Also, this is the profile output of this Volume:
gluster> volume profile mariadb_gluster_volume info cumulative
Brick: laeft-dccdb01p.core.epay.us.loc:/export/mariadb_backup/brick
-------------------------------------------------------------------
Cumulative Stats:
Block Size: 16384b+ 32768b+ 65536b+
No. of Reads: 0 0 0
No. of Writes: 83391 465750 102911
Block Size: 131072b+
No. of Reads: 33
No. of Writes: 8551
%-latency Avg-latency Min-Latency Max-Latency No. of calls Fop
--------- ----------- ----------- ----------- ------------ ----
0.00 0.00 us 0.00 us 0.00 us 26 FORGET
0.00 0.00 us 0.00 us 0.00 us 33 RELEASE
0.00 0.00 us 0.00 us 0.00 us 92 RELEASEDIR
0.00 102.25 us 70.00 us 155.00 us 4 SETXATTR
0.00 159.80 us 97.00 us 207.00 us 5 RMDIR
0.00 266.75 us 121.00 us 675.00 us 4 SETATTR
0.00 100.62 us 29.00 us 700.00 us 16 INODELK
0.00 84.33 us 29.00 us 477.00 us 33 FLUSH
0.00 68.16 us 34.00 us 165.00 us 92 OPENDIR
0.01 88.35 us 21.00 us 608.00 us 92 STAT
0.01 754.04 us 262.00 us 10104.00 us 25 CREATE
0.02 169.04 us 27.00 us 997.00 us 179 READDIRP
0.04 150.90 us 43.00 us 1867.00 us 365 LOOKUP
0.04 16330.75 us 297.00 us 46360.00 us 4 MKDIR
0.18 7896.70 us 71.00 us 256814.00 us 33 READ
0.68 466.93 us 19.00 us 1848.00 us 2119 STATFS
2.39 151339.17 us 227.00 us 540998.00 us 23 UNLINK
5.49 320155.28 us 2035.00 us 1273394.00 us 25 FSYNC
31.28 69.00 us 30.00 us 11447.00 us 660603 WRITE
59.84 10899586.88 us 5827342.00 us 13921169.00 us 8 OPEN
Duration: 71425 seconds
Data Read: 4325376 bytes
Data Written: 29195534336 bytes
As you can see OPEN Fop take the most amount of time and they normally timeout, even with default value for networking.frame-timeout
Thanks
On Tue, Jun 13, 2017 at 3:43 PM, Julio Guevara <julioguevara150@xxxxxxxxx> wrote:
I'm having a hard time trying to get a gluster volume up and running. I have setup other gluster volumes on other systems without much problems but this one is killing me.The gluster vol was created with the command:gluster volume create mariadb_gluster_volume laeft-dccdb01p:/export/mariadb/brick
I had to lower frame-timeout since the system would become unresponsive until the frame failed by timeout:gluster volume set mariadb_gluster_volume networking.frame-timeout 5running gluster version: glusterfs 3.8.12The workload i'm using is: sysbench --test=fileio --file-total-size=4G --file-num=64 preparesysbench version: sysbench 0.4.12-5.el6kernel version: 2.6.32-696.1.1.el6centos: 6.8Issue: Whenever I run the sysbench over the mount /var/lib/mysql_backups I get the error that is shown on the log output.It is a constant issue, I can reproduce it when I start increasing the --file-num for sysbench above 3.
_______________________________________________ Gluster-users mailing list Gluster-users@xxxxxxxxxxx http://lists.gluster.org/mailman/listinfo/gluster-users
