Merge pull request #26 from swisspost/develop

PR for release

README.md

@@ -4,7 +4,7 @@ vertx-cluster-watchdog
 [![Build Status](https://travis-ci.com/swisspush/vertx-cluster-watchdog.svg?branch=master)](https://travis-ci.com/swisspush/vertx-cluster-watchdog)
 [![codecov](https://codecov.io/gh/swisspost/vertx-cluster-watchdog/branch/master/graph/badge.svg?token=nbrYxHDMmJ)](https://codecov.io/gh/swisspost/vertx-cluster-watchdog)
 
-Checks if all your hazelcast cluster members are receiveing published messages over the bus.
+Checks if all your hazelcast cluster members are receiving published messages over the bus.
 
 How to run the watchdog
 -----------------------
@@ -63,3 +63,22 @@ Tests
 -----
 
 The tests try to simulate the cluster with multiple instances of the verticle. The amount of cluster members is injected over the config. 
+
+Micrometer metrics
+------------------
+When enabled, `vertx-cluster-watchdog` is monitored with micrometer. The following metrics are available:
+* cluster_watchdog_members
+* cluster_watchdog_members_responded
+
+Example metrics:
+
+```
+# HELP cluster_watchdog_members Amount of members visible to the cluster
+# TYPE cluster_watchdog_members gauge
+cluster_watchdog_members 2.0
+# HELP cluster_watchdog_members_responded Amount of cluster members responded when accessed
+# TYPE cluster_watchdog_members_responded gauge
+cluster_watchdog_members_responded 2.0
+```
+
+To enable the metrics, set a `MeterRegistry` instance by calling `setMeterRegistry(MeterRegistry meterRegistry)` method in `ClusterWatchdog` class.

pom.xml

@@ -2,7 +2,7 @@
     <modelVersion>4.0.0</modelVersion>
     <groupId>org.swisspush</groupId>
     <artifactId>cluster-watchdog</artifactId>
-    <version>3.1.1-SNAPSHOT</version>
+    <version>3.1.2-SNAPSHOT</version>
     <name>cluster-watchdog</name>
     <packaging>jar</packaging>
     <description>
@@ -79,6 +79,11 @@
             <artifactId>slf4j-simple</artifactId>
             <version>${slf4j.version}</version>
         </dependency>
+        <dependency>
+            <groupId>io.micrometer</groupId>
+            <artifactId>micrometer-core</artifactId>
+            <version>${micrometer.version}</version>
+        </dependency>
         <!-- TEST dependencies -->
         <dependency>
             <groupId>junit</groupId>
@@ -399,6 +404,7 @@
     <properties>
         <vertx.version>4.5.2</vertx.version>
         <slf4j.version>2.0.10</slf4j.version>
+        <micrometer.version>1.12.13</micrometer.version>
         <commons-lang.version>2.6</commons-lang.version>
         <commons-io.version>2.15.1</commons-io.version>
         <commons-collections4.version>4.4</commons-collections4.version>

src/main/java/org/swisspush/vertx/cluster/ClusterWatchdog.java

@@ -1,5 +1,7 @@
 package org.swisspush.vertx.cluster;
 
+import io.micrometer.core.instrument.Gauge;
+import io.micrometer.core.instrument.MeterRegistry;
 import io.vertx.core.AbstractVerticle;
 import io.vertx.core.Handler;
 import io.vertx.core.Promise;
@@ -11,6 +13,7 @@
 
 import java.text.SimpleDateFormat;
 import java.util.*;
+import java.util.concurrent.atomic.AtomicLong;
 
 public class ClusterWatchdog extends AbstractVerticle {
 
@@ -32,6 +35,9 @@ public class ClusterWatchdog extends AbstractVerticle {
     private Map<String,List<JsonObject>> healthCheckResponses;
     private ClusterWatchdogHttpHandler clusterWatchdogHttpHandler;
 
+    private final AtomicLong atomicClusterMemberCountRequired = new AtomicLong(0);
+    private final AtomicLong atomicClusterMemberRespondersCount = new AtomicLong(0);
+
     @Override
     public void start(Promise<Void> startPromise) {
 
@@ -50,6 +56,7 @@ public void start(Promise<Void> startPromise) {
         } else {
             clusterMemberCount = clusterMemberCountFromConfig;
         }
+        atomicClusterMemberRespondersCount.set(0);
 
         int resultQueueLength = config.getInteger("resultQueueLength", 100);
         log.info("ClusterWatchdog used resultQueueLength: " + resultQueueLength);
@@ -111,6 +118,15 @@ public void start(Promise<Void> startPromise) {
         });
     }
 
+    public void setMeterRegistry(MeterRegistry meterRegistry) {
+        if(meterRegistry != null) {
+            Gauge.builder("cluster.watchdog.members", atomicClusterMemberCountRequired, AtomicLong::get)
+                    .description("Amount of members visible to the cluster").register(meterRegistry);
+            Gauge.builder("cluster.watchdog.members.responded", atomicClusterMemberRespondersCount, AtomicLong::get)
+                    .description("Amount of cluster members responded when accessed").register(meterRegistry);
+        }
+    }
+
     class ClusterCheckHandler implements Handler<Long> {
 
         public void handle(Long event) {
@@ -135,6 +151,8 @@ public void handle(Long event) {
                 return;
             }
 
+            atomicClusterMemberCountRequired.set(clusterMemberCount);
+
             // publish the broadcast event which will us get the response of all the registered handlers
             eb.publish(BROADCAST, testpayload);
 
@@ -148,6 +166,9 @@ public void handle(Long event) {
                 watchdogResult.time = time;
                 watchdogResult.verticleId = uniqueId;
                 watchdogResult.clusterMemberCount = clusterMemberCount;
+
+                atomicClusterMemberRespondersCount.set(responses != null ? responses.size() : 0);
+
                 if(responses == null) {
                     log.error("ClusterWatchdog found no responses for timestamp: " + timestamp);
                     watchdogResult.status = ClusterHealthStatus.INCONSISTENT;