19:00:48 <rbergeron> #startmeeting Cloud SIG
19:00:48 <zodbot> Meeting started Fri Jul  1 19:00:48 2011 UTC.  The chair is rbergeron. Information about MeetBot at http://wiki.debian.org/MeetBot.
19:00:48 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
19:00:52 <rbergeron> #meetingname Cloud SIG
19:00:52 <zodbot> The meeting name has been set to 'cloud_sig'
19:00:57 <rbergeron> #topic Roll call
19:01:00 * jsmith is here
19:01:03 <rbergeron> Pipe up, kids.
19:01:03 * kkeithley is here
19:01:16 * clalance here
19:01:22 <gholms> bacon
19:01:27 <msavy> hello
19:01:28 * mgoldmann is here!
19:01:30 <rbergeron> #chair gholms
19:01:30 <zodbot> Current chairs: gholms rbergeron
19:01:43 <rbergeron> Okay. let's see.
19:01:47 <rbergeron> #topic EC2 status
19:01:58 <rbergeron> #info jforbes and dgilmore out this week
19:02:11 <rbergeron> So FYI, for a long fun story made significantly shorter:
19:02:16 <gholms> Heh
19:02:22 <rbergeron> SHUSH YOU :)
19:02:33 <rbergeron> As you may remember from our last episode,
19:02:35 * mgoldmann looks up "shush"
19:02:55 <rbergeron> We need to get signoff from multiple teams to have a "spin" be official.
19:03:25 <mgoldmann> awesome :)
19:03:35 <rbergeron> QA has no time now, nor any prospective time in the future. (And that includes come every release time.)
19:03:41 <rbergeron> So where that leaves us is:
19:03:47 <rbergeron> (a) We have to make our own test plan
19:04:01 <rbergeron> (b) we have to gtet community folks to test it somehow.
19:04:01 <jsmith> rbergeron: It's not a spin, it's an image.
19:04:09 <rbergeron> #chair jsmith
19:04:09 <zodbot> Current chairs: gholms jsmith rbergeron
19:04:19 <athmane> rbergeron: I can help in some case (I'm from qa)
19:04:31 <gholms> #info QA does not have time to sign off on cloud images, so community members need to do it.
19:04:46 <rbergeron> athmane: that would be much appreciated.
19:05:21 <rbergeron> jsmith: my apologies.
19:05:28 <rbergeron> s/spin/image.
19:05:52 <rbergeron> Except it's still called "the spins process", so don't s/spin/image that one.
19:05:53 <mgoldmann> so, what a test plan really should look like?
19:05:56 <rbergeron> Anyway.
19:06:03 <mgoldmann> is "it started and network connectivity is available" sufficient?
19:06:15 <rbergeron> mgoldmann: So - j_dulaney from QA land has offered to help a bit here.
19:06:18 <mgoldmann> or we need to test processes owners and count? :)
19:06:21 <rbergeron> and perhaps athmane as well.
19:06:23 <gholms> http://fedoraproject.org/wiki/QA:Test_Plan_Template
19:06:30 <rbergeron> mgoldmann: things like - can people log in, etc.
19:06:51 <rbergeron> We don't need to be insane, but we want to make sure that Bad Things aren't happening.
19:07:08 <mgoldmann> yah, this was included in my sentence - start, log in, test network
19:07:24 <gholms> Install httpd and make sure it works...
19:07:34 <mgoldmann> the difference between spins and or images is fundamental - in 95% cases it'll be jeos :)
19:07:43 <mgoldmann> not a full graphical system
19:08:00 <rbergeron> Right.
19:08:18 <msavy> do QA not have a battery of automated tests that we could use, presumably they have some that cover JEOS?
19:08:53 <rbergeron> msavy: I don't know.
19:09:06 <rbergeron> QA as in Fedora QA?
19:09:12 <rbergeron> or rht qa?
19:09:21 <mgoldmann> I think Fedora
19:09:43 <gholms> I wouldn't mind rht qa, of course. :)
19:09:43 <rbergeron> We don't have anything in jeos land, afaik. I'm sure they would have said something when I spoke with them.
19:09:52 <rbergeron> well, I have to think SOMEONE HAS DONE THIS
19:10:03 <rbergeron> So:
19:10:06 <gholms> "Install the Minimal package profile" isn't a test?
19:10:16 <mgoldmann> I think we should ask - how Fedora QA tests things
19:10:21 <rbergeron> I'm supposed to write an email trying to motivate people to pull something together, but my time sucks.
19:10:29 <rbergeron> mgoldmann: with test matrices and human-powerz
19:11:00 <mgoldmann> from our (BG) POV - we can add any tests to our integration tests
19:11:21 <mgoldmann> so build this, then start on EC2, log in, do some stuff, check, shutdown
19:11:23 <rbergeron> So I am going to beg and plead for someone to help me, since my time is thin since getting back, and also, it's helpful when someone who knows the difference between a spin and an image is working on the test plan.
19:11:36 <msavy> yeah, human-powerz sucks, slow and not repeatable a lot of the time. +1 to automated tests
19:11:45 <msavy> ask rht QA too maybe
19:12:15 * rbergeron nods
19:12:19 <gholms> #idea Add some acceptance tests to the BoxGrinder integration tests
19:12:28 <rbergeron> msavy, mgoldman: maybe you could send some info about what you do to the list?
19:12:39 <mgoldmann> sure
19:12:39 <rbergeron> at least as far as the JEOS testing goes?
19:12:45 <rbergeron> that would be super helpful.
19:12:49 <mgoldmann> btw, I'm writing a post on this NOW
19:13:00 <rbergeron> Can i get a volunteer to maybe make a test plan page? :D
19:13:11 <gholms> #action mgoldmann to mail the cloud list about BoxGrinder's JEOS testing
19:13:31 <mgoldmann> rbergeron: our tests currently does not involve launching on EC2, we test appliances in an offline mode
19:13:55 <mgoldmann> but there is no reason we couldn't extend this
19:14:24 <mgoldmann> and yes, this is the beginning with integration testing in BG, but we feel the power :)
19:14:33 * rbergeron nods
19:15:03 <rbergeron> gholms: you have alluded to something that ... essentially emulates ec2 env on desktop, correct?
19:15:12 <rbergeron> so that someone wouldn't need to "get into EC2" and test?
19:15:35 <gholms> Eucalyptus can do it, but that isn't in Fedora.
19:15:58 <rbergeron> and isn't going to be anytime soon, i presume. ;)
19:16:02 <mgoldmann> gholms: my posts is not about JEOS testing specifically - it's now testing BG features - mostly appliance definition sections, inclusions, etc
19:16:02 * jsmith would prefer to do the testing on "real" EC2 for now
19:16:05 * rbergeron poutyfaces
19:16:43 <gholms> #undo
19:16:43 <zodbot> Removing item from minutes: <MeetBot.items.Action object at 0x1ae68bac>
19:16:58 <rbergeron> mgoldmann: but you at least make sure it boots up, etc. - could that be extended?
19:17:01 <gholms> #action mgoldmann to mail the cloud list with info about BoxGrinder's test suite
19:17:08 <rbergeron> ;)
19:17:25 <mgoldmann> rbergeron: definitely - like I said - it can be extended
19:17:40 <mgoldmann> we just stop with sending the appliance to S3
19:17:49 <rbergeron> mgoldmann: most excellent.
19:17:52 <mgoldmann> we don't launch it on Ec2
19:17:55 <mgoldmann> EC2*
19:18:10 <gholms> Realistically it isn't super difficult to emulate a simple EC2 environment if you have a system that runs httpd and kvm.
19:18:35 <gholms> Then again it also isn't super simple. ;)
19:18:40 <msavy> probably a bit harder to emulate the  edge cases though
19:18:52 * gholms nods
19:18:54 <msavy> for instance when AWS components get stuck in weird states etc
19:18:56 <jsmith> Those edge cases are the ones I'm worried about
19:19:01 <gholms> If we *can* test in EC2, we should.
19:19:03 <jsmith> As we saw with the F14 images, etc.
19:19:07 <jsmith> gholms: +1
19:19:41 <gholms> If we're stuck with community-only testing that means that we're asking people to spend real money to do this.
19:19:53 <gholms> As opposed to just spending time and lending a machine for a day
19:19:55 <msavy> that's true, although they can use micro instances
19:20:10 <msavy> micro is free, just the cost of an EBS volume (transiently, presumably)
19:20:28 <gholms> If your account is new enough, yes.
19:20:31 <msavy> downside: slower than a 3 legged donkey
19:20:32 <rbergeron> gholms: yeah. I've asked the Fedora person over there to get back to me with any suggestions on it.
19:20:35 * gholms 's is too old for that :(
19:20:37 <rbergeron> We'll see.
19:20:48 <mgoldmann> I'm sure 80% of all eager testers will fail with help on testing EC2 instances
19:20:51 * gholms hands rbergeron an #info
19:21:21 <rbergeron> :)
19:21:44 <msavy> dont suppose amazon donate any EC2 time for OSS causes?
19:21:45 <rbergeron> #info rbergeron has asked Fedora Dude at amzn about suggestoins for community testing.
19:21:45 <mgoldmann> ...because of money, fear, learing curve
19:21:54 <rbergeron> i don't know that they do.
19:21:59 <rbergeron> ONE CAN HOPE.
19:22:15 * rbergeron sighs
19:22:17 <gholms> mgoldmann: We would have to counter that with really good test case instructions. :-\
19:22:50 <rbergeron> Meanwhile: It probably wouldn't be impossible to come up with some sort of solution, even if it's like... giving people 5 bucks for gift card on amazon, if we have a reasonable idea for how much it costs, etc.
19:22:55 <rbergeron> I don't think it's impossible.
19:23:02 <rbergeron> Need test plan first.
19:23:38 <fenrus02> ec2 has a free level of service
19:23:42 <msavy> i think micro instances should really help with that
19:23:59 <msavy> fenrus02: yah, i mentioned that earlier, but apparently some accounts dont support it. nothing to stop them signing up for a new one afaik?
19:24:05 <rbergeron> Ah, yeah.
19:24:16 <fenrus02> https://aws.amazon.com/free/ -- no idea why an account would not support it.
19:24:31 <mgoldmann> it's only for new accounts IIRC
19:24:32 <gholms> fenrus02: Accounts older than a certain date don't qualify.
19:24:36 <rbergeron> yeah, new accounts.
19:24:43 <rbergeron> Most people don't want to keep recreating, etc.
19:24:49 <fenrus02> gholms, I see.  new to ec2 or new to amazon in general?  (shared account?)
19:24:59 <gholms> New to EC2, AFAIK
19:25:08 <msavy> oh that is slightly different. i think all micro instances are free anyway(?) just pay for EBS etc
19:25:45 <gholms> Even free micro instances don't get people over the hurdle of signing up for an account with a third party and registering a credit card number.
19:25:51 <rbergeron> #info if you have suggestions on how to ease costs of testing on testers on ec2, by all means, send mail to the list.
19:25:58 <rbergeron> (Cuz I gotta move on.)
19:26:07 <rbergeron> DO I HAVE ANY TAKERS on trying to drive a test plan page?
19:26:09 <gholms> And even beyond that, we need testing for stuff that isn't EBS-backed.
19:26:32 <rbergeron> gholms: yeah.
19:26:36 <rbergeron> Le sigh.
19:27:06 <rbergeron> Or an alternate solution altogether.
19:27:07 <rbergeron> Teehee.
19:27:10 <rbergeron> Anyway.
19:27:21 <rbergeron> Can we wrap up ec2 discussion for this week?
19:27:36 * rbergeron is looking at busy feature folks who also need attention
19:27:43 <mgoldmann> rbergeron: create a portal where people would sign up for testing, we'll be pay for AWS usage, they will get creds and address to isntance
19:28:07 <mgoldmann> just an idea
19:28:09 * mgoldmann hides
19:28:14 <rbergeron> mgoldmann: *nod*
19:28:19 <rbergeron> Okay.
19:28:19 <clalance> Actually, that can work with the shared stuff.
19:28:29 <clalance> (we have accounts setup that way internal to RedHat)
19:28:39 <clalance> The problem with that scheme is that EC2's controls are crap.
19:28:42 <rbergeron> clalance: who owns those accounts, so i can go harass them?
19:28:47 <clalance> So anyone can terminate anyone else's instances.
19:28:57 <mgoldmann> clalance: which controls?
19:28:59 <clalance> rbergeron: Matt Hicks.
19:29:03 <clalance> mgoldmann: The account controls.
19:29:14 <clalance> Basically everyone gets assigned to the same account (with different credentials).
19:29:17 <rbergeron> clalance: ah. thanks. I'll attempt to procure 20 seconds of his time. ;)
19:29:26 <clalance> But it means that anyone can do anything to all of the instances.
19:29:30 <clalance> Which is a dangerous situation.
19:29:32 <rbergeron> Right. that's the problem
19:29:34 <mgoldmann> ouch
19:30:13 <mgoldmann> clalance: but we wouldn't expose anything besides: keypair and IP, right?
19:30:13 <rbergeron> Okay.
19:30:14 <clalance> They also have a new account system, which is supposed to fix some of this stuff, but is still immature.
19:30:18 <rbergeron> mmm
19:30:21 <clalance> So the options are limited.
19:30:21 <rbergeron> interesting
19:30:27 <mgoldmann> and this isn sufficient to log in or do mess
19:30:28 <clalance> rbergeron: Anyway, you should talk to matt hicks.
19:30:34 <clalance> I ran away screaming after he told me some of htis.
19:30:38 <clalance> s/htis/this/
19:30:40 <rbergeron> clalance: will do. lol
19:30:56 <rbergeron> #topic Aeolus
19:31:08 <rbergeron> clalance: I know you have a boatload of packags available for review.
19:31:10 <mgoldmann> and we could have even one account on EC 2 for this
19:31:23 <clalance> rbergeron: Yeah, that is pretty much what I have to report.
19:31:24 <rbergeron> THAT NEED LOVE, because it's a Feature, yo.
19:31:34 <clalance> We are still hardening up 0.3.0, development wise.
19:31:44 <clalance> And I'm working on getting the dependencies in, as you may have seen.
19:31:56 <rbergeron> #link http://aeolusproject.org/page/Packages_Missing_From_Fedora
19:31:56 <clalance> (it's a tough slog, though, because rawhide is in pitiful shape at the moment)
19:32:01 <rbergeron> (should be current)
19:32:01 <rbergeron> yeah.
19:32:21 <clalance> rbergeron: One question I have for you.
19:32:37 <rbergeron> clalance: yessir
19:32:41 <clalance> rbergeron: Some of the problems stem from packages that I (or someone on my team) don't own.
19:32:53 <clalance> And I've tried emailing the package owners, with very little success.
19:32:59 <rbergeron> Okay. Sooooooo
19:33:12 <clalance> So what do we do there?
19:33:13 <fenrus02> "non-responsive packager" ?
19:33:18 <gholms> Did you file bugs?
19:33:24 <rbergeron> There is a ... non-responsive packager process.
19:33:27 <clalance> gholms: I did not, maybe that is the right thing to do.
19:33:33 <clalance> I'm not freaking out yet.
19:33:38 <clalance> (it has only been a few days)
19:33:40 <rbergeron> just being proactive ;)
19:33:49 <gholms> I would file bugs before worrying about non-responsiveness.
19:33:55 <clalance> But if we get towards July 21, I might freak out more :).
19:34:01 <gholms> Then at least things are tracked better.
19:34:02 <rbergeron> http://fedoraproject.org/wiki/Policy_for_nonresponsive_package_maintainers
19:34:04 <clalance> gholms: OK, yeah.  I should do it that way, in addition to the email.
19:34:07 <clalance> gholms: Yeah, point.
19:34:38 <rbergeron> clalance: that link describes more or less the process.
19:35:02 <clalance> rbergeron: Ouch, OK.
19:35:09 <clalance> So it seems like a minimum of 24 days.
19:35:40 <gholms> If the fixes are small and you have patches you could seek out a provenpackager to help you out.
19:35:41 <clalance> All right.  Well, I'll do as gholms suggests, at least start it that way.
19:35:53 <clalance> gholms: Ah, right.  OK, another avenue to keep in mind.
19:36:31 <gholms> #action clalance to file bugs against aeolus-blocking stuff in rawhide
19:36:42 * mgoldmann just learned about vacation page
19:36:49 <clalance> rbergeron: So other than that, I have lots of need for review.
19:36:51 <rbergeron> what's a vacation? :)
19:36:52 <clalance> (and willing to swap)
19:36:54 <rbergeron> yeah.
19:37:07 <mgoldmann> rbergeron: dunno, /me looks up
19:37:11 <rbergeron> #info please, please, please, need reviews on aeolus, please.
19:37:27 * rbergeron may try some blog advertising on Monday.
19:38:12 <rbergeron> clalance: thanks. :)
19:38:19 <clalance> rbergeron: Thank you.
19:38:21 * rbergeron knows you guys are busybusy right now.
19:38:32 <clalance> Man, I need this long weekend ;).
19:38:34 <rbergeron> #topic The feature formerly known as CloudFS
19:38:40 <mgoldmann> lol
19:38:47 <gholms> Haha
19:38:52 <kkeithley> Got a special symbol just for that
19:38:58 <rbergeron> HAH
19:39:07 <rbergeron> kkeithley: SO DO TELL
19:39:16 <kkeithley> ;-)
19:39:50 <kkeithley> er, where did jdarcy go
19:39:58 <rbergeron> i don't see him here.
19:40:02 <rbergeron> It's all you. ;)
19:40:16 <rbergeron> he went to cancun.
19:40:21 <rbergeron> or the USPTO. Unknown.
19:40:32 <rbergeron> want me to summarize?
19:40:49 <kkeithley> yeah. anyway. summarize the Feature Formerly Known as CloudFS?
19:40:54 <kkeithley> sure, you summarize
19:41:24 <rbergeron> #info CloudFS is getting a new name due to trademark / legal FUN STUFF
19:41:34 <rbergeron> Did you guys pick a name yet?
19:41:48 * msavy reckons it should be CumulusFS, or something ;-)
19:41:51 <rbergeron> I think you should call it PrinceFS
19:41:59 <clalance> rbergeron: Love it ;).
19:42:02 <rbergeron> (artist formerly known as, feature formerly known as...)
19:42:09 <msavy> better idea, F-FS
19:42:10 <kkeithley> I made a bunch of suggestions. PrinceFS/Princess
19:42:30 <kkeithley> GNATFS (good names all taken FS)
19:42:36 <rbergeron> yeah, FFS was my suggestion.
19:42:44 <rbergeron> Because FFS was pretty much what I said. ;)
19:42:44 <gholms> FFS already exists. :(
19:42:50 <msavy> darn it!
19:42:50 <clalance> jdarcy mentioned Block Storage FS (BSFS) to me ;).
19:42:58 <rbergeron> lol
19:43:04 <gholms> I like that one.
19:43:10 <kkeithley> I made a bunch of suggestions, but I think the ball is really in jdarcy's court.
19:43:14 <rbergeron> Okay.
19:43:21 * gholms wonders how many filesystems do not have block storage
19:43:22 <rbergeron> #info jdarcy is working on a new name.
19:43:33 <clalance> gholms: Hence the acronym ;).
19:43:45 <kkeithley> and I've been working on packaging, don't want to go down to the wire
19:43:51 <rbergeron> kkeithley: and you need ke4qqq to get you sponsored.
19:44:00 <rbergeron> and probably jdarcy as well.
19:44:01 <kkeithley> clalance did an unofficial review of my first stab at it
19:44:05 <rbergeron> awesome.
19:44:13 <rbergeron> kkeithley: have you done any reviews of anyone else's packages?
19:44:21 <kkeithley> yes, I poked ke4qqq by email earlier, he's starting the sponsor stuff later today
19:44:24 <kkeithley> nope, I haven't
19:44:55 <rbergeron> kkeithley: that might help too. i odn't know what ke4qqq's sponsoring policy is, but sponsoring often includes having you do informal reviews of other packages.
19:45:05 <rbergeron> (So the sponsor can get a grip on your package-fu.)
19:45:07 <kkeithley> yeah, I'm cool with that
19:45:34 <rbergeron> kkeithley: so I'd suggest that maybe being uber-proactive might include informally reviewing an aeolus package.
19:45:37 <rbergeron> ;)
19:45:39 <rbergeron> unofficially
19:45:47 <rbergeron> or you can wait to see what ke4 says.
19:45:51 <rbergeron> Just a thought.
19:46:02 <rbergeron> I'm sure he'll get you taken care of.
19:46:43 <kkeithley> where do I sign up? ;-)
19:47:01 <rbergeron> for... informal reviews?
19:47:15 <kkeithley> #aeolus-whatever, I suppose
19:47:23 <rbergeron> pick something off the list...
19:47:23 <clalance> kkeithley: Yeah, #aeolus.
19:47:25 * rbergeron grabs link
19:47:29 <kkeithley> clalance has some that need reviewed?
19:47:30 <rbergeron> yeah, or pop into #aeolus
19:47:39 <rbergeron> kkeithley: "some" is a gentle word
19:47:43 <clalance> kkeithley: There's this list: http://aeolusproject.org/page/Packages_Missing_From_Fedora
19:47:45 <rbergeron> kkeithley: "many" is more like it
19:47:49 <rbergeron> ;)
19:48:04 * rbergeron is gonna move on, unless you have anything else kkeithley?
19:48:25 <kkeithley> nope not from my end
19:48:28 <rbergeron> #topic OpenStack
19:48:31 <rbergeron> thanks ;)
19:48:33 <rbergeron> Okay.
19:48:39 <gholms> Nova is up for review.
19:48:41 <rbergeron> So we've been trying to get this other openstack package in for a while.
19:49:17 <rbergeron> https://bugzilla.redhat.com/show_bug.cgi?id=707199
19:49:30 <rbergeron> mdomsch has taken some lead on it, which is great.
19:49:46 <rbergeron> and i quote:
19:49:47 <rbergeron> 11:59 < mdomsch> I can't stay long.  I did a first pass review of openstack-nova, gave a patch to help it along, still needs a little more love.
19:50:08 <rbergeron> #info OpenStack package still needs a little love, has a patch to help move it along.
19:50:22 <gholms> There's confusion about whether or not the package is allowed to spit out a message that tells people to disable SELinux upon installation.
19:50:44 <rbergeron> Is that something we need to go visit fesco about?
19:51:04 <rbergeron> or is that something we can just sort out on our own
19:51:07 <gholms> If anything it's a FPC question: whether or not stuff has to work with SELinux on.
19:51:15 <rbergeron> that was my next listing
19:51:20 <rbergeron> err, suggestion
19:51:36 <rbergeron> Okay. well....
19:51:50 <gholms> If it doesn't then they just need to drop that echoing.
19:52:17 <rbergeron> Yeah.
19:52:19 <gholms> But I would really like to know whether or not we allow packages that require one to shut SELinux off.
19:52:37 <gholms> The guidelines don't require it, but I'm not sure if that is a conscious decision or not.
19:52:39 <rbergeron> gholms: would you like to chase that down? :)
19:52:59 <rbergeron> in your vast spare time? :)
19:53:03 <gholms> Shur
19:53:09 <rbergeron> thanks :)
19:53:20 <clalance> gholms: Another avenue is to open up a bug against the policy and get dan walsh to look at it.
19:53:31 <clalance> He might be able to whip you up some policy, if he has time.
19:53:34 <rbergeron> #action gholms to investigate kosherness of a package requirement to disable SELinux.
19:53:39 <rbergeron> clalance: good point
19:53:44 <gholms> clalance: Goot point
19:53:46 <gholms> *good
19:53:50 * gholms can't type today
19:53:59 <rbergeron> #idea open up a bug against the policy and see if dan walsh can help out.
19:54:20 <rbergeron> clalance: thanks for that tip.
19:54:28 <rbergeron> s/tip/reminder ;)
19:54:35 * rbergeron looks around.
19:54:47 <rbergeron> #topic BoxGrinder
19:54:47 <gholms> [Ceiling Cat watches silently.]
19:55:02 <rbergeron> msavy, mgoldmann: anything brief and interesting? :)
19:55:07 <rbergeron> is ceiling cat related to keyboard cat?
19:55:08 <mgoldmann> definitely
19:55:09 <mgoldmann> :)
19:55:18 <mgoldmann> msavy where are you?! :)
19:55:18 <msavy> only brief, jees?
19:55:23 <gholms> Man, rbergeron is in a hurry today. :-\
19:55:27 <rbergeron> Hey.
19:55:29 <rbergeron> Three day weekend, yo.
19:55:34 <gholms> lol
19:55:35 <rbergeron> TICK TOCK ;)
19:55:38 * clalance is taking off after this ;).
19:55:46 <msavy> ok, so, we released 0.9.3 finally, huzzah!
19:55:50 * rbergeron claps
19:56:04 <msavy> features in brief:
19:56:13 <msavy> scientific linux build support
19:56:21 <gholms> #info BoxGrinder 0.9.3 released; see http://boxgrinder.org/blog/2011/06/28/boxgrinder-0-9-3-released/ for details
19:56:45 <msavy> you can now build and deliver EBS backed AMIs to any Amazon region
19:57:06 <msavy> we've resolved loads of concurrency/timing issues, and so far we seem to be in the clear
19:57:25 <msavy> a new meta appliance is out (boxgrinder meta 1.6), for people that wanna try BoxGrinder Build locally or on EC2
19:57:34 <msavy> http://boxgrinder.org/download/boxgrinder-build-meta-appliance/
19:58:03 <mgoldmann> it's EBS based and available in all EC2 regions of course!
19:58:34 <rbergeron> :)
19:58:35 <msavy> yes, and we have a user who's been using it in Singapore
19:58:42 <rbergeron> ah, cool.
19:58:44 <msavy> ap-southeast-1
19:59:14 <msavy> mgoldmann has added a nice new validation structure to BG, meaning a plugin can validate _before_ the build process begins
19:59:41 <msavy> which saves a lot of pain, as people would do the first few stages of building then it would explode because their parameters were wrong, or some other validation issue
19:59:56 <rbergeron> Ah. that's a good idea to add in.
20:00:19 <mgoldmann> and we plan next big thing, maybe for 0.9.4 even
20:00:34 <msavy> oh and also mgoldmann's fancy new integ tests, which we've heard about :D
20:00:45 <mgoldmann> you will be able to upload files to appliance
20:00:46 <rbergeron> nice.
20:00:53 <gholms> msavy: If you're interested I did a scratch build of euca2ools that supports S3 location constraints.
20:01:02 * gholms needs testes
20:01:05 <gholms> *testers
20:01:09 <msavy> rofl
20:01:12 <gholms> >.<
20:01:13 <rbergeron> lololol
20:01:14 <clalance> LOL
20:01:18 <clalance> That was a great one.
20:01:28 <gholms> Stupid laptop keyboards...
20:01:35 * rbergeron is trying to not cry from laughing
20:01:42 <mgoldmann> rotfl, /me just looked up
20:01:50 <rbergeron> THIS IS WHY YOU STAY ON FRIDAYS, FOLKS
20:01:51 <rbergeron> ;)
20:02:08 <gholms> rbergeron: *You're* the one who wants to leave.
20:02:18 <rbergeron> Not when there's all these testers here.
20:02:22 <rbergeron> C'mon.
20:02:33 <msavy> gholms: teehee.. anyway, we use amazon-ec2 gem for s3/ebs stuff
20:02:33 <rbergeron> ;)
20:02:58 <gholms> Ok
20:03:01 * gholms has to go
20:03:06 <rbergeron> gholms: thanks. :)
20:03:18 <msavy> however, i'll take a look, as amazon-ec2 gem is not being maintained much
20:03:18 <rbergeron> unles you're being facetious
20:03:26 * rbergeron nods
20:03:30 <msavy> s/as/as the
20:03:32 <msavy> so cheers for that
20:04:10 <msavy> but yeah, we squashed loads of bugs in this release, and despite me screwing up and saying it was release 0.8.3 in our mailing list (it was 9pm, jees!)
20:04:17 <rbergeron> ;)
20:04:25 <rbergeron> very nice, guys. ;)
20:04:28 <mgoldmann> msavy: it's 10pm now :)
20:04:40 <mgoldmann> yah, that's all folks
20:04:48 <rbergeron> Cool. thank you, as always :)
20:04:52 <rbergeron> OH
20:05:35 * rbergeron notes that there is apparently some sort of F15 video coming out at some point, and there is some talk about BG in it.
20:05:47 <mgoldmann> oh, nice!
20:05:50 <msavy> sweet
20:05:51 <rbergeron> (Note that I am the one talking, so..........................)
20:06:09 <rbergeron> We should see it eventually. :)
20:06:13 <msavy> oh darn, you dont want our dulcet tones :(
20:06:15 * mgoldmann eagerly waits for link when it will be ready
20:06:18 <rbergeron> lol
20:06:54 <gholms> [There is the sound of a banjo falling hundreds of feet and hitting the ground]
20:06:55 <rbergeron> msavy: I would have preferred you guys, but sadly I got kind of roped into it while within filming range on RDU campus
20:07:02 <gholms> Whoops, wrong channel
20:07:04 <rbergeron> gholms: something like that. ;)
20:07:12 <rbergeron> It's not *that* bad.
20:07:21 <msavy> its ok, mgoldmann and i will travel to RDU to do it ;-)
20:07:32 * msavy snickers
20:07:36 <rbergeron> Anyway. Something to look forward to, I suppose, except for me, becuase I'm self-conscious.
20:07:37 <msavy> ok, let us know how it goes :)
20:07:39 <rbergeron> Anyway!
20:07:44 <rbergeron> #topic Open Floor
20:08:24 <rbergeron> #info for those playing along at home, this weekend is a 3-day weekend in US, so people may be missing on Monday.
20:08:29 <rbergeron> Anything else, anyone?
20:08:38 <gholms> Today is Canada Day, anyway.
20:08:47 <rbergeron> hoser
20:08:50 <gholms> If you play your cards right you can have a four-day weekend.
20:08:53 <rbergeron> I could.
20:09:04 <rbergeron> I could just take all of next week off, too.
20:09:10 <rbergeron> But I'll probably be working on monday instead. ;)
20:09:14 <gholms> havefunwiththat.jpg
20:09:23 * rbergeron has SEKRIT PLANS TO TAKE OER THE WORLD THAT NEED ATTENTION
20:09:32 <rbergeron> Anyway.
20:09:38 <rbergeron> I think that's about all, unless someone else has something.
20:10:10 <msavy> nada
20:10:15 * rbergeron looks at the countdown ticker
20:10:17 <gholms> [You hear some noises in the distance]
20:10:26 <rbergeron> Thanks for coming, everyone :)
20:10:27 <clalance> Was that a gunshot?
20:10:29 <clalance> :)
20:10:38 <mgoldmann> yeah, kitten is dead
20:10:40 * rbergeron encourages people to make noise on the list if they need feature-related help.
20:10:41 <mgoldmann> http://fuckyeahnouns.com/weekend
20:10:45 <clalance> rbergeron: As usual, thanks :).
20:10:59 <rbergeron> clalance: I try. Thank you ;)
20:11:27 * rbergeron waves
20:11:31 <rbergeron> OH LOOK FIREWORKS
20:11:33 <rbergeron> #endmeeting