Re: [Code] [Stock bug?] write_to_ouput oops. :-)

From: Tony Robbins (tonyr@NWPACLINK.COM)
Date: 07/15/98

At 06:04 PM 7/15/98 EDT, you wrote:
>In a message dated 98-07-15 17:09:49 EDT, jelson@CIRCLEMUD.ORG writes:
><< This is confusing, I admit, because CREATE() is a macro and not a
> function.  The second argument to create is supposed to be a type so
> that the return value of malloc() can be typecast.  If CREATE() is
> failing it might indicate you're out of memory. >>
>I thought bpl14 was supposed to be more efficient though. MUD worked fine
>bpl13a, but now with bpl14 people can't create characters? Any way I can get
>around this?

Did you hand-patch?  I haven't gone up to bpl14 yet, because of extra work
at work, but I'm pretty sure that's a bug and not a feature.  :P

BTW, this apparently is a bug in DG pl4.

For room reset triggers, I was getting crashes every few hours, so I put in
log statements to find out what numbers it was crashing on.  It cycled
through the room numbers for reset triggers, but apparently this:

if (room_rnum) reset_wtrigger(&world[room_rnum]);

...does not account for rnum of -1 (NOWHERE).  What fixed it for me (no
more crashes from it, been up for nearly two days), was to change it to:

if (room_rnum != NOWHERE) reset_wtrigger(&world[room_rnum]);

...which as repaired it and the triggers still work.  However, in hindsight
I wonder, is a room_rnum of 0 legal?  Seems to me it would be, and point to
Limbo or the Void or whatever.


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

This archive was generated by hypermail 2b30 : 12/15/00 PST