Nick (and other module developers), I am still struggling with some Seg Faults. They no longer occur from directory/file/location contexts. Now, I have determined that it only occurs on one specific directive-handler and only if the value of the flag is "On" (not "Off") and called from a server/vhost context. Detecting the context (server or directory) of a directive-handler call seemed to fix half (the directory-context calls) of the Seg Fault occurances.
Am I trying to write to a read-only struct? The part that stumps me about that is there is another int in the same struct that is taking directive-value changes just fine. I have also noticed what seems to me to be an anomaly in my code: When I create the server structs, I set a particular flag (an integer member) to be 0. By the time the directive-handling call is made, the value is something I have not set it to (usually around 135,xxx,xxx). This only seems to happen in my SERVER config struct and for only ONE of its members. I have scoured my code for every possible place where I set this value and I don't ever set it to be anything other than (0|1). I never use it as a string or a pointer (from what I can tell). The other values in the server config struct are set correctly, and this particular value is set by the end of the create_server_config struct. I know I am working with the correct server struct. Any idea on what could be setting the server config struct variable? Thanks, Dave
