Strange behavior

My Squid (1.1.11 on Linux 2.0.27) suddenly tied up all CPU time and did
not respond to any request or signal. The last cache log message was
"rotate_logs: SIGUSR1" received, precisely at the time when it ceased
service. There was definitely no such signal sent. When restarting I saw
about 5.500 swapfile clashes in contrast to just a few I get on regular
restarts twice a week.

I supposedly had this once before with an older version, but then I was
on vacation and had no chance to look at the logs, when my staff restated
the entire machine.

Any advice ?

Regards, Gero
Received on Sat Jun 21 1997 - 12:31:11 MDT