FastCGI: continual process regeneration

I have a RT installation where the FastCGI children have been observed to
be dying and restarting themselves very rapidly, e.g.

[Sun Jul 31 00:28:57 2005] [warn] FastCGI: server
"/usr/local/rt3/bin/mason_handler.fcgi" (pid 23631) terminated by calling
exit with status ‘0’
[Sun Jul 31 00:28:57 2005] [warn] FastCGI: server
"/usr/local/rt3/bin/mason_handler.fcgi" restarted (pid 23642)
[Sun Jul 31 00:29:08 2005] [warn] FastCGI: server
"/usr/local/rt3/bin/mason_handler.fcgi" (pid 23634) terminated by calling
exit with status ‘0’
[Sun Jul 31 00:29:08 2005] [warn] FastCGI: server
"/usr/local/rt3/bin/mason_handler.fcgi" has remained running for more than
30 seconds, its restart interval has been restored to 5 seconds
[Sun Jul 31 00:29:08 2005] [warn] FastCGI: server
"/usr/local/rt3/bin/mason_handler.fcgi" restarted (pid 23644)

Although RT still “works” (functionally) this keeps the system load at 4.

Is there any way to fix or diagnose this? I have turned up RT’s own
debugging to “debug” but nothing particularly weird shows up.

The version of Perl on the system is:

This is perl, v5.8.5 built for i386-freebsd-thread-multi-64int

Would the threaded Perl have anything to do with it?

  • Julian

– Julian C. Dunn Julian_Dunn@cbc.ca jdunn@nm.cbc.ca
– Platform Administrator, CBC.ca Production & Operations
– Office: 2C310-Q * Tel.: (416) 205-3311 x5592