[3096] in Kerberos-V5-bugs
pending/737: krb_sendauth_hidden_tkt_len
daemon@ATHENA.MIT.EDU (Brad Thompson)
Mon Aug 9 21:32:06 1999
Resent-From: gnats@rt-11.MIT.EDU (GNATS Management)
Resent-To: gnats-admin@rt-11.MIT.EDU
Resent-Reply-To: krb5-bugs@MIT.EDU,
Brad Thompson <yak@point-and-drool.mit.edu>
Message-Id: <199908100131.VAA01730@point-and-drool.mit.edu>
Date: Mon, 9 Aug 1999 21:31:28 -0400 (EDT)
From: Brad Thompson <yak@point-and-drool.mit.edu>
To: krb5-bugs@MIT.EDU
>Number: 737
>Category: pending
>Synopsis: krb_sendauth_hidden_tkt_len
>Confidential: yes
>Severity: serious
>Priority: medium
>Responsible: gnats-admin
>State: open
>Class: sw-bug
>Submitter-Id: unknown
>Arrival-Date: Mon Aug 09 21:32:01 EDT 1999
>Last-Modified:
>Originator:
>Organization:
>Release:
>Environment:
>Description:
>How-To-Repeat:
>Fix:
>Audit-Trail:
>Unformatted:
So, there's this variable krb_sendauth_hidden_tkt_len in lib/krb4/sendauth.c.
It's making my life unhappy for uninteresting reasons. It seems that it is
not actually used, though. The comments reference kcmd.c, which does not
touch it, or any of the things #define'd to it. I think we should make
the following patch:
*** sendauth.c.old Mon Aug 9 21:19:54 1999
--- sendauth.c Mon Aug 9 21:22:06 1999
***************
*** 107,124 ****
*
* See FIXME KLUDGE code in appl/bsd/kcmd.c.
*/
- KRB4_32 __krb_sendauth_hidden_tkt_len;
- #define raw_tkt_len __krb_sendauth_hidden_tkt_len -
/*
* Read a server's sendauth response out of a file descriptor.
* Returns a Kerberos error code.
- *
- * Note sneaky code using raw_tkt_len to stash away a bit of info
- * for use by appl/bsd/kcmd.c. Now that krb_net_rd_sendauth is
- * a separate function, kcmd should call it directly to get this
- * sneaky info.
*/
int
krb_net_rd_sendauth (fd, reply, raw_len)
--- 107,116 ----
***************
*** 230,235 ****
--- 222,229 ----
/* mutual authentication, if desired */
if (options & KOPT_DO_MUTUAL) {
+ KRB4_32 raw_tkt_len;
+
/* get credentials so we have service session
key for decryption below */
if (cc = krb_get_cred(service, srv_inst, realm, cred))
yak
(can't seem to get send-pr working on this box)
