[1196] in Kerberos_V5_Development
Re: krb5_db_entry and kadm5 info
daemon@ATHENA.MIT.EDU (Christopher Provenzano)
Thu May 16 13:44:43 1996
Reply-To: proven@cygnus.com
To: Sam Hartman <hartmans@MIT.EDU>
Cc: "Barry Jaspan" <bjaspan@MIT.EDU>, krbdev@MIT.EDU, proven@pbi.proven.org
In-Reply-To: Your message of "16 May 1996 02:26:22 EDT."
<tslivdxb0c1.fsf@tertius.mit.edu>
Date: Thu, 16 May 1996 13:42:03 -0400
From: Christopher Provenzano <proven@proven.org>
>
> bjaspan> int version; /* database entry vno */
>
> It is intended that the length field at the beginning of the
> db_entry serve as a version number and length; the length will only be
> increased. This was done for compactness on disk. I am not convinced
> that I agree with this decision, but I figured that noting it couldn't
> hurt.
This was again done with the idea of forward compatibility. If for some
reason a field is needed in the base db_entry it could be added to the
end of the database record and the size can be incremented. Current
implementation could then still read/write to the new database and would
only be unable to modify the new fields. Obviously the tl_data is a
much better mechanizm for forward compatibility but on the off chance that
the kdc absolutly_positivly_needs_this_new_field_in_the_base_struture, it
can be added.
CAP
