, and nova-compute was failing to operate upon restart since ironic had not been updated yet
17:18:26 <pas-ha> :)
17:18:27 <rloo> mgoddard: are you planning on taking over?
17:18:33 <mgoddard> I've not done any work on traits and John didn't mention having done any before he left
17:18:54 <rloo> mgoddard: ok
17:19:01 <jroll> dtantsur: I know you're joking, but not sure I have time to pick up big things yet. would like to pick it up if I can find time but no promises
17:19:17 <dtantsur> no problem man, I was joking indeed :)
17:19:23 <dtantsur> I can pick it up, but not in Dec
17:19:25 <mgoddard> It's not currently lined up for me but I'm aware it is slipping so I'll talk to stig and see if I can get some time for it
17:19:34 <TheJulia> hmmmm
17:20:00 <Grenth> hey
17:20:31 <Grenth> is inspection_dhcp_wait_timeout still a working kernel param during introspection? :x
17:20:37 <jroll> TheJulia: gotcha
17:20:48 <dtantsur> Grenth: it should look differently, but we have a meeting here right now :)
17:20:55 <dtantsur> please wait some more time
17:21:08 <Grenth> alright! sorry for bothering...
17:22:36 <TheJulia> I keep retyping thoughts on the traits stuff, I'm not sure we have much of a choice other than just getting it done asap
17:22:54 <dtantsur> yep, I'd not delay it forever
17:23:02 <dtantsur> if not Queens, then early Rocky at the latest
17:23:10 <rloo> TheJulia: i think we'd like the basic node traits in Queen's, right?
17:23:36 <dtantsur> yes
17:23:37 <TheJulia> well, when exactly is nova planning on removing processing of properties?
17:23:45 <TheJulia> for scheduling purposes?
17:24:00 <dtantsur> jaypipes: up for a quick question ^^^?
17:24:01 <rloo> i don't think they've announced that.
17:24:14 <TheJulia> I understood it to be Rocky timeframe
17:24:22 <TheJulia> but, I don't think I've heard anything official either
17:24:25 <jroll> I suspect it's waiting on us
17:24:53 <rloo> i thought there was a comment in the traits spec, from tubafather, about that not going away until ... i forgot what.
17:25:06 <jaypipes> TheJulia: yeah, Rocky is when the ComputeCapabilitiesFilter will be replaced with the traits processing in placement API.
17:25:18 <dtantsur> thanks jaypipes
17:25:23 <jaypipes> no problemo.
17:25:31 <dtantsur> so we have a hard deadline of Rocky. I'd prefer the bits to be in place a bit earlier though..
17:25:45 <rloo> yup.
17:26:15 <TheJulia> So basically we need client api and rest api and storage done by what, week of January 22nd?
17:26:24 <rloo> the node traits bit should be 'easy', just a matter of coding :)
17:26:34 <dtantsur> TheJulia: ideally, yes
17:26:53 <TheJulia> I think this just became the top priority
17:27:01 <TheJulia> because otherwise, we're not going tbe bale to deliver it
17:27:17 <TheJulia> to be able, silly browser locking up for a moment
17:27:18 <dtantsur> I'm all for it, if we find someone with free time to work on it
17:27:22 <mgoddard> ok, I've just got the go ahead for some time on traits
17:27:31 <dtantsur> nice! mgoddard++
17:27:36 <rloo> mgoddard to the rescue! thx! :)
17:27:45 <dtantsur> should we vote on making this an essentialy priority for the cycle?
17:27:51 * dtantsur knows rloo likes voting
17:28:15 <rloo> it is definitely high, is it essential?
17:28:43 <dtantsur> it's already high, but I think TheJulia means it's essential
17:28:46 <rloo> well, yeah, i suppose it is essential.
17:28:48 <TheJulia> without it, things are likely goign to be a very hard breaking upgrade if someone goes to queens -> rocky
17:28:52 <jroll> +1
17:28:57 <rloo> if they really are going to change filters in rocky.
17:29:08 <rloo> (let's bet on whether the nova change happens in rocky... just joking)
17:29:12 <rloo> ok, essential...
17:29:20 <rloo> let's vote cuz i love that
17:29:22 <openstackgerrit> Hugo Nicodemos proposed openstack/ironic master: Introduce hpOneView and ilorest to OneView  https://review.openstack.org/523943
17:29:48 <dtantsur> #startvote Should we raise the priority of the traits for to essential? Yes, No
17:29:49 <openstack> Begin voting on: Should we raise the priority of the traits for to essential? Valid vote options are Yes, No.
17:29:50 <openstack> Vote using '#vote OPTION'. Only your last vote counts.
17:29:55 <dtantsur> #vote yes
17:30:00 <sambetts> #vote yes
17:30:02 <jroll> #vote yes
17:30:02 <TheJulia> #vote yes
17:30:03 <rloo> #vote yes
17:30:13 <jroll> mgoddard: fwiw I should be able to help with reviews on this, if not code
17:30:17 <jroll> feel free to ping
17:30:22 <jlvillal> #vote yes
17:30:28 <vdrok> #vote yes
17:30:34 * dtantsur will review as well, me likes API additions :)
17:30:47 <mgoddard> jroll: fantastic, thanks
17:31:11 <dtantsur> any more votes?
17:31:22 <dtantsur> c'mon, somebody vote no just for fun :D
17:31:25 <etingof> #vote yes
17:31:29 <etingof> ;)
17:31:41 <NobodyCam> #vote maybe (for dtantsur)
17:31:41 <dtantsur> okay, okay
17:31:41 <openstack> NobodyCam: maybe (for dtantsur) is not a valid option. Valid options are Yes, No.
17:31:48 <dtantsur> lol
17:31:51 <TheJulia> lol
17:31:53 <NobodyCam> :)
17:31:57 <Nisha_Agarwal> #vote yes
17:31:57 <rloo> #vote no cuz I'm not convinced nova will actually make the switch in rocky
17:31:58 <openstack> rloo: no cuz I'm not convinced nova will actually make the switch in rocky is not a valid option. Valid options are Yes, No.
17:32:08 <rloo> i tried :)
17:32:11 <dtantsur> that bot is picky :D
17:32:13 <dtantsur> anyway
17:32:13 <TheJulia> heh
17:32:16 <dtantsur> #endvote
17:32:17 <openstack> Voted on "Should we raise the priority of the traits for to essential?" Results are
17:32:18 <openstack> Yes (9): TheJulia, vdrok, rloo, jlvillal, jroll, Nisha_Agarwal, etingof, dtantsur, sambetts
17:32:32 <dtantsur> #agreed The traits work priority is raised from High to Essential
17:32:40 <dtantsur> finished with the statuses now?
17:32:55 <rloo> +1 finished, that was stressful... :)
17:33:16 <dtantsur> it should be, it's hard work!
17:33:24 <dtantsur> #topic Deciding on priorities for the coming week
17:33:30 <dtantsur> let's see where we are
17:33:43 <dtantsur> #link https://etherpad.openstack.org/p/IronicWhiteBoard line 109
17:34:03 <rloo> the tempest migration. do the rest of us need to do anything or it is in jlvillal and dtantsur ballcourt?
17:34:18 <dtantsur> rloo: review things when needed
17:34:25 <jlvillal> rloo, Well the inspector stuff needs to get fixed.
17:34:31 <dtantsur> esp. since only you, TheJulia and me have core on stable branches
17:34:42 <rloo> ok
17:34:45 <jlvillal> I will work on the new tempest plugin repository when I get the word.
17:35:08 <rloo> i don't see any changes to what's there for priorities then
17:35:25 <rloo> but next week, i hope to see traits patches :)
17:35:37 <dtantsur> I'm about to block any patches changing the tempest plugin after this meeting
17:35:52 <rloo> dtantsur: thx
17:35:54 <dtantsur> so, how's the list looking? no new additions, except for the split
17:36:31 <TheJulia> stepping outside
17:36:47 <rloo> dtantsur: i think it is good
17:37:02 <dtantsur> we need a vote each time :D
17:37:07 * dtantsur is kidding, as usual
17:37:10 <rloo> yay, more votes!
17:37:16 <rloo> boo
17:37:20 <dtantsur> moving on in 3..
17:37:22 <dtantsur> 2..
17:37:25 <dtantsur> 1..
17:37:32 <dtantsur> #topic Appointing a bug triaging lead for the coming week
17:37:41 <dtantsur> who's up?
17:37:57 * dtantsur wants to appoint jroll :D but he's a kind PTL
17:38:05 <milan_> crickets? :D
17:38:14 <jroll> don't make me quit already, I just started!
17:38:17 <dtantsur> lol
17:38:18 <TheJulia> lol
17:38:27 * milan_ can take it again
17:38:33 <dtantsur> jroll: you're stronger than that, you survived being a PTL for 2 cycles
17:38:36 <dtantsur> thanks milan_
17:38:37 <TheJulia> I can also do it
17:38:43 <dtantsur> #action milan_ to lead bug triaging again
17:38:46 <dtantsur> mm
17:38:48 <dtantsur> #undo
17:38:49 <openstack> Removing item from minutes: #action milan_ to lead bug triaging again
17:38:55 <dtantsur> #action milan_ and TheJulia to lead bug triaging again
17:39:00 <milan_> \0/ :D
17:39:01 <dtantsur> meh, "again"
17:39:07 <dtantsur> okay, I won't fix it the 2nd time :)
17:39:14 <milan_> lol
17:39:17 <dtantsur> #topic Lets discuss a bug: Delete/Creation of ports on active nodes
17:39:24 <dtantsur> #link https://bugs.launchpad.net/ironic/+bug/1736373
17:39:26 <openstack> Launchpad bug 1736373 in Ironic "succeed to delete port and create port when node is active" [High,In progress] - Assigned to wangzhengwei (wangsansui)
17:39:28 <dtantsur> TheJulia: the mic is yours
17:39:46 <TheJulia> So I ran across this bug in launchpad ?friday? and I'm thinking we need to discuss it
17:40:18 <TheJulia> specifically should we be allowing port deletion/creation when a node is in an active state
17:40:32 * TheJulia thoughts we had it blocked long ago, but couldn't remember
17:40:42 <dtantsur> I read it as deletion is already allowed
17:40:46 <dtantsur> but update is not
17:41:18 <rloo> i think the bug is 'we currently allow create/delete of ports when a node is active; the bug is that we shouldn't
17:41:41 <sambetts> +1 I agree we shouldn't (unless in maintenance)
17:41:56 <vdrok> yeah, at least deleting should be forbidden
17:42:06 <rloo> as an aside, i wonder what they were trying to update
17:42:14 <dtantsur> given that it's a breaking change, it has to be at least microversioned..
17:42:31 <sambetts> right I think that is the reason we never did it for delete/create
17:42:40 <rloo> argh, love/hate microversion
17:42:49 <sambetts> because it only applies from a new microversion onwarfd
17:42:50 <milan_> nova allows that on an instance?
17:43:09 <dtantsur> milan_: nova operates on virtual interfaces, it's natural to create/delete them
17:43:29 <dtantsur> most hardware still does not allow creating/deleting interfaces on demand, though we're moving there..
17:43:32 <sambetts> nova allows attach/detach of neutron ports, but this is for ironic phyical devices which means someone physically added a new NIC card
17:43:40 <jroll> that makes me wonder if this should depend on the network driver, since some hardware can do that thing, right?
17:43:47 * jroll looks at sam
17:43:54 <TheJulia> My concern was the fact that we would have to guard it with a microversion, and I'm not sure we really want to do this at this point
17:43:54 <sambetts> true
17:43:55 * milan_ was just wondering, reconfiguring a switch might not be that hard to imagine
17:43:56 <dtantsur> even if they can, doesn't it make sense to put a node in maintenance?
17:44:13 <dtantsur> milan_: you're confusing VIF attach/detach with port creation/deletion
17:44:23 <sambetts> dtantsur: I have a case where on vif attach I'll want to create a port in Ironic (I think)
17:44:26 <milan_> dtantsur, yeah maybe O:-)
17:44:55 <dtantsur> sambetts: this is from inside of ironic, right?
17:45:01 <dtantsur> not like an operator does it?
17:45:08 <sambetts> dtantsur: yeah from the inside
17:45:14 <sambetts> so not through the API
17:45:32 <dtantsur> that's different in my opinion. from inside we can make sure it happens in the right way and with the right hardware
17:45:43 <sambetts> +1 I think that makes sense
17:45:54 <rloo> what about portgroups
17:46:20 <TheJulia> I would htink the vif would have to be detached, ports updated, or portgroup updated, then the vif reattached
17:46:24 <TheJulia> in order to do it properly
17:46:25 <dtantsur> deleting an active port/port group may make a VIF ID to be lost.. not sure how bad it is
17:46:44 <TheJulia> dtantsur: they are stored on the portgroup, not on the port
17:46:46 <dtantsur> I think we at least should not allow deleting ports with a VIF attached
17:46:50 <dtantsur> TheJulia: on either
17:47:10 <sambetts> dtantsur: right but we can't block that on old API versions right?
17:47:12 <dtantsur> depending on what entity a VIF is attached to
17:47:23 <dtantsur> sambetts: well, it depends on how broken it is
17:47:23 <TheJulia> dttrue, true for ports as well, I was just referring to only portgroups
17:47:40 <dtantsur> yep
17:48:31 <dtantsur> I think we can and should prevent deletion/update of attached ports/portgroups, even without a microversion
17:48:43 <sambetts> unless in maintenance
17:48:44 <dtantsur> I'm not sure it's worth bothering with non-attached ones at all
17:48:54 <dtantsur> well, yes. in maintenance you're free to screw up yourself :)
17:49:41 <TheJulia> I can agree with this without microversion because it is kind of an edge case where someone is wanting to manipulate a node directly
17:50:29 <TheJulia> 10 minute warning
17:50:43 <dtantsur> are we in agreement? do we want a ML thread? a vote? :)
17:50:54 <sambetts> yeah, I've just thought, do we block someone deleting a port/portgroup while a node is in deploying / other transition states? because that is a seriously breaking change because it might end up with different lists of ports at different point suring deployment
17:51:12 <dtantsur> I hope we do...
17:51:18 <TheJulia> I feel like we are in agreement, but now is the time for someone to speak up if we are not
17:51:35 <rloo> maybe before we vote/decide, someone could go through and just verify when we allow what to ports/portgroups...
17:51:37 <TheJulia> sambetts: we do not, but interestingly enough the patch also looks at the presence of an instance_uuid
17:51:43 <TheJulia> s/patch/proposed patch/
17:52:55 <rloo> the instance uuid is only useful if using nova, or does bifrost also use that field?
17:53:34 <TheJulia> rloo: it is settable via the ansible modules, it is not used, but bifrost's port support is extremely lacking right now, so I really wouldn't worry about it. It is on my todo list to update support for it all
17:53:42 <dtantsur> a standalone deployment does not have to use instance_uuid
17:53:58 <TheJulia> ^^^ that
17:54:07 <rloo> so it makes me wonder about the submitted patch
17:54:29 <rloo> it shouldn't be looking for presence of the instance uuid
17:54:46 <dtantsur> yep
17:54:53 <TheJulia> I agree it likely shount, it is an easy way to know if nova is touching the node, but we should be using a list of states
17:54:56 <dtantsur> and it should be looking at attached VIF IMO
17:54:57 <sambetts> they are also specifically looking at state == active and blocking it, instead of what we do in port update which is allow if its in manageable/enroll or maintenance
17:55:04 <TheJulia> dtantsur: ++
17:55:23 <rloo> so i think we need to decide/agree on behaviour. then look at how it is coded :)
17:55:29 <milan_> submitted patch aside, bug it is, no microversion needed, all agreed? ;)
17:55:34 <milan_> rloo,++
17:55:40 * milan_ being productive :P
17:55:49 <rloo> thx milan_
17:56:05 <TheJulia> I feel like we're kind of agreeing given the riskof a port being deleted while a vif is attached
17:56:15 <rloo> if i understand, we are relaxing the restriction on 'no updates to ports and possibly portgroups' to allowing if not attached
17:56:33 <dtantsur> this is a separate question :) we probably should, but I'm not sure
17:56:35 <rloo> and we are restricting deletion of ports/portgroups, allowed only if not attached
17:56:41 <rloo> or maintenance ^^
17:56:48 <rloo> and creation of port/portgroups?
17:56:55 <TheJulia> rloo: more thought required!
17:56:59 <sambetts> we allow some port updates already, mac address works
17:57:11 <dtantsur> can someone please port to openstack-operators/oepnstack-dev?
17:57:13 <TheJulia> 3 Minute warning
17:57:16 <dtantsur> and let's get back to this topic next time
17:57:19 <dtantsur> OUCH
17:57:22 <rloo> so that goes back to it being useful for me to know what we allow for ports & portgroups, and when
17:57:31 <dtantsur> #info dtantsur is very likely to be out next Monday (and the Monday after that)
17:57:35 <dtantsur> forgot ^^^
17:57:50 <rloo> dtantsur: only monday?
17:57:57 <dtantsur> rloo: yep, next week only Monday
17:58:06 <dtantsur> then out of x-mas starting with 24th
17:58:09 <dtantsur> s/of/for/
17:58:11 <rloo> TheJulia: you avail to chair next Monday?
17:58:16 <TheJulia> rloo: I am
17:58:20 <rloo> TheJulia: thx
17:58:36 <rloo> i expect the dec 25 meeting to be cancelled
17:58:43 <rloo> oh, and jan 1...
17:58:47 <TheJulia> yup
17:58:52 <TheJulia> Both fall on a monday
17:58:53 <jlvillal> Agreed
17:59:05 <openstackgerrit> Pavlo Shchelokovskyy proposed openstack/ironic master: Use adapters for neutronclient  https://review.openstack.org/476170
17:59:06 <openstackgerrit> Pavlo Shchelokovskyy proposed openstack/ironic master: Finalize migration to keystoneauth adapters  https://review.openstack.org/478825
17:59:09 <rloo> ok, end of year party next monday then :D
17:59:14 <TheJulia> heh
17:59:15 <dtantsur> #info TheJulia will chair the meeting on Dec 18th, the ones on Dec 25th and Jan 1st are cancelled
17:59:18 <jlvillal> Well likely very little attendance from North America and Europe, I imagine
17:59:36 <jlvillal> And probably South America too ;)
17:59:36 <pas-ha> +1
17:59:40 <dtantsur> anything else before we wrap up?
17:59:54 <milan_> rloo, crickets ;)
17:59:59 <TheJulia> nothing from me
17:59:59 <jlvillal> And Oceania
18:00:02 <dtantsur> #endmeeting