This sample application implements a simple caching solution using a HashMap.

For example, see following code.

Map<Key, String> map = new HashMap<>();
while (true) {
    for (int i = 0; i < max; i++) {
        Key key = new Key(i);
        if (!map.containsKey(key)) {
            map.put(key, "Number:" + i);
        }
    }
}

The map is not expected to grow beyond max value as all keys will be there in the map after the first while loop.

However, if the Key class does not contain a proper equals() implementation along with the hashCode() implementation, the key elements will be keep getting added as the !map.containsKey(key) will always return false. This is a good example of a memory leak and this sample was inspired by the blog post: "How to create a memory leak"

The application will throw Out of Memory error after some time when you run following command

java -Xms1g -Xmx1g -XX:+PrintGCDetails -XX:+PrintGC -XX:+PrintGCDateStamps -Xloggc:gc.log -XX:+HeapDumpOnOutOfMemoryError -XX:HeapDumpPath="/tmp/heap-dump.hprof" -XX:+UnlockDiagnosticVMOptions -XX:+DebugNonSafepoints -XX:+UnlockCommercialFeatures -XX:+FlightRecorder -XX:StartFlightRecording=settings=profile,duration=5m,name=MemoryLeak,filename=memoryleak.jfr -XX:FlightRecorderOptions=loglevel=info -jar target/memoryleak.jar

In Memory -> Memory tab, you should see a high memory growth rate.

Also see Memory -> Garbage Collections tab and check the frequency of GC events.

You can analyze the GC logs using GCViewer

Run the application with --key-type GOOD parameter, which will use a Key object with proper equals() implementation.