[442] in Kerberos-V5-bugs
Re: Possible memory leak in get_in_tkt.c
daemon@ATHENA.MIT.EDU (Theodore Ts'o)
Fri Mar 11 23:01:57 1994
Date: Fri, 11 Mar 94 23:00:24 EST
From: tytso@MIT.EDU (Theodore Ts'o)
To: mullan_s@apollo.hp.com
Cc: krb5-bugs@MIT.EDU, mullan_s@apollo.hp.com (Sean Mullan),
In-Reply-To: "[424] in Kerberos-V5-bugs"
[0424] daemon@ATHENA.MIT.EDU (mullan_s@apollo.hp.com) Kerberos-V5-bugs 02/25/94 15:52 (23 lines)
Date: Fri, 25 Feb 94 15:50:25 -0500
From: mullan_s@apollo.hp.com
In get_in_tkt.c / get_in_tkt(), storage is allocated for
request.padata but is only freed once, on an error return
from krb5_obtain_padata().
Shouldn't this memory be freed on all returns (including
success) from get_in_tkt()?
Yup. get_in_tkt() was missing a lot of allocated memory cleanup
handling on different for error returns. I've fixed this. Thanks for
pointing this out!
- Ted
