Small explanation for why MS created their own KRB5 OID value.

git-svn-id: http://anonsvn.wireshark.org/wireshark/trunk@7160 f5534014-38df-0310-8fa8-9805f1628bb7

diff --git a/packet-spnego.c b/packet-spnego.c
index 1ce86ea7491bbfb8936479e52d4dac70b374b5e3..aab1093e2dde9ad32b2f7ed0f7bae4486bb11948 100644 (file)
--- a/packet-spnego.c
+++ b/packet-spnego.c
@@ -4,7 +4,7 @@
  * Copyright 2002, Tim Potter <tpot@samba.org>
  * Copyright 2002, Richard Sharpe <rsharpe@ns.aus.com>
  *
- * $Id: packet-spnego.c,v 1.41 2003/01/10 23:47:30 guy Exp $
+ * $Id: packet-spnego.c,v 1.42 2003/02/17 17:32:59 sharpe Exp $
  *
  * Ethereal - Network traffic analyzer
  * By Gerald Combs <gerald@ethereal.com>
@@ -1427,6 +1427,14 @@ proto_reg_handoff_spnego(void)
            "SPNEGO - Simple Protected Negotiation");
 
        /* Register both the one MS created and the real one */
+       /*
+        * Thanks to Jean-Baptiste Marchand and Richard B Ward, the
+        * mystery of the MS KRB5 OID is cleared up. It was due to a library
+        * that did not handle OID components greater than 16 bits, and was
+        * fixed in Win2K SP2 as well as WinXP.
+        * See the archive of <ietf-krb-wg@anl.gov> for the thread topic
+        * SPNEGO implementation issues. 3-Dec-2002.
+        */
        spnego_krb5_handle = create_dissector_handle(dissect_spnego_krb5,
                                                     proto_spnego_krb5);
        spnego_krb5_wrap_handle = new_create_dissector_handle(dissect_spnego_krb5_wrap,