12:01:14 <ndevos> #startmeeting
12:01:14 <zodbot> Meeting started Wed Apr  8 12:01:14 2015 UTC.  The chair is ndevos. Information about MeetBot at http://wiki.debian.org/MeetBot.
12:01:14 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
12:01:22 <ndevos> #info Agenda for today: https://public.pad.fsfe.org/p/gluster-community-meetings
12:01:23 * hagarth waves .. partially here
12:01:31 <ndevos> #topic Roll Call
12:01:36 * jimjag is here
12:01:37 <ndevos> hello hagarth!
12:01:43 * ndevos _o/ jimjag
12:01:44 * kshlm is here
12:01:50 * JustinClift waves
12:02:20 * kkeithley_ is here
12:02:22 * overclk is here
12:02:25 * raghu is here
12:02:32 * jdarcy is
12:02:58 <ndevos> looks like a good show up today, thanks!
12:03:06 * hchiramm_ is here
12:03:10 <ndevos> #topic Last weeks action items
12:03:15 <ndevos> #topic tigert to continue investigating the blog post duplication
12:03:27 <ndevos> tigert: hows the progress?
12:03:59 * spot is here
12:04:06 <ndevos> #info ENOTIGERT
12:04:15 <ndevos> #topic hchiramm will share the outcome of the non-mailinglist packagng discussions on the mailinglist (including the Board)
12:04:23 <hchiramm_> ndevos, TBD
12:04:34 <ndevos> #info TBD
12:04:45 <ndevos> #topic Pranith will complete the feature page for (k)vm hyperconvergence
12:04:58 <ndevos> hmm, pk isnt there...
12:05:16 <ndevos> http://www.gluster.org/community/documentation/index.php/Features/Ovirt_Hyperconvergence for the people who have not heard about it
12:05:47 <ndevos> its about deploying oVirt and Gluster on the same physical servers
12:06:02 <JustinClift> hagarth: ^ any idea? (it used to be your AI)
12:06:08 <tigert> ergh
12:06:15 * tigert is present 5 minutes late :P
12:06:19 <ndevos> I know pranith got involved in more discussions about it this week
12:06:31 <ndevos> so, it seems to make progress
12:06:33 * Debloper is present, sorry for the delay
12:07:03 <ndevos> #action Pranith should report progress on the oVirt Hyperconvergence feature page
12:07:26 <ndevos> #topic spot to reach out to community about website messaging
12:07:44 <ndevos> spot did send an email about this, and there were some nice responses
12:07:50 <spot> :D
12:08:06 <ndevos> spot: got a summary/collection/tracking somewhere? what are the next steps?
12:08:07 <spot> as i said in the email, i'll circle back to the community about this in about a month
12:08:22 <spot> i need to sit down with the RH marketing folks too
12:08:41 <spot> but there is definitely some good stuff already.
12:08:46 <tigert> okay
12:09:07 <ndevos> sounds good
12:09:12 <tigert> this would definitely be something we should sync the web efforts with
12:09:32 * tigert will reply to deblopers mail thread in users list,
12:09:52 <tigert> but the messaging does have a rather big say in how the site should be shaped too
12:09:58 * Debloper looks forward for that :D
12:10:02 <tigert> yeah
12:10:11 * hchiramm_ reg#hyperconvergence - http://www.ovirt.org/Features/GlusterFS-Hyperconvergence
12:11:49 <ndevos> spot: any thougts on how to combine the website efforts with the messaging?
12:12:23 <JustinClift> Awesome, my RealSense 3D camera just got delivered. (way off topic ;>)
12:12:32 <spot> Not immediately, but its not really my area of expertise. I'll work with folks who have that skill. :)
12:12:32 <ndevos> okay, I guess not
12:12:40 <ndevos> :)
12:12:50 <tigert> messaging defines what we want to say
12:13:06 <tigert> what we want to say has an impact on how to put that into a website form
12:13:12 <tigert> so its all related
12:13:25 <tigert> if I understand "messaging" correctly of course :)
12:13:53 <shaunm> and in the big pool of webbiness we have to fit good docs
12:13:54 <tigert> words are sometimes interesting when you have different areas of expertise working together
12:14:03 <tigert> those docs yeah
12:14:37 <tigert> they are a big part in the welcoming new folks part
12:14:47 <tigert> erm, that has one part too many
12:14:53 <Debloper> Please correct me if I'm wrong, spot - are we talking about the slogan thread?
12:15:03 <spot> Debloper: i thought we were
12:15:15 <spot> but this action item is painfully vague/broad
12:15:18 <Debloper> ah, right - great :0
12:15:19 <ndevos> yes, I think this is mostly about the slogan
12:15:20 <Debloper> :)
12:16:12 <ndevos> #topic tigert to continue investigating the blog post duplication
12:16:19 <Debloper> spot: yep, that's what I was having trouble defining the scope/expanse of messaging (and the thorough impact on the content)
12:16:20 <ndevos> lets try that one again :)
12:16:46 <tigert> ndevos: so I looked into reimplementing the whole planet with planetplanet.org code
12:16:55 <tigert> but did not yet get far enough
12:16:58 <tigert> will continue
12:17:14 <Debloper> if it's mostly about the slogan, then there's practically very less blocker that both the gathering a good slogan, as well as making the site better can't go in parallel.
12:17:20 <tigert> I guess the only list of included blogs is in the wordpress plugin=?
12:17:50 <ndevos> ~.
12:18:08 * Debloper can help managing WordPress, if volunteer required.
12:19:57 <ndevos> tigert: I dont have an idea about the blog, not the slightest
12:20:36 <ndevos> so, any progress you make is appreciated, jus tkeep us informed :)
12:20:39 <ndevos> #topic hagarth to carry forward discussion on automated builds for various platforms in gluster-infra ML
12:21:00 <ndevos> hagarth sent the email, I did read it, but have not seen any replies
12:21:13 <tigert> ndevos: yeah, will look. my idea is to recreate the same feed syndication with planetplanet and see if the error persists
12:21:26 <ndevos> #info hagarth did sent the email, waiting for replies
12:21:36 <tigert> ndevos: so if that is gone, then the issue is in wordpress which should die some day after we make it redundant
12:21:37 <ndevos> tigert: sounds like a plan :)
12:21:55 <ndevos> #topic telmich will send an email to the gluster-users list about Gluster support in QEMU on Debian/Ubuntu
12:22:09 <ndevos> telmich isnt here, JustinClift you wanted to contach him?
12:22:23 <JustinClift> Yeah
12:22:28 <ndevos> did you?
12:22:31 <JustinClift> I'm pretty sure I emailed him actually
12:22:32 * JustinClift checks
12:23:36 <JustinClift> Hmmmm, unsure.  I'll find the email / and if not then I'll email him by next meeting
12:24:05 <ndevos> #action JustinClift will send telmich an email (or find the one he sent)
12:24:13 <ndevos> #topic hagarth will post more details about an overlay xlator (container targetted) when he has a bit more clarity (and post his hacks)
12:24:39 <ndevos> I do not think he sent anything yet
12:25:35 <ndevos> #action hagarth should pass more details about the overlay xlator on to someone attending the meeting
12:25:39 <ndevos> #topic firemanxbr to submit a patch for 3.6.3, change IP addr.
12:25:54 <ndevos> hmm, this should have been moved, I think?
12:26:00 <ndevos> raghu: do you remember?
12:26:31 <raghu> ndevos: I think it was decided in last meeting that, the patch will be considered for 3.6.4
12:26:51 <hchiramm_> yeah, thats what we decided  in last meeting
12:27:11 <ndevos> #info firemanxbr will work on his script to rename servers or change ip-addresses for 3.6.4
12:27:19 <raghu> apart from that, that patch has not been sent for review yet
12:27:30 <ndevos> okay
12:27:36 <ndevos> #topic jdarcy to schedule a 4.0 catch up meeting next week
12:28:14 <ndevos> jdarcy: did you plan a meeting yet?
12:28:23 <jdarcy> Nope.
12:28:44 <JustinClift> Note - Email to Nico (telmich) sent
12:28:47 <JustinClift> Just now. ;)
12:28:49 <ndevos> okay, we'll get back to 4.0 later
12:28:55 <ndevos> #topic kshlm to propose gluster winter of code ideas on blog/mailing lists
12:29:48 <kshlm> No progress on this.
12:29:49 <ndevos> I dont think I have seen anything about this yet, but I've been a little disconnected the last days
12:30:05 <kshlm> I've been busy with 3.7 deliverables.
12:30:17 <ndevos> sure, no problem
12:30:24 <ndevos> you think you get something out for next week?
12:30:58 <kshlm> Hopefully, if hagarth and I get some time to proceed with the discussion.
12:31:19 <ndevos> okay, fingers crossed!
12:31:43 <ndevos> #topic GlusterFS 3.7
12:32:04 <ndevos> hagarth: you're up, if you're there?
12:32:23 <ndevos> list of bugs, and hopefully their correct status: https://bugzilla.redhat.com/show_bug.cgi?id=glusterfs-3.7.0
12:32:47 <ndevos> #info once all patches for a bug have been merged, the developer should move the status of the bug to MODIFIED
12:33:33 <ndevos> hagarth plans to branch 3.7 this weekend
12:33:41 <hchiramm_> as u know hagarth has sent a mail to gluster MLs about 3.7 branching
12:34:16 <ndevos> that hopefully contains all the important fixes and features, only important bugs will then get merged in release-3.7
12:34:43 <ndevos> and, that means tat all patches go to the master branch first, and need a backport to the release-3.7 branch
12:35:22 <ndevos> actually, this link shows the 3.7.0 bugs better: https://bugzilla.redhat.com/showdependencytree.cgi?id=glusterfs-3.7.0&hide_resolved=1
12:36:07 <ndevos> we also need to fix the spurious failures listed on this etherpad:  https://public.pad.fsfe.org/p/gluster-spurious-failures
12:36:28 <jdarcy> +100
12:36:36 <ndevos> the main failure that I now see, is a core dump (on cleanup?)
12:36:46 <ndevos> jdarcy: you were looking into that issue?
12:37:00 <jdarcy> ndevos: Yes, the core files from pl_inodelk_something_cleanup are the leading cause of failures.
12:37:22 <jdarcy> I looked into it yesterday, nothing leapt out at me.  I've heard Pranith was also looking.
12:37:48 <ndevos> ah, I do not think pranith had much time for that today :-/
12:38:02 <jdarcy> What do people think about disabling the core-file check until this is resolved?  It doesn't seem to be telling us anything we don't know.
12:38:03 <JustinClift> Hmmm, we haven't really been updating that Etherpad as well as we should
12:38:19 <jdarcy> JustinClift: I update it, but I think I'm almost the only one.
12:38:28 <JustinClift> k.  Keep updating it.
12:38:29 <JustinClift> ;)
12:38:34 <ndevos> JustinClift: actually, it should have all been single bug reports :-/
12:38:45 <JustinClift> I'll hassle the mailing list about that shortly
12:38:55 <JustinClift> "BAD mailing list" ;)
12:39:05 <ndevos> jdarcy: were you able to identify which test-case causes the issue?
12:39:25 <jdarcy> ndevos: No, that was the original purpose of http://review.gluster.org/#/c/10157/
12:39:27 * ndevos did see some run-tests.sh change that could help with identifying that
12:39:58 <jdarcy> It would be *really* helpful to know which tests trigger it, but we don't have that information currently.
12:40:00 <ndevos> okay, I thought you might have used that to run on a different system
12:40:03 <JustinClift> jdarcy: If we disable the core file check, is there any chance that will cause us to lessen the priority of fixing this?
12:40:30 <jdarcy> JustinClift: I can't say it wouldn't.
12:40:31 <ndevos> JustinClift: this is an issue that needs to be fixed
12:40:52 <ndevos> running without the coredump check would surely help getting patches in faster
12:41:15 <JustinClift> Understood.  I'm kinda on the fence about it myself...
12:41:22 <JustinClift> ndevos: What're your thoughts?
12:41:37 <jdarcy> We *could* try to write a gdb script that will separate out these crashes from others we haven't seen before (and such do occur), but that would take someone some time.
12:42:07 <JustinClift> jdarcy: Probably better to spend the time on getting it fixed. ;)
12:42:07 <jdarcy> Run "thread apply bt all" on it, look for a few recognizable strings.
12:42:11 <ndevos> but, we also need to be careful to not introduce other segfaults - I do not know how often we get patches that actually get segfaults and no test-failures
12:42:22 <hagarth> jdarcy: that looks like a potentially good approach
12:42:40 <jdarcy> ndevos: Segfaults are the leading cause of failures right now.  Most, but not all, are this one bug.
12:42:47 <JustinClift> "corecheck.sh"? ;)
12:43:27 <ndevos> okay, who wants to write a script to match this segfault/core pattern?
12:43:29 <jdarcy> I'll spend half an hour today seeing if I can make such a gdb script work.
12:43:39 <ndevos> thanks jdarcy!
12:43:43 <hagarth> Pranith has a potential root cause for this core .. I'll remind him again about sending a note on -devel
12:44:22 <JustinClift> hagarth: If Pranith does that soonish, maybe the 1/2 hour of jdarcy's time is better spent on the potential root cause?
12:44:23 <ndevos> #action jdarcy will briefly look into writing a script to match the known segfault issue
12:44:39 <jdarcy> hagarth: I'd love to hear his theory.
12:45:02 <hagarth> jdarcy, JustinClift: will try reminding him this evening, if I cannot it will be tomorrow.
12:45:16 <ndevos> hagarth, jdarcy: could one of you send him an email and put gluster-devel on cc?
12:45:24 <jdarcy> ndevos: Something like this will undoubtedly be needed at some point in the future.  Good to have the script in our quiver for that day.
12:45:39 <hagarth> ndevos: I already did that
12:45:43 <hagarth> jdarcy: +1
12:45:56 <ndevos> jdarcy: yes, definitely
12:46:01 <ndevos> hagarth: ok, thanks!
12:46:14 <ndevos> hagarth: anything else you want to say about 3.7?
12:46:42 <hagarth> ndevos: haven't been paying lot of attention here .. but I captured most of the todo in the email I sent yesterday
12:46:59 <hagarth> I am still looking for an alpha and branching later this week
12:47:03 <ndevos> hagarth: okay, thats all then :)
12:47:13 <ndevos> #topic GlusterFS 3.6
12:47:21 <ndevos> raghu: up to you
12:47:45 <raghu> not much from me. 3.6.3beta2 is already done. I am going to make 3.6.3 tomorrow
12:48:24 <ndevos> #info 3.6.3 is planned to be released tomorrow
12:48:29 <JustinClift> Cool :)
12:48:33 <ndevos> anything else?
12:48:40 <raghu> Some patches that need a review, will be considered for 3.6.4
12:48:53 <hagarth> JustinClift: let us use 3.6.3 for our gerrit and Jenkins backend in b.g.o ;)
12:49:12 <ndevos> raghu: can you get the 3.6.4 tracker bug filed and move any outstandng 3.6.3 to the new bug?
12:49:33 <JustinClift> hagarth: Maybe
12:49:34 <JustinClift> ;)
12:49:36 <raghu> ndevos: sure. Will do it before making 3.6.3
12:49:46 <ndevos> thanks raghu!
12:50:00 <JustinClift> hagarth: We'd need to buy two more servers though, for quorum... hmmm...
12:50:16 <hagarth> JustinClift: we can spin up VMs but more later.
12:50:20 * JustinClift nods
12:50:24 <ndevos> #topic GlusterFS 3.5
12:50:40 <JustinClift> Just emailed Pranith btw
12:50:59 <ndevos> a beta tag has been created, hchiramm_ and lala were looking into getting packages out, I think
12:51:28 <hchiramm_> ndevos, yes .
12:51:32 <ndevos> 3.5.4beta1 should become available soon(ish?)
12:51:58 <ndevos> once we have some test results, we'll think about fixing or releasing 3.5.4
12:52:11 <ndevos> #topic GlusterFS 3.4
12:52:24 <ndevos> kkeithley_: hows the 3.4.7 build doing?
12:52:35 <kkeithley_> na na na na, hey hey, good bye
12:52:44 <kkeithley_> 3.4.7 was released earlier today
12:52:54 <kkeithley_> email to the lists, blog posts
12:52:55 <ndevos> wohoo! https://www.gluster.org/pipermail/gluster-users/2015-April/021433.html
12:52:56 * JustinClift cheers
12:53:06 <hchiramm_> \0/
12:53:17 <kkeithley_> rpms for a variety of linux distros on download.gluster.org
12:53:25 <hagarth> congratulations on getting 3.4.7 out!
12:53:40 <hchiramm_> http://download.gluster.org/pub/gluster/glusterfs/3.4/LATEST/
12:53:47 <spot> hooray!
12:53:50 <kkeithley_> .debs for wheezy and squeeze too
12:54:03 <ndevos> kkeithley_: and arm packages?
12:54:18 <kkeithley_> For Raspbian, yes. Pidora later, maybe
12:54:26 <ndevos> okay, nice
12:54:33 <kkeithley_> Fedora ARM too, yes
12:54:48 * ndevos runs 3.5 on his arm systems
12:55:06 <ndevos> any last words on 3.4?
12:55:21 <kkeithley_> 3.4 is EOL effective with the branch of 3.7
12:55:24 <ndevos> we'll scratch the topic from next weeks agenda, if you dont mind?
12:55:32 <kkeithley_> be my guest
12:55:41 <hchiramm_> kkeithley++
12:56:19 <ndevos> kkeithley_: wait, "branch" of 3.7, or the release of 3.7.0?
12:56:28 * kkeithley_ is pretty sure there will not be a 3.4.8
12:56:34 * ndevos wants to #info that
12:56:50 <ndevos> #info kkeithley_ is pretty sure there will not be a 3.4.8
12:56:59 <ndevos> #topic GlusterFS 4.0
12:57:07 <ndevos> jdarcy: you again?
12:57:27 <jdarcy> Not for long.  Nothing to report, since everyone has been dealing with 3.7 stuff.
12:57:27 <ndevos> https://bugzilla.redhat.com/show_bug.cgi?id=glusterfs-4.0.0 does not exist yet
12:58:00 <ndevos> but anybody is able to create that, just set Alias=glusterfs-4.0.0 in the bug
12:58:22 <hagarth> I think all of us would need to start ramping up 4.0 activities soon so that we can have more meaningful exchanges at the design summit
12:58:29 <ndevos> jdarcy, you'll create that bug and set the Alias field?
12:58:59 <jdarcy> ndevos: OK.  Add an AI so I don't forget.
12:59:27 <ndevos> #action jdarcy will create the glusterfs-4.0.0 tracker bug and sets the Alias field to glusterfs-4.0.0
12:59:41 <ndevos> #topic Open Floor
13:00:11 <ndevos> there are no additional topics in the agenda, anyone wants to discuss someting else?
13:00:22 * JustinClift shakes head
13:00:31 <spot> want some good news on summit?
13:00:38 <JustinClift> Yep
13:00:38 <hagarth> spot: yes, please!
13:00:39 <jdarcy> Yes, please.
13:00:54 <spot> As of about an hour ago, we have dates and a venue
13:01:07 <ndevos> wow, thats great!
13:01:09 <spot> Dates are May 12 - May 13
13:01:12 <hagarth> nice!
13:01:14 <spot> Venue is https://www.h10hotels.com/en/barcelona-hotels/h10-marina-barcelona/
13:01:29 <spot> Please do not go book rooms yet, we need to get our block setup
13:01:42 <spot> i will email everyone with the hotel details.
13:01:54 <spot> I'm also going to start emailing people about flights today
13:01:57 <jdarcy> Sweet.
13:02:06 <ndevos> #info Gluster Summit takes place in Barcelona, Spain from May 12 - May 13
13:02:37 <jdarcy> What's the visa situation for US/Spain or India/Spain?
13:02:51 <spot> Spain is EU, so no visa for the US
13:03:42 <jdarcy> :)
13:03:43 <sankarshan_> reasonably easy from India as well
13:03:53 <JustinClift> :)
13:03:57 <spot> sankarshan_: good to know
13:04:16 <jdarcy> sankarshan_: Does "reasonably easy" mean required but not hard to get?
13:04:30 <spot> jdarcy: a quick google seems to imply that is the case.
13:04:34 <hagarth> jdarcy: yes
13:04:35 <hchiramm_> jdarcy, yes :)
13:04:52 <hagarth> I've had a bit of fun in getting a spanish visa in the past
13:05:34 <hagarth> you are required to take copies of all pages in your passport(s) including blank ones :)
13:05:40 <JustinClift> Heh
13:05:46 <ndevos> lol
13:06:27 <JustinClift> k, we're over time...
13:06:37 <jdarcy> There's a severe scratch-paper shortage in Spain, this is how they address it.
13:06:49 <ndevos> I'm much looking forward meeting everyone in Barcelone
13:07:04 <hagarth> jdarcy: lol
13:07:36 <ndevos> #info until the Summit in Barcelona we have at least 2 weekly meetings, one on Tuesdays and one on Wednesday - JOIN THEM
13:07:54 <ndevos> thanks all for joining today, and talk to you soon again!
13:07:59 <ndevos> #endmeeting