Perl 6 - the future is here, just unevenly distributed

IRC log for #gluster-dev, 2014-03-06

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

All times shown according to UTC.

Time Nick Message
03:05 bharata-rao joined #gluster-dev
03:35 raghu joined #gluster-dev
03:57 itisravi joined #gluster-dev
04:15 shubhendu joined #gluster-dev
04:20 ndarshan joined #gluster-dev
04:26 hagarth joined #gluster-dev
04:39 ppai joined #gluster-dev
04:41 Yuan_ joined #gluster-dev
04:57 bala joined #gluster-dev
05:03 kdhananjay joined #gluster-dev
05:12 deepakcs joined #gluster-dev
05:23 kanagaraj joined #gluster-dev
05:28 mohankumar__ joined #gluster-dev
05:35 aravindavk joined #gluster-dev
06:19 JoeJulian jclift: yes. /etc/rc.d/init.d/glusterfsd should be getting installed.
06:21 JoeJulian jclift: Without it (and the patch for bug 1073217) when the server reboots, the clients will experience ping-timeout because the network will stop without the bricks getting killed. No TCP FIN.
06:21 glusterbot Bug https://bugzilla.redhat.com:​443/show_bug.cgi?id=1073217 medium, unspecified, ---, vraman, NEW , /var/lock/subsys/glusterfsd missing
06:35 glusterbot joined #gluster-dev
06:59 kdhananjay joined #gluster-dev
07:05 ndarshan joined #gluster-dev
07:14 kshlm joined #gluster-dev
07:20 lalatenduM joined #gluster-dev
07:56 kdhananjay joined #gluster-dev
08:14 an_ joined #gluster-dev
08:53 Frankl joined #gluster-dev
08:58 kdhananjay joined #gluster-dev
09:01 an_ joined #gluster-dev
09:12 an__ joined #gluster-dev
09:14 shubhendu joined #gluster-dev
09:18 an_ joined #gluster-dev
09:20 bharata-rao joined #gluster-dev
09:22 an_ joined #gluster-dev
09:26 an__ joined #gluster-dev
09:37 shubhendu joined #gluster-dev
09:38 kanagaraj joined #gluster-dev
09:46 Frankl joined #gluster-dev
09:53 ndarshan joined #gluster-dev
10:05 an_ joined #gluster-dev
10:07 lalatenduM joined #gluster-dev
10:10 Frankl joined #gluster-dev
10:45 Frankl joined #gluster-dev
10:53 ndarshan joined #gluster-dev
10:54 kanagaraj joined #gluster-dev
11:42 edward1 joined #gluster-dev
11:44 hagarth joined #gluster-dev
12:00 kkeithley1 joined #gluster-dev
12:00 kkeithley1 left #gluster-dev
12:01 kkeithley1 joined #gluster-dev
12:03 lpabon joined #gluster-dev
12:12 ppai joined #gluster-dev
12:14 social oh hay, I'd love to ask, did anyone consider stuff like lxc in tests for jenkins? It seems that lately I hit more and more bugs that cannot be simulated on one node setup of gluster and I started to use lxc there
12:23 mohankumar__ joined #gluster-dev
12:28 an_ joined #gluster-dev
12:41 itisravi joined #gluster-dev
12:57 an_ joined #gluster-dev
12:59 purpleidea joined #gluster-dev
13:00 mohankumar__ joined #gluster-dev
13:14 an_ joined #gluster-dev
13:29 an_ joined #gluster-dev
13:34 kkeithley_ jclift: not so fast on that gNFS slowdown fix backport to 3.4.  The source is very different from 3.4 to 3.5-and-master.
14:01 badone joined #gluster-dev
14:20 jclift kkeithley_: No worries.  Should we let the guy know?
14:20 jclift kkeithley_: Hmmm, when you say 3.5-and-master, is the 3.5 source similar/same to master, so it'll prob be ok for 3.5.x
14:20 jclift ?
14:25 kkeithley_ yes, 3.5 is close
14:26 kkeithley_ http://review.gluster.org/#/c/7200/
14:30 jclift Cool. :)
14:32 kkeithley_ I'll look at 3.4 some more but not promising anything
14:42 jobewan joined #gluster-dev
14:54 aravindavk joined #gluster-dev
14:55 jclift No worries.  If it's a real pain, we'll let him know the patches will be in 3.5 so he'll need to wait for that.  At least we looked into it properly, etc. :)
15:15 jobewan joined #gluster-dev
15:45 kkeithley_ jclift: yeah, there's new  throttling functionality in the rpc transport that's not in 3.4.x. That's major work, not a bugfix. Too much for  x.y.Z release.
16:04 jclift kkeithley_: No worries.  I'll email him back on list to say we've looked into it and it's not feasible for 3.4.x.
16:04 jclift Point him at 3.5 which is in beta atm, etc.
16:05 jclift kkeithley_: Thanks for looking into this. :)
16:08 jclift ndevos: How many people do you reckon attended the Gluster Event the other day in Amsterdam?
16:17 ndevos jclift: 58 showed up
16:19 hagarth joined #gluster-dev
16:19 aravindavk joined #gluster-dev
16:21 jclift ndevos: Cool. :)
16:23 kkeithley_ from the pictures johnmark posted it looked like a good turnout. 58 is awesome.
16:25 lpabon joined #gluster-dev
16:44 ndevos http://review.gluster.org/7202 contains a fix for users/processes with > 93 groups (nfs or gfapi only, not fuse)
16:44 ndevos in case anyone is interested in reviewing this rpc fix/workaround, please do!
17:08 JoeJulian jclift: Did you see my comments regarding the glusterfsd init file?
17:15 johnmark kkeithley: it was an awesome turnout
17:15 johnmark the dutch team gets high marks for making that event awesome
17:21 JoeJulian Hmm, now I have to remember how to update my gerrit submissions.
17:27 ndk joined #gluster-dev
17:38 jclift JoeJulian: No?  I can see review comments from ndevos, but not from you.  You are meaning for http://review.gluster.org/#/c/7192/ aren't you?
17:40 jclift ndevos: With the /etc/init.d/glusterfsd file, I wasn't sure if we want it or not, as when I was looking through a related BZ there were people strongly arguing about having vs not having it present.
17:41 jclift There didn't seem to be a clear result there.
17:41 ndevos jclift: we want it, but it should be possible to disable restarts of glusterfsd (sysconfig file) when doing an update
17:42 jclift Cool
17:42 * ndevos is in the train, connections will die soon
17:42 jclift np
17:42 jclift I'm take a look at it properly after teh glupy stuff.
17:43 jclift I'm having trouble getting a full run of the regression test passing with normal upstream git master, let alone adding Glupy stuff.
17:43 jclift Still working through potential causes, etc. ;)
17:54 badone joined #gluster-dev
17:59 lpabon joined #gluster-dev
18:03 ndevos kkeithley: whats your take on testcases that create new users/groups, should they delete them afterwards?
18:06 ndevos jclift: when you file bugs for your patches, feel free to assign those bugs to yourself, like bug 1073168 ;)
18:06 glusterbot Bug https://bugzilla.redhat.com:​443/show_bug.cgi?id=1073168 low, unspecified, ---, kparthas, NEW , The Gluster Test Framework could use some initial sanity checks
18:09 jclift Good point. :)
18:10 ndevos jclift: and, if it is not needed, please do not rebase patches between submitting, if there is no rebase, gerrit can show the delta between patch versions, with a rebase there is a lot of noise
18:15 primechuck joined #gluster-dev
18:20 JoeJulian jclift: I wasn't aware of a gerrit patch, I was just referring to in-channel comments.
18:21 JoeJulian [22:19] <JoeJulian> jclift: yes. /etc/rc.d/init.d/glusterfsd should be getting installed.
18:21 JoeJulian [22:21] <JoeJulian> jclift: Without it (and the patch for bug 1073217) when the server reboots, the clients will experience ping-timeout because the network will stop without the bricks getting killed. No TCP FIN.
18:21 glusterbot Bug https://bugzilla.redhat.com:​443/show_bug.cgi?id=1073217 medium, unspecified, ---, vraman, NEW , /var/lock/subsys/glusterfsd missing
18:26 danikov joined #gluster-dev
18:40 jclift JoeJulian: Cool, that makes sense. :)
18:48 JoeJulian jclift, kkeithley : Would you like to take a look at http://review.gluster.org/#/q/status:op​en+project:glusterfs+topic:bug-1073217,n,z
18:49 JoeJulian ndevos: oops, I just did that too.
18:50 JoeJulian Though it doesn't seem to have broken anything that I can see.
18:51 * JoeJulian shrugs
19:15 badone joined #gluster-dev
19:50 vilenchic joined #gluster-dev
20:06 ndevos JoeJulian: when I review a change for the 1st time, and have some points that I'd like to see changed, I can check the changes by selecting the "Old version history" in the patch to filter any unchanged chunks
20:08 ndevos if there is a rebase, the diff between the 1st patch and the 2nd submission is cluttered with unrelated changes (the rebase changes), and that makes it more timeconsuming to review the change
20:13 JoeJulian ndevos: Ah, I see that now. I'll try to avoid that.
20:14 ndevos JoeJulian: it's not a big deal, but sometimes it's annoying if the patch is big an touches multiple files
20:14 JoeJulian Mine are not likely to ever be big. :D
20:15 ndevos :)
20:20 kanagaraj joined #gluster-dev
21:22 badone joined #gluster-dev

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