I experienced zombies on a system that had too many requests (httpd, 
mysqld, and then analog, i.e the disaster happened at 4 am) for too 
little memory (the system swapped itself to death).
To understand what was going on I ran a cron job at every 15 min that
logged:
a) date (for time stamp)
b) w    (for cpu load (1,5,15 min average)
c) pstree -l -n (for compact view)
d) ps -Afl      (for the details)
(e) top (?forgot switch for 1 event?) - but top from non-konsole jobs 
(cron and at) didn't work on older distros(couldn't get default terminal
info for non-existing terminal).

top also shows zombies in the header.

I noticed the zombies are all 'sh' , some with pretty close PID
 - that's a hint.

Good lueck ....................... Horst.

 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

On Thu, 23 May 2002, Bob Crandell wrote:

> Hi,
> I have a server with a growing number of these:
> 20344 ?        Z      0:00 [sh <defunct>]
> 20354 ?        Z      0:00 [sh <defunct>]
> 20355 ?        Z      0:00 [sh <defunct>]
> 20363 ?        Z      0:00 [sh <defunct>]
> 
> How do I get rid if them?  How do I find out what's causing them?
> --
> Bob Crandell
> Assured Computing
> When you need to be sure.
> Cell 541-914-3985
> FAX  240-371-7237
> [EMAIL PROTECTED]
> www.assuredcomp.com
> Eugene, Or. 97402
> 

Reply via email to