Hi,
I have around 100 servers running same java applications as clusters.
BUt only on one server when i check for page in and page out values its very
high comparatively. The swap memory used is low though. Why does it happen
so?
=> free
total used free shared buffers cached
Mem: 33276632 26206184 7070448 0 252748 22345152
-/+ buffers/cache: 3608284 29668348
Swap: 33551744 0 33551744
=> sar -B
06:20:01 AM pgpgin/s pgpgout/s fault/s majflt/s
06:30:01 AM 2.72 704.72 3025.80 0.00
06:40:01 AM 2.75 758.59 3010.29 0.00
06:50:01 AM 0.05 700.88 2420.78 0.00
07:00:01 AM 0.07 672.92 2420.77 0.00
07:10:01 AM 34800.40 1111.32 16698.30 0.20
07:20:01 AM 92274.63 825.56 4488.70 0.09
07:30:01 AM 36624.02 900.63 3094.32 0.05
07:40:01 AM 48.38 903.60 3359.76 0.07
07:50:01 AM 2296.02 860.05 3691.91 0.04
08:00:01 AM 22.74 715.82 14809.04 0.00
08:10:01 AM 18361.58 1174.03 6149.49 0.05
Average: 3806.72 1468.74 15781.81 0.02
On all the other servers the swao usage is same and pgpgouts never cross
200. What is the reason for such huge page in and page out?
Regards
Jibs
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Linux Users Group.
To post a message, send email to [email protected]
To unsubscribe, send email to [email protected]
For more options, visit our group at
http://groups.google.com/group/linuxusersgroup
-~----------~----~----~----~------~----~------~--~---
