Hi Francis,

Thanks for your feedback.
Since Traceroute is not a standard, but rather an application, it has several 
implementations.
Indeed, the UNIX implementation uses UDP messages - this is also described in 
RFC 2151 (informational).
The windows implementation of Traceroute, on the other hand, uses ICMP Echo 
messages, see:
http://technet.microsoft.com/en-us/library/bb491018.aspx

So I think maybe the draft should explain that Traceroute is an application 
that has more than one implementation, and that it can be implemented with 
either UDP or ICMP messages. Does this make sense?

Thanks.
Tal.

-----Original Message-----
From: [email protected] [mailto:[email protected]] 
Sent: Saturday, July 23, 2011 2:44 PM
To: [email protected]
Cc: [email protected]
Subject: (partial) review of draft-ietf-opsawg-oam-overview-05.txt

I am the assigned Gen-ART reviewer for this draft. For background on Gen-ART, 
please see the FAQ at <http://wiki.tools.ietf.org/area/gen/trac/wiki/GenArtfaq>.

Please resolve these comments along with any other Last Call comments you may 
receive.

Document: draft-ietf-opsawg-oam-overview-05.txt
Reviewer: Francis Dupont
Review Date: 20110723
IETF LC End Date: 20110720
IESG Telechat date: unknown

Summary: Not Ready

Major issues: 4.1 page 13 explains the use of ICMP in Traceroute: this is 
plainly wrong: ICMP can't be used in this way because no ICMP error can be 
triggered by an ICMP. BTW:
 - traceroute uses UDP probes just for this reason
 - I put this as a major issue because this must not be published in
  a document from the IETF

Regards

[email protected]

PS: I stopped the review at this point but I'll resume and send a full review 
ASAP (I like the document, OAM is something really useful and to try to 
organize it a great idea).
_______________________________________________
Gen-art mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/gen-art

Reply via email to