[486] in Kerberos-V5-bugs
Re: new krb5_get_in_TKT
daemon@ATHENA.MIT.EDU (Theodore Ts'o)
Wed Jun 1 15:24:56 1994
Date: Wed, 1 Jun 94 15:24:25 EDT
From: tytso@MIT.EDU (Theodore Ts'o)
To: georgesr@wrq.com (georges rahbani)
Cc: krb5-bugs@MIT.EDU, georgesr@elmer.wrq.com
In-Reply-To: "[482] in Kerberos-V5-bugs"
Date: Wed, 25 May 1994 14:56:50 -0700
From: georgesr@wrq.com (georges rahbani)
* request.addresses: This field is either set to the addrs passed as
a parameter in which case everything is OK. Or krb5_os_localaddr
allocates. In this case, I seem to have missed where to free this
address array. And if we do (?) We should make sure not to free the
parameter addrs memory.
You're right. I've added the following code to the cleanup code:
if (!addrs)
krb5_free_addresses(request.addresses);
* The cleanup stuff seems ok but in reverse order. If we get any errors
after we start to fill the creds struct, we jump to cred_cleanup. This is
fine, but in this case we don't free the other stuff that cleanup frees and
we don't return any error codes.
What I have done is add a jump to cleanup and that (I think) should solve
the problem).
Yup, your solution is the right one. I forgot to include the jump when
I was coding it up.
Thanks for bringing these two things up!
- Ted
