reverseproxy: Improve hashing LB policies with HRW (#4724)

* reverseproxy: Improve hashing LB policies with HRW

Previously, if a list of upstreams changed, hash-based LB policies
would be greatly affected because the hash relied on the position of
upstreams in the pool. Highest Random Weight or "rendezvous" hashing
is apparently robust to pool changes. It runs in O(n) instead of
O(log n), but n is very small usually.

* Fix bug and update tests

2 files changed, 46 insertions, 41 deletions

diff --git a/modules/caddyhttp/reverseproxy/selectionpolicies.go b/modules/caddyhttp/reverseproxy/selectionpolicies.go
index 001f7f8..125a07f 100644
--- a/modules/caddyhttp/reverseproxy/selectionpolicies.go
+++ b/modules/caddyhttp/reverseproxy/selectionpolicies.go
@@ -514,21 +514,26 @@ func leastRequests(upstreams []*Upstream) *Upstream {
 	return best[weakrand.Intn(len(best))]
 }
 
-// hostByHashing returns an available host
-// from pool based on a hashable string s.
+// hostByHashing returns an available host from pool based on a hashable string s.
 func hostByHashing(pool []*Upstream, s string) *Upstream {
-	poolLen := uint32(len(pool))
-	if poolLen == 0 {
-		return nil
-	}
-	index := hash(s) % poolLen
-	for i := uint32(0); i < poolLen; i++ {
-		upstream := pool[(index+i)%poolLen]
-		if upstream.Available() {
-			return upstream
+	// Highest Random Weight (HRW, or "Rendezvous") hashing,
+	// guarantees stability when the list of upstreams changes;
+	// see https://medium.com/i0exception/rendezvous-hashing-8c00e2fb58b0,
+	// https://randorithms.com/2020/12/26/rendezvous-hashing.html,
+	// and https://en.wikipedia.org/wiki/Rendezvous_hashing.
+	var highestHash uint32
+	var upstream *Upstream
+	for _, up := range pool {
+		if !up.Available() {
+			continue
+		}
+		h := hash(s + up.String()) // important to hash key and server together
+		if h > highestHash {
+			highestHash = h
+			upstream = up
 		}
 	}
-	return nil
+	return upstream
 }
 
 // hash calculates a fast hash based on s.
diff --git a/modules/caddyhttp/reverseproxy/selectionpolicies_test.go b/modules/caddyhttp/reverseproxy/selectionpolicies_test.go
index 7175f77..aa001e4 100644
--- a/modules/caddyhttp/reverseproxy/selectionpolicies_test.go
+++ b/modules/caddyhttp/reverseproxy/selectionpolicies_test.go
@@ -22,9 +22,9 @@ import (
 
 func testPool() UpstreamPool {
 	return UpstreamPool{
-		{Host: new(Host)},
-		{Host: new(Host)},
-		{Host: new(Host)},
+		{Host: new(Host), Dial: "0.0.0.1"},
+		{Host: new(Host), Dial: "0.0.0.2"},
+		{Host: new(Host), Dial: "0.0.0.3"},
 	}
 }
 
@@ -95,13 +95,13 @@ func TestIPHashPolicy(t *testing.T) {
 	// We should be able to predict where every request is routed.
 	req.RemoteAddr = "172.0.0.1:80"
 	h := ipHash.Select(pool, req, nil)
-	if h != pool[1] {
-		t.Error("Expected ip hash policy host to be the second host.")
+	if h != pool[0] {
+		t.Error("Expected ip hash policy host to be the first host.")
 	}
 	req.RemoteAddr = "172.0.0.2:80"
 	h = ipHash.Select(pool, req, nil)
-	if h != pool[1] {
-		t.Error("Expected ip hash policy host to be the second host.")
+	if h != pool[0] {
+		t.Error("Expected ip hash policy host to be the first host.")
 	}
 	req.RemoteAddr = "172.0.0.3:80"
 	h = ipHash.Select(pool, req, nil)
@@ -117,13 +117,13 @@ func TestIPHashPolicy(t *testing.T) {
 	// we should get the same results without a port
 	req.RemoteAddr = "172.0.0.1"
 	h = ipHash.Select(pool, req, nil)
-	if h != pool[1] {
-		t.Error("Expected ip hash policy host to be the second host.")
+	if h != pool[0] {
+		t.Error("Expected ip hash policy host to be the first host.")
 	}
 	req.RemoteAddr = "172.0.0.2"
 	h = ipHash.Select(pool, req, nil)
-	if h != pool[1] {
-		t.Error("Expected ip hash policy host to be the second host.")
+	if h != pool[0] {
+		t.Error("Expected ip hash policy host to be the first host.")
 	}
 	req.RemoteAddr = "172.0.0.3"
 	h = ipHash.Select(pool, req, nil)
@@ -138,7 +138,7 @@ func TestIPHashPolicy(t *testing.T) {
 
 	// we should get a healthy host if the original host is unhealthy and a
 	// healthy host is available
-	req.RemoteAddr = "172.0.0.1"
+	req.RemoteAddr = "172.0.0.4"
 	pool[1].setHealthy(false)
 	h = ipHash.Select(pool, req, nil)
 	if h != pool[2] {
@@ -147,16 +147,16 @@ func TestIPHashPolicy(t *testing.T) {
 
 	req.RemoteAddr = "172.0.0.2"
 	h = ipHash.Select(pool, req, nil)
-	if h != pool[2] {
-		t.Error("Expected ip hash policy host to be the third host.")
+	if h != pool[0] {
+		t.Error("Expected ip hash policy host to be the first host.")
 	}
 	pool[1].setHealthy(true)
 
 	req.RemoteAddr = "172.0.0.3"
 	pool[2].setHealthy(false)
 	h = ipHash.Select(pool, req, nil)
-	if h != pool[0] {
-		t.Error("Expected ip hash policy host to be the first host.")
+	if h != pool[1] {
+		t.Error("Expected ip hash policy host to be the second host.")
 	}
 	req.RemoteAddr = "172.0.0.4"
 	h = ipHash.Select(pool, req, nil)
@@ -167,29 +167,29 @@ func TestIPHashPolicy(t *testing.T) {
 	// We should be able to resize the host pool and still be able to predict
 	// where a req will be routed with the same IP's used above
 	pool = UpstreamPool{
-		{Host: new(Host)},
-		{Host: new(Host)},
+		{Host: new(Host), Dial: "0.0.0.2"},
+		{Host: new(Host), Dial: "0.0.0.3"},
 	}
 	req.RemoteAddr = "172.0.0.1:80"
 	h = ipHash.Select(pool, req, nil)
-	if h != pool[0] {
-		t.Error("Expected ip hash policy host to be the first host.")
-	}
-	req.RemoteAddr = "172.0.0.2:80"
-	h = ipHash.Select(pool, req, nil)
 	if h != pool[1] {
 		t.Error("Expected ip hash policy host to be the second host.")
 	}
-	req.RemoteAddr = "172.0.0.3:80"
+	req.RemoteAddr = "172.0.0.2:80"
 	h = ipHash.Select(pool, req, nil)
 	if h != pool[0] {
 		t.Error("Expected ip hash policy host to be the first host.")
 	}
-	req.RemoteAddr = "172.0.0.4:80"
+	req.RemoteAddr = "172.0.0.3:80"
 	h = ipHash.Select(pool, req, nil)
 	if h != pool[1] {
 		t.Error("Expected ip hash policy host to be the second host.")
 	}
+	req.RemoteAddr = "172.0.0.4:80"
+	h = ipHash.Select(pool, req, nil)
+	if h != pool[0] {
+		t.Error("Expected ip hash policy host to be the first host.")
+	}
 
 	// We should get nil when there are no healthy hosts
 	pool[0].setHealthy(false)
@@ -252,14 +252,14 @@ func TestURIHashPolicy(t *testing.T) {
 
 	request := httptest.NewRequest(http.MethodGet, "/test", nil)
 	h := uriPolicy.Select(pool, request, nil)
-	if h != pool[0] {
-		t.Error("Expected uri policy host to be the first host.")
+	if h != pool[2] {
+		t.Error("Expected uri policy host to be the third host.")
 	}
 
-	pool[0].setHealthy(false)
+	pool[2].setHealthy(false)
 	h = uriPolicy.Select(pool, request, nil)
 	if h != pool[1] {
-		t.Error("Expected uri policy host to be the first host.")
+		t.Error("Expected uri policy host to be the second host.")
 	}
 
 	request = httptest.NewRequest(http.MethodGet, "/test_2", nil)