> From the man page for bridge (4): > > If an IP packet is too large for the outgoing interface, the bridge will > perform IP fragmentation. This can happen when bridge members have > different MTUs or when IP fragments are reassembled by pf. Non-IP > packets which are too large for the outgoing interface will be dropped. > > Is it somehow possible (maybe with different features?) to fragment a > layer2 frame for reassembly on the other side of a bridge?
If it is IP, it will be re-assembled at the other end. pf's fragment cache can even help this further. It works, and if the network you use is (a) short [for instance, a metro network], (b) non-lossy and (c) non-reordering, you will find that it works fantastically. > This would > be for use with MPLS, so using pf's scrub directive would not be > applicable. The desired behavior would be to take in 1500 bytes of IP > payload with an added MPLS label and transport this inside a gif/ipsec > tunnel over the internet (with a wan-link MTU of 1500). For mpls, I cannot comment.