19:01:41 <jeblair> #startmeeting infra
19:01:42 <openstack> Meeting started Tue Jul  8 19:01:41 2014 UTC and is due to finish in 60 minutes.  The chair is jeblair. Information about MeetBot at http://wiki.debian.org/MeetBot.
19:01:43 <krtaylor> o/
19:01:43 <anteaya> I am
19:01:43 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
19:01:45 <openstack> The meeting name has been set to 'infra'
19:02:00 <jeblair> #link https://wiki.openstack.org/wiki/Meetings/InfraTeamMeeting
19:02:06 <jeblair> #link http://eavesdrop.openstack.org/meetings/infra/2014/infra.2014-07-01-19.02.html
19:02:25 <jeblair> drat, i think most of the agenda is left over from last week
19:02:45 <jeblair> #topic  Actions from last meeting
19:02:50 <jeblair> nibalizer and jeblair to spin up a puppet3 master
19:02:54 <jeblair> this did not happen
19:02:56 <pleia2> o/
19:03:03 <jeblair> several changes were still outstanding and needed to land
19:03:20 <jeblair> nibalizer: i think they have landed now, and we're ready to actually try booting one?
19:04:43 <jeblair> that's my story and i'm sticking to it
19:04:46 <jeblair> #topic  Zanata/Pootle updates and time frame (pleia2)
19:04:56 <pleia2> so I have a couple updates here
19:05:41 <pleia2> I met with the translatiosn team at their meeting last week, they're still keen to try Zanata "soon" since they're having trouble with Transifex support (creating new projects is hard right now, takes a lot of time since they need to go to support)
19:05:47 <nibalizer> jeblair: very quickly, yes we can, but im dealing with some other stuff right now, so  i'll emerge from the cloud... eventually
19:06:10 <nibalizer> cloud was perhaps not the right word in this context
19:06:21 <Ajaeger1> pleia2: Here's an update from transifex: They fixed the problem, we can create new projects again!
19:06:26 <pleia2> the Zanata folks told us 5-6 weeks before they have Zanata running on the WildFly application server, which we can run on Centos7 - no licenses
19:06:32 * Ajaeger1 created 5+ new projects the last few days
19:06:36 <pleia2> Ajaeger1: ah good, so less urgent :)
19:06:43 <Ajaeger1> pleia2: exactly!
19:07:10 <Ajaeger1> still, the long time it took to resolve this, means we have to switch eventually.
19:07:28 <jeblair> pleia2: does wildfly have packages in centos?
19:07:33 <pleia2> so we can set up basic Zanata + AS7 pre-packaged thing in the -infra so the Translations folks can test it + upgrade Pootle and schedule a demo with a Pootle contributor so they can compare side-by-side
19:07:41 <pleia2> jeblair: no, it's still heavily in development
19:07:57 <jeblair> pleia2: how will be deploy wildfly?
19:07:57 <pleia2> jeblair: it's pretty much jboss, we'd manage from tarballs, not awesome tbh :\
19:08:36 <jeblair> pleia2: isn't that the same problem with jboss?
19:09:00 <pleia2> jeblair: jboss has that problem, plus the community version having known security problems and the enterprise version requiring a paid license
19:09:19 <pleia2> so with wildfly we only solve the licensing issues
19:09:41 <jeblair> pleia2: wildfly community version is expected to get security updates?
19:09:51 <pleia2> jeblair: that's the plan
19:10:05 <jeblair> i have to admit, i'm extremely skeptical
19:10:20 <jeblair> i believe redhat chooses to make jboss installation difficult as a business decison
19:10:29 <jeblair> and i don't see why that would be different with wildfly
19:10:40 <pleia2> as I understand it, it's not jboss, but it is a java application server that will have the ability to run zanata
19:10:56 <pleia2> but this is all new to me :)
19:10:58 <jeblair> not to mention, the number of hoops we have to go through to manage installing software from tarballs
19:11:29 <fungi> well, red hat traditionally made their linux distro security updates hard too, as part of their business model. hence centos. then they acknowledged that people want to run community versions of server things, so they absorbed centos
19:11:30 <zaro> pleia2: can't zanata run in tomcat?
19:11:32 <pleia2> I also have very little expertise managing java/jboss applications, not sure how we're doing talent-wise with the rest of the team
19:11:44 <pleia2> zaro: I don't think so
19:11:50 <clarkb> I have never done jboss before either :)
19:12:09 <zaro> i have *some* experience with jboss
19:12:20 <jeblair> pleia2: i've only used tomcat, because, well, the R1 university i was at couldn't afford a jboss license.  :)
19:13:20 <jeblair> pleia2: so it seems like the plan is to set up both systems un-managed for further evaluation
19:13:28 <pleia2> jeblair: yeah
19:13:59 <jeblair> i think that the puppeting, etc, should be part of the evaluation
19:14:23 <clarkb> ++
19:14:37 <jeblair> i don't think the systems are equivalent from an installation/upgrade/management pov
19:15:06 <pleia2> agreed
19:15:24 <pleia2> Carlos of the Zanata team sent me some incomplete puppet configs for zanata running on jboss+RHEL and they are keen to see improvements, so I'll take another look at those
19:15:34 <pleia2> I added this topic for our sprint next week, so I'll do some prep for that
19:15:51 <fungi> i agree that the deployment automation of a system is part of that system from an evaluation perspective. if it runs well but is impossible to deploy, then it may be less preferable to something which runs okay and is easy to deploy
19:16:13 <jeblair> pleia2: would you be interested in working with the translation team on the evaluation, to make sure that these concerns are shared?
19:16:27 <pleia2> jeblair: yeah, I'll handle that
19:16:51 <pleia2> the real concern for them is that if it's tough for us to maintain, it'll be a poor experience + have downtime as we upgrade, etc
19:16:51 <fungi> s/deploy/deploy and manage/g
19:16:52 <zaro> IIRC deploying jboss and tomcat are pretty similiar. there's lots of automation around those
19:17:17 <jeblair> zaro: deploying tomcat is "yum install tomcat"
19:17:43 <pleia2> puppet require package :)
19:18:08 <jeblair> pleia2: indeed; but i never would have gotten the rocketships aligned in irc :)
19:18:14 <pleia2> hehe
19:18:34 <pleia2> anyway, that's all I've got
19:18:58 <jeblair> cool, thanks.  i'm happy to help out with any pootle hacking that needs doing too.
19:19:09 <pleia2> great
19:19:21 <jeblair> #topic  Replacement for docs.openstack.org (AJaeger)
19:19:39 <Ajaeger1> This is a followup to the Atlanta summit session https://etherpad.openstack.org/p/summit-b301-ci-doc-automation
19:20:09 <Ajaeger1> We talked briefly about the need to replace docs.openstack.org with another system - one of the issues was a system where we can actually delete files ;)
19:20:29 <Ajaeger1> I wanted to followup on this and see what the next steps are here.
19:21:04 <clarkb> I think there are a few options open to us. like rsync or using 'slots' to host the content
19:21:08 <Ajaeger1> jeblair: I took as note after the meeting "Jim Blair will brainstorm with team on how to do it best, will need at
19:21:08 <Ajaeger1> least a month, so get back to him after mid-June." - and now I'm back ;)
19:21:17 <clarkb> I briefly looked at read the docs but that is far too complicated for our needs
19:21:19 <jeblair> Ajaeger1: we may not need a terribly complicated system, possibly just a different way of copying them
19:21:36 <Ajaeger1> jeblair: copying and deleting ;) yeah
19:21:51 <jeblair> Ajaeger1: yeah, i'd actually like to push that back another 2 weeks if we can... i'd like to do some brainstorming at our mid-cycle in germany
19:21:57 <Ajaeger1> It's serving static pages, so shouldn't be something complicated.
19:22:11 <Ajaeger1> jeblair: fine with me!
19:22:22 <fungi> in particular, knowing what "features" of rackspace cloudsites the current docs hosting might be taking for granted would be helpful
19:22:54 <wenlock> *for us
19:22:58 <Ajaeger1> fungi: if you have some questions, I can figure them out with annegentle
19:22:59 <wenlock> mt
19:23:24 <fungi> i suspect that just cramming those files onto an basic apache server may not get you everything you require. i know in the past there were content search libraries and other things provided by the cloudsites platform
19:23:41 <fungi> so knowing what they are up front will help speed the design along
19:24:00 <Ajaeger1> fungi: I can check, I'm not aware of these yet.
19:24:02 <jeblair> hrm.  current search seems to use google
19:24:22 <fungi> though with the docs-draft deployment we do at least serve up the basic content on a bog-standard apache server, so might just work
19:24:38 <fungi> jeblair: oh, they might have changed that
19:24:47 <Ajaeger1> fungi: yeah, just see http://docs-draft.openstack.org/63/105263/4/check/gate-openstack-manuals-tox-doc-publish-checkbuild/94f3e9d/publish-docs/
19:24:54 <Ajaeger1> works fine for publishing manuals...
19:25:01 <Ajaeger1> just click on the link
19:25:18 <jeblair> yeah, i think the basic question is how to delete the files we want to (obsolete files), but not the ones we don't (files that come from a different build, eg, a stable branch)
19:25:42 <Ajaeger1> jeblair: indeed.
19:26:15 <Ajaeger1> we publish to single directories for each book
19:26:20 <Ajaeger1> but only publish the changed manuals
19:26:20 <jeblair> so we'll have a bunch of people in the room that understand what we're doing now, and what we're trying to do with swift, so i bet we can come up with something
19:26:28 <fungi> cool--all good stuff to discuss next week in darmstadt
19:27:07 <Ajaeger1> if you have questions for me, just ping me - I'm in Germany as well so in your timezone next week ;)
19:27:23 <jeblair> Ajaeger1: thanks, will do!
19:27:28 <fungi> the search i was thinking of was the stuff like http://docs.openstack.org/admin-guide-cloud/content/#searchDiv btw
19:27:55 <Ajaeger1> fungi: try http://docs-draft.openstack.org/63/105263/4/check/gate-openstack-manuals-tox-doc-publish-checkbuild/94f3e9d/publish-docs/high-availability-guide/content/index.html#searchDiv
19:28:08 <Ajaeger1> works on docs-draft as well
19:28:28 <Ajaeger1> Thanks for putting this on the agenda for Darmstadt!
19:28:54 <fungi> right, dita but i forget what provides it
19:29:00 <jeblair> Ajaeger1: thank you!
19:29:07 <jeblair> what a perfect transition...
19:29:12 <jeblair> #topic Meetup
19:29:17 <jeblair> #link https://wiki.openstack.org/wiki/Qa_Infra_Meetup_2014
19:29:20 <jeblair> that's next week
19:29:32 <jeblair> we're out of room and registration is closed
19:29:33 <clarkb> yes I arrive sunday morning local time
19:29:49 <pleia2> me too
19:30:05 <zaro> clarkb is my flying buddy
19:30:10 <anteaya> me too
19:30:13 <pleia2> I get in at 10:55am
19:30:22 <jeblair> there's a waitlist, so if anyone on the registration list is not going, please let me know
19:30:36 <jeblair> also, if you are only attending certain days, please annotate that in the comments section next to your name
19:30:49 <fungi> looks like i get to frankfurt at noon
19:30:52 <anteaya> I will create an etherpad for commits to infra-manual, I'd like everyone to pick a section and offer at least one paragraph of content
19:31:02 <anteaya> I'll link the etherpad to the wikipage
19:31:13 <jeblair> anteaya: good one
19:32:43 <anteaya> I also have assorted third party things for beer discussion
19:33:15 <fungi> third-party beer
19:33:20 <pleia2> and assorted beer for third party discussion
19:33:27 <pleia2> (woo germany!)
19:33:35 <anteaya> ha ha ha
19:33:47 <clarkb> and world cup
19:33:48 <krtaylor> anteaya, take good notes on the third-party beer
19:33:52 <Ajaeger1> No need to bring your own beer to Germany ;)
19:34:16 <fungi> Ajaeger1: yeah, i expect to be able to find plenty there
19:34:20 <anteaya> krtaylor: I will be able to tell many stories
19:34:38 <anteaya> krtaylor: I still want to you be there, but can't figure out how to shoe horn you in
19:35:00 <krtaylor> anteaya, it all good, I'll stay here and get some patches done
19:35:06 <anteaya> kk
19:35:16 <anteaya> you can review many infra-manual patches
19:35:21 <anteaya> and assorted others
19:35:42 <krtaylor> yep, that too
19:35:47 <anteaya> :D
19:36:40 <jeblair> anyone have anything else?
19:36:53 <clarkb> I sort of do
19:37:23 <clarkb> it being hosted at corporate land is there a secret handshake we need to use to get into the building?
19:37:52 <clarkb> looks like we have an address to go to. And how early should we show up?
19:37:53 <jeblair> btw, i've been trying to avoid reviewing changes that don't have a +/-1 code review; so non-core infra reviewers, your help is much appreciated.  :)
19:38:10 <Ajaeger1> clarkb: last time I visited Deutsche Telekom they had a reception where I just told them who I am and what I wanted.
19:38:16 <clarkb> Ajaeger1: perfect
19:38:19 <jeblair> clarkb: i think the registration list is being used for that purpose
19:38:31 <anteaya> jeblair: I hear that
19:38:32 <Ajaeger1> But there are many buildings of DT in Darmstadt...
19:38:59 <Ajaeger1> Your host is Marc Koderer - remember his name ;)
19:39:05 <jeblair> the wiki page says " Mina-Rees-Str. 12"  hopefully that's specific enough :)
19:39:14 <fungi> if i end up at the wrong building, i'll just recruit some new dt employees to work on infra for us
19:39:34 * Ajaeger1 just looks it up
19:39:47 <anteaya> fungi: go you
19:39:50 <Ajaeger1> The whole area is Telekom buildings, so take the proper entry ;)
19:39:57 <anteaya> we need more eu representation
19:40:12 * SergeyLukjanov hopes that will be able to participate somehow through irc
19:40:30 <clarkb> I am guessing there will be enough of us weird people hanging around if we can't figure it out that someone will help us :)
19:41:01 <krtaylor> SergeyLukjanov, I was hoping for some google hangouts
19:41:20 <wenlock> +1 for a google hangout, but it's EU timezone eh?
19:41:27 <Ajaeger1> wenlock: yeah, UTC+2
19:41:38 <Ajaeger1> wenlock: there's more than one timezone in EU ;)
19:41:40 * fungi was hoping to use free software, not google
19:41:46 <wenlock> :D
19:42:03 <wenlock> fungi, would be nice for something that records the session
19:42:15 <fungi> wenlock: for example, an irc bot!
19:42:34 <wenlock> intreasting
19:42:37 * fungi thinks we already have one in the #openstack-infra channel
19:43:20 <fungi> and we should take good notes for the benefit of those not able to participate with us real-time
19:43:34 <anteaya> etherpads will be created
19:44:09 <pleia2> if we do it summit-style, etherpads were sufficient for following up after the summit I missed
19:44:28 <anteaya> cool
19:44:54 <anteaya> I don't think it will be a hardship for us to work summit style
19:44:58 <Ajaeger1> great, thanks!
19:45:05 <wenlock> take your go-pros :D
19:45:09 <fungi> yep. agreed, etherpads for taking notes
19:45:29 <anteaya> we can add links to the sprint wikipage
19:45:41 <fungi> perfect
19:45:52 <jeblair> wenlock: now i'm going to search for "extreme daredevil hacking videos" on youtube
19:46:04 <wenlock> jeblair haha
19:46:38 <pleia2> hehe
19:46:58 <jeblair> oh, one last thing...
19:47:08 <jeblair> i reckon we should probably cancel this meeting next week :)
19:47:15 <anteaya> ha ha ha
19:47:17 <pleia2> yeah
19:47:23 <fungi> i reckon so
19:47:49 <fungi> or, rather than cancel it, hold it in person, and hack on things during it
19:47:57 <jeblair> should we also cancel the week after (oscon)?
19:48:03 <clarkb> jeblair: possibly
19:48:10 <jeblair> i honestly don't know what i'm going to be doing at that time
19:48:10 <clarkb> I will be there as will you and fungi iirc
19:48:15 <pleia2> I'm skipping oscon
19:48:17 <clarkb> I won't be at oscon proper though
19:48:24 <jeblair> i think mordred will be in a board meeting...
19:48:31 <fungi> i think i can make the oscon week meeting, but it would be from a hotel room, or mobile modem or crappy conference wifi
19:48:32 <jeblair> come to think of it, that might be were some of us may be too
19:48:37 <fungi> oh, right
19:48:48 <fungi> i meant to show up for the board meeting too
19:48:51 <clarkb> fungi: we might need to do backyard hacking with kegerator
19:48:57 <clarkb> and non crappy wifi
19:48:59 <anteaya> I'm not at oscon
19:49:13 <fungi> clarkb: i'll be happy to help you hack on your kegerator
19:49:29 <fungi> clearly it needs its own microcontrolled sensor array
19:49:35 <clarkb> brewpi is a thing
19:49:42 <clarkb> occasionally I am tempted to do something like that
19:49:52 <clarkb> but then I would just worry about it more
19:49:56 <jeblair> i think we may want to declare it an opportunistic meeting... if people show up and feel strongly about having a meeting, go for it.  but expect many absences, or no meeting at all.
19:49:58 <pleia2> ok, between oscon and board things... next meeting on July 29th
19:49:59 <ttx> o/
19:50:01 <clarkb> jeblair: ++
19:50:12 <fungi> wfm
19:50:12 <clarkb> ttx: ohai
19:50:19 <fungi> ttx: talk went well?
19:50:25 <jeblair> #agreed no meeting on July 22 due to in-person meetup
19:50:37 <pleia2> jeblair: 15th
19:50:38 <jeblair> oops
19:50:46 <jeblair> is there an undo?
19:50:49 <pleia2> #undo
19:50:57 <fungi> oho, neat!
19:51:11 <jeblair> could that be more intuitive?
19:51:13 <fungi> i figured we were just going to have to #agree calendars are hard
19:51:24 <jeblair> #agreed no meeting on July 15 due to in-person meetup
19:51:32 <jeblair> #agreed possibly no meeting on July 22 due to oscon
19:51:42 <jeblair> #agreed next scheduled meeting July 29
19:51:56 <ttx> fungi: a bit empty room, but our numbers attracted the usual "wow"
19:52:08 <jeblair> ttx: what was the talk?
19:52:18 <ttx> jeblair: infra, same as my oscon one
19:52:22 <jeblair> ttx: what event?
19:52:24 <ttx> err
19:52:26 <ttx> fosdem one
19:52:31 <ttx> which you attended
19:52:42 <ttx> Libre software meeting
19:52:48 <jeblair> oh cool
19:53:02 <jeblair> that reminds me:  lca cfp closes soon.
19:53:13 <clarkb> I still haven't come up with something good for that
19:53:16 <clarkb> I should think harder
19:53:19 * ttx still tries to find a way to jutify cfping for it
19:53:24 <ttx> +s
19:53:31 <anteaya> jeblair: oh, thanks, I didn't know it was open
19:53:35 <fungi> do we have a nodepool presentation yet? i couldn't find one in the usual place
19:53:52 <fungi> would probably be something work knocking up if nobody's done one yet
19:54:01 <clarkb> oh wow really soon
19:54:01 <fungi> s/work/worth/
19:54:07 <jeblair> i'm going to submit some stuff about gertty and gear.
19:54:13 <jeblair> fungi: i have not done one yet
19:54:38 <clarkb> I had kciked around something along the lines of 100k changes there and back again
19:54:46 <clarkb> because as you know those films were filmed in that country <_<
19:54:55 <jeblair> clarkb: nice
19:54:55 <anteaya> clarkb: ohhh, I like it
19:55:02 <clarkb> maybe I will go ahead and do that
19:55:06 <anteaya> you are waxing poetic on the titles
19:55:14 * clarkb does this
19:55:17 <anteaya> :D
19:55:25 <anteaya> work ghost tooth into it
19:55:46 <jeblair> clarkb: statistically speaking, several people in the audience will have worked on those films.  :)
19:55:52 <clarkb> jeblair: :)
19:56:14 <clarkb> ok I can't not submit that paper now
19:56:17 <jeblair> i just hope we don't revert all 100k changes
19:56:40 <clarkb> that was my secret plan
19:56:47 <fungi> statistically speaking, several people in the audience will have reverted some of those 100k changes
19:56:55 <jeblair> i mean, i'm trying to :)
19:57:00 <clarkb> no I wanted to look forward to the next 100k
19:57:09 <anteaya> This year the papers committee is going to be focused on open source in education -- hey I can talk about mentoring and mentoring mentors
19:57:17 <clarkb> so a this is how we got here and it was crazy. This is how much crazier its going to be
19:57:18 <anteaya> I wonder if they will like that proposal
19:58:29 <jeblair> well, thanks everyone!  see many of you next week!
19:58:39 <Ajaeger1> Have a good trip to Germany!
19:58:41 <fungi> looking forward to it!
19:58:42 <anteaya> thanks jeblair
19:58:42 <pleia2> thanks jeblair
19:58:42 <krtaylor> safe travel everyone
19:58:48 <jeblair> #endmeeting