tedcheng@symas.com wrote:
The following patch fixed the issue:
Index: servers/slapd/back-ldap/bind.c
RCS file: /var/CVSROOT/ldap24/servers/slapd/back-ldap/bind.c,v retrieving revision 1.12 retrieving revision 1.13 diff -u -u -r1.12 -r1.13 --- servers/slapd/back-ldap/bind.c 28 Jul 2011 09:31:34 -0000 1.12 +++ servers/slapd/back-ldap/bind.c 17 Nov 2011 21:35:46 -0000 1.13 @@ -2056,7 +2056,7 @@ struct berval ndn; int dobind = 0;
if ( op->o_conn == NULL || op->o_do_not_cache ) {
if ( op->o_conn == NULL ) { goto done; }
In your original post you state:
The ldap_back_dobind() function calls ldap_back_dobind_int() for bind, back-ldap/bind.c. The following ldap_back_dobind_int() code is destined for assertion failure, if op->o_do_not_cache flag is set and there is no valid binddn and bindcred returned by ldap_back_getconn(). Setting an invalid LDAP URI for the remote AD Windows box is such a case. <<<
So it sounds like this assert only triggers because you have an invalid configuration. I don't believe your patch is the correct solution if that's the only error case.
-
hyc＠symas.com