> > Hi i have recently moved to a new server, and when the mud crashes it isnt
> > producing a core file to help me debug.
> > Is this because of the nature of the crash or is it something i have missed
> > when compiling the mud?
> > Any help would be greatly appreciated.
> [snip]
>
> The host machine can be configured (under Linux at least, and presumably
> in other *NIX OSes as well) to refuse non-root user processes the right to
> create core files. It is often seen as a potential security hole...
>
> First rule out that you're possibly trying to dump core into a folder that
> you don't actually have write permissions for or are deleting it with a
> script of some kind (for example, try to get another program to segfault
> and see if it creates a core), then talk to the server admins to find out
> if they've taken any actions that may be keeping your processes from being
> able to dump corefiles.
>
You can also scan the lists for 'core dump' and you'll see quite a
few other options. The most common of these is to set your user limits
(using the 'limit' or 'ulimit' command depending on your shell), and save
the limit changing in your startup file (.cshrc, .login, .profile, etc).
If I remember my syntax correcly,
ulimit -Hc 10000000
or
limit coredumpsize 10000000
Should do the same things. You'll want to test of course.
Worse comes to worse, you can always go and modify the default
values for these in the kernel headers and recompile your kernel. I did
that after a while, when I got pissy :)
PjD
+------------------------------------------------------------+
| Ensure that you have read the CircleMUD Mailing List FAQ: |
| http://qsilver.queensu.ca/~fletchra/Circle/list-faq.html |
+------------------------------------------------------------+
This archive was generated by hypermail 2b30 : 04/11/01 PDT