Tag Info

New answers tagged

0

It appears that the kernel processes may have been stealing CPU time during transfers to/from swap. The server's cache settings had somehow been reset without my knowledge, setting swappiness to 60. From the output of "sar -W", the hang-ups seemed to coincide with high load periods during which pswpin/s and pswpout/s were large (greater than 2.00 or so, ...


1

I've tracked issues with migration kernel process reported here. It seems that linux kernel prior to 3.6.11 are affected. The link show similar symptom where migration process take large CPU time. If possible, you might want to upgrade the kernel to see if the problem persists.


0

migration is the kernel process that handles moving processes from one CPU to another. So, for some reason your Linux scheduler decides that processes need moving to another CPU, and the migration process eats the CPU time. You could try pinning processes to specific CPUs, or try different schedulers with your kernel. Maybe some other scheduler isn't so ...



Top 50 recent answers are included