20:03:18 <pcalarco> #startmeeting Fedora Insight
20:03:18 <zodbot> Meeting started Tue Sep  6 20:03:18 2011 UTC.  The chair is pcalarco. Information about MeetBot at http://wiki.debian.org/MeetBot.
20:03:18 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
20:03:30 <pcalarco> #chair asrob
20:03:30 <zodbot> Current chairs: asrob pcalarco
20:03:55 <pcalarco> #topic Fedora Insight: rollcall
20:04:29 <pcalarco> #meetingname Insight
20:04:29 <zodbot> The meeting name has been set to 'insight'
20:04:58 <pcalarco> is here
20:05:39 * asrob is here as well
20:05:48 <pcalarco> I see stickster in the room, but he may be busy
20:05:57 * stickster here!
20:06:09 <stickster> sorry, was replying to an urgent $DAYJOB thingie
20:06:10 <pcalarco> stickster: Hi!
20:06:19 <stickster> Hi pcalarco, asrob :-)
20:06:20 <pcalarco> yep, no problem
20:06:27 <pcalarco> #chair stickster
20:06:27 <zodbot> Current chairs: asrob pcalarco stickster
20:06:27 <asrob> stickster, hi :)
20:06:53 <pcalarco> nirik is in the room too, he may join us as he is able
20:07:08 <pcalarco> so, review of last week's items
20:07:08 <stickster> Great
20:07:11 * nirik is around, but as usually busy. ;)
20:07:19 <stickster> Always busy!
20:07:20 <pcalarco> #topic Review last week's action items
20:07:32 <stickster> #link http://meetbot.fedoraproject.org/fedora-meeting/2011-08-30/insight.2011-08-30-20.00.html
20:07:45 <pcalarco> #link http://meetbot.fedoraproject.org/fedora-meeting/2011-08-30/insight.2011-08-30-20.00.html
20:07:49 <stickster> Heh
20:07:51 <pcalarco> whoops :0
20:07:54 <stickster> #undo
20:07:54 <zodbot> Removing item from minutes: <MeetBot.items.Link object at 0x17ad8c8c>
20:08:04 <stickster> No worries, I skunked you silently pcalarco :-)
20:08:16 <pcalarco> ty :)
20:08:17 * stickster has done a fairly successful D7 port of the authfas module
20:08:20 <stickster> oops
20:08:25 <stickster> #info stickster has done a fairly successful D7 port of the authfas module
20:09:14 <stickster> A couple remaining issues: http://git.fedorahosted.org/git/?p=drupal6-authfas.git;a=blob;f=TODO.txt;h=df523fa72bf725b0312b017d09884f2f1599a054;hb=d7work
20:09:45 <pcalarco> great, thanks for that link
20:10:03 <pcalarco> anything to report on Finish setting up pull of fedora-insight-features on stg and prod hosts
20:10:05 <stickster> #info stickster still needs to set up feature pulls, only reason he hasn't is because he ran into a local issue on his system during his appointed time
20:10:14 <stickster> #action stickster Finish setting up pull of fedora-insight-features on stg and prod hosts (30 min for stg, 4 hrs for prod)
20:10:45 <stickster> And we can take up the issue with the D7 exemplar after this #topic if you guys like
20:10:45 <pcalarco> great, ok
20:10:51 * asrob will help stickster in porting
20:11:35 <pcalarco> any other items to update on before we hit D7 exemplar?
20:11:58 <asrob> nothing
20:11:59 <stickster> -
20:12:06 <pcalarco> #topic Drupal 7 exemplars
20:12:36 <stickster> So asrob was looking to get a D7 host running so he could test migration and see if it's possible for us to move without significant pain to D7
20:13:10 <stickster> asrob: I posted to the list earlier about this. Having D7 and D6 live on the same host, as you know, isn't impossible. But it's a little more difficult for us given the way we set up D6 in puppet.
20:13:20 <asrob> yep, I read it
20:13:48 <stickster> So it would be best if we could get smooge or nirik or someone to set up an insight02.dev box for us. I would assume that starts with us filing a request ticket in the fedora-infrastructure Trac.
20:14:15 <nirik> we could... might take some shuffling as the site we usually make the dev machines is pretty full...
20:14:21 <nirik> but yeah, a ticket would be good.
20:15:07 <stickster> asrob: OK, since you're the guy who knows the most about this migration effort, can you take the action of filing the ticket and following up with the Infra team as it goes?
20:15:55 <asrob> stickster, you mean, we migrate our d6 instance to d7?
20:16:26 <stickster> asrob: The ticket is only to request an insight02.dev.fp.o host so we can install D7 there for you to work with
20:16:55 <asrob> ah, yes, okay
20:16:56 <stickster> asrob: In actuality most of the testing you should be able to do locally, since we already have backup_migrate for the data and you can simply test on a local sandbox
20:17:07 <asrob> ++
20:17:40 <asrob> can I use action command?
20:17:47 <stickster> asrob: Of course!
20:17:49 <stickster> :-)
20:18:01 <pcalarco> :)
20:18:08 <stickster> The insight02.dev host gives us a way to make sure our puppetization works properly, and allows us to fiddle with the theme and so forth as we tweak
20:18:54 <asrob> #action asrob will be filling an item for migrating, requesting an insight02.dev.fp.o host
20:19:25 <jsmith> stickster: Mind if I chime in with a D7 topic?
20:19:55 <pcalarco> #chair jsmith nirik
20:19:55 <zodbot> Current chairs: asrob jsmith nirik pcalarco stickster
20:20:05 <stickster> jsmith: go for it
20:20:36 <jsmith> So, I've been running Drupal 7 packages on my own servers, and noticed a lot of apache threads exiting unexpectedly
20:21:03 <jsmith> I haven't tracked down the cause yet, but there seems to be something "wonky" in the php53 packages (which drupal7 depends on)
20:21:16 <jsmith> Just thought I'd give you a heads up, in case you see the same thing
20:21:21 <jsmith> eof
20:21:21 <stickster> jsmith: Those are EPEL 5 running hosts?
20:21:34 <jsmith> stickster: Yes, that's correct.
20:21:59 <jsmith> stickster: EPEL 5 + php53-* + drupal7 + postgresql
20:22:00 <stickster> jsmith: I think we're running on RHEL 6 so we're using a newer htpd
20:22:02 <asrob> jsmith, you can use php 5.2.6 or later as well, see drupal 7 minimum requirements, but thank you for the notice ;)
20:22:03 <stickster> oops, httpd
20:22:15 <jsmith> Ah, you're on EPEL 6... if only I were so lucky :-p
20:22:32 <jsmith> Hopefully you don't have any problems then :-)
20:22:53 <stickster> I'd have to ask nirik or someone more Apache-savvy how we would monitor for any such problems
20:23:10 <jsmith> The dead threads show up in the apache error log
20:23:32 <stickster> jsmith: OK, interesting.
20:23:41 <nirik> yeah, we do have monitoring, but not of dev instances usually.
20:23:51 <nirik> and we dont have drupal7 anywhere yet. ;)
20:24:09 <jsmith> [Fri Aug 26 16:54:55 2011] [notice] child pid 10523 exit signal Segmentation fault (11)
20:24:14 <jsmith> That's what I'm seeing :-)
20:24:20 <stickster> nirik: Fair enough -- but asrob and I have shell access on insight01.dev and could check on this.
20:24:53 <jsmith> Grepping for "child pid \d+ exit signal" should be enough to find it, if it's happening
20:24:58 <stickster> Yup
20:26:35 <stickster> Anything else on D7?
20:26:44 <stickster> Oh wait, I remember, I have something else
20:26:51 <stickster> I heard from someone else at Red Hat who'
20:26:55 <stickster> oops
20:27:28 <stickster> I heard from someone else at Red Hat who's working on a Drupal project and may want to get modules packaged for Drupal 7. I've asked him if we can maybe team up so that we'll have other people helping with that module packaging.
20:27:52 <pcalarco> that's great
20:27:57 <asrob> cool :)
20:28:05 <stickster> jsmith: Also, I think you have a couple bugs open where rpmlint got hung up on some errors in those packages you posted, I set them NEEDINFO -> you
20:28:37 <stickster> pcalarco: asrob: A bunch of the packages he needs are the same ones we'd be using too, backup_migrate, cck, ctools, date, devel, features, etc.
20:28:37 <jsmith> stickster: Aye, I'll be working on those this weekend at OLF :-)
20:28:51 <stickster> Guess I'll see you there!
20:29:08 <jsmith> stickster: If I get some free time (yeah right!) before then, I'll take a stab at fixing them up
20:29:28 <jsmith> Unfortunately, Drupal stuff is strictly personal for me -- can't justify doing it on work hours
20:29:58 <stickster> Yup
20:30:41 <pcalarco> okay, any more D7 items on this topic?
20:30:50 <stickster> That was it for me, f'real, thanks pcalarco :-)
20:31:17 * jsmith has to leave early... later!
20:31:42 <pcalarco> do we have any of the critical path items documented yet, or is it early days for that a bit?
20:31:58 <pcalarco> as regards D7
20:32:58 <stickster> pcalarco: Do you mean, critical path to migrate to D7?
20:33:26 <pcalarco> yes -- IDEA: Continue to look at D7 migration, but base decision on critical path needs, followed by "nice to have" things
20:33:27 <asrob> pcalarco, well, I don't know what is the best for us now, but what do you think whether we can create a D7 feature set and can we "migrate"(copy & paste) from old contents into new site?
20:35:24 <pcalarco> I'm thinking like: here are the D7 modules we need, here is what is packaged, and a testing workflow to ensure D6 and D7 modules work nominally?
20:35:38 <stickster> asrob: Have you tried a migration already, including data? Do you feel good that this will be fairly easy and painless to do?
20:35:52 * stickster wasn't sure we had answered that completely yet
20:36:58 <asrob> stickster, yeah, I have, so-so, I think it was better as well
20:37:39 <stickster> asrob: Sorry, not sure I understood that part correctly -- my fault. When you say "so-so," do you mean it was somewhat difficult?
20:38:10 * stickster is trying to figure out how much this affects our ability to do other things like the calendar ideas.
20:38:13 <asrob> stickster, correct, I updated a locelized site from 6 to 7
20:38:20 <asrob> localized
20:38:37 <stickster> asrob: On a scale of 1-10, where 1 is "very easy" and 10 is "very difficult," how difficult would you say it was?
20:39:12 <asrob> hm, it was easy, so 3 or 4
20:39:17 <stickster> OK, not bad
20:39:27 <stickster> asrob: Was that a local copy of our site, with our data?
20:40:00 <asrob> stickster, no, it was a hungarian sysadmin site, http://sys-admin.hu
20:40:21 <stickster> OK, cool
20:40:44 <stickster> So, pcalarco had a good idea which was to make a list of critical path tasks needed to do a migration
20:41:25 <asrob> but we decided that we re-create that site, because the translation ratio is bad for example
20:41:34 <asrob> stickster, okay, I got it
20:41:37 <stickster> Ah
20:42:03 <stickster> What would we list in the critical path? It seems to me that would include:
20:42:20 <stickster> (1) Figure out all modules needed, and get them all packaged (set a date deadline)
20:42:26 <stickster> (2) Recreate all views and menus
20:42:35 <stickster> (3) Migrate the theme
20:43:17 <stickster> (4) Finish AuthFAS migration (we're at ~90%)
20:43:17 <pcalarco> Test critical functionality
20:43:25 <stickster> (5) Test, test, test! (right on pcalarco)
20:43:45 <stickster> What else did I miss?
20:43:54 <asrob> (4a) Mediawiki_api migration ;)
20:43:55 <stickster> oh yeah
20:44:09 <stickster> (2.5) Recreate all features and deploy in a D7 branch
20:44:14 <pcalarco> 6) document
20:44:19 <stickster> pcalarco: +1
20:45:13 <stickster> Well... I guess I was not going to be a big fan of moving to D7 because it seems like something we are doing instead of other development. But on the other hand, having used D6 & D7 now for a bit, I see D7 is a BIG improvement in administration and it buys us several extra years in lifetime
20:45:32 <pcalarco> Since you guys are doing the heavy lifting, I'll take documenting this on the Insight wiki for us
20:45:47 <stickster> pcalarco: That would be awesome
20:46:07 <pcalarco> #action pcalarco start documenting what critical path items we have for D7 migration on Insight wiki
20:46:19 <stickster> Since the nirik and the Infra team have agreed to keep us on the "value add" (?) list, it means we don't have to wait for freeze periods, so we could conceivably start working on this right away
20:46:32 <stickster> with the idea being we would start on calendar stuff right afterward
20:47:13 <pcalarco> sounds fine to me, and we may even gain some functionality we're looking for with the calendar uses
20:47:24 <stickster> pcalarco: True
20:48:18 <asrob> understood
20:49:01 <stickster> So pcalarco is going to draw up that critpath list, can you send that to the list?
20:49:14 <stickster> er, send the critpath list to the mailing list, that is. :-)
20:49:15 <pcalarco> sure, absolutely, to logistics
20:49:20 <stickster> superb!
20:50:33 <stickster> So are we agreed that this is going to be our focus for the next few weeks?
20:50:48 <asrob> yep
20:50:49 <pcalarco> +1
20:50:57 * stickster just trying to plan what he's going to do with his few hours of Insight time in the next week
20:51:27 <stickster> #action stickster continue conversation with internal Red Hat guy who might want to help package stuff, will report more next week
20:52:44 <pcalarco> ready to move on to Puppetizing?
20:53:04 <asrob> I'm ready
20:53:19 <pcalarco> #topic Puppetizing
20:53:24 <pcalarco> IDEA: stickster would like to start by puppetizing the Features stuff on staging next. The dev host is done already
20:53:59 <stickster> Yeah, this was the #action I set for myself earlier
20:54:32 <pcalarco> ah, ok, so we're covered, perfect
20:54:36 <stickster> Part of deploying this is being able to fix or yell for help if it breaks when I try it on staging. And the only time I had available was yesterday when a lot of people were gone for Labor Day in the USA
20:55:04 <stickster> I know what needs to be done though -- just need to find some more amenable day time to do it -- Ohio LF may give me that opportunity :-)
20:55:11 <asrob> :))
20:55:20 <pcalarco> that is last week's items, anything new to add?
20:55:23 <pcalarco> :)
20:55:24 <stickster> -
20:55:34 <asrob> -
20:56:13 <pcalarco> I tweaked the calendar use cases page a bit, adding in the zodbot automation we'd like to see in the longer time frame
20:56:38 <pcalarco> that's about it, so we can finish a couple minutes early! :)
20:57:01 <pcalarco> #info pcalarco tweaked the calendar use cases page a bit, adding in the zodbot automation we'd like to see in the longer time frame
20:57:09 <stickster> Thanks for doing that pcalarco
20:57:13 <pcalarco> ending in 5
20:57:15 <pcalarco> 4
20:57:17 <pcalarco> 3
20:57:19 <asrob> ok, thank you guys
20:57:19 <pcalarco> 2
20:57:21 <pcalarco> 1
20:57:27 <pcalarco> #endmeeting