Re: [DNSOP] Priming query transport selection

[Sebastian Castro <[e-mail]>]

[e-mail] wrote:
> 
>> The text in RFC 2671, presented as a hint, could deal to similar issues
>> with the TCP transport for DNS (working to change SHOULD for MUST).
> 
> Can you elaborate on what you mean?
> 
> I presume you're aware of my draft-ietf-dnsext-dns-tcp-requirements ?

Yes, I'm aware of your draft and I meant to support Patrik Falstrom
comment: we have to be very careful with the wording if we don't want to
 be working in future years to update/correct the limit for UDP buffer
size in EDNS.


> 
>> From BIND ARM 9.7.0
>>
>> ----------------------
>> edns-udp-size
>>    Sets the advertised EDNS UDP buffer size in bytes to control the size
>> of packets received.
>>    Valid values are 1024 to 4096 (values outside this range will be
>> silently adjusted)
>> ----------------------
> 
> Yes, that's the one.  I was sat on a train with a flakey 3G connection
> when I sent the last message so couldn't check it, but that confirms my
> recollection.
> 
> I've already submitted to ISC that the choice of value should be left
> entirely to the sysadmin, and not restricted to an arbitrary lower value
> by their software.
> 
> kind regards,
> 
> Ray
> 
> -- 
> Ray Bellis, MA(Oxon) MIET
> Senior Researcher in Advanced Projects, Nominet
> e: [e-mail], t: [phone]
> 

Cheers
Sebastian Castro
NZRS


> 
>