Please recompile with make debug and provide a full backtrace, as adviced in doc/BUG-REPORT. We need a backtrace in order to find the problem, since the DEBUG file is not useful by itself.
I was asking you for a full gdb backtrace, in order to figure out what's your problem. Make debug is to make the debug information more useful, did you actually thought it will fix your bot?same [censored] Smile different day
[22:03:07] CTCP DCC: CHAT chat 1835584260 1024 from k (kiril@kiril.FreeBSD.org)
[22:03:07] DCC connection: CHAT (Kiril!kiril@kiril.FreeBSD.org)
[22:03:12] Logged in: Kiril (kiril@kiril.FreeBSD.org/1024)
[22:03:12] * Last context: tclhash.c/710 [Tcl proc: *chon:console:chon, param: $_chonof1 $_chonof2]
[22:03:12] * Please REPORT this BUG!
[22:03:12] * Check doc/BUG-REPORT on how to do so.
[22:03:12] * Wrote DEBUG
[22:03:12] * SEGMENT VIOLATION -- CRASHING!
Although you didn't provide any useful information, I can guess what's your problem. You're using the unofficial, unsupported and buggy SSL patch, which has already two serious problems and that one is the third.[22:03:12] * Last context: tclhash.c/710 [Tcl proc: *chon:console:chon, param: $_chonof1 $_chonof2]
It shifts down the module symbols lookup table in src/modules.c, without updating properly the corresponding definitions in src/mod/module.h, so instead of the console_chon function, a completely different one is called, which logically leads to a crash.
The patch is very buggy and obviously the author didn't bother to test it. Here are the other problems, I've found so far (we find a new one almost every day):It's the patch author's job to fix these problems and do his homework before submitting patches. All of these should have been obvious at the moment of writing the patch.
Statistics: Posted by pseudo — Thu Aug 12, 2010 5:24 pm
]]>