Irssi 0.8.9 core dump

[irssibot]

• • • Issue #64 opened by @Puh

Sudden core dump after several days or weeks of use. I've seen this several times:

(gdb) bt
#0  0x3f256058 in ?? ()
#1  0x40272625 in _IO_vsprintf (string=0x8475dd8 "Ð\\f3@", 
    format=0x81aa6f0 ">> %s", args=0xbffff224) at iovsprintf.c:46
#2  0x400bfff0 in g_strdup_vprintf (format=0x81aa6f0 ">> %s", args1=0xbffff224)
    at gstrfuncs.c:158
#3  0x400c001c in g_strdup_printf (format=0x81aa6f0 ">> %s") at gstrfuncs.c:172
#4  0x080cfe70 in rawlog_input (rawlog=0x823e638, 
    str=0xbffff110 "\\001\\200­ûØ]G\\bØ]G\\bØ]G\\bØ]G\\bØ]G\\b×]G\\fØ]G\\b×]G\\f")
    at rawlog.c:83
#5  0x080a7da5 in irc_parse_incoming (server=0x8422090) at irc.c:379
#6  0x080ca4ed in irssi_io_invoke (source=0x823e668, condition=138894808, 
    data=0x846e3d0) at misc.c:56
#7  0x400b5a7a in g_io_unix_dispatch (source_data=0x846e3e0, 
    current_time=0xbffff330, user_data=0x846e3d0) at giounix.c:137
#8  0x400b7055 in g_main_dispatch (dispatch_time=0xbffff330) at gmain.c:656
#9  0x400b7659 in g_main_iterate (block=1, dispatch=1) at gmain.c:877
#10 0x400b7725 in g_main_iteration (block=1) at gmain.c:907
#11 0x0807ea05 in main (argc=0, argv=0xbffff3f4) at irssi.c:385
#12 0x4021e336 in __libc_start_main (main=0x807e960 <main>, argc=1, 
    ubp_av=0xbffff3f4, init=0x806ac38 <_init>, fini=0x8193c30 <_fini>, 
    rtld_fini=0x4000d25c <_dl_fini>, stack_end=0xbffff3ec)
    at ../sysdeps/generic/libc-start.c:129

Loaded modules: Module Type Submodules core static core irc fe fe-irc fe-text dcc static irc fe-irc flood static irc notifylist static irc fe-irc perl static core fe icq dynamic core fe mobile dynamic core idea dynamic core

Loaded scripts: mailcount /home/stdan/.irssi/scripts/mailcount.pl splitlong /home/stdan/.irssi/scripts/splitlong.pl burl /home/stdan/.irssi/scripts/burl.pl

I can provide more information if necessary, just ask.

[irssibot]

• • • Comment 113 by @cras

Heap corruption somewhere. The backtrace doesn't really help as the bug happened sometimes before it. These ones are difficult to track down, except by running irssi via "valgrind irssi 2>log" and sending the log file when it crashes (or before, if it gets any entries).

Do you use SSL?

[irssibot]

• • • Comment 176 by @guly

irssi just crashed me too, re-run with valgrind --tool=corecheck hope if crash again this will help you.

[irssibot]

• • • Comment 207 by @Puh

Crashes stopped or became very rare as hardware changed. I don't really know why as the old hardware had no memory or heat issues -- other than irssi crashes it was rock solid, but slow.

[irssibot]

• • • Closed by @coekie