From: Etienne Dechamps <etienne@edechamps.fr>
Date: Sun, 21 Jul 2013 12:05:42 +0000 (+0100)
Subject: Introduce lightweight PMTU probe replies.
X-Git-Tag: release-1.1pre8~21
X-Git-Url: https://git.tinc-vpn.org/git/browse?a=commitdiff_plain;h=e3c763eae89df9a69bb2d611238ef18f78de311f;hp=e3a4672afb8eb341b380e74b2bf6d098f61c08a3;p=tinc

Introduce lightweight PMTU probe replies.

When replying to a PMTU probe, tinc sends a packet with the same length
as the PMTU probe itself, which is usually large (~1450 bytes). This is
not necessary: the other node wants to know the size of the PMTU probes
that have been received, but encoding this information as the actual
reply length is probably the most inefficient way to do it. It doubles
the bandwidth usage of the PMTU discovery process, and makes it less
reliable since large packets are more likely to be dropped.

This patch introduces a new PMTU probe reply type, encoded as type "2"
in the first byte of the packet, that indicates that the length of the
PMTU probe that is being replied to is encoded in the next two bytes of
the packet. Thus reply packets are only 3 bytes long.

(This also protects against very broken networks that drop very small
packets - yes, I've seen it happen on a subnet of a national ISP - in
such a case the PMTU probe replies will be dropped, and tinc won't
enable UDP communication, which is a good thing.)

Because legacy nodes won't understand type 2 probe replies, the minor
protocol number is bumped to 3.

Note that this also improves bandwidth estimation, as it is able to
measure bandwidth in both directions independently (the node receiving
the replies is measuring in the TX direction) and the use of smaller
reply packets might decrease the influence of jitter.
---

diff --git a/src/net_packet.c b/src/net_packet.c
index 4bbb2d6f..d700fd6b 100644
--- a/src/net_packet.c
+++ b/src/net_packet.c
@@ -179,12 +179,20 @@ void send_mtu_probe(node_t *n) {
 }
 
 static void mtu_probe_h(node_t *n, vpn_packet_t *packet, length_t len) {
-	logger(DEBUG_TRAFFIC, LOG_INFO, "Got MTU probe length %d from %s (%s)", packet->len, n->name, n->hostname);
-
 	if(!packet->data[0]) {
+		logger(DEBUG_TRAFFIC, LOG_INFO, "Got MTU probe request %d from %s (%s)", packet->len, n->name, n->hostname);
+
 		/* It's a probe request, send back a reply */
 
-		packet->data[0] = 1;
+		/* Type 2 probe replies were introduced in protocol 17.3 */
+		if ((n->options >> 24) == 3) {
+			uint8_t* data = packet->data;
+			*data++ = 2;
+			uint16_t len16 = htons(len); memcpy(data, &len16, 2); data += 2;
+		} else {
+			/* Legacy protocol: n won't understand type 2 probe replies. */
+			packet->data[0] = 1;
+		}
 
 		/* Temporarily set udp_confirmed, so that the reply is sent
 		   back exactly the way it came in. */
@@ -194,6 +202,16 @@ static void mtu_probe_h(node_t *n, vpn_packet_t *packet, length_t len) {
 		send_udppacket(n, packet);
 		n->status.udp_confirmed = udp_confirmed;
 	} else {
+		length_t probelen = len;
+		if (packet->data[0] == 2) {
+			if (len < 3)
+				logger(DEBUG_TRAFFIC, LOG_WARNING, "Received invalid (too short) MTU probe reply from %s (%s)", n->name, n->hostname);
+			else {
+				uint16_t probelen16; memcpy(&probelen16, packet->data + 1, 2); probelen = ntohs(probelen16);
+			}
+		}
+		logger(DEBUG_TRAFFIC, LOG_INFO, "Got type %d MTU probe reply %d from %s (%s)", packet->data[0], probelen, n->name, n->hostname);
+
 		/* It's a valid reply: now we know bidirectional communication
 		   is possible using the address and socket that the reply
 		   packet used. */
@@ -203,7 +221,7 @@ static void mtu_probe_h(node_t *n, vpn_packet_t *packet, length_t len) {
 		/* If we haven't established the PMTU yet, restart the discovery process. */
 
 		if(n->mtuprobes > 30) {
-			if (len == n->maxmtu + 8) {
+			if (probelen == n->maxmtu + 8) {
 				logger(DEBUG_TRAFFIC, LOG_INFO, "Increase in PMTU to %s (%s) detected, restarting PMTU discovery", n->name, n->hostname);
 				n->maxmtu = MTU;
 				n->mtuprobes = 10;
@@ -218,10 +236,10 @@ static void mtu_probe_h(node_t *n, vpn_packet_t *packet, length_t len) {
 
 		/* If applicable, raise the minimum supported MTU */
 
-		if(len > n->maxmtu)
-			len = n->maxmtu;
-		if(n->minmtu < len)
-			n->minmtu = len;
+		if(probelen > n->maxmtu)
+			probelen = n->maxmtu;
+		if(n->minmtu < probelen)
+			n->minmtu = probelen;
 
 		/* Calculate RTT and bandwidth.
 		   The RTT is the time between the MTU probe burst was sent and the first
@@ -238,7 +256,7 @@ static void mtu_probe_h(node_t *n, vpn_packet_t *packet, length_t len) {
 			n->rtt = diff.tv_sec + diff.tv_usec * 1e-6;
 			n->probe_time = now;
 		} else if(n->probe_counter == 3) {
-			n->bandwidth = 2.0 * len / (diff.tv_sec + diff.tv_usec * 1e-6);
+			n->bandwidth = 2.0 * probelen / (diff.tv_sec + 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);
 		}
 	}
diff --git a/src/protocol.h b/src/protocol.h
index 93ef3f23..1a685554 100644
--- a/src/protocol.h
+++ b/src/protocol.h
@@ -26,7 +26,7 @@
 /* Protocol version. Different major versions are incompatible. */
 
 #define PROT_MAJOR 17
-#define PROT_MINOR 2 /* Should not exceed 255! */
+#define PROT_MINOR 3 /* Should not exceed 255! */
 
 /* Silly Windows */