I've been having some freeze problems with my "managed" freebsd server
that my host has been less than helpful with; I hope that this is the
right place to ask the questions I have.

os: freebsd 4.8-stable
major applications: apache 1.3.29 + php 4.3.10 , mysql 4.1.18-log,
dirvish, riff-backup

  Machine freezes with nothing written to the logs or console - if you
happen to be logged in and running top when it "starts" to freeze your
top session will run completely normally and without lag (spacebar
refreshes display, you can resort on size or cpu, etc), but no other
processes can start - typing a command into another open shell will not
start that program.  Until it fully freezes it will echo characters back
in the shell - and top will continue to run as normal.  Top always shows
a load of <0.1, there's always 5MB to 50MB of ram free.

  All of the hardware has been replaced (motherboard, cpu, ram, power
supply, hard drive)

  I can't make it freeze on demand by replaying the web hits or database
queries that occurred before the crash.

  I am able to make it freeze on demand by slurping down a particular
dirvish vault with rsync.  The freeze symptoms are the same as the
random freeze symptoms (top responds normally, new processes can't start)

  The random freezes occur whether or not I'm running dirvish on a schedule.

  The rsync freezing I can work around if needed, the random freezes I
cannot.  Does anybody have any suggestions on how I might track down the


freebsd-questions@freebsd.org mailing list
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Reply via email to