17:00:19 <smooge> #startmeeting EPEL
17:00:19 <zodbot> Meeting started Fri Nov  7 17:00:19 2014 UTC.  The chair is smooge. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:00:19 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
17:00:22 <smooge> #meetingname epel
17:00:23 <zodbot> The meeting name has been set to 'epel'
17:00:34 <bstinson> hi all
17:00:43 <smooge> #chairs smooge Evolution bstinson nirik dgilmore
17:00:53 <smooge> #topic Meat and Greets
17:01:09 <smooge> porterhouse
17:01:15 <smooge> and hello
17:02:21 <smooge> nirik, is dealing with a pileup in other channels
17:02:27 <smooge> dgilmore, is in Australia I think
17:02:33 <smooge> Evolution, is ...
17:02:35 * nirik is sorta kinda around, but a bit swamped
17:03:10 <Evolution> I'm here
17:03:42 <smooge> nirik, I have decided that whenever I call this meeting.. it is a sign for the universe to hit you with 10k IRQs.. I was watching all the channels.. quiet.. 9:50 BOOM HEY NIRIK
17:04:52 <smooge> #topic Set date for upcoming Bug Squash Day
17:04:59 * maxamillion is here
17:05:08 <bstinson> https://fedoraproject.org/wiki/EPEL_Bug_Day_2014
17:05:11 <bstinson> started a wiki page
17:05:15 <smooge> Ok last meeting we said we would like to have a bug squash day
17:05:20 <smooge> Yay! Thanks bstinson
17:05:26 <maxamillion> bstinson: oh cool, +1
17:05:38 <smooge> that was on my todo and I have one less thing
17:06:27 <Evolution> smooge: don't worry. I'm sure we can find another task to fill the gap.
17:06:52 <smooge> Evolution, it means I get to work on my presentation so it is already filled
17:07:12 <Evolution> you and me both.
17:07:40 <nirik> smooge: no kidding. ;)
17:07:44 <smooge> OK so in Fedora land we have a beta that just got shipped and then final prep for release.
17:08:12 <smooge> in CentOS there are various events and other SIG stuff coming up.
17:08:36 <smooge> so I was wondering if we should look at the first bug squashing day to be in say the first week of December?
17:09:29 <Evolution> I'm fine with it. so long as we've got 2 weeks lead to plan/advertise
17:09:31 <smooge> it misses the November holidays and should be during a freeze in infrastructure
17:09:56 <smooge> so what day works best for people?
17:10:01 <bstinson> i think that would be a good time, especially so we can hit some of the orphaned packages before the retire date
17:10:21 * nirik notes he and smooge will be traveling to RDU that week, but eariler in the week might work
17:10:27 <smooge> oh fudge
17:10:36 <smooge> I forgot about that ..
17:10:45 <smooge> thanks for reminding me.
17:10:52 <rsc> Yes, but having the EPEL bug squash day before the orphaning in EL-5 would make sense.
17:10:54 <smooge> Monday and Tuesday look like my free days that week
17:10:57 <nirik> 1-3rd should be ok... travling on the 4th I think
17:11:42 <smooge> I will be travelling on the 3rd to be able to pay for stuff for people who get there that day.
17:12:05 <smooge> so how does Tuesday sound?
17:12:14 <bstinson> wfm
17:12:41 <smooge> start at 16:00 UTC and go to 22:00 UTC. [I think 6 hours is about all I can concentrate on bugzilla for.. but we could do longer]
17:12:44 <maxamillion> +1
17:13:07 <smooge> nirik, how does your Tuesdays hate you usually?
17:13:12 <Evolution> smooge: pretty sure going over 6hrs of bugzilla time qualifies as a war crime
17:14:39 <smooge> Evolution, can you do Dec 2nd for a couple of hours ?
17:14:47 <Evolution> yep.
17:14:57 <smooge> doesn't need to be the full set
17:15:19 <Evolution> well, let me caveat that.
17:15:34 <smooge> ...
17:15:37 <Evolution> I'm traveling Dec1, but father-in-law is scheduled for surgery
17:15:46 <Evolution> I *should* be clear for dec2.
17:16:00 <Evolution> but won't know for certain for a week or so yet
17:17:02 <nirik> 2nd should be ok
17:17:27 <smooge> ok thanks.. I hope all goes well with that
17:17:43 <smooge> OK I think we have enough people to help run it so
17:17:59 <Evolution> smooge: I'll make sure we have a couple folks from the centos side available as well.
17:18:08 <smooge> #agreed EPEL Bug Squashing Day 2014-12-02
17:18:44 <smooge> so next thing.. what do we want people to do during that time? This is more of a mailing list thing but it helps to frame it so that it doesn't go all over the place.
17:18:59 <smooge> bstinson, you were talking about orphans
17:20:07 <Evolution> I think there are 3 separate goals here that all circle around bugs.
17:20:17 <Evolution> feedback for packages in testing.
17:21:00 <Evolution> various bug reports to validate/comment/update/close, and finding owners for orphaned packages.
17:21:17 * nirik nods.
17:21:24 <nirik> the three pronged attack!
17:21:49 <Evolution> is that too broad for one day? did I leave something out?
17:22:08 <smooge> ok do people think it would help if each of us concentrated on one that way volunteers can know who to point questions to?
17:23:51 * bstinson can coordinate finding owners
17:23:53 <smooge> but I agree Evolution that covers the spectrum
17:24:07 <smooge> I can coordinate bugzilla or package testing
17:24:53 <smooge> and someone can take the third.
17:25:30 <smooge> So we need to start advertising this. Does anyone want to take that after the meeting?
17:25:37 <Evolution> I'll tentatively take 3rd. will be definite next week or so when I find out from the wife/family.
17:26:01 <smooge> yeah.. I understand completely
17:26:09 <Evolution> smooge: I can post to the centos and centos-devel lists, and will poke twitter as well.
17:26:16 <smooge> okie dokie
17:26:22 <Evolution> can you email the epel lists?
17:26:29 <smooge> Anything else people have on this topic?
17:26:52 <smooge> #topic Upcoming Stuff
17:27:10 <smooge> #info Bug Squashing Day 2014-12-02
17:27:33 <smooge> #info Orphan Removal Day 2014-12-17
17:27:48 <smooge> #info Dependency Cleanup Day 2014-12-19
17:28:32 <smooge> #info Infrastructure Freezes: Final Fedora Release 2014-11-19->2014-12-??
17:28:53 <nirik> 09 hopefully. ;)
17:29:02 <smooge> #info Infrastructure Freezes: Final Fedora Release 2014-11-09->2014-12-?? [corrected]
17:29:16 <nirik> no, unfreeze 2014-12-09
17:29:16 <rsc> How does infrastructure freeze for Fedora impact EPEL?
17:29:26 <smooge> #info Infrastructure Freezes: Final Fedora Release 2014-11-19->2014-12-09 [really corrected]
17:29:56 <smooge> #Infrastructure Slush: Red Hat Employee Enforced Holiday [2014-12-23->2015-01-02]
17:31:11 <smooge> The majority of items with the Infrastructure freeze would be anything that we wanted to do that affects koji, bodhi, etc
17:31:53 <smooge> A couple of the things we have been talking about to discuss but haven't gotten very far into would be affected if decided on during that time
17:31:55 <rsc> Does it also affect pushes from koji to bodhi? Because it felt like that some weeks ago...
17:32:46 <smooge> rsc, I don't know the answer on that. nirik or dgilmore would have to answer
17:33:31 <nirik> rsc: only in the sense that people who do updates pushes are busy. ;)
17:33:42 <nirik> it shouldn't hopefully affect epel much at all.
17:34:26 <maxamillion> bleh, $dayjob stuff keeps creeping up
17:34:29 * maxamillion reads backscroll
17:36:25 <smooge> OK so I didn't have anything to put for a 2nd topic for the meeting. We have a ton on the todo pile but I think a couple of them need to have  a complete proposal versus my tossed out ideas on IRC.
17:36:54 <smooge> I was thinking of moving to open flood and let people bring up a second topic and if nothing we call it a meeting.
17:37:45 <smooge> #topic Open Flood
17:37:49 <smooge> Ok open floor.
17:37:52 <Evolution> I've got nothing. my focus has been a bit elsewhere this week.
17:38:30 <maxamillion> same
17:38:38 <rsc> Is there a chance to "get" testers for bodhi karma somehow?
17:38:39 <smooge> mine has been on being sick and then recovering
17:38:52 <smooge> rsc, what do you mean
17:38:55 <maxamillion> this week has been purely fighting fires at $dayjob ... haven't had a chance to look at EPEL stuff I wanted to do ... maybe this weekend
17:39:20 <rsc> I have with phpMyAdmin usually the issue that upstream releases security updates more often than 14 days (or 14 days + delay for push from koji to bodhi)
17:39:40 <rsc> Right now this ends for me with bothering some people to test packages...
17:39:55 <rsc> (Or to delay a security update by replacing/obsoleting it by the new one in testing)
17:41:00 <rsc> (But if I do last of it a couple of times no security update goes to stable for weeks)
17:42:01 <Evolution> <insert snark about php code here>
17:42:10 <smooge> rsc, well a lot of things need people who know the product to test the product. I wouldn't know what to do with phpmyadmin if you gave it to me.. so we need to figure out how to get people who know the stuff and care for the stuff to get out of the woodwork and help you
17:42:57 <smooge> rsc, I actually will put that in my talk for Monday at LISA. Thanks.
17:42:59 <Evolution> so, in theory automated testing could help here.
17:43:13 <rsc> Evolution: that would be great. Maybe a defined test set to check some basic functions.
17:43:32 <rsc> phpMyAdmin (as an example) is one of the packages that seems to be used but less loved (for good reasons)
17:43:36 <Evolution> right. but I don't think that functionality exists in epel.
17:44:09 <smooge> well testing webapps automatically was a whole can of hell for a long time :)
17:44:23 <nirik> possibly testing security fixes is something epel wranglers could do?
17:44:31 <nirik> (time and ability to test permitting)
17:44:52 <Evolution> nirik: I have been for some. but as smooge said, for some things, I have 0 idea.
17:44:53 <smooge> I think that would be a good idea
17:45:04 <Evolution> having a defined set of tests for a package would be nice.
17:45:06 <smooge> nirik, you have boxes that can be used ofr this still?
17:45:15 <Evolution> does <package> do <list>
17:46:10 <nirik> smooge: sure.
17:46:25 <nirik> https://fedoraproject.org/wiki/Test_Machine_Resources_For_Package_Maintainers
17:46:56 <smooge> ok will add that to the list
17:47:57 <smooge> rsc, so to come back to your question. We should have a process for testing security stuff in place. I will look at writing out the blind monkey version of what needs to be done so that people can follow it and do so.
17:49:05 <rsc> That would be cool. Let me know if I could help here (especially from the package maintainer side)
17:49:40 <smooge> I think the best thing would be a quick "how can a blind monkey test this package and know it covers the bare minimal things I need it to."
17:50:30 <smooge> Most of the people I talk about testing usually see it as a "I need to comprehensively check every line of this package before I can say its tested." and we want to say "nope you just need to do X, Y, and Z (and maybe A if needed)."
17:51:16 <rsc> Do we have a namespace/section in the wiki already?
17:52:44 <smooge> rsc, we have many many many.. it is on my Xmas list to clean up
17:53:21 <smooge> rsc, actually write something up in wiki format and send to the mailing list. It will at least get it in front of eyes
17:53:39 <smooge> s/wiki/whatever floats your boat/
17:54:36 <rsc> Understood. So basically the commands to run phpMyAdmin properly and then some test suggestions for the webbased part (in this case).
17:55:49 <smooge> yep pointers to webpages written by others if needed (eg you need to set up a MySQL database.. follow this howto. You need to give 2 pints of blood.. follow this FAQ.)
17:56:57 <smooge> Ok we are coming up to the hour.
17:57:04 <smooge> #topic Next Meeting?
17:57:57 <smooge> Next week Evolution and I will be at USENIX (or travelling from USENIX) on Friday. I was thinking of doing the meeting from the hotel room but further thought was wondering how feasible it is.
17:58:36 <smooge> bstinson, Evolution nirik .. want to try it or push to Nov 21?
17:58:57 <bstinson> ok by me
17:59:53 <bstinson> maybe we can take care of some more trac tickets this week
18:00:05 * Evolution checks friday flight time
18:00:40 <nirik> either way.
18:00:52 <nirik> I'd say I could run things, but then given how this friday turned out...
18:00:53 <Evolution> hmmm, in theory I should be good.
18:00:59 <Evolution> I'm fine with trying.
18:03:38 <smooge> ok we will try for next week. If we have problems we will abort later in the week
18:06:13 <smooge> #endmeeting