[bind10-dev] Solaris doesn't honor IPV6_USE_MIN_MTU?

[Shane Kerr]

Michal,

On Tuesday, 2012-02-28 15:51:59 +0100, 
Michal 'vorner' Vaner <michal.vaner at nic.cz> wrote:
> Hello
> 
> On Tue, Feb 28, 2012 at 03:42:32PM +0100, Jelte Jansen wrote:
> > > Hmm, so we need to work around this somehow. Maybe not using the
> > > IPV6_MTU (setting the size explicitly) as a fallback, but setting
> > > it whenever it is available? So we set both if we have a chance
> > > and hope at least one of them work?
> > > 
> > 
> > would not help in this case, IPV6_MTU isn't defined on solaris.
> > 
> > Btw, it's not the only flag that gets ignored on setsockopt
> 
> Ah, stupid solaris. We still have the raw interface and could forge
> the packets ourself, but I'd rather avoid that. Could we just say if
> it doesn't work on solaris, it's their problem? O:-)
> 
> Considering we did a mostly copy of what both bind9 and NSD does, and
> nobody complained to them (I assume), does the problem actually exist?

The thing about DNS is that it is so damn awesome that it tends to work
even in the face of problems.

So, in this case, if a server was replying and the answer did not make
it back because of MTU issues, the client would try another server. If
the client is a BIND 9 resolver, and runs out of other servers I think
that it will try a lower packet size on EDNS(0) and then the server
should limit the reply packet to one that fits in the client-specified
space.

Given the small portion of traffic that is IPv6-enabled and the fact
that it will actually work eventually, it is quite possible that nobody
ever noticed that this option doesn't work on Solaris. :)

--
Shane
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <https://lists.isc.org/pipermail/bind10-dev/attachments/20120228/3bf8b0f4/attachment.bin>