On Mon, 6 Jan 1997, Daniel Koepke wrote: - On Tue, 7 Jan 1997, Nic Suzor wrote: - - > i have looked through the vnum command, and cant see anything that would - > cause a crash. If anyone can see what i missed, please tell me - - Debuggers are your friend. If you don't have one, get one. It comes - in use when you have odd crashes. - - Now, as for this, it may be that you have mobiles without names or - keywords, thus causing a null pointer problem in vnum_mobile(). I - would suggest, first, though, that you try a debugger (gdb) on the - core and executable. Chances are that it'll point you in the correct - direction. Yeah, what he said :) First figure out how to use your debugger and see where it's crashing (is_name) for yourself. You can learn a lot that way. THen you can fix this problem for good by adding a few checks in parse_mobile and parse_object (woe to those who carry nameless objects in their inventories). The way I did it was to check for NULL name lists, and if any are found, it sets their namelist to EMPTY and notes which vnums need names in the syslog, so you can either track them down using vnum <o|m> empty, or check the syslog. Sam +-----------------------------------------------------------+ | Ensure that you have read the CircleMUD Mailing List FAQ: | | http://cspo.queensu.ca/~fletcher/Circle/list_faq.html | +-----------------------------------------------------------+
This archive was generated by hypermail 2b30 : 12/18/00 PST