On 14/11/2012 08:09, Hannes Frederic Sowa wrote:
> On 11/13/2012 12:27 AM, Mark Andrews wrote:
>> In message <[email protected]>, Hannes Frederic Sowa writes:
>>> On 11/07/2012 02:30 AM, Mark Andrews wrote:
>>>> Should setting IPV6_USE_MIN_MTU to one (1) result in fragmented TCP
>>>> packets?
>>>> Should setting IPV6_DONTFRAG to one (1) work on TCP sockets?
>>> As I have not read otherwise I would treat IPV6_USE_MIN_MTU as
>>> transport-agnostic.
>> The problem is that there are at least 2 OS's with TCP layers that
>> do not use IPV6_USE_MIN_MTU when calculating the TCP segment size to
>> send. I would have thought that this would have been obvious to do.
>>
>> Do we need update RFC3542 to say that IPV6_USE_MIN_MTU is to be used
>> when calculating the TCP segment size?
>
> Do we know of any use cases for IPV6_USE_MIN_MTU on TCP sockets?
I think the use case is when there is a user option to use 1280 bytes
because of a path MTU limitation due to a tunnel, *and* PMTUD and/or
fragmentation fails. (Read draft-taylor-v6ops-fragdrop and be scared.)
So using it to trigger fragmentation is pointless; to me the only thing
that makes sense is if it reduces the MSS accordingly. I've seen
that fail in the wild, when running 6to4.
> Will
> DNS server software vendors would use it as fallback to UDP?
Huh? They are in bad trouble with UDP if running DNSSEC and fragmentation
doesn't work.
Brian
>
>>> The usage of IPV6_DONTFRAG is only specified on UDP and raw sockets (see
>>> RFC3542 11.2).
>> Which is short sighted. There is no reason not to apply it to all
>> transports as all transports have to handle PTB and this is just
>> a locally generated PTB.
>>
>> And while we are talking about RFC3542 how to we get POSIX updated to
>> include the advanced half of the API as it is a pain in the butt for
>> application developers.
>
> Seems this is already the case. I looked at some usees of
> IPV6_USE_MIN_MTU and i.e. bind is already working around some strange
> setsockopt interactions with other options on FreeBSD (bind specifies
> IPV6_USE_MIN_MTU as a per packet control message there).
>
--------------------------------------------------------------------
IETF IPv6 working group mailing list
[email protected]
Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------
