I followed this up mentioning it wasn't Apache's fault. That was quite a while ago. You can delete the bug report. ;)
There was a problem with FastCGI that it wasn't configuring itself properly, to lock/unlock around accept(), as well as an accept() that was never locked. Multiple accept()'s on Solaris == dead fastcgi manager process. We've corrected both errors (and a 3rd these fixes revealed), and sent them into the FCGI team for inclusion in future versions. -Jonathan At 08:59 PM 6/22/97 -0700, [EMAIL PROTECTED] wrote: >Synopsis: fastcgi processes dying off over time, not restarting. > >State-Changed-From-To: open-analyzed >State-Changed-By: dgaudet >State-Changed-When: Sun Jun 22 20:59:56 PDT 1997 >State-Changed-Why: >What was the last version that fastcgi worked with? Have you >asked folks on the fastcgi mailing list? It'd be a lot easier for >someone more familiar with fastcgi internals to diagnose this... > >Dean > > -- Jonathan Roy - [EMAIL PROTECTED] - Idle Communications, Inc. Idle Communications accepts contract work in Perl/C/C++ programming, CGI development, and general system admin.
