Sabuj, please try with a lower aggregate-size in write-behind. avati 2008/7/21 Sabuj Pattanayek <sabujp@xxxxxxxxx>: > Hi > > On Sun, Jul 20, 2008 at 9:56 PM, Anand Babu Periasamy <ab@xxxxxxxxxx> > wrote: > > If you got some spare time, can you also benchmark this version.. > > > http://ftp.zresearch.com/pub/gluster/glusterfs/1.4-qa/glusterfs-1.4.0qa29.tar.gz > > It's busted. Tried to do: > > df -h > cd /export/glfs/tmp > ls -l > time dd if=/dev/zero of=testFile bs=4k count=20000 > dd: writing `testFile': Software caused connection abort > dd: closing output file `testFile': Transport endpoint is not connected > 0.001u 0.003s 0:00.12 0.0% 0+0k 0+0io 0pf+0w > > It immediately causes an EOF from peer error and then socket read > failed (Transport endpoint is not connected) in state 1 and a crash in > the glusterfs client, You can find the client and server logs in > http://structbio.vanderbilt.edu/~pattans/glfs/<http://structbio.vanderbilt.edu/%7Epattans/glfs/>. > > HTH, > Sabuj Pattanayek > > > _______________________________________________ > Gluster-devel mailing list > Gluster-devel@xxxxxxxxxx > http://lists.nongnu.org/mailman/listinfo/gluster-devel > -- If I traveled to the end of the rainbow As Dame Fortune did intend, Murphy would be there to tell me The pot's at the other end.
