To see if we have disk contention problems we’d take a look at an iostat. if asvc_t >50 for sustained periods, the system
has disk contention problems.
[root@ehrpitlin001 ~]# iostat -x 3 10000 | awk '$11>=50' Device: rrqm/s wrqm/s r/s w/s rsec/s wsec/s avgrq-sz avgqu-sz await svctm %util
we can also check if the %wio cpu is > 5% for more than a minute constantly
[root@ehrpitlin001 ~]# vmstat 1 3000 | awk '$16>=5' r b swpd free buff cache si so bi bo in cs us sy id wa st r b swpd free buff cache si so bi bo in cs us sy id wa st r b swpd free buff cache si so bi bo in cs us sy id wa st r b swpd free buff cache si so bi bo in cs us sy id wa st
if we have SAR setup we could check it
[root@ehrpitlin001 ~]# sar | awk '$7>=3{print}' | head -5
12:10:01 AM all 4.93 0.00 2.00 11.84 0.00 81.23
12:20:01 AM all 22.83 0.00 4.96 14.90 0.00 57.31
12:30:01 AM all 37.14 0.00 3.83 11.84 0.00 47.19
12:40:01 AM all 44.26 0.00 3.88 12.71 0.00 39.16
12:50:01 AM all 48.83 0.03 3.57 9.25 0.00 38.32