I've not used the later code revisions yet, but my experience is that
this is normal. I'd have a mostly idle F5 unit with a load of 2. As long
as performance is good I wouldn't worry too much about it. I remember
talking to a few F5 engineers about this. Something about how the kernel
reports load isn't right in their modified kernel.
When the load gets to 6.0 or higher, however, I've seen it do some strange
things. This only happens though when the unit is pushed to it's limits,
such many thousands of connections per second.
31 Aug 2001, Chris Kirby wrote:
> We have a pair of clustered F5 BIGIP HA+ boxes. Just looked at
> the load average today, first time in a long while, and shows
> the following:
> up 80 days, 2:49, 1 user, load averages: 3.64, 3.49, 3.44
> Seems a bit high I thought and our LB's are not very heavily
> used. Wanted to find out if this was normal.
-- -------------- -- ---- ---- --- - - - - - -- - - - - - - Tony Bourke tonyIZZATvegan.net
This archive was generated by hypermail 2b30 : Fri Aug 31 2001 - 16:30:03 EDT