read back in as same.
I have kept the size as a long to be compatible with the Cygnus V4
distribution. The problem was introduced when we changed the include/kerberosIV
structures to use 32 bit ints for timestamps. So, under OSF/1, tf_util would
write out a 64 bit issue date, and then try reading back in a 32 bit one. Since
Cygnus uses sizeof(long) we will too.
git-svn-id: svn://anonsvn.mit.edu/krb5/trunk@7667
dc483132-0cff-0310-8789-
dd5450dbe970
+Tue Mar 19 11:23:13 1996 Ezra Peisach <epeisach@kangaroo.mit.edu>
+
+ * tf_util.c (tf_get_cred): Issue date is written out as a long,
+ read back in as same.
+
Sat Feb 24 09:27:08 1996 Ezra Peisach <epeisach@kangaroo.mit.edu>
* g_svc_in_tkt.c, put_svc_key.c, rd_req.c, rd_svc_key.c: Declare
{
KTEXT ticket = &c->ticket_st; /* pointer to ticket */
int k_errno;
+ long issue_date;
if (fd < 0) {
if (krb_debug)
/* don't try to read a silly amount into ticket->dat */
ticket->length > MAX_KTXT_LEN ||
tf_read((char *) (ticket->dat), ticket->length) < 1 ||
- tf_read((char *) &(c->issue_date), sizeof(c->issue_date)) < 1
+ tf_read((char *) &(issue_date), sizeof(issue_date)) < 1
) {
tf_close();
return TKT_FIL_FMT;
}
+ c->issue_date = issue_date;
#ifdef TKT_SHMEM
memcpy(c->session, tmp_shm_addr, KEY_SZ);
tmp_shm_addr += KEY_SZ;