On Tue, May 22, 2012 at 5:41 AM, Chris Angelico <rosuav@gmail.com> wrote:
>
> On Tue, May 22, 2012 at 8:23 PM, Jayashankar K B
> <Jayashankar.KB@lnties.com> wrote:
> > But here, the crash is happening right at the insert statement. That is
> > insert itself is failing.
> > Unless the insert is successful, stored procedure is not triggered.
>
> Hmm. I wonder is it possible that going past ID 999 and into a
> four-digit number is causing stack damage that crashes the server a
> few iterations later... many things are possible. I'd look at the code
> of the procedure and see if there's any possible memory/stack issues.
Hm, on linux you check stack size with ulimit -s? If stack is set too
low, a lower setting of max_stack_depth should prevent the crash.
It's pretty hard to hit that unless you have extraordinarily complex
and/or recursive functions though.
Any chance of seeing the function source?
merlin
