Develop Reference

RxNetty reuse the connection

I want to use Netflix-Ribbon as TCP client load balancer without Spring Cloud,and i write test code.
public class App implements Runnable
{
public static String msg = "hello world";
public BaseLoadBalancer lb;
public RxClient<ByteBuf, ByteBuf > client;
public Server echo;
App(){
lb = new BaseLoadBalancer();
echo = new Server("localhost", 8000);
lb.setServersList(Lists.newArrayList(echo));
DefaultClientConfigImpl impl = DefaultClientConfigImpl.getClientConfigWithDefaultValues();
client = RibbonTransport.newTcpClient(lb, impl);
}
public static void main( String[] args ) throws Exception
{
for( int i = 40; i > 0; i--)
{
Thread t = new Thread(new App());
t.start();
t.join();
}
System.out.println("Main thread is finished");
}
public String sendAndRecvByRibbon(final String data)
{
String response = "";
try {
response = client.connect().flatMap(new Func1<ObservableConnection<ByteBuf, ByteBuf>,
Observable<ByteBuf>>() {
public Observable<ByteBuf> call(ObservableConnection<ByteBuf, ByteBuf> connection) {
connection.writeStringAndFlush(data);
return connection.getInput();
}
}).timeout(1, TimeUnit.SECONDS).retry(1).take(1)
.map(new Func1<ByteBuf, String>() {
public String call(ByteBuf ByteBuf) {
return ByteBuf.toString(Charset.defaultCharset());
}
})
.toBlocking()
.first();
}
catch (Exception e) {
System.out.println(((LoadBalancingRxClientWithPoolOptions) client).getMaxConcurrentRequests());
System.out.println(lb.getLoadBalancerStats());
}
return response;
}
public void run() {
for (int i = 0; i < 200; i++) {
sendAndRecvByRibbon(msg);
}
}
}
i find it will create a new socket everytime i callsendAndRecvByRibbon even though the poolEnabled is setting to true. So,it confuse me,i miss something?
and there are no option to configure the size of the pool,but hava a PoolMaxThreads and MaxConnectionsPerHost.
My question is how to use a connection pool in my simple code, and what's wrong with my sendAndRecvByRibbon,it open a socket then use it only once,how can i reuse the connection?thanks for your time.
the server is just a simple echo server writing in pyhton3,i comment outconn.close() because i want to use long connection.
import socket
import threading
import time
import socketserver
class ThreadedTCPRequestHandler(socketserver.BaseRequestHandler):
def handle(self):
conn = self.request
while True:
client_data = conn.recv(1024)
if not client_data:
time.sleep(5)
conn.sendall(client_data)
# conn.close()
class ThreadedTCPServer(socketserver.ThreadingMixIn, socketserver.TCPServer):
pass
if __name__ == "__main__":
HOST, PORT = "localhost", 8000
server = ThreadedTCPServer((HOST, PORT), ThreadedTCPRequestHandler)
ip, port = server.server_address
server_thread = threading.Thread(target=server.serve_forever)
server_thread.daemon = True
server_thread.start()
server.serve_forever()
and the pom of mevan,i just add two dependency in IED's auto generated POM.
<dependency>
<groupId>commons-configuration</groupId>
<artifactId>commons-configuration</artifactId>
<version>1.6</version>
</dependency>
<dependency>
<groupId>com.netflix.ribbon</groupId>
<artifactId>ribbon</artifactId>
<version>2.2.2</version>
</dependency>
the code for printing src_port
#Sharable
public class InHandle extends ChannelInboundHandlerAdapter {
public void channelRead(ChannelHandlerContext ctx, Object msg) throws Exception {
System.out.println(ctx.channel().localAddress());
super.channelRead(ctx, msg);
}
}
public class Pipeline implements PipelineConfigurator<ByteBuf, ByteBuf> {
public InHandle handler;
Pipeline() {
handler = new InHandle();
}
public void configureNewPipeline(ChannelPipeline pipeline) {
pipeline.addFirst(handler);
}
}
and change the client = RibbonTransport.newTcpClient(lb, impl);to Pipeline pipe = new Pipeline();client = RibbonTransport.newTcpClient(lb, pipe, impl, new DefaultLoadBalancerRetryHandler(impl));

So, your App() constructor does the initialization of lb/client/etc.
Then you're starting 40 different threads with 40 different RxClient instances (each instance has own pool by default) by calling new App() in the first for loop. To make things clear - the way you spawn multiple RxClient instances here does not allow them to share any common pool. Try to use one RxClient instance instead.
What if you change your main method like below, does it stop creating extra sockets?
public static void main( String[] args ) throws Exception
{
App app = new App() // Create things just once
for( int i = 40; i > 0; i--)
{
Thread t = new Thread(()->app.run()); // pass the run()
t.start();
t.join();
}
System.out.println("Main thread is finished");
}
If above does not help fully (at least it will reduce created sockets count in 40 times) - can you please clarify how exactly do you determine that:
i find it will create a new socket everytime i call sendAndRecvByRibbon
and what are your measurements after you update constructor with this line:
DefaultClientConfigImpl impl = DefaultClientConfigImpl.getClientConfigWithDefaultValues();
impl.set(CommonClientConfigKey.PoolMaxThreads,1); //Add this one and test
Update
Yes, looking at the sendAndRecvByRibbon it seems that it lacks marking the PooledConnection as no longer acquired by calling close once you don't expect any further reads from it.
As long as you expect the only single read event, just change this line
connection.getInput()
to the
return connection.getInput().zipWith(Observable.just(connection), new Func2<ByteBuf, ObservableConnection<ByteBuf, ByteBuf>, ByteBuf>() {
#Override
public ByteBuf call(ByteBuf byteBuf, ObservableConnection<ByteBuf, ByteBuf> conn) {
conn.close();
return byteBuf;
}
});
Note, that if you'd design more complex protocol over TCP, then input bytebuf can be analyzed for your specific 'end of communication' sign which indicates the connection can be returned to the pool.

OkHttp MockWebServer fails to accept connections when restarted

I'm using the OkHttp MockWebServer to mock my server responses for unit tests.
It works great for the first test, but on the 2nd test my client fails with:
Failed to connect to localhost/0:0:0:0:0:0:0:1:63631
This happens even if the 2nd test is exactly the same as the 1st one.
Here's what I'm doing:
#RunWith(RobolectricTestRunner.class)
#Config(shadows = MyClassTest.MyNetworkSecurityPolicy.class,
manifest = "src/main/AndroidManifest.xml",
constants = BuildConfig.class,
sdk = 16)
public class MyClassTest {
private MockWebServer mockServer;
private MyServerApi serverApi;
#Before
public void setUp() throws Exception {
System.out.println("\ntest start");
this.mockServer = new MockWebServer();
this.mockServer.start();
this.serverApi = new MyServerApi(this.mockServer.url("/").toString());
}
#Test
public void testOne() throws Exception {
final String responseBody = // read response from file
this.mockServer.enqueue(new MockResponse().setResponseCode(200).setBody(responseBody));
final Waiter waiter = new Waiter();
this.serverApi.getData("some_id", new Callback<MyResponseData> {
#Override
public void onResponse(final Call<MyResponseData> call, final Response<MyResponseData> response) {
waiter.assertEquals("some_value", response.body().getValue());
waiter.resume();
}
#Override
public void onFailure(final Call<T> call, final Throwable error) {
waiter.fail(error);
}
});
waiter.await();
final RecordedRequest recordedRequest = this.mockServer.takeRequest();
assertEquals("GET", recordedRequest.getMethod());
}
#Test
public void testTwo() throws Exception {
final String responseBody = // read response from file
this.mockServer.enqueue(new MockResponse().setResponseCode(200).setBody(responseBody));
final Waiter waiter = new Waiter();
this.serverApi.getData("some_id", new Callback<MyResponseData> {
#Override
public void onResponse(final Call<MyResponseData> call, final Response<MyResponseData> response) {
waiter.assertEquals("some_value", response.body().getValue());
waiter.resume();
}
#Override
public void onFailure(final Call<T> call, final Throwable error) {
waiter.fail(error);
}
});
waiter.await();
final RecordedRequest recordedRequest = this.mockServer.takeRequest();
assertEquals("GET", recordedRequest.getMethod());
}
#After
public void tearDown() throws Exception {
System.out.println("test end\n");
this.mockServer.shutdown();
}
#Implements(NetworkSecurityPolicy.class)
public static class MyNetworkSecurityPolicy {
#Implementation
public static NetworkSecurityPolicy getInstance() {
try {
Class<?> shadow = MyNetworkSecurityPolicy.class.forName("android.security.NetworkSecurityPolicy");
return (NetworkSecurityPolicy) shadow.newInstance();
} catch (Exception e) {
throw new AssertionError();
}
}
#Implementation
public boolean isCleartextTrafficPermitted() {
return true;
}
}
}
The first test passes as it should, but the second one fails with the message I wrote above.
The output in the console is:
test start
okhttp3.mockwebserver.MockWebServer$3 execute
INFO: MockWebServer[63631] starting to accept connections
WARNING: no system properties value for gsm.sim.operator.alpha
okhttp3.mockwebserver.MockWebServer$4 processOneRequest
INFO: MockWebServer[63631] received request: GET REQUEST_PATH HTTP/1.1 and responded: HTTP/1.1 200 OK
okhttp3.mockwebserver.MockWebServer$3 acceptConnections
test end
INFO: MockWebServer[63631] done accepting connections: Socket closed
test start
okhttp3.mockwebserver.MockWebServer$3 execute
INFO: MockWebServer[63649] starting to accept connections
okhttp3.mockwebserver.MockWebServer$3 acceptConnections
INFO: MockWebServer[63649] done accepting connections: Socket closed
on error: Failed to connect to localhost/0:0:0:0:0:0:0:1:63631
test end
(The Waiter object thing is from the concurrentunit lib)
Any idea why this happens?

Your second request is using the URL of the first MockWebServer instance. (Each instance has a distinct URL.)

This localhost connection fails because it does not belong tot he mock server. Use this instead
(mockHelper.getHost()).willReturn(okhttpMockWebServer.url("/").toString())

Broadcasting with Jersey SSE: Detect closed connection

I believe this question is not a duplicate of Server sent event with Jersey: EventOutput is not closed after client drops, but probably related to Jersey Server-Sent Events - write to broken connection does not throw exception.
In chapter 15.4.2 of the Jersey documentation, the SseBroadcaster is described:
However, the SseBroadcaster internally identifies and handles also client disconnects. When a client closes the connection the broadcaster detects this and removes the stale connection from the internal collection of the registered EventOutputs as well as it frees all the server-side resources associated with the stale connection.
I cannot confirm this. In the following testcase, I see the subclassed SseBroadcaster's onClose() method never being called: not when the EventInput is closed, and not when another message is broadcasted.
public class NotificationsResourceTest extends JerseyTest {
final static Logger log = LoggerFactory.getLogger(NotificationsResourceTest.class);
final static CountingSseBroadcaster broadcaster = new CountingSseBroadcaster();
public static class CountingSseBroadcaster extends SseBroadcaster {
final AtomicInteger connectionCounter = new AtomicInteger(0);
public EventOutput createAndAttachEventOutput() {
EventOutput output = new EventOutput();
if (add(output)) {
int cons = connectionCounter.incrementAndGet();
log.debug("Active connection count: "+ cons);
}
return output;
}
#Override
public void onClose(final ChunkedOutput<OutboundEvent> output) {
int cons = connectionCounter.decrementAndGet();
log.debug("A connection has been closed. Active connection count: "+ cons);
}
#Override
public void onException(final ChunkedOutput<OutboundEvent> chunkedOutput, final Exception exception) {
log.trace("An exception has been detected", exception);
}
public int getConnectionCount() {
return connectionCounter.get();
}
}
#Path("notifications")
public static class NotificationsResource {
#GET
#Produces(SseFeature.SERVER_SENT_EVENTS)
public EventOutput subscribe() {
log.debug("New stream subscription");
EventOutput eventOutput = broadcaster.createAndAttachEventOutput();
return eventOutput;
}
}
#Override
protected Application configure() {
ResourceConfig config = new ResourceConfig(NotificationsResource.class);
config.register(SseFeature.class);
return config;
}
#Test
public void test() throws Exception {
// check that there are no connections
assertEquals(0, broadcaster.getConnectionCount());
// connect subscriber
log.info("Connecting subscriber");
EventInput eventInput = target("notifications").request().get(EventInput.class);
assertFalse(eventInput.isClosed());
// now there are connections
assertEquals(1, broadcaster.getConnectionCount());
// push data
log.info("Broadcasting data");
String payload = UUID.randomUUID().toString();
OutboundEvent chunk = new OutboundEvent.Builder()
.mediaType(MediaType.TEXT_PLAIN_TYPE)
.name("message")
.data(payload)
.build();
broadcaster.broadcast(chunk);
// read data
log.info("Reading data");
InboundEvent inboundEvent = eventInput.read();
assertNotNull(inboundEvent);
assertEquals(payload, inboundEvent.readData());
// close subscription
log.info("Closing subscription");
eventInput.close();
assertTrue(eventInput.isClosed());
// at this point, the subscriber has disconnected itself,
// but jersey doesnt realise that
assertEquals(1, broadcaster.getConnectionCount());
// wait, give TCP a chance to close the connection
log.debug("Sleeping for some time");
Thread.sleep(10000);
// push data again, this should really flush out the not-connected client
log.info("Broadcasting data again");
broadcaster.broadcast(chunk);
Thread.sleep(100);
// there is no subscriber anymore
assertEquals(0, broadcaster.getConnectionCount()); // FAILS!
}
}
Maybe JerseyTest is not a good way to test this. In a less ... clinical setup, where a JavaScript EventSource is used, I see onClose() being called, but only after a message is broadcasted on the previously closed connection.
What am I doing wrong?
Why doesn't SseBroadcaster detect the closing of the connection by the client?
Follow-up
I've found JERSEY-2833 which was rejected with Works as designed:
According to the Jersey Documentation in SSE chapter (https://jersey.java.net/documentation/latest/sse.html) in 15.4.1 it's mentioned that Jersey does not explicitly close the connection, it's the responsibility of the resource method or the client.
What does that mean exactly? Should the resource enforce a timeout and kill all active and closed-by-client connections?

In the documentation of the constructor org.glassfish.jersey.media.sse.SseBroadcaster.SseBroadcaster(), it says:
Creates a new instance. If this constructor is called by a subclass, it assumes the the reason for the subclass to exist is to implement onClose(org.glassfish.jersey.server.ChunkedOutput) and onException(org.glassfish.jersey.server.ChunkedOutput, Exception)methods, so it adds the newly created instance as the listener. To avoid this, subclasses may call SseBroadcaster(Class) passing their class as an argument.
So you should not leave default constructor and try implementing your constructor invoking super with your class:
public CountingSseBroadcaster(){
super(CountingSseBroadcaster.class);
}

I believe it might be better to set a timeout on your resource and kill only that connection, for example:
#Path("notifications")
public static class NotificationsResource {
#GET
#Produces(SseFeature.SERVER_SENT_EVENTS)
public EventOutput subscribe() {
log.debug("New stream subscription");
EventOutput eventOutput = broadcaster.createAndAttachEventOutput();
new Timer().schedule( new TimerTask()
{
#Override public void run()
{
eventOutput.close()
}
}, 10000); // 10 second timeout
return eventOutput;
}
}

Im wondering if by subclassing you may have changed the behaviour.
#Override
public void onClose(final ChunkedOutput<OutboundEvent> output) {
int cons = connectionCounter.decrementAndGet();
log.debug("A connection has been closed. Active connection count: "+ cons);
}
In this you don't close the ChunkedOutput so it won't release the connection. Could this be the problem?

How to get an existing websocket instance

I'm working on an application that uses Websockets (Java EE 7) to send messages to all the connected clients asynchronously. The server (Websocket endpoint) should send these messages whenever a new article (an engagement modal in my app) is created.
Everytime a connection is established to the websocket endpoint, I'm adding the corresponding session to a list, which I could be able to access outside.
But the problem I had is, when I'm accessing this created websocket endpoint to which all the clients connected from outside (any other business class), I've get the existing instance (like a singleton).
So, can you please suggest me a way I can get an existing instance of the websocket endpoint, as I can't create it as new MyWebsocketEndPoint() coz it'll be created by the websocket internal mechanism whenever the request from a client is received.
For a ref:
private static WebSocketEndPoint INSTANCE = null;
public static WebSocketEndPoint getInstance() {
if(INSTANCE == null) {
// Instead of creating a new instance, I need an existing one
INSTANCE = new WebSocketEndPoint ();
}
return INSTANCE;
}
Thanks in advance.

The container creates a separate instance of the endpoint for every client connection, so you can't do what you're trying to do. But I think what you're trying to do is send a message to all the active client connections when an event occurs, which is fairly straightforward.
The javax.websocket.Session class has the getBasicRemote method to retrieve a RemoteEndpoint.Basic instance that represents the endpoint associated with that session.
You can retrieve all the open sessions by calling Session.getOpenSessions(), then iterate through them. The loop will send each client connection a message. Here's a simple example:
#ServerEndpoint("/myendpoint")
public class MyEndpoint {
#OnMessage
public void onMessage(Session session, String message) {
try {
for (Session s : session.getOpenSessions()) {
if (s.isOpen()) {
s.getBasicRemote().sendText(message);
}
} catch (IOException ex) { ... }
}
}
But in your case, you probably want to use CDI events to trigger the update to all the clients. In that case, you'd create a CDI event that a method in your Websocket endpoint class observes:
#ServerEndpoint("/myendpoint")
public class MyEndpoint {
// EJB that fires an event when a new article appears
#EJB
ArticleBean articleBean;
// a collection containing all the sessions
private static final Set<Session> sessions =
Collections.synchronizedSet(new HashSet<Session>());
#OnOpen
public void onOpen(final Session session) {
// add the new session to the set
sessions.add(session);
...
}
#OnClose
public void onClose(final Session session) {
// remove the session from the set
sessions.remove(session);
}
public void broadcastArticle(#Observes #NewArticleEvent ArticleEvent articleEvent) {
synchronized(sessions) {
for (Session s : sessions) {
if (s.isOpen()) {
try {
// send the article summary to all the connected clients
s.getBasicRemote().sendText("New article up:" + articleEvent.getArticle().getSummary());
} catch (IOException ex) { ... }
}
}
}
}
}
The EJB in the above example would do something like:
...
#Inject
Event<ArticleEvent> newArticleEvent;
public void publishArticle(Article article) {
...
newArticleEvent.fire(new ArticleEvent(article));
...
}
See the Java EE 7 Tutorial chapters on WebSockets and CDI Events.
Edit: Modified the #Observer method to use an event as a parameter.
Edit 2: wrapped the loop in broadcastArticle in synchronized, per #gcvt.
Edit 3: Updated links to Java EE 7 Tutorial. Nice job, Oracle. Sheesh.

Actually, WebSocket API provides a way how you can control endpoint instantiation. See https://tyrus.java.net/apidocs/1.2.1/javax/websocket/server/ServerEndpointConfig.Configurator.html
simple sample (taken from Tyrus - WebSocket RI test):
public static class MyServerConfigurator extends ServerEndpointConfig.Configurator {
public static final MyEndpointAnnotated testEndpoint1 = new MyEndpointAnnotated();
public static final MyEndpointProgrammatic testEndpoint2 = new MyEndpointProgrammatic();
#Override
public <T> T getEndpointInstance(Class<T> endpointClass) throws InstantiationException {
if (endpointClass.equals(MyEndpointAnnotated.class)) {
return (T) testEndpoint1;
} else if (endpointClass.equals(MyEndpointProgrammatic.class)) {
return (T) testEndpoint2;
}
throw new InstantiationException();
}
}
You need to register this to an endpoint:
#ServerEndpoint(value = "/echoAnnotated", configurator = MyServerConfigurator.class)
public static class MyEndpointAnnotated {
#OnMessage
public String onMessage(String message) {
assertEquals(MyServerConfigurator.testEndpoint1, this);
return message;
}
}
or you can use it with programmatic endpoints as well:
public static class MyApplication implements ServerApplicationConfig {
#Override
public Set<ServerEndpointConfig> getEndpointConfigs(Set<Class<? extends Endpoint>> endpointClasses) {
return new HashSet<ServerEndpointConfig>
(Arrays.asList(ServerEndpointConfig.Builder
.create(MyEndpointProgrammatic.class, "/echoProgrammatic")
.configurator(new MyServerConfigurator())
.build()));
}
#Override
public Set<Class<?>> getAnnotatedEndpointClasses(Set<Class<?>> scanned) {
return new HashSet<Class<?>>(Arrays.asList(MyEndpointAnnotated.class));
}
Of course it is up to you if you will have one configurator used for all endpoints (ugly ifs as in presented snippet) or if you'll create separate configurator for each endpoint.
Please do not copy presented code as it is - this is only part of Tyrus tests and it does violate some of the basic OOM paradigms.
See https://github.com/tyrus-project/tyrus/blob/1.2.1/tests/e2e/src/test/java/org/glassfish/tyrus/test/e2e/GetEndpointInstanceTest.java for complete test.

How to pass application/user data to the ChannelHandler Netty

When you design an a client that is going to connect to a lot of servers, like a crawler.
You will code something like that :
// the pipeline
public class CrawlerPipelineFactory implements ChannelPipelineFactory {
public ChannelPipeline getPipeline() throws Exception {
return Channels.pipeline(new CrawlerHandler());
}
}
// the channel handler
public class CrawlerHandler extends SimpleChannelHandler {
#Override
public void channelConnected(ChannelHandlerContext ctx, ChannelStateEvent e) throws Exception {
// ...
}
}
// the main :
public static void main(){
ChannelFactory factory = new NioClientSocketChannelFactory(Executors.newCachedThreadPool(),Executors.newCachedThreadPool());
ClientBootstrap scannerBootstrap = new ClientBootstrap(factory);
scannerBootstrap.setPipelineFactory(new CrawlerPipelineFactory());
while(true){
MyURL url = stack.pop();
ChannelFuture connect = scannerBootstrap.connect(url.getSocketAddress());
}
}
Now when you are in your ApplicationHandler, the stuff that implements your SimpleChannelHandler or WhatEverStreamHandler, (CrawlerHander in the example) the only piece of information you get is the socketAdress you are connecting to that you can recover in "public void channelConnected()" function.
Ok but what if I want to recover some user data, like the MyURL object you see in my code example ?
I use a dirty hack, I use a Map<"ip:port",MyURL> so I can retrieve the associated data in channelConnected because I know ip:port i'm connected on.
This hack is really dirty, it won't work if you are connecting simultaneously to the same server (or you'll have to bind to a local port and use a key like "localport:ip:remoteport" but it's so dirty).
So I'm seeking what is the good way to pass data the the CrawlerHander ?
It would be cool if we could pass this data via the connect() method of the bootstrap. I know I can pass argument in my ChannelPipelineFactory.getPipeline() because it's invoked via connect(). But now we can't, so here is another dirty hack I use :
EDIT:
// the main
while(!targets.isEmpty()){
client.connect("localhost",111); // we will never connect to localhost, it's a hack
}
// the pipleline
public ChannelPipeline getPipeline() throws Exception {
return Channels.pipeline(
new CrawlerHandler(targets.pop()) // I specify each new host to connect here
);
}
// in my channel handler
// Now I have the data I want in the constructor, so I m sure I get them before everything is called
public class CrawlerHandler extends SimpleChannelHandler {
ExtraParameter target;
public CrawlerHandler(ExtraParameter target) {
this.target = target;
// but, and it's the most dirty part, I have to abort the connection to localhost, and reinit a new connection to the real target
boolean bFirstConnect=true;
#Override
public void connectRequested(ChannelHandlerContext ctx, ChannelStateEvent e) throws Exception {
if(bFirstConnect){
bFirstConnect = false;
ctx.getChannel().connect(target.getSocketAddr());
}

You can pass variables to Channel via Bootstrap.
Netty.io 4.1 & SO - Adding an attribute to a Channel before creation
Update to this answer while very late.

You can pass the data to the newly connected channel/channel handler using ChannelLocal or in ChannelHandlerContext (or in the Channel it self in latest Netty 3.x) using a connect future listener. In below example, ChannelLocal is used.
public class ChannelDataHolder {
public final static ChannelLocal<String> CHANNEL_URL = new ChannelLocal<String>(true);
}
// for each url in bootstrap
MyURL url = ....;
ChannelFuture cf = scannerBootstrap.connect(url.getSocketAddress());
final String urlString = url.getUrl();
cf.addListener(new ChannelFutureListener() {
#Override
public void operationComplete(ChannelFuture future) throws Exception {
ChannelDataHolder.CHANNEL_URL.set(future.getChannel(), urlString);
}
});
//In the handler
public class CrawlerHandler extends SimpleChannelHandler {
#Override
public void channelConnected(ChannelHandlerContext ctx, ChannelStateEvent e) throws Exception {
String urlString = ChannelDataHolder.CHANNEL_URL.get(ctx.getChannel());
// ...use the data here
}
}
Note: instead of ChannelLocal, you can set and get the data using
ChannelHandlerContext.setAttachment()/getAttachment()
Channel.setAttachment()/getAttachment() in latest 3.x version of Netty
but both approaches does not support type safety.