20:00:27 <mmcgrath> #startmeeting Infrastructure
20:00:27 <zodbot> Meeting started Thu Aug 26 20:00:27 2010 UTC.  The chair is mmcgrath. Information about MeetBot at http://wiki.debian.org/MeetBot.
20:00:27 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
20:00:30 <mmcgrath> #meetingname infrastructure
20:00:30 <zodbot> The meeting name has been set to 'infrastructure'
20:01:05 <mmcgrath> #topic who's here?
20:01:15 * nirik is lurking around.
20:01:15 <smooge> here
20:01:18 * wulfin is here
20:02:57 <mmcgrath> Ok, lets get started
20:03:01 <mmcgrath> #topic Infrastructure -- The release
20:03:07 <mmcgrath> I think everything went well
20:03:10 <mmcgrath> had a few link misshaps
20:03:26 <mmcgrath> sijis: did we get whatever docs updated that we needed to to make sure the link thing doesn't happen again?
20:04:33 * jsmith was very happy with the job infrastructure did with the alpha release
20:04:58 * mmcgrath assumes sijis is busy at the moment
20:05:00 <mmcgrath> jsmith: thanks
20:05:06 <mmcgrath> anyone have anything else related to the release?
20:05:53 <mmcgrath> ok
20:05:55 <mmcgrath> #topic FAS update
20:06:01 <mmcgrath> I'm working on deploying the new fas version
20:06:10 <mmcgrath> everything has been fine so far except bodhi, which errors on login for some reason.
20:06:20 <mmcgrath> I just pinged lmacken to see if he has any ideas (like just seconds ago pinged him)
20:06:24 <jsmith> mmcgrath: What's the reason for the change, for those of us who haven't been following?
20:06:33 <mmcgrath> jsmith: the fas update?
20:06:51 <jsmith> mmcgrath: Yes
20:07:15 <mmcgrath> the older version of sqlalchemy is no longer going to be supported so we've had to change to a newer version, which caused a massive rewrite of the data layer of the app
20:07:28 <mmcgrath> which also caused alchemy issues with older versions of turbogears
20:07:42 <mmcgrath> so to fix it without being too hackish we're updating RHEL6 and fas at the same time on the fas servers.
20:08:03 <mmcgrath> this new version also has some basic fixes as well as the invite-only stuff tibbs put together
20:08:09 <jsmith> Cool
20:08:17 <CodeBlock> oh meeting. I'm here-ish
20:08:19 <CodeBlock> I'm in class
20:08:25 <krich> is the plan to move fas to tg2 at some point?
20:08:52 <mmcgrath> krich: it is
20:09:16 <mmcgrath> though we're in research mode on that.  we have a custom identity provider since fas is an identity provider which has made that a little difficult
20:09:21 <mmcgrath> but I'm confident we'll get there
20:09:42 <krich> cool
20:09:53 <mmcgrath> anyone have any questions or concerns there?
20:10:09 * mmcgrath notes krich has started helping me with FAS
20:10:14 <mmcgrath> much appreciated krich
20:10:31 <krich> no prob ... look forward to getting into it
20:10:48 * smooge wonders if tg3 will occur just as we get to tg2
20:10:48 <mmcgrath> ok
20:10:57 <mmcgrath> #topic staging
20:11:01 <smooge> cool thanks krich
20:11:01 <mmcgrath> I rebuilt all of staging last week
20:11:05 <lmacken> smooge: no, I'm quite confident that it won't :)
20:11:06 <smooge> yeah!
20:11:24 <mmcgrath> the build system still doesn't quite work but I'm hoping to get to that this week / next week sometime.
20:11:30 <mmcgrath> all in all it went well
20:11:32 <smooge> mmcgrath, thanks
20:11:34 <mmcgrath> I was pleased by that.
20:11:50 <mmcgrath> As part of that, we also have xen13 running RHEL6(ish) and is running kvm
20:11:54 <mmcgrath> so all of staging is in kvm
20:11:59 <mmcgrath> so far so good, we've hit a few bumps here and there
20:12:03 <smooge> mmcgrath, I was wondering.. could we rename app01 to bapp01 since its more bapp01 than app01?
20:12:07 <mmcgrath> a couple of weird oops's we didn't see before for example.
20:12:13 <smooge> oh wait, what were the bumbps?
20:12:18 <smooge> kernel oops?
20:12:36 <mmcgrath> smooge: it's just as much an app1 as a bapp1 so no, lets just keep it.  If we feel we need a bapp1 (people have requested it) then we can build a bapp1.
20:12:43 <mmcgrath> smooge: yeah
20:12:52 * sijis is around.. sorry
20:12:53 <mmcgrath> it's happened to app1, proxy1 and I think the db server.
20:13:00 <mmcgrath> though it hasn't happened recently
20:13:06 <mmcgrath> and one morning I came in and app1 was paused
20:13:15 <mmcgrath> sijis noticed it was down so I went to look and sure enough, paused :)
20:13:28 <mmcgrath> so just stuff to keep an eye on
20:13:39 <mmcgrath> we're both new to KVM and RHEL6 and since RHEL6 itself is new... there you have it
20:13:48 <mmcgrath> any questions / comments on that?
20:13:57 <smooge> mmcgrath, ok the big issue was dealing with updates on app01 but since app01.stg is probably EL6 now... it is probably not a big deal
20:14:14 <lmacken> app01.stg is RHEL5.5
20:14:18 <smooge> ah ok
20:14:24 <smooge> so it has the funky rpm on it
20:14:29 <mmcgrath> naw only fas01.stg has been upgraded to RHEL6
20:14:48 <mmcgrath> we're generally not upgrading to RHEL6 unless there's a reason to at this point (like fas and it's alchemy dep
20:14:53 <smooge> ok sorry.
20:15:01 <mmcgrath> any other questions / comments on staging?
20:15:13 <smooge> none. I will work on a bapp01 next week
20:15:38 <mmcgrath> #topic zarafa
20:15:44 <mmcgrath> so I got zarafa up and running for our calendaring
20:15:52 <mmcgrath> I've run into an issue making it highly available.
20:16:03 <mmcgrath> it's a multi part issue and I'm not 100% sure what I'm going to do to fix it at this point.
20:16:28 <mmcgrath> but I'll keep testing things and see how it goes.
20:16:43 <smooge> how does one access it ? via admin like community?
20:16:47 <mmcgrath> https://admin.fedoraproject.org/zarafa/
20:16:55 <mmcgrath> that's where it is in production, staging in its normal spot.
20:17:05 <mmcgrath> don't rely on it for anything yet, I may wipe the database at any moment.
20:17:14 <smooge> no problem.
20:17:25 <mmcgrath> but do poke around and see what all you can see
20:17:45 <lmacken> ugh, gives me the Outlook chills
20:17:46 <smooge> will not have all my fedora mail go there either :)
20:18:03 <mmcgrath> lmacken: :) that is part of their design
20:18:14 <mmcgrath> it's got full outlook support and all that AFAIK
20:18:18 <mmcgrath> I haven't tested that though
20:18:22 <mmcgrath> it's also got an ical and cal-dav interface
20:18:38 <mmcgrath> which is exposed right now but I need to put docs together on how to create a URI and point to it
20:18:47 <mmcgrath> Anyone have any questions or comments on that?
20:19:22 <mmcgrath> ok
20:19:26 <mmcgrath> well that's all I had for this meeting
20:19:30 <mmcgrath> #topic Open Floor
20:19:36 <mmcgrath> anyone have anything they'd like to discuss?
20:19:40 <sijis> let me follow up on your initial question
20:19:46 <mmcgrath> sijis: sure
20:20:00 <sijis> i haven't updated the websites_release_SOP
20:20:06 <sijis> i will do it tonight
20:20:16 <mmcgrath> sijis: much appreciated
20:20:26 <sijis> sorry about those busted links on release day
20:20:27 <sijis> my bad
20:20:28 <mmcgrath> sijis did a good job with the release
20:20:37 <smooge> yes
20:20:37 <mmcgrath> sijis: no worries, your first time though went pretty darn well IMHO
20:20:51 <sijis> thanks :)
20:21:13 <sijis> no other websites issues either, except updating the release notes to the proper location
20:21:43 <mmcgrath> cool
20:21:51 <mmcgrath> anyone have anything else they'd like to discuss?
20:22:16 <krich> just a quick question ... does anyone have some time right after this to talk about fas
20:22:33 <krich> looking for a high level arch or just a quick q&a
20:22:38 <mmcgrath> krich: I can, lmacken and I will be talking about other fas stuff.  sure
20:22:45 <mmcgrath> Ok, if no one has anything else we'll close in 30
20:23:46 <mmcgrath> #endmeeting