GSS krb5 mech and ticket expiration

[Henry B. Hotz]

On Jun 9, 2010, at 9:04 AM, krbdev-request at mit.edu wrote:

> I recently committed a change to stop checking for context expiration
> times in the krb5 GSS mech's wrap and unwrap functions.  From the
> commit message:


In an ideal world, I would argue that app's should renew tgt's and rekey sessions transparently as needed.  However in practice it seems to me that most services will allow a session or operation to continue to completion, and at most enforce ticket expiration for new operations.

+1, since it seems to match current practice better.

------------------------------------------------------
The opinions expressed in this message are mine,
not those of Caltech, JPL, NASA, or the US Government.
Henry.B.Hotz at jpl.nasa.gov, or hbhotz at oxy.edu