On 03/06/2015 17:22, Joe Abley wrote:

> I think there's a baked-in expectation that OPT pseudo-RR is included in
> every DNS message, not on every connection (where the transport is
> connection-oriented).

Joe,

Part of the reason this came up is this text in
draft-ietf-edns-tcp-keepalive:

"DNS clients MAY include the edns-tcp-keepalive option in the first
 query sent to a server using TCP transport to signal their desire
 that that specific TCP session be used for multiple DNS transactions."

> While the conventional case of resolver-talks-to-authority makes this
> seem like a highly pedantic observation, we can never be sure of what is
> happening behind the curtain; ALGs that bridge between transports
> (providing a UDP interface on one side but managing a TCP connection
> pool on the other, for example) exist, for example.

Yes, an ALG such as this was the most compelling (albeit hypothetical)
reason that I could think of for not making this assumption, but then
EDNS is *supposed* to be hop-by-hop.

kind regards,

Ray

_______________________________________________
DNSOP mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dnsop

Reply via email to