Use a smarter algorithm for choosing MTU discovery probe sizes.

[tinc] / src / net_packet.c

diff --git a/src/net_packet.c b/src/net_packet.c
index 9bebca4..2f15bb4 100644 (file)
--- a/src/net_packet.c
+++ b/src/net_packet.c
@@ -157,10 +157,9 @@ static void udp_probe_h(node_t *n, vpn_packet_t *packet, length_t len) {
                        try_fix_mtu(n);
                }
 
                        try_fix_mtu(n);
                }
 

-               /* Calculate RTT and bandwidth.
+               /* Calculate RTT.

                   The RTT is the time between the MTU probe burst was sent and the first
                   The RTT is the time between the MTU probe burst was sent and the first

-                  reply is received. The bandwidth is measured using the time between the
-                  arrival of the first and third probe reply (or type 2 probe requests).
+                  reply is received.

                 */
 
                struct timeval now, diff;
                 */
 
                struct timeval now, diff;


@@ -180,12 +179,7 @@ static void udp_probe_h(node_t *n, vpn_packet_t *packet, length_t len) {
                if(n->probe_counter == 1) {
                        n->rtt = diff.tv_sec + diff.tv_usec * 1e-6;
                        n->probe_time = probe_timestamp;
                if(n->probe_counter == 1) {
                        n->rtt = diff.tv_sec + diff.tv_usec * 1e-6;
                        n->probe_time = probe_timestamp;

-               } else if(n->probe_counter == 3) {
-                       /* TODO: this will never fire - we're not sending batches of three anymore. */
-                       struct timeval probe_timestamp_diff;
-                       timersub(&probe_timestamp, &n->probe_time, &probe_timestamp_diff);
-                       n->bandwidth = 2.0 * probelen / (probe_timestamp_diff.tv_sec + probe_timestamp_diff.tv_usec * 1e-6);
-                       logger(DEBUG_TRAFFIC, LOG_DEBUG, "%s (%s) RTT %.2f ms, burst bandwidth %.3f Mbit/s, rx packet loss %.2f %%", n->name, n->hostname, n->rtt * 1e3, n->bandwidth * 8e-6, n->packetloss * 1e2);
+                       logger(DEBUG_TRAFFIC, LOG_DEBUG, "%s (%s) RTT %.2f ms, rx packet loss %.2f %%", n->name, n->hostname, n->rtt * 1e3, n->packetloss * 1e2);

                }
        }
 }
                }
        }
 }


@@ -928,13 +922,30 @@ static void try_mtu(node_t *n) {
                if(n->maxmtu + 8 < MTU)
                        send_udp_probe_packet(n, n->maxmtu + 8);
        } else {
                if(n->maxmtu + 8 < MTU)
                        send_udp_probe_packet(n, n->maxmtu + 8);
        } else {

-               /* Probes are sent with random sizes between the
-                  lower and upper boundaries for the MTU thus far discovered. */
-               int len = n->maxmtu;
-               if(n->minmtu < n->maxmtu)
-                       len = n->minmtu + 1 + rand() % (n->maxmtu - n->minmtu);
-               send_udp_probe_packet(n, MAX(len, 64));
-
+               /* Decreasing the number of probes per cycle might make the algorithm react faster to lost packets,
+                  but it will typically increase convergence time in the no-loss case. */
+               const length_t probes_per_cycle = 8;
+
+               /* This magic value was determined using math simulations.
+                  It will result in a 1339-byte first probe, followed (if there was a reply) by a 1417-byte probe.
+                  Since 1417 is just below the range of tinc MTUs over typical networks,
+                  this fine-tuning allows tinc to cover a lot of ground very quickly. */
+               const float multiplier = 0.982;
+
+               const float cycle_position = probes_per_cycle - (n->mtuprobes % probes_per_cycle) - 1;
+               const length_t minmtu = MAX(n->minmtu, 64);
+               const float interval = n->maxmtu - minmtu;
+
+               /* The core of the discovery algorithm is this exponential.
+                  It produces very large probes early in the cycle, and then it very quickly decreases the probe size.
+                  This reflects the fact that in the most difficult cases, we don't get any feedback for probes that
+                  are too large, and therefore we need to concentrate on small offsets so that we can quickly converge
+                  on the precise MTU as we are approaching it.
+                  The last probe of the cycle is always 1 byte in size - this is to make sure we'll get at least one
+                  reply per cycle so that we can make progress. */
+               const length_t offset = powf(interval, multiplier * cycle_position / (probes_per_cycle - 1));
+
+               send_udp_probe_packet(n, minmtu + offset);

                if(n->mtuprobes >= 0)
                        n->mtuprobes++;
        }
                if(n->mtuprobes >= 0)
                        n->mtuprobes++;
        }