Re: [CODE] tracking down an infinite loop

From: R. E. Paret (
Date: 01/31/97

On Thu, 30 Jan 1997, RHS Linux User wrote:
> ERm this is kinda weird.  How do i go about tracking down an infinite loop
> (or when the mud just hangs... doesnt core dump).  I do a kill -11 PID
> and this creates a core file that looks like this:
> Program terminated with signal 11, Segmentation fault.
> #0  0x4003af86 in ?? ()

Once again gdb comes to the rescue. :P
start circle using autorun and then run gdb in your bin directory like 
this: gdb circle <pid number of /bin/circle -q 4000 or whatever>
Then use step, next, or inext to step through the program up to the crash 
(you're going to need multiple windows for this) The idea is you log on 
to your mud and step through it until it seg faults, then go and look at 
gdb... It should tell you exactly where its seg faulting.


| Ensure that you have read the CircleMUD Mailing List FAQ: |
|   |

This archive was generated by hypermail 2b30 : 12/18/00 PST