Re: [tsvwg] draft-ietf-udp-options issues from IETF 104

["Black, David" <[e-mail]>]

Inline …

Thanks, --David

From: Joe Touch <[e-mail]>
Sent: Friday, July 12, 2019 10:43 AM
To: Black, David
Cc: Tom Herbert; tsvwg
Subject: Re: [tsvwg] draft-ietf-udp-options issues from IETF 104


[EXTERNAL EMAIL]



On Jul 12, 2019, at 7:09 AM, Black, David <[e-mail]<mailto:[e-mail]>> wrote:

I think I’m over on the “FORCE” side of that dichotomy (i.e., I don’t agree with “easily optional”), although we can split hairs on what degree of “FORCE” is appropriate.   My underlying reasons are these two:
                1) strong "running code" evidence of what breaks when the OCS is omitted,
                2) helps defend against … possibility [of] … other uses of surplus space

IMHO, OCS being the “default” is a good starting point, but … my current view is that on its own, item 1) justifies “MUST implement/SHOULD use” language for OCS (and “SHOULD use” is complementary to and stronger than “default”) and setting the space aside for OCS, at least when LITE is not present.  That view is reinforced by item 2), and I note neither of those two items are within the control of the endpoint implementer or application that uses the endpoint’s UDP option support.

Also note that the “running code” evidence is based on an error. That’s why I worry about designing overhead we live with forever only to overcome a bug that may be ephemeral.
[David>] Albeit a widely deployed error.  I acknowledge the worry, and suggest that it deserves further discussion on what can reasonably be expected in deployed Internet infrastructure.

That said, again, is there some reason you think we CANNOT allow the user to turn off OCS (in this case, set to zero)? Again, this is UDP. The risk - and the decision - is up to the user, IMO.
[David>] Asked in that form, I think my “SHOULD use” is in rough agreement with your views.  I suspect that we may not agree on the appropriate tone/emphasis of the text, in that I believe that “SHOULD use” provides an important warning to those who are inclined not to use OCS, and ought to be backed by more specific discussion of the potential things that can go wrong.   I’m also concerned that this discussion may be oversimplifying a subtle/complex area – e.g., RFC 8085 devotes about 3 pages of text to this topic, including UDP-Lite and IPv6 considerations for UDP checksum usage.

Joe