How to determine the process to swapping-out

From: Simon (simon.zcyuan@gmail.com)
Date: Mon Sep 17 2007 - 02:19:50 EDT


Gurus,

Our solaris 9 is runing into trouble,that we found the "w" column in
'vmstat' output always kept very high,and reach to 70,sometimes up to
78,as:

# vmstat 2
 kthr memory page disk faults cpu
 r b w swap free re mf pi po fr de sr s0 s1 s2 s3 in sy cs us sy id
 0 0 70 63627184 31472752 326 2238 0 56 56 0 0 12 0 0 12 16323 76495
20587 18 15 66
 0 0 70 63627104 31472968 1015 6954 0 171 171 0 0 21 0 0 21 16351
84333 20565 16 17 67
 0 2 70 63627176 31473432 832 5714 0 143 143 0 0 18 0 0 18 15583 77438
19683 16 17 67

# sar -q 1 2
14:13:29 runq-sz %runocc swpq-sz %swpocc
14:13:30 0.0 0 70.0 94
14:13:31 0.0 0 70.0 95

Generally,higher "w" indicates the physical memory shortage,but it's
not the case in fact,we have 48G memory populated,and abundant free
memory:

# netstat -k system_pages |grep pagesfree
pp_kernel 344545 pagesfree 3933796 pageslocked 1192472 pagestotal 6038433

I suspected it may be caused by an unkown process,it locked something
that should have freed,but how to know the procee ?

Any helps are appciated.

Thanks.
Simon

Average 0.0 0 70.0 95
_______________________________________________
sunmanagers mailing list
sunmanagers@sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



This archive was generated by hypermail 2.1.7 : Wed Apr 09 2008 - 23:42:19 EDT