fork() problem?

[Mike Culbertson]

I am running Nagios on a P4 1.4 running RedHat 7.3, and also at another location on a Sun Ultra 10 running Debian Sparc/Linux.
The versions are 1.0b6 and 1.0b5 respectvely.  The problem I am encountering is that on the RedHat box, when I set max_concurrent_checks to 0, nagios appears to start forking infinitely (so to speak), I believe at the same rate that it is receiving passive checks (approx 1 new fork every 1-3 minutes).  The only reason I referenced the Sparc machine is because it does not suffer from this same problem, although it too accepts passive checks and shares an identical nagios.cfg as well as identical build options.

I found a simple fix to be to set max_concurrent_checks to a static number, and the forked processes stay within this boundary.  I am still can't figure out, though, why the forks seem to run rampant when not restrained.  

What might the condition be that prevents the forks from completing? 

Not being much of a c coder, stracing isn't showing me much I can work with but I'll be happy to attach logs on request. Thanks in advance.

Mike Culbertson
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://www.monitoring-lists.org/archive/users/attachments/20021021/a551cbf1/attachment.html>