PingTimeout

Etienne Dechamps etienne at edechamps.fr
Thu Jul 25 10:44:47 CEST 2013


On 07/25/2013 12:36 AM, Cédric Lemarchand wrote:
>> If your nodes are getting disconnected because of timeouts, it's
>> because tinc's PING/PONG messages are not getting through on your TCP
>> metaconnections.
> Hum very interesting, so if I *only* shape the tinc's udp flows (aka
> data channels) between my nodes, tinc daemons will not get disconnected
> by the PingTimeout, even if the data channel become very busy, thus get
> hight latency (except of course in the case of a real links congestion).
> Sounds good on the paper, but what happens with UDP probes ? are they
> only used for MTU path discovery or there is a similar PING/PONG checks ?

UDP probes are used both for discovering MTU and to check if a node is 
reachable over UDP. If tinc doesn't receive any UDP packets from a node 
for a while, it will fall back to indirect communication (i.e. it will 
send UDP packets to an intermediate node which will forward it). If it 
is unable to reach that intermediate node over UDP as well, it will 
ultimately fall back to sending packets over its TCP metaconnections.

So if you shape UDP to the point where it becomes so congested that your 
tinc nodes are unable to talk to each other using UDP, they will adapt 
and send packets over TCP, which is probably not what you want.

-- 
Etienne Dechamps


More information about the tinc mailing list