Hi Olivier, The paris traceroute implementation is intersting.
However, to use the UDP checksum field (instead of src port) to convey the offset value (per our proposal) may consideriably increase the complexity. The reason is that it will require reverse engineering (after all, checksum is calculated after the payload is in place, not the other way around) and require stuffing/padding data in the UDP payload. Wrt TCPtraceroute, using a TCP option for conveying the offset value is much easier than reusing the seq# (as it is done in paris traceroute) for our purpose. Cheers, Rajiv > -----Original Message----- > From: Olivier Bonaventure [mailto:[EMAIL PROTECTED] > Sent: Tuesday, November 25, 2008 2:32 AM > To: Rajiv Asati (rajiva) > Cc: [email protected]; Enke Chen (enkechen) > Subject: Re: [Int-area] draft-shen-udp-traceroute-ext - Discussion > > Rajiv, > > > When this draft-shen-udp-traceroute-ext was presented last week > during > > the int-area meeting, there seems to be a consensus about the problem > > that the draft was solving, however, there was a minor concern about > the > > solution. > > > > Specifically, the concern was that using last 4 bits in the > > UDP src port# will result in varying the src port#, hence, > > traceroute probe may not really test the actual path (if > > multiple paths exist) that the application traffic may take. > > The solution used for Paris traceroute (changing the checksum instead > of > changing the last 4 bits of the src port) could be applicable as well. > See http://www.paris-traceroute.net/ and > http://www.imconf.net/imc-2006/papers/p15-augustin.pdf > > Olivier > > -- > http://inl.info.ucl.ac.be , UCLouvain, Belgium _______________________________________________ Int-area mailing list [email protected] https://www.ietf.org/mailman/listinfo/int-area
