Ethereal-dev: Re: [Ethereal-dev] DNS flags are not parsed completely

Note: This archive is from the project's previous web site, ethereal.com. This list is no longer active.

From: Yaniv Kaul <ykaul@xxxxxxxxxxxxxx>
Date: Mon, 13 Jan 2003 09:26:29 +0200
Lets show them all. What is not needed, can be divided to two:
1. reserved - need to be 0 by RFC - check that indeed it's zero'ed, if it's not, annotate it as such!
2. Value ignored - just display the value, if it does not have a meaning at all, in query for example, annotate it as such ('meaningless in queries' / 'valid in response only')

Brad Hards wrote:
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Fri, 10 Jan 2003 01:44, Yaniv Kaul wrote:
  
In DNS, some flags are omitted. I tend to believe this causes some
inaccuracies in the parsing of the flags.
See atfached.
1. The AA flag is missing from a query (which is theoretically fine).
2. The location of the AA flag - just over where the Z flag bits are.

Similiar issues are in the response.
The Z flag is not shown it all. I think it should be, even (and actually
because) it has to be 0!

In general, I think we should show even fields that have to be 0 (or any
fixed value) - even if just to verify they are!
    
I'm not sure I understand what you think ethereal should show. If I can 
understand, I think I can create a patch, and then we can discuss it better.

Lets deal with AA flag first. It would be no problem to show the value of the 
bit, but what should it be annotated to say in a query? What would you like 
the text to show if it is a zero? And if it is a one?

Now lets do Z. Are we talking about the version in RFC1035 or the version in 
RFC2535? I assume the latter, although you refer to "flag bits", while there 
is only one Z bit in RFC2535. What would the annotation show in a query for 
each of one and zero?
And for a response containing a one or a zero?

You haven't requested RA in a query. Why do you think AA should be shown and 
not RA?

This is not to say that a patch that implemented these changes would be 
accepted anyway. It would be interesting to get a ruling from the Ethereal 
demigods on policy for display of reserved and undefined message elements.

Brad
- -- 
http://linux.conf.au. 22-25Jan2003. Perth, Aust. I'm registered. Are you?
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iD8DBQE+HpSHW6pHgIdAuOMRAm+2AJ4i5HNJBEH9ndPQxE9Y82kLy93y0gCgqRs2
HzWYWTb0OQcjBSkLSENwkGY=
=erp0
-----END PGP SIGNATURE-----