> 2- Section 4: Related to the "leaf port".
> This draft removed the YANG default statement for the port leaf
> (previously default 49) to accommodate the future IANA-assigned port
> for TACACS+ over TLS, which is currently referenced as TBD in the
> description. Thus, my understanding is the following:
>
> * In the absence of a "default" or "mandatory true" statements, the
> port leaf becomes optional;
>
> * YANG does not interpret prose in the description field for default
> behavior;
>
> * Therefore, if the port is not configured explicitly by the user,
> the system behavior is undefined by the schema, and implementations
> may diverge.
>
> * Is this interpretation correct?
>
> Thus, to avoid ambiguity at runtime and to enforce consistent
> behavior across implementations, what about?
>
> leaf port {
>
> type inet:port-number;
>
> mandatory true;
>
> "The port number of TACACS+ server. Default port number for legacy
> TACACS+ is
> 49, while it is TBD for TACACS+TLS.";
>
> }
>
> OR
>
> leaf port {
>
> type inet:port-number;
>
> default "TBD"; // Replace with the actual IANA-assigned port
>
> "The port number of TACACS+ server. Default port number for legacy
> TACACS+ is
> 49, while it is TBD for TACACS+TLS.";
>
> }
[Med] You have a valid point here. Went with a default with TBD.
[JMC] IMHO I’d like to see “mandatory true” since this is an extension on top of a well-entrenched protocol and defaulting to the new port feels
like it may cause some troubleshooting frustration.
Joe
|
--
last-call mailing list -- last-call@xxxxxxxx
To unsubscribe send an email to last-call-leave@xxxxxxxx