Configure HAProxy to check health every 200 milliseconds and very quickly add/remove (un)healthy nodes

Author: ramonsmits

haproxy.cfg

@@ -28,9 +28,9 @@ listen rabbitmq
         timeout client  3h
         timeout server  3h
         option          clitcpka
-        server          rabbitmq1 rabbitmq1:5672  check inter 5s rise 2 fall 3
-        server          rabbitmq2 rabbitmq2:5672  check inter 5s rise 2 fall 3
-        server          rabbitmq3 rabbitmq3:5672  check inter 5s rise 2 fall 3
+        server          rabbitmq1 rabbitmq1:5672  check inter 200ms rise 1 fall 1
+        server          rabbitmq2 rabbitmq2:5672  check inter 200ms rise 1 fall 1
+        server          rabbitmq3 rabbitmq3:5672  check inter 200ms rise 1 fall 1
 
 listen mgmt
         bind *:15672
@@ -39,6 +39,6 @@ listen mgmt
         timeout client  3h
         timeout server  3h
         option          clitcpka
-        server          rabbitmq1 rabbitmq1:15672  check inter 5s rise 2 fall 3
-        server          rabbitmq2 rabbitmq2:15672  check inter 5s rise 2 fall 3
-        server          rabbitmq3 rabbitmq3:15672  check inter 5s rise 2 fall 3
+        server          rabbitmq1 rabbitmq1:15672  check inter 200ms rise 1 fall 1
+        server          rabbitmq2 rabbitmq2:15672  check inter 200ms rise 1 fall 1
+        server          rabbitmq3 rabbitmq3:15672  check inter 200ms rise 1 fall 1