Comment 2 for bug 1187876

The process somehow gets wedged in libc's malloc error handler. gdb probably isn't going to pinpoint the source of the corruption. valgrind instead shows:

# valgrind rsyslogd -c5
==5249== Memcheck, a memory error detector
==5249== Copyright (C) 2002-2012, and GNU GPL'd, by Julian Seward et al.
==5249== Using Valgrind-3.8.1 and LibVEX; rerun with -h for copyright info
==5249== Command: rsyslogd -c5
==5249==
==5249== Invalid write of size 1
==5249== at 0x557BA89: __vsprintf_chk (vsprintf_chk.c:86)
==5249== by 0x557B9CC: __sprintf_chk (sprintf_chk.c:32)
==5249== by 0x40A981: queryLocalHostname (stdio2.h:33)
==5249== by 0x40D11E: realMain (syslogd.c:2761)
==5249== by 0x548DEA4: (below main) (libc-start.c:260)
==5249== Address 0x58423b8 is 0 bytes after a block of size 24 alloc'd
==5249== at 0x4C2CD7B: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==5249== by 0x40A956: queryLocalHostname (syslogd.c:2423)
==5249== by 0x40D11E: realMain (syslogd.c:2761)
==5249== by 0x548DEA4: (below main) (libc-start.c:260)
==5249==
<snip>