Re: Byte counting in draft-winter-radext-fancyaccounting-00

[Alan DeKok <aland@deployingradius.com>]

Peter Deacon wrote:
> There is sometimes confusion (For me anyway) regarding what counters
> actually count.  Traditionally Input/Output octets are protocol agnostic
> and just count low layer bits shuffled between interfaces.
> 
> Lets say we run IPv6 over PPP or Ethernet do the byte counts apply only
> to the higher layer IP frames and not a lower layer ethernet or PPP frame?

  I'd say it's still bits over the interface, including ethernet or PPP.

> Alternately if there was a traffic class for even higher layer messages
> such as HTTP..does it count the HTTP layer only or the IP header and
> possibly ethernet/PPP header?

  We can define traffic classes for all of that, I think.

> I don't really have an opinion on the behavior only that it would be
> best if all implementations stand a good chance of counting the same way.

  The specification needs to be clear on what is being counted.  It
would help to have specific examples of this in the doc.

  e.g. If the description is "HTTP traffic on port 80", then the data
being counted should be *only* HTTP, not lower-layer TCP, IP, etc.

  Alan DeKok.

--
to unsubscribe send a message to radiusext-request@ops.ietf.org with
the word 'unsubscribe' in a single line as the message text body.
archive: <http://psg.com/lists/radiusext/>