m***t
发帖数: 22 | 1 以下是最近解决不了的几个debug问题
//help
Too long evaluation. Execution aborted.
arg 0)
program: adm/daemons/channeld.c, object: adm/daemons/channeld, file: /adm/daemons/channeld.c:112
obj = all_inventory(environment(ob));
for(i=0; i
if( !living(obj[i]) || !userp(obj[i]) || obj[i]==man ) continue;
if( userp(obj[i]) ) {
man->set_leader(obj[i]);
man->kill_ob(obj[i]);
obj[i]->fight_ob(man);
}
}
return ;
} | p**f
发帖数: 49 | 2 It is the MUDOS's limitation about how many instruction can execute in a
single function. It is how MudOS detect the endless loop.
U can rebuild ur mudos to increase this number
but maybe this will waste ur CPU when real endless loop happens.
it is your choice :-) | g****s
发帖数: 306 | 3
but some other bug really not due to this, like the crash
in logind, it has no records, seems the instruction pointer
goes to somewhere unknown and causes the crash, but how
can I find when or in which function it
goes crazy?
【在 p**f 的大作中提到】
: It is the MUDOS's limitation about how many instruction can execute in a
: single function. It is how MudOS detect the endless loop.
: U can rebuild ur mudos to increase this number
: but maybe this will waste ur CPU when real endless loop happens.
: it is your choice :-)
| p**f
发帖数: 49 | 4
hehe, a big question
There is not special way that can resolve these kinds of crash.
From my experience, when the system is not stable,
open the debug option of mudos, and try to get rid of all the error
msg in logfile.
【在 g****s 的大作中提到】
:
: but some other bug really not due to this, like the crash
: in logind, it has no records, seems the instruction pointer
: goes to somewhere unknown and causes the crash, but how
: can I find when or in which function it
: goes crazy?
|
|
