Acceptable load average on SPLAT/RHEL

[kaldek]

Hi ppls,

For those of you that may ever be curious about it, you may notice that the load average of a SPLAT or RHEL Linux box never goes below 1.00. Some of you may be concerned by this, however the following information comes direct from Check Point:

Background
The load average is the average number of processes that are in state of running, in the run queue or waiting for I/O (e.g. reading packets from the network). On Linux kernel that includes kernel threads a.k.a tasks.

What
If you run ps or top on a Check Point gateway the output will include:

root       566  0.0  0.0     0    0 ?        DW   Mar28   0:00 [fw1-wt]

Why?
This is a FireWall-1 worker thread kernel thread constantly waiting on I/O from network, hence a load average output of 1.0 on idle a Check Point gateway is indeed normal.


There ya go - nothing to worry about.