On Tue, Dec 17, 2013 at 05:14:47PM +0100, Florent Bautista wrote: > More informations : > > root@node1:~# tinc -n coppint info node2 > Node: sllm2 > Reachability: directly with UDP > PMTU: 1451 > > root@node2:~# tinc -n coppint info node1 > Node: sllm1 > Reachability: directly with UDP > PMTU: 1518 > > Why PMTU is different ?? They are on a same switch, with same network > configuration (MTU 1500)...
Indeed, they should both be smaller than 1500 (because of the overhead of
tunneling the packets). If one node thinks the PMTU is 1518 that means packets
will be fragmented, which might very well explain the much lower throughput you
have observed. Could you run tinc with the options "-d5 -D", so you get all
debug messages in a terminal, and then check whether both nodes send PMTU
probes to each other, and whether there really are MTU probes larger than 1500
being sent AND responded to?
--
Met vriendelijke groet / with kind regards,
Guus Sliepen <[email protected]>
signature.asc
Description: Digital signature
_______________________________________________ tinc mailing list [email protected] http://www.tinc-vpn.org/cgi-bin/mailman/listinfo/tinc
