Re: [ippm] draft-ietf-ippm-owdp-08.txt

[stanislav shalunov <shalunov at internet2.edu>]

To make comparison with the old version easier:

The new version of the draft incorporates a change in treatment of the
TTL field.  While previously it was not specified, it now allows to
measure the decrement of TTL as follows:

* The sender SHOULD set TTL of OWAMP-Test packets to 255 (the maximum
  value); the only reason not to set it to 255 is inability to change
  the value because of limitations of the socket interface that the
  implementation uses.

* The receiver SHOULD record the value of the TTL field in OWAMP-Test
  packets it receives; the only reason not to record it is inability
  to access the field value due to socket interface limitations (if a
  receiver cannot record the actual value, it MUST record 255---a
  value that is otherwise impossible in a received packet).

* The session results, when transmitted, now include TTL for each
  packet.

The uses of this technique remain unspecified in the document, but
include correlating delay changes to routing changes.

	------------------------------------------------------------

One change remains pending for the specification document: free-text
error messages in addition to error codes, as suggested by Al Morton
and others in the past.

-- 
Stanislav Shalunov		http://www.internet2.edu/~shalunov/

This message is designed to be viewed at 600dpi.

_______________________________________________
ippm mailing list
ippm@ietf.org 
https://www1.ietf.org/mailman/listinfo/ippm