On 20 Jan 2010, at 09:36, Lars Eggert wrote:
> thanks for the summary.
>
> On 2010-1-20, at 11:28, Arnt Gulbrandsen wrote:
>
>> Lars Eggert writes:
>>> I'm not on the yam list and this is the first message in this thread
>>> that I was CC'ed on. It seems like you are suggesting changes to
>>> draft-ietf-tsvwg-iana-ports, but due to the lack of context I'm at a
>>> loss as to what they are...
>>
>> Tthe thead in a nutshell: The iana-ports document says "one port for
>> each purpose" etc.
>
> yes - as a guideline for *new* assignments. No existing assignment is
> modified by this document. (Maybe we need to explicitly say this if we aren't
> already.)
>
>> IMAP, POP3 and perhaps SMTP, however, exist in two.
>> 993 ("imaps") and 143 ("imap") for IMAP. Everyone on the list dislikes
>> port 993. The question is whether port 993 should survive in the IANA
>> registry.
>
> It should and it will.
>
>> My suggestion for the iana-ports document is to permit more than one
>> port in those cases where that's currently deployed, and apply the
>> one-port-per-purpose rule only to new allocations.
>
> Exactly. The intent is not to revisit past assignments and check whether they
> comply with the new rules.
>
>> My rationale for that is that the extra ports/service names aren't
>> really free. You can't use port 993 or service name imaps for anything
>> else, and IMO that's reason enough to keep it in the registry.
>
> Exactly.
Deployment consideration is AFAIK the one and only reason to specify this
usage. I'm not asking for a definitive answer, but does the revocation
procedure not give us a reason to discourage its usage altogether using at
least some warning language? It might be a flight of fancy, but if in a
decade's time the world has migrated to IPSec, for instance, we'll be left with
lots of obsolete port assignments and no documentation for them, which IMO is
not so good. Would like to think that, although we must not break existing
deployments, we should try to observe current and future reservation guidelines.
Cheers,
Sabahattin
_______________________________________________
yam mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/yam
