HI, On one of the server we are noticing high %sy usage due to which server is very slow. We have configured one of file system with xfs which is used to store mail queue. Below is the iostat output on one of the instance when there is high %sy usage avg-cpu: %user %nice %system %iowait %steal %idle 0.01 0.00 48.26 0.01 0.00 51.73
Device: rrqm/s wrqm/s r/s w/s rkB/s wkB/s avgrq-sz avgqu-sz await svctm %util
cciss/c0d0 0.00 16.99 0.00 8.36 0.00 414.90 99.30 0.77 92.57 46.30 38.69
cciss/c0d0p1 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
cciss/c0d0p2 0.00 0.00 0.00 6.13 0.00 338.02 110.32 0.77 126.18 63.09 38.66
cciss/c0d0p3 0.00 0.28 0.00 0.84 0.00 4.46 10.67 0.00 0.33 0.33 0.03
cciss/c0d0p4 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
cciss/c0d0p5 0.00 0.56 0.00 0.28 0.00 3.34 24.00 0.00 0.00 0.00 0.00
cciss/c0d0p6 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
cciss/c0d0p7 0.00 11.42 0.00 0.56 0.00 47.91 172.00 0.00 0.00 0.00 0.00
cciss/c0d0p8 0.00 4.74 0.00 0.56 0.00 21.17 76.00 0.00 0.00 0.00 0.00
~]# xfs_info /dev/cciss/c0d0p2
meta-data="" isize=256 agcount=16, agsize=163787 blks
= sectsz=512 attr=1
data = "" blocks=2620592, imaxpct=25
= sunit=0 swidth=0 blks, unwritten=1
naming =version 2 bsize=4096
log =internal bsize=4096 blocks=2560, version=1
= sectsz=512 sunit=0 blks, lazy-count=0
realtime =none extsz=4096 blocks=0, rtextents=0
Are we facing any performance issue with xfs. Please let me know -- |
_______________________________________________ xfs mailing list xfs@xxxxxxxxxxx http://oss.sgi.com/mailman/listinfo/xfs