SIGSEGV slapd 2.0.21

[Matthew Palmer <mjp16@ieee.uow.edu.au>]

OK, who else is getting these?  Using GSSAPI over a Heimdal KDC, the last
bit of -d -1 seems OK:

do_sasl_bind: dn (uid=mjp16+realm=IEEE.UOW.EDU.AU) mech GSSAPI
conn=0 op=2 BIND dn="UID=MJP16+REALM=IEEE.UOW.EDU.AU" method=163
==> sasl_bind: dn="uid=mjp16+realm=IEEE.UOW.EDU.AU" mech=<continuing>
datalen=0

and then it segfaults and all is no good.

Mr. strace gives nothing which seems useful:

rt_sigprocmask(SIG_SETMASK, NULL, [RT_0], 8) = 0
rt_sigsuspend([] <unfinished ...>
--- SIGRT_0 (Real-time signal 0) ---
<... rt_sigsuspend resumed> )           = -1 EINTR (Interrupted system call)
sigreturn()                             = ? (mask now [])
rt_sigprocmask(SIG_SETMASK, NULL, [RT_0], 8) = 0
rt_sigsuspend([] <unfinished ...>
--- SIGSEGV (Segmentation fault) ---

I guess the next step is recompiling with -g and going digging with gdb,
which, quite honestly, doesn't thrill me to bits.  I'll do it if it's
absolutely necessary, but I don't relish the thought.  Anyone got any nicer
suggestions for trying to find the problem, and helping out the OpenLDAP
developers?


-- 
-----------------------------------------------------------------------
#include <disclaimer.h>
Matthew Palmer
mjp16@ieee.uow.edu.au