2.0.27 slapadd segfaults

[John Morrissey <jwm@horde.net>]

I'm trying to slapadd a large ldif into a gdbm backend under OpenLDAP
2.0.27. slapadd segfaults about 25-30% through the process. Here's a
backtrace:

#0  0x4207abf7 in chunk_alloc () from /lib/i686/libc.so.6
#1  0x4207a158 in malloc () from /lib/i686/libc.so.6
#2  0x40068609 in _gdbm_read_entry () from /usr/lib/libgdbm.so.2
#3  0x40068848 in _gdbm_findkey () from /usr/lib/libgdbm.so.2
#4  0x400666a1 in gdbm_fetch () from /usr/lib/libgdbm.so.2
#5  0x0808cc16 in ldbm_fetch (ldbm=0x941e548, key=
      {dptr = 0x1038b2e0 "xKo?=\030V\210\003\231F{o?=o?=o?=co?=m", dsize = 16})
    at ../../../libraries/libldbm/ldbm.c:656
#6  0x08080874 in ldbm_cache_fetch (db=0x80cbd94, key=
      {dptr = 0x1038b2e0 "xKo?=\030V\210\003\231F{o?=o?=o?=co?=m", dsize = 16})
    at ../../../../servers/slapd/back-ldbm/dbcache.c:329
#7  0x08085760 in idl_fetch_one (be=0x80cb9a0, db=0x80cbd94, key=
      {dptr = 0x1038b2e0 "xKo?=\030V\210\003\231F{o?=o?=o?=co?=m", dsize = 16})
    at ../../../../servers/slapd/back-ldbm/idl.c:138
#8  0x08085f0f in idl_insert_key (be=0x80cb9a0, db=0x80cbd94, key=
      {dptr = 0x1038b2e0 "xKo?=\030V\210\003\231F{o?=o?=o?=co?=m", dsize = 16},
    id=137844) at ../../../../servers/slapd/back-ldbm/idl.c:403
#9  0x08088493 in key_change (be=0x80cb9a0, db=0x80cbd94, k=0x11536848,
    id=137844, op=1) at ../../../../servers/slapd/back-ldbm/key.c:68
#10 0x08087e61 in indexer (be=0x80cb9a0, dbname=0x80b4568 "uid",
    atname=0x80b4568 "uid", vals=0x117d2698, id=137844, op=1, mask=532)
    at ../../../../servers/slapd/back-ldbm/index.c:212
#11 0x08087f69 in index_at_values (be=0x80cb9a0, type=0x80b4950, lang=0x0,
    vals=0x117d2698, id=137844, op=1, dbnamep=0xbfffe874, maskp=0xbfffe870)
    at ../../../../servers/slapd/back-ldbm/index.c:254
#12 0x080880a6 in index_values (be=0x80cb9a0, desc=0x116451d0,
    vals=0x117d2698, id=137844, op=1)
    at ../../../../servers/slapd/back-ldbm/index.c:304
#13 0x080881a6 in index_entry (be=0x80cb9a0, op=1, e=0x10460000, ap=0x103602d0)
    at ../../../../servers/slapd/back-ldbm/index.c:329
#14 0x080790b5 in ldbm_tool_entry_put (be=0x80cb9a0, e=0x10460000)
    at ../../../../servers/slapd/back-ldbm/tools.c:180
#15 0x0804fb90 in main (argc=1, argv=0xbfffeac4)
    at ../../../../servers/slapd/tools/slapadd.c:112
#16 0x42017589 in __libc_start_main () from /lib/i686/libc.so.6

I'm not sure why the segfault would be happening there. There don't seem to
be any rlimits that slapadd is hitting. Any thoughts?

thanks,
john
-- 
John Morrissey          _o            /\         ----  __o
jwm@horde.net        _-< \_          /  \       ----  <  \,
www.horde.net/    __(_)/_(_)________/    \_______(_) /_(_)__