[bind10-dev] #1534, IPV6_USE_MIN_MTU and similar

[Mark Andrews]

In message <m2boozan8k.wl%jinmei at isc.org>, JINMEI Tatuya / =?ISO-2022-JP?B?GyRC
P0BMQEMjOkgbKEI=?= writes:
> At Wed, 15 Feb 2012 12:17:39 +0100,
> Michal 'vorner' Vaner <michal.vaner at nic.cz> wrote:
> 
> > > IMO using the minimum MTU for UDP/IPv6 is the most reasonable default
> > > behavior (I don't know how a tftp could be an exception, but even if
> > > it could be it seems to be very unlikely we have it in BIND 10), so it
> > > makes more sense to do it by default.  If an application specifically
> > > wants to perform PMTU discovery for UDP/IPv6 itself (IMO which will be
> > > very, very unlikely), that specific application can then disable
> > > USE_MIN_MTU, set IPV6_RECVPATHMTU and follow MTU change from ancillary
> > > data, etc (again, only very, very few apps would bother to do that).
> > 
> > Well, I remember Shane telling us about dream where it would contain tftp a
> lso,
> > so it could do netbooting completely (with assigning names, addresses, prov
> iding
> > the correct kernels).
> > 
> > And tftp usually runs on one network (I don't think tftp is ever used for
> > anything else but network booting), so the fragment size would not change o
> n the
> > path at all, and having larger packets would be better for performance.
> 
> I don't remember the dream, but in any case the "dream" sounds nearly
> equal to "very, very, unlikely".  And, as far as I know (some?) Cisco
> phones use tftp to reach a server in a remote network.
> 
> And, most important, the question in this context is which is the more
> reasonable default, whether to use minimal MTU for UDP/IPv6.  Even if
> we really have one minor example usage of UDP/IPv6 that doesn't need
> it, that doesn't automatically support the argument of using it
> optionally since we already have a very important usage that
> requires it, i.e, DNS.

TFTP across a anything but a LAN needs it as well.  PMTUD really doesn't
work well.  TCP can be made to work when ICMP PTB messages don't make it
though.  Other protocols are sometime a lot harder / impossible.
 
> ---
> JINMEI, Tatuya
> Internet Systems Consortium, Inc.
> _______________________________________________
> bind10-dev mailing list
> bind10-dev at lists.isc.org
> https://lists.isc.org/mailman/listinfo/bind10-dev
-- 
Mark Andrews, ISC
1 Seymour St., Dundas Valley, NSW 2117, Australia
PHONE: +61 2 9871 4742                 INTERNET: marka at isc.org