Camelia, the Perl 6 bug

IRC log for #gluster-dev, 2013-03-29

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

All times shown according to UTC.

Time Nick Message
00:22 foster joined #gluster-dev
00:57 yinyin joined #gluster-dev
01:10 jules_ joined #gluster-dev
01:12 rastar joined #gluster-dev
01:27 yinyin joined #gluster-dev
01:39 portante joined #gluster-dev
01:41 a2 portante, ping?
01:42 a2 portante, in your ufo tests on XFS, was the entire file getting written in a single write(), or in pieces?
01:42 yinyin joined #gluster-dev
01:43 portante a2: hello
01:43 portante depends on the file
01:44 portante you can log into gprfs013.sbu.lab.eng.bos.redhat.com and strace the object servers to verify
01:44 a2 for the fragmentation test what were the file size?
01:44 portante but the buffers are 2.6MB, so most files are one write system call
01:44 yinyin joined #gluster-dev
01:44 portante all over the place
01:44 a2 gigabytes?
01:45 a2 ok.. if most of the files could be fit in a single write() call, that mostly answers my question - whether fragmentation is still in issue if the files could be completely written in a single write syscall
01:46 portante check out the bz I filed, it has the .zip file that lists all the files names and their sizes
01:47 a2 i somewhat recollected dchinner saying that if a file could be completely written in a single write() then preallocation would not have kicked in just yet (and hence you shouldn't have the freespace fragmentation) looks like that might not be the case
01:48 portante there is something else going on
01:49 portante we need to get control of the behavior before we can make progress
01:49 a2 yeah..
01:49 a2 i'm headed home, be back online later
01:49 a2 ttyl
01:49 portante ttyl
02:03 bala1 joined #gluster-dev
02:15 yinyin joined #gluster-dev
02:53 bharata joined #gluster-dev
03:02 yinyin joined #gluster-dev
04:03 sripathi joined #gluster-dev
04:07 yinyin joined #gluster-dev
04:08 rastar joined #gluster-dev
04:21 pai joined #gluster-dev
04:27 bala1 joined #gluster-dev
04:33 pai joined #gluster-dev
04:46 rastar joined #gluster-dev
05:08 hagarth joined #gluster-dev
05:11 yinyin joined #gluster-dev
05:15 lalatenduM joined #gluster-dev
05:19 pai joined #gluster-dev
05:21 mohankumar joined #gluster-dev
05:35 yinyin joined #gluster-dev
05:36 aravindavk joined #gluster-dev
06:04 hagarth joined #gluster-dev
06:17 vshankar joined #gluster-dev
07:25 hagarth joined #gluster-dev
08:16 sripathi joined #gluster-dev
09:14 yinyin joined #gluster-dev
11:33 badone joined #gluster-dev
13:09 wushudoin joined #gluster-dev
13:20 yinyin_ joined #gluster-dev
14:18 rastar joined #gluster-dev
14:29 awheeler_ Anybody want to take a quick look at my patch for bug 924792 before I submit it for review via rfc.sh:   https://bugzilla.redhat.com/attach​ment.cgi?id=718091&action=diff
14:29 glusterbot Bug http://goo.gl/Smv7Z medium, unspecified, ---, junaid, ASSIGNED , Gluster-swift does not allow operations on multiple volumes concurrently.
14:49 awheeler_ Updated the patch: https://bugzilla.redhat.com/attach​ment.cgi?id=718100&action=diff
15:08 awheeler_ Submitted
15:14 awheeler_ I'm looking at the output from ufounit, the ring unit tests don't apear to be getting executed: http://build.gluster.org/job/ufounit/1034/console
15:16 awheeler_ It looks very different from the output when I call unittest.sh -- 130 tests for me vs 104 for ufounit.
15:19 awheeler_ It appears to be checking out master, rather than the bug branch I created.
16:25 hagarth joined #gluster-dev
16:31 hagarth joined #gluster-dev
18:27 awheeler_ joined #gluster-dev
19:57 __Bryan__ joined #gluster-dev

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