Opened 8 years ago
Closed 8 years ago
#189 closed task (fixed)
SRV protocol registry
Reported by: | jouni.nospam@… | Owned by: | jouni.nospam@… |
---|---|---|---|
Priority: | minor | Milestone: | |
Component: | dynamic-discovery | Version: | |
Severity: | - | Keywords: | |
Cc: |
Description
Do we actually need this registry asked for in IANA considerations?
This document requests the creation of a new IANA registry named
"RADIUS/TLS SRV Protocol Registry" with the following initial
entries:
o _tcp
o _udp
One is already asking them as part of the service name or port number assignment request. Thus those are "registered" for SRV usage implicitly, right?
Change History (3)
comment:1 Changed 8 years ago by stefan.winter@…
comment:2 Changed 8 years ago by stefan.winter@…
In fact, I didn't want to leave this out completely, so added my reasoning in an extra paragraph:
"This specification makes use of the SRV Protocol identifiers "_tcp"
and "_udp" which are mentioned as early as [RFC2782] but do not
appear to be assigned in an actual registry. Since they are in wide-
spread use in other protocols, this specification refrains from
requesting a new registry "RADIUS/TLS SRV Protocol Registry" and
continues to make use of these tags implicitly."
comment:3 Changed 8 years ago by jouni.nospam@…
- Resolution set to fixed
- Status changed from new to closed
I am OK with the proposed solution in -12.
This is about the "Proto" field for the SRV RR. I see that the original RFC2782 already uses _tcp and _udp in its examples and considers them useful, but this has never made its way into an actual registry entry at IANA as far as I can see.
Other (new) protocol tags have been defined properly, e.g. https://www.iana.org/assignments/im-srv-labels/im-srv-labels.xhtml#im-srv-labels-1
In the absence of an existing IANA registry specifically for _tcp and _udp, I was trying to create one.
If we can consider those two existing by legacy or via some implicit rule (which I still don't "get" after reading your comment), I can also drop the paragraph.
For the sake of releasing -12 before the cutoff, I'll comment out the section and we can think plentiful about whether it should rather be in at a later point.