Perl 6 - the future is here, just unevenly distributed

IRC log for #gluster-dev, 2014-11-21

| Channels | #gluster-dev index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:17 nishanth joined #gluster-dev
01:46 bala joined #gluster-dev
02:18 topshare joined #gluster-dev
02:45 kdhananjay joined #gluster-dev
03:16 hagarth joined #gluster-dev
03:44 bharata-rao joined #gluster-dev
03:45 kanagaraj joined #gluster-dev
04:10 soumya joined #gluster-dev
04:17 atinmu joined #gluster-dev
04:20 bala joined #gluster-dev
04:24 shubhendu joined #gluster-dev
04:24 nishanth joined #gluster-dev
04:31 ndarshan joined #gluster-dev
04:32 aravindavk joined #gluster-dev
04:35 anoopcs joined #gluster-dev
04:40 kshlm joined #gluster-dev
04:41 Gaurav__ joined #gluster-dev
04:52 rafi1 joined #gluster-dev
04:54 lalatenduM joined #gluster-dev
04:55 lalatenduM_ joined #gluster-dev
05:01 atalur joined #gluster-dev
05:08 atinmu joined #gluster-dev
05:09 kdhananjay joined #gluster-dev
05:19 ppai joined #gluster-dev
05:26 jiffin joined #gluster-dev
05:27 spandit joined #gluster-dev
05:31 nkhare joined #gluster-dev
05:50 deepakcs joined #gluster-dev
06:16 shubhendu joined #gluster-dev
06:22 nishanth joined #gluster-dev
06:30 hagarth joined #gluster-dev
06:32 bharata-rao joined #gluster-dev
06:44 kdhananjay joined #gluster-dev
06:53 ppai joined #gluster-dev
06:56 soumya joined #gluster-dev
07:10 hagarth joined #gluster-dev
07:12 kdhananjay joined #gluster-dev
07:29 shubhendu joined #gluster-dev
07:30 nishanth joined #gluster-dev
07:43 rafi1 joined #gluster-dev
07:43 topshare joined #gluster-dev
07:53 ndarshan joined #gluster-dev
07:56 ppai joined #gluster-dev
08:13 hagarth joined #gluster-dev
08:19 ndarshan joined #gluster-dev
08:21 ndevos who wants to do a simple review change? I can offer http://review.gluster.org/9035 and can review a small change in return!
08:21 ndevos offer is only valid until that one got merged :)
08:22 hagarth ndevos: I will take that up if nobody else does it in 4 hours
08:22 hagarth (heading to the GlusterFS meetup soon)
08:22 ndevos hagarth: cool, thanks
08:23 ndevos there is a GlusterFS meetup? was that not sent to a mailinglist?
08:23 hagarth ndevos: this is a glusterfs-india meetup happening in Bangalore
08:24 hagarth http://www.meetup.com/glusterfs-India/
08:24 hagarth ndevos: the meetup is already full .. so we did not have to advertise it more :)
08:24 ndevos hagarth: great, but I would really like to see things like that (and maybe a report afterwards) on the mailinglist(s)
08:24 hagarth report will definitely appear on the ML
08:24 ndevos nice!
08:24 hagarth ndevos: noted
08:25 hagarth feel free to join the meetup group
08:25 hagarth we can feature you as a remote speaker in one of our meetups :)
08:25 ndevos hagarth: I think we can/should show more on what the community is doing, these meetups and all get almost no attention
08:26 * ndevos actually wanted to ask davemc to post a report of the Paris meetup too <- there is the request!
08:27 hagarth ndevos: yes, let us discuss this and possibly have a better process for these things
08:27 * hagarth off to the meetup now
08:27 ndevos hagarth: lead by example! :D
08:36 sharknardo joined #gluster-dev
08:43 vimal joined #gluster-dev
08:44 ppai joined #gluster-dev
08:45 kshlm joined #gluster-dev
08:49 atinmu joined #gluster-dev
08:52 nishanth joined #gluster-dev
08:57 spandit joined #gluster-dev
09:08 atalur joined #gluster-dev
09:09 kshlm joined #gluster-dev
09:09 kshlm joined #gluster-dev
09:34 jiffin joined #gluster-dev
09:37 Gaurav__ joined #gluster-dev
09:37 hagarth joined #gluster-dev
09:37 ppai joined #gluster-dev
09:37 lalatenduM_ joined #gluster-dev
09:37 nkhare joined #gluster-dev
09:41 nishanth joined #gluster-dev
09:41 spandit joined #gluster-dev
09:41 atalur joined #gluster-dev
09:46 atinmu joined #gluster-dev
09:48 kanagaraj joined #gluster-dev
09:53 kdhananjay1 joined #gluster-dev
09:57 pp joined #gluster-dev
10:04 aravindavk joined #gluster-dev
10:13 ppai joined #gluster-dev
10:22 soumya joined #gluster-dev
10:27 kshlm joined #gluster-dev
10:49 kdhananjay joined #gluster-dev
10:50 atinmu joined #gluster-dev
10:51 nkhare_ joined #gluster-dev
10:53 ppai joined #gluster-dev
11:40 lalatenduM joined #gluster-dev
11:41 soumya_ joined #gluster-dev
11:47 ndarshan joined #gluster-dev
12:18 Debloper joined #gluster-dev
12:22 soumya_ joined #gluster-dev
12:27 edward1 joined #gluster-dev
13:05 anoopcs joined #gluster-dev
13:19 sharknardo joined #gluster-dev
13:38 tdasilva joined #gluster-dev
14:13 shyam joined #gluster-dev
14:19 ws2k3 joined #gluster-dev
14:27 topshare joined #gluster-dev
14:49 lalatenduM joined #gluster-dev
14:51 kkeithley lalatenduM: ping. Should we submit something to the CentOS Distro DevRoom at FOSDEM?
14:54 lalatenduM kkeithley, we can actually , will catch u after some time , have a meeting in few mins
15:14 wushudoin joined #gluster-dev
15:17 jobewan joined #gluster-dev
15:21 _Bryan_ joined #gluster-dev
15:23 bala joined #gluster-dev
15:41 hagarth joined #gluster-dev
16:01 shubhendu joined #gluster-dev
16:13 ndevos kkeithley: did you close the 3.4.6 bugs already, or do you want me to do that?
16:26 soumya_ joined #gluster-dev
17:00 hagarth joined #gluster-dev
17:36 lalatenduM joined #gluster-dev
17:46 kkeithley ndevos: probably not.
17:55 hagarth joined #gluster-dev
17:59 lalatenduM kkeithley, should we discuss abt devroom
18:04 lpabon joined #gluster-dev
18:13 kkeithley sure
18:13 kkeithley lalatenduM: ^^^
18:14 lalatenduM kkeithley, a centos related topic or packaging topic seems to be fitting the bill
18:14 lalatenduM in my opinion
18:15 shyam joined #gluster-dev
18:17 kkeithley okay, yes. That seems reasonable.  Did we hit any Hard Problems repackaging for the SIG? Do we know how many downloads there have been from the SIG? Are there any interesting take-aways or lessons learned from setting up the SIG?
18:18 lalatenduM kkeithley, couple of issues, redoing stuffs , like getting pkgs from epel to sig
18:19 lalatenduM maintaining same stuff
18:19 kkeithley what should we have done differently? Did the {RHEL,CentOS}6.6 and 7.0 releases that included 3.6.0.x have been handled better?
18:19 kkeithley please elaborate on getting pkgs from epel to sig?
18:20 lalatenduM e.g. building nfs-ganesha for sig
18:21 lalatenduM cmake28 was not avbl in sig
18:21 lalatenduM for building samba , we had to bring extra 4 pkgs from upstream
18:21 kkeithley because nfs-ganesha in EPEL doesn't have FSAL_GLUSTER, but nfs-ganesha in the SIG does?  Oh, right, cmake28; CentOS isn't quite an exact clone of RHEL
18:22 lalatenduM centos just builds RHEL mainline
18:22 lalatenduM not other channels
18:22 lalatenduM samba is another beast
18:23 kkeithley hmm, so, which things weren't in RHEL mainline? is cmake/cmake28 one of those?
18:23 lalatenduM yes cmake28 was in epel
18:24 kkeithley why is samba a beast? Because of dependencies?
18:24 lalatenduM for Samba, these lib pkgs http://download.gluster.org/pub/gluster/​glusterfs/samba/EPEL.repo/epel-6/x86_64/ are neither in EL mainline or epel
18:24 lalatenduM these are brought from upstream AFAIR
18:25 lalatenduM kkeithley, yes bcz of dependencies
18:25 lalatenduM and SIG supposed to provide upgrade path to all these pkgs always
18:25 kkeithley I see.
18:26 kkeithley All those package dependencies are in some other RHEL channel then, not mainline?
18:27 lalatenduM kkeithley, yes thats right
18:27 lalatenduM for example rhs
18:28 lalatenduM kkeithley, whats the future of a SIG
18:29 lalatenduM can it evolve as a distribution
18:29 lalatenduM ?
18:29 lalatenduM Will have its life cycle
18:30 lalatenduM or replacement of multiple of repos
18:30 davemc ndevos, you really don't want to know how it went
18:30 lalatenduM kkeithley, may be we should go to PM
18:31 kkeithley sure
18:31 kkeithley although I think this is fine, unless you're going to say mean things about someone. ;-)
18:32 kkeithley I think ndevos has left for the day.
18:32 lalatenduM kkeithley, no prob we can continue then here
18:32 lalatenduM :)
18:35 lalatenduM kkeithley, but in reality I hope SIG dont bring lots of fragmentation
18:36 kkeithley yup. which brings me back to one of my questions, which is how many downloads have there been?
18:36 kkeithley do we know?
18:36 kkeithley Otherwise how do we measure the success of the SIG
18:36 lalatenduM kkeithley, we have not announced it actually
18:36 lalatenduM as we are waiting for some infrastructure from the project side
18:37 lalatenduM how ever the testing pkgs are ready
18:37 kkeithley that's fine, but from an internal perspective, do we know if it's a success because there have been x00,000 downloads, o
18:37 lalatenduM i can announce that
18:37 _Bryan_ joined #gluster-dev
18:37 kkeithley So it's too early to say
18:38 lalatenduM kkeithley, i think noone knows that we have built pkgs
18:38 lalatenduM yeah
18:41 kkeithley Besides Gluster, nfs-ganesha+gfapi, and samba+gfapi, what else are we shipping in the Storage SIG, anything?  At least on other SIG has oVirt with gfapi; correct?
18:42 lalatenduM kkeithley, we have openafs
18:42 lalatenduM ovirt is not packaged till now
18:43 lalatenduM but we can do that
18:43 lalatenduM with negios too
18:43 kkeithley we're doing oVirt in the Storage SIG?
18:43 kkeithley openafs, that's cool.
18:44 kkeithley who is doing what with Nagios?
18:44 lalatenduM kkeithley, we have not decided abt ovirt , but why not
18:44 lalatenduM same with Nagios
18:44 kkeithley Well, your time is finite, right?
18:45 lalatenduM kkeithley, yes :)
18:46 lalatenduM kkeithley, hchiramm_ is joining the effort :)
18:47 kkeithley oh, that's good
18:47 lalatenduM kkeithley, hopefully the initial stuff will inspire others to make it better
18:47 kkeithley that's always the hope, yes. ;-)
18:48 JustinClift lalatenduM: Has KB been in contact with you, to help the SCST guys get onboard with the SIG?
18:49 lalatenduM JustinClift, yes
18:49 JustinClift lalatenduM: Cool :)
18:49 JustinClift lalatenduM: Do you have time/inclination for it?
18:49 lalatenduM we were discussing that
18:49 * kkeithley wonders what SCST means
18:50 JustinClift kkeithley: http://scst.sourceforge.net/
18:50 lalatenduM kkeithley, SCSI Target Subsystem for Linux
18:50 JustinClift kkeithley: Alternate SCSI stack fo the Linux kernel
18:50 JustinClift Very high performance, etc
18:50 kkeithley ooooh
18:50 JustinClift (most of the storage vendors with appliances based on Linux use it)
18:51 * JustinClift thinks it'd be a good fit for the Storage SIG
18:51 JustinClift :)
18:52 JustinClift ... it also means the storage vendors would (pretty likely) be able to then base their stuff on the Storage SIG version of CentOS, instead of having to roll their own *nix distribution, just to get SCST
18:52 lalatenduM JustinClift, we are starting irc meetings for storage sig
18:52 lalatenduM JustinClift, right
18:53 lalatenduM JustinClift, the challenge is with Koji, i.e. how to build kernel everytime with these things e.g. openafs kmod when base distribution a new version
18:54 JustinClift Hmmm, use magic. :D
18:54 lalatenduM Thats something , we have to fix , so kernel can be rebuild for that openafs and SCST automatically
18:55 lalatenduM kkeithley, thats means SIG will provide a diff kernel
18:55 lalatenduM JustinClift, yeah magic fascinates me :)
18:56 JustinClift lalatenduM: :D It's not a blocker for getting some initial releases ready for people to try out is it?
18:56 lalatenduM JustinClift, yeah
18:57 JustinClift lalatenduM: Put it this way... there are storage companies around who already have staff for doing their custom *nix releases.  Those people may be _very_ interested in putting time into making this work, as it could potentially reduce their workload
18:58 JustinClift Given a few manually built releases to prove the concept, they could pick up the effort
18:58 lalatenduM JustinClift, yes , my previous company used to build Linux kernel with SCST for a internal product
18:58 JustinClift Just thinking out load tho ;)
18:58 JustinClift loud even
18:58 JustinClift lalatenduM: Ahhh cool.  I didn't know that. :D
18:58 lalatenduM they used to emulate LUNs
18:59 JustinClift Yeah, I had a dev/test setup a few years ago doing that (emulating LUNS) and exporting them with SRP
18:59 JustinClift Worked very well.  Fast. :D
18:59 lalatenduM JustinClift, cool
18:59 JustinClift Anyway, back to work... :)
19:00 lalatenduM JustinClift, :)
19:00 lalatenduM kkeithley, should we brainstorm more on this, or we should move to another topic I have in mind
19:04 kkeithley next topic?
19:04 kkeithley unless you have more on this one?
19:06 lalatenduM kkeithley, nope i dont have anything more
19:06 lalatenduM On for next topic
19:07 lalatenduM how can we make packaging same for all Linux/GNU distributions?
19:07 lalatenduM I think it is extremely difficult thing to do
19:08 lalatenduM but what are the difficulties
19:08 kkeithley RPM packaging? Or are you trying to include DPKG too?
19:08 lalatenduM yes
19:08 lalatenduM all included
19:10 kkeithley That is a big big challenge
19:11 lalatenduM kkeithley, may be big+1000 :)
19:11 kkeithley Isn't there some python (magic) class, and associated subclasses, that will create packages for the
19:12 lalatenduM kkeithley, dont know
19:12 kkeithley packages for the host system
19:15 lalatenduM kkeithley, are you talking abt  pip or easy_install
19:15 kkeithley Might be pip
19:15 lalatenduM hmm
19:18 kkeithley but...
19:18 * kkeithley wonders how this is interesting to CentOS and the Storage SIG. Please explain?
19:19 lalatenduM kkeithley, nope it is not related to CentOS and the Storage SIG
19:20 lalatenduM but a generic topic
19:20 kkeithley oh, okay
19:20 kkeithley Are we just talking about a general problem you'd like to solve, or a topic for a paper?
19:23 JustinClift lalatenduM: Just be careful with doing this.  If you make a meta approach to packaging for *nix distro's, this happens: https://xkcd.com/927/
19:24 lalatenduM JustinClift, yes haha
19:25 lalatenduM kkeithley, may be BoF, i just wanted to bring the reasons why it can't be done to notice of all distributions
19:26 lalatenduM and hopefully multiple minds can come up with some idea
19:27 lalatenduM if we can put the idea in to others subconscious , I mean inception , that would be good :) lol
19:27 * kkeithley thinks they already know.....  and don't care
19:27 lalatenduM kkeithley, hmm
19:28 kkeithley but hey, you never know
19:29 lalatenduM kkeithley, might become a deja vu moment for others :)
19:29 lalatenduM lol
19:29 lalatenduM all Hollywood movie names :)
19:29 kkeithley sorry, I don't mean to rain on your idea
19:29 lalatenduM kkeithley, hey it is fine
19:30 JustinClift lalatenduM: A lot of people have put time into solving this problem.
19:30 lalatenduM kkeithley, I know it is very large problem to solve
19:30 kkeithley The best way to get someone to do something it to make them feel like it was their idea in the first place.
19:30 JustinClift lalatenduM: It seems to be highly political, more than technical
19:30 lalatenduM JustinClift, yeah I expect so
19:30 JustinClift lalatenduM: If you want to spend several years on this, then go ahead
19:30 kkeithley yeah, lots of NIH too
19:30 * JustinClift suggests you put the time into something else
19:30 JustinClift Just saying. ;)
19:31 lalatenduM JustinClift, yeah i understand
19:34 lalatenduM kkeithley, right
19:41 vipulnayyar joined #gluster-dev
22:57 badone joined #gluster-dev

| Channels | #gluster-dev index | Today | | Search | Google Search | Plain-Text | summary