^^^^^The basic idea is that a source host initially assumes that the PMTU of a path is the (known) MTU of its first hop, and sends all datagrams on that path with the DF bit set.
And this is where all the problems start. This flies directly in the face of the the robustness principle. If the host would just set the DF bit in one in 1000, one in 100 or one in 10 packets, we wouldn't be in this mess.
Note that the RFC doesn't say the DF bit must be set on ALL packets. But then, it also doesn't say this shouldn't be done either.
_______________________________________________ Ietf@xxxxxxxx https://www1.ietf.org/mailman/listinfo/ietf