[Standards-JIG] [Fwd: [interop] day 1 summary]

[Peter Saint-Andre]

Albert Holm wrote:
> Thursday 27 July 2006 22:53 skrev Gary Burd:
>> On 7/27/06, Albert Holm <albert+sjig at cdr.se> wrote:
>>> I would believe that most DNS hosts that believe SRV is
>>> difficult/hard/not supported also believe that everything else except MX,
>>> A, CNAME and maybe even NS is equally difficult/hard/not supported and
>>> thus does not support TXT.
>> We found that many hosts support TXT records. We suspect that TXT
>> support is driven by SPF.
> 
> I am still not convinced it is a good way and currently SRV is not required 
> which might seem discouraging to future implementors (even though current 
> software seems to handle it).

Not required by what? RFC 3920?

> It feels a bit like if SMTP would use formulations like SHOULD when it comes 
> to using the MX record as the first way to resolve the domain name.
> 
> Ensuring that no fallback mechanism is used if SRV is found is something that 
> I believe the XMPP RFC should define. SMTP does in section 5 of 2821 for 
> example. 

I think that having no fallback mechanism would be a bad idea, given
that many (most?) deployments will not have access to SRV records.

> I also think that SRV MUST be the first way to resolve the domain for XMPP 
> just like MX is the first way to resolve the mail host for a domain. Right 
> now it is only SHOULD/recommendation.

I agree that SRV MUST be attempted first.

Peter

-- 
Peter Saint-Andre
Jabber Software Foundation
http://www.jabber.org/people/stpeter.shtml

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 7358 bytes
Desc: S/MIME Cryptographic Signature
Url : http://mail.jabber.org/pipermail/standards/attachments/20060803/55ba9b04/smime.bin