Re: [tcpm] draft-ietf-tcpm-tcp-uto-02

Gorry,

Thanks so much for your comments. You'll find my responses inline....

>It seems you have responded to most (if not all) my substantial 
>comments,  and also tightened the
>language, I like these improvements, and to me, this seems much 
>safer in this respect.
>
>I would like to check one point:
>Section 3.1:
>
>/Very short user timeout values can affect TCP transmissions over
>high-delay paths.  If the user timeout occurs before an
>acknowledgment for an outstanding segment arrives, possibly due to ..../
>
>- This rev of the draft also includes a change to mandate 
>the  minimum value to be at least the RTO value, which should at 
>least help in this case?

Agreed. However, the document argues to enforce a minimum *L_LIMIT* 
value of at least one RTO, with L_LIMIT being a parameter of the 
RECOMMENDED (ie, not mandatory) scheme for selecting the user timeout.

So two possible ways to fix this would be:

a) State that the minimum allowed value for the user timeout is one 
RTO (in any case.... not just in the recommended scheme)
b) Make the recommended scheme mandatory.

I would personally argue in favour of (a). But well, opinions are welcome.

>End of section 1:
>
>I think phrase below is good guidance, but could be strengthened:
>/If such a system-wide toggle were provided, it would default to "off".
>
>Perhaps some text something like:
>
>An implementation that provides a system-wide toggle, should assign 
>the default value to disable this feature.

How about:
"An implementation that provides a system-wide toggle, SHOULD assign 
the default value to disable this feature"
(ie, RFC2119 speak)

>/It must be noted that the two endpoints of the connection will not
>necesarilly adopt the same user timeout./
>
>- Could be better worded, but agreed.

Any suggestions?

Thanks so much!

Kindest regards,

--
Fernando Gont
e-mail: fernando@gont.com.ar || fgont@acm.org
PGP Fingerprint: 7809 84F5 322E 45C7 F1C9 3945 96EE A9EF D076 FFF1

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