spring-amqp: Channel shutdown with NACKS RECEIVED message - spring-amqp

Using spring-amqp with heavy load on RabbitMQ bus, we sometimes get logs from org.springframework.amqp.rabbit.connection.CachingConnectionFactory saying :
Channel shutdown: clean channel shutdown; protocol method: #method<channel.close>(reply-code=200, reply-text=NACKS RECEIVED, class-id=0, method-id=0)
Can you explain this log, please, and why is it at ERROR level?
Do we have any adjustments to make?
Thanks in advance for your answer.

The channel throws an exception if all publisher confirms are not returned with the timeout...
#Override
public void waitForConfirmsOrDie(long timeout)
throws IOException, InterruptedException, TimeoutException
{
try {
if (!waitForConfirms(timeout)) {
close(AMQP.REPLY_SUCCESS, "NACKS RECEIVED", true, null, false);
throw new IOException("nacks received");
}
} catch (TimeoutException e) {
close(AMQP.PRECONDITION_FAILED, "TIMEOUT WAITING FOR ACK");
throw(e);
}
}
The DefaultChannelCloseLogger will only skip normal closes (200) if the reply text is OK...
/**
* Return true if the {#link ShutdownSignalException} reason is AMQP.Channel.Close and
* the reply code was AMQP.REPLY_SUCCESS (200) and the text equals "OK".
* #param sig the exception.
* #return true for a normal channel close.
*/
public static boolean isNormalChannelClose(ShutdownSignalException sig) {
Method shutdownReason = sig.getReason();
return isNormalShutdown(sig) ||
(shutdownReason instanceof AMQP.Channel.Close
&& AMQP.REPLY_SUCCESS == ((AMQP.Channel.Close) shutdownReason).getReplyCode()
&& "OK".equals(((AMQP.Channel.Close) shutdownReason).getReplyText()));
}
If you want to ignore these errors, you can configure a custom close exception logger:
/**
* Set the strategy for logging close exceptions; by default, if a channel is closed due to a failed
* passive queue declaration, it is logged at debug level. Normal channel closes (200 OK) are not
* logged. All others are logged at ERROR level (unless access is refused due to an exclusive consumer
* condition, in which case, it is logged at INFO level).
* #param closeExceptionLogger the {#link ConditionalExceptionLogger}.
* #since 1.5
*/
public void setCloseExceptionLogger(ConditionalExceptionLogger closeExceptionLogger) {
Assert.notNull(closeExceptionLogger, "'closeExceptionLogger' cannot be null");
this.closeExceptionLogger = closeExceptionLogger;
if (this.publisherConnectionFactory != null) {
this.publisherConnectionFactory.setCloseExceptionLogger(closeExceptionLogger);
}
}

Related

netty-readtimeout and return customized response to front end

I have a question regarding configuration of timeouts on a netty TCP server.
Currently we have configured readTimeOut as 120s. Set the connect timout like this:
socketChannel.pipeline().addLast(new ReadTimeoutHandler(120, TimeUnit.SECONDS));
But if the read time exceeds 120s, service doesn't response to front end correctly. If tested from postman, got the "Could not get any response" as response.
Following is the netty config we using:
public class EventLoopNettyCustomizer implements NettyServerCustomizer {
#Override
public HttpServer apply(HttpServer httpServer) {
EventLoopGroup bossGroup = new NioEventLoopGroup(1);
EventLoopGroup workGroup = new NioEventLoopGroup();
return httpServer.tcpConfiguration(tcpServer -> tcpServer
.bootstrap(serverBootstrap -> serverBootstrap
.group(bossGroup, workGroup)
.option(ChannelOption.SO_BACKLOG, 10000)
.option(ChannelOption.CONNECT_TIMEOUT_MILLIS, 30000)
.childHandler(new ChannelInitializer<SocketChannel>() {
#Override
protected void initChannel(SocketChannel socketChannel) throws Exception {
socketChannel.pipeline().addLast(new ReadTimeoutHandler(120, TimeUnit.SECONDS));
socketChannel.pipeline().addLast(new WriteTimeoutHandler(120, TimeUnit.SECONDS));
}
})
.channel(NioServerSocketChannel.class)));
}
}
How can I config the netty so that it is able to return customized response? Including http status and message.

Solace Client Acknowledgement Replay

I'm putting together a Solace Point-to-point solution in C#.
In my subscriber/listener, I am using ClientAck mode to ensure messages are successfully processed before being removed from the queue.
My question (probably due to my limited experience in messaging) is regarding failed messages, e.g. if I cannot process the message and hence not send the Ack, how is the message replayed?
An example of what I have is as follows:
using (ISession session = context.CreateSession(sessionProperties, null, null))
{
ReturnCode returnCode = session.Connect();
if (returnCode == ReturnCode.SOLCLIENT_OK)
{
var endpointProps = new EndpointProperties()
{
Permission = EndpointProperties.EndpointPermission.Consume,
AccessType = EndpointProperties.EndpointAccessType.Exclusive
};
using (IQueue queue = ContextFactory.Instance.CreateQueue(queueName))
{
session.Provision(queue, endpointProps,
ProvisionFlag.IgnoreErrorIfEndpointAlreadyExists | ProvisionFlag.WaitForConfirm, null);
_flow = session.CreateFlow(new FlowProperties { AckMode = MessageAckMode.ClientAck }, queue, null, HandleMessageEvent, HandleFlowEvent);
_flow.Start();
do { WaitEventWaitHandle.WaitOne(); } while (!cancellationToken.IsCancellationRequested);
};
return Task.CompletedTask;
}
else
{
throw new Exception($"Connection failed, return code: {returnCode}");
}
}
and then handling incoming messages
void HandleMessageEvent(object sender, MessageEventArgs args)
{
using (IMessage message = args.Message)
{
try
{
_handler(message.ApplicationMessageType, message.BinaryAttachment);
_flow.Ack(message.ADMessageId);
}
finally
{
WaitEventWaitHandle.Set();
}
}
}
So, if I don't Ack, message remains on queue as expected (and required), however, how (best-practice) can I re-process it without manual intervention?
After a message has been delivered to a consumer from a Solace PubSub+ queue, the message will only be resent if the client unbinds before sending an acknowledgement back. The exception to this is specific to JMS clients with the session.recover() action.
If a message needs to be re-delivered to a C# application after it has already been sent but not acknowledged, the client will need to unbind and rebind to the queue. Note that if there are other clients also bound to the queue, the message may be re-sent to those clients before your client rebinds.

No errors are being raised when unsuccessfully writing to Azure service bus

When writing a message to the Azure Service Bus (using Microsoft.Azure.ServiceBus standard library, not the .Net Framework version) it works fine. However, when switching networks to a network that blocks that traffic and running it again I would expect an error being raised by SendAsync yet no error is thrown, therefor the function considers the send successful even though it is not.
Am I missing some logic to make sure that errors do get raised and trapped, it seems to be inline with all the examples I have seen.
I have tried this possible solution ..
Trouble catching exception on Azure Service Bus SendAsync method
.ContinueWith(t =>
{
Console.WriteLine(t.Status + "," + t.IsFaulted + "," + t.Exception.InnerException);
}, TaskContinuationOptions.OnlyOnFaulted);
.. and at no point does ContinueWith get hit.
[HttpPost]
[Consumes("application/json")]
[Produces("application/json")]
public ActionResult<Boolean> Post(Contract<T> contract)
{
Task.Run(() => SendMessage(contract));
// Other stuff
}
private async Task<ActionResult<Boolean>> SendMessage(Contract<T> contract)
{
JObject json = JObject.FromObject(contract);
Message message = new Message();
message.MessageId = Guid.NewGuid().ToString();
message.ContentType = ObjectType;
message.PartitionKey = ObjectType;
message.Body = Encoding.UTF8.GetBytes(JsonConvert.SerializeObject(contract));
foreach (KeyValuePair<String, String> route in DataRouting)
{
JToken jToken = json.SelectToken(route.Value);
if (jToken != null)
{
message.UserProperties[route.Key] = jToken.Value<String>();
}
else
{
String routeError = $"Could not find routing information in request for: {route.Key} in {route.Value}";
Logger.LogError(routeError);
return new UnprocessableEntityObjectResult(routeError);
}
}
// Send the message
try
{
await topicClient.SendAsync(message);
}
catch(Exception ex)
{
return new UnprocessableEntityObjectResult($"'Could not transmit message to service bus - {ex.Message}'");
}
return new OkObjectResult(true);
}
I expect that the error trap would be hit if the SendAsync fails to send the message. However it essentially fire and forgets, the message send is blocked by the firewall but is never reported to the caller by throwing an error.
Ok, found the answer, but I will leave this out there in case anyone else does this to themselves. It was down to my general muppetry when putting the MVC Controller together. Set async on the Post action and configure the await on the send. Obvious really but I missed it.
public virtual async Task<ActionResult<Boolean>> Post(Contract<T> contract){}
...
// Send the message
try
{
await topicClient.SendAsync(message).ConfigureAwait(false);
return new OkObjectResult(true); // Success if we got here
}
catch(Exception ex)
{
return new UnprocessableEntityObjectResult($"'Could not transmit message to service bus - {ex.Message}'");
}

How to retrieve exact reason of the error from async HttpRequest?

I am trying to figure out how to find out exact reason of (async) HttpRequest (from 'dart:html') failure, and, to be honest, I am a bit lost here.
The onError callback receives only HttpRequestProgressError object, which doesn't have anything useful, and the HttpRequest object itself has "status" set to "0" in case of failure, even console shows "Failed to load resource" with no details.
What I want is to know the exact reason - like "connection refused" or "host name not resolved".
Is this possible at all?
Thank you!
Unfortunately, there is no property to report the error as detailed as you'd like. The reason is that JavaScript doesn't support this.
There are the properties status and statusText on the HttpRequest object (which you could get from your HttpRequestProgressEvent with evt.target, but those represent HTTP status codes. Every other error has the status code 0 - request failed. This could be anything, and the only place to look at is the browser's console, because this is an Exception thrown by the browser.
If your request was synchronous, you could surround the send() with a try-catch. If your request is async, this won't work.
See here
#library('Request');
#import('dart:html');
#import("dart:json");
typedef void RequestHandler(String responseText);
typedef void ErrorHandler(String error);
class ResourceRequest {
XMLHttpRequest request;
RequestHandler _callbackOnSuccess;
ErrorHandler _callbackOnFailure;
ResourceRequest.openGet(String url, RequestHandler callbackOnSuccess, [ErrorHandler callbackOnFailure])
: request = new XMLHttpRequest(),
_callbackOnSuccess = callbackOnSuccess,
_callbackOnFailure = callbackOnFailure {
request.open("GET", url, async : true);
request.on.loadEnd.add((XMLHttpRequestProgressEvent e) => onLoadEnd(e));
}
void send() {
request.send();
}
void onLoadEnd(XMLHttpRequestProgressEvent event) {
if (request.readyState == 4 && request.status == 200) {
_callbackOnSuccess(request.responseText);
} else if (_callbackOnFailure != null) {
_callbackOnFailure(request.statusText);
}
}
}

Blackberry: Make a iterative HTTP GET petition using Comms API

I want to store position coords (latitude, longitude) in a table in my MySQL DB querying a url in a way similar to this one: http://locationstore.com/postlocation.php?latitude=var1&longitude=var2 every ten seconds. PHP script works like a charm. Getting the coords in the device ain't no problem either. But making the request to the server is being a hard one. My code goes like this:
public class LocationHTTPSender extends Thread {
for (;;) {
try {
//fetch latest coordinates
coords = this.coords();
//reset url
this.url="http://locationstore.com/postlocation.php";
// create uri
uri = URI.create(this.url);
FireAndForgetDestination ffd = null;
ffd = (FireAndForgetDestination) DestinationFactory.getSenderDestination
("MyContext", uri);
if(ffd == null)
{
ffd = DestinationFactory.createFireAndForgetDestination
(new Context("MyContext"), uri);
}
ByteMessage myMsg = ffd.createByteMessage();
myMsg.setStringPayload("doesnt matter");
((HttpMessage) myMsg).setMethod(HttpMessage.POST);
((HttpMessage) myMsg).setQueryParam("latitude", coords[0]);
((HttpMessage) myMsg).setQueryParam("longitude", coords[1]);
((HttpMessage) myMsg).setQueryParam("user", "1");
int i = ffd.sendNoResponse(myMsg);
ffd.destroy();
System.out.println("Lets sleep for a while..");
Thread.sleep(10000);
System.out.println("woke up");
} catch (Exception e) {
// TODO Auto-generated catch block
System.out.println("Exception message: " + e.toString());
e.printStackTrace();
}
}
I haven't run this code to test it, but I would be suspicious of this call:
ffd.destroy();
According to the API docs:
Closes the destination. This method cancels all outstanding messages,
discards all responses to those messages (if any), suspends delivery
of all incoming messages, and blocks any future receipt of messages
for this Destination. This method also destroys any persistable
outbound and inbound queues. If Destination uses the Push API, this
method will unregister associated push subscriptions. This method
should be called only during the removal of an application.
So, if you're seeing the first request succeed (at least sometimes), and subsequent requests fail, I would try removing that call to destroy().
See the BlackBerry docs example for this here
Ok so I finally got it running cheerfully. The problem was with the transport selection; even though this example delivered WAP2 (among others) as an available transport in my device, running the network diagnostics tool showed only BIS as available. It also gave me the connection parameters that I needed to append at the end of the URL (;deviceside=false;ConnectionUID=GPMDSEU01;ConnectionType=mds-public). The code ended up like this:
for (;;) {
try {
coords.refreshCoordinates();
this.defaultUrl();
this.setUrl(stringFuncs.replaceAll(this.getUrl(), "%latitude%", coords.getLatitude() + ""));
this.setUrl(stringFuncs.replaceAll(this.getUrl(), "%longitude%", coords.getLongitude() + ""));
cd = cf.getConnection(this.getUrl());
if (cd != null) {
try {
HttpConnection hc = (HttpConnection)cd.getConnection();
final int i = hc.getResponseCode();
hc.close();
} catch (Exception e) {
}
}
//dormir
Thread.sleep(15000);
} catch (Exception e) {
} finally {
//cerrar conexiones
//poner objetos a null
}
Thanks for your help #Nate, it's been very much appreciated.

Resources