Saku Ytti wrote:
> Any opinions on SVCINFO[0]. I was really big fan of SRV until I read the
> draft, it made really compelling arguments to me.
> Goal should be that 1 query returns all the information client
> needs to decide
The problem of URI and SVCINFO RRs is that they return too much
information.
With SVCINFO, browsers don't know how to interprete "InstanceId".
Note that port number returned from SRV is already too much.
Though SVCINFO draft says:
There are two concerns with SRV. First, one
must indicate the transport protocol as part of the QNAME. This
means that discovery of multiple transport protocols requires
multiple queries.
for browsers,
_proto.example.com. SRV ...
should be better. The second problem in the draft merely
means that, with SRVINFO, when there are a lot of
SRVINFO RRs at "example.com", "example.com" will be a zone.
Masataka Ohta
_______________________________________________
DNSOP mailing list
DNSOP@ietf.org
https://www.ietf.org/mailman/listinfo/dnsop
