Update, just recompiled with bdb 4.2.52 w/ 4 patches from sleepycat and
the patch supplied in openldap 2.3 and got the same problem.
Is this useful? I don't want to be spamming the list with useless
debugging junk.
Here is what it shows when I run slapadd under gdb.
(gdb) run -q -w -l /root/BUILD/import/dump.ldif
Starting program: /usr/local/openldap23_bdb42/sbin/slapadd -q -w -l
/root/BUILD/import/dump.ldif
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 2 (LWP 100278)]
0x080efe9c in hdb_dn2id (op=0xbfbfe7f0, txn=0x0, in=0x820dba0,
ei=0xbfbfe750)
at dn2id.c:647
647 BDB_ID2DISK( ei->bei_parent->bei_id, &idp );
(gdb) thread apply all bt
Thread 3 (Thread 2 (LWP 100278)):
#0 0x080efe9c in hdb_dn2id (op=0xbfbfe7f0, txn=0x0, in=0x820dba0,
ei=0xbfbfe750) at dn2id.c:647
#1 0x080cc86e in hdb_tool_dn2id_get (be=0x81b6c00, dn=0x820dba0)
at tools.c:155
#2 0x080b4384 in slapadd (argc=5, argv=0xbfbfebf0) at slapadd.c:326
#3 0x0804c6f1 in main (argc=5, argv=0xbfbfebf0) at main.c:279
Thread 2 (Thread 1 (LWP 100288)):
#0 0x283df2fb in pthread_testcancel () from /usr/lib/libpthread.so.1
#1 0x283d7902 in pthread_mutexattr_init () from /usr/lib/libpthread.so.1
#2 0x00000000 in ?? ()
0x080efe9c 647 BDB_ID2DISK( ei->bei_parent->bei_id, &idp
);