On Thu, May 29, 2003 at 09:12:41AM -0400, [EMAIL PROTECTED] wrote:
> Has there been any further thought to moving the DOCSIS encapsulation
> from a frame level decoding to a packet level decoding?

What do you mean by "frame level decoding" and "packet level decoding"?

We currently treat DOCSIS as a link-layer protocol, as that's how it
appears in the capture files we support; there's no other protocol in
which it's encapsulated.

> Has anyone ran across IP / DOCSIS packets in the same packet stream?

One of the sample captures that were sent out by, I think, the author of
the DOCSIS dissector contains IP packets that have DOCSIS link-layer
headers, as well as other non-IP packets (such as MAC management
packets) that also have DOCSIS link-layer headers.

> Any suggestions on how to decode them both currently?

Ethereal decodes that capture correctly (if you've turned on the "treat
everything as DOCSIS" flag).

Reply via email to