On 05/21/04 at 12:31, 'twas brillig and Laura Ramirez scrobe:
> Hello Ole,
> from the fact that the loadavg is increasing, i would assume
> some process is spinning, on some lock probably...the problem
> is to find the culprit. A bta at the kdb prompt gives a back trace
> of all the processes. However, you probably want to start looking
> at the back trace of a hung process. It would indicate what
> lock its wants. Then depending on the type of lock, (some locks
> indicate who the current user is) we can find out who has that
> lock and why it hasnt released it.
Thanks, Laura. Unfortunately, I had to reboot it; one of the
grad students was desperate for cycles in order to make a deadline
early next week. If/when it happens again I'll try to find a suitable
process to backtrace.
NB: the load average went up to 48.00 but no further; it
stayed there for a few hours before I hit the Big Red Switch.
Ole Craig * UNIX, linux, SMTP-ninja; news, web; SGI martyr * CS Computing
Facility, UMass * <www.cs.umass.edu/~olc/pgppubkey.txt> for public key
Where are the missing deficit-reduction program-related activities?