This is the scenario - There are multiple app servers. Browser can connect via websocket to any app server.
The app servers (consumers) are all listening on a particular queue. As soon as a web socket connection is received, the particular app server binds the queue with a routing key {userId} to a direct exchange.
I want a message sent to the direct exchange with the routing key {userId} to be received by only the particular app server where the binding has occured.
Is a direct exchange the right exchange to use in this case? Or should some other type of exchange be used?
I'm using spring-amqp to create dynamic bindings when a websocket comes in
// create the RabbitMq queue and bind to it
String routingKey = MessageConstants.getRoutingKeyForUserRecommendationQueue(user);
Binding userRecommendationBinding = BindingBuilder.bind(userRecommendationsQueue).
to(directExchange).with(routingKey);
amqpAdmin.declareBinding(userRecommendationBinding);
Send message to a particular consumer in a queue
this is not possible. any consumer connected to a queue has a chance of consuming any given message in the queue
I want a message sent to the direct exchange with the routing key {userId} to be received by only the particular app server where the binding has occured.
you can do this by creating exclusive / autoDelete queues for your consumer, with a binding that directs all messages for that consumer to that queue.
Is a direct exchange the right exchange to use in this case?
either a direct exchange or a topic exchange is fine. direct exchange is slightly easier to understand, but topic exchange is more flexible
Actually you go right way.
And yes: Direct Exchange with an appropriate binding should save you.
See more info in the RabbitMQ Tutorial: http://www.rabbitmq.com/tutorials/tutorial-four-java.html
Also take a look into Spring AMQP Samples on the matter: https://github.com/spring-projects/spring-amqp-samples/tree/master/rabbitmq-tutorials
UPDATE
Unfortunately that is not what is happening. The messages seem to go randomly to any consumer, and not just the consumer that created the binding.
M-m-m. That's possible, because we route only my the key, but after that the message is placed to the queue, which may have several consumers on different machines.
In this case yes: the dynamic binding doesn't help.
You should consider to create an unique new queue (auto-deleted is fine) and bind and listen exactly from that. The SimpleMessageListenerContainer supports addQueues() at runtime to start a new consumer for a new queue.
I think that should work for you.
You still shouldn't do anything on the producer side: the same exhchange and routingKey logic.
Related
Using RabbitMQ 3.7.16, with spring-amqp 2.2.3.RELEASE.
Multiple clients publish messages to the DataExchange topic exchange in our RabbitMQ server, using a unique routing key. In the absence of any bindings, the exchange will route all the messaged to the data.queue.generic through the AE.
When a certain client (client ID 1 and 2 in the diagram) publishes lots of messages, in order to scale the consumption of their messages independently from other clients, we are starting consumers and assign them to only handle a their client ID. To achieve this, each client-consumer is defining a new queue, and it binds it to the topic exchange with the routing key events.<clientID>.
So scaling up is covered and works well.
Now when the messages rate for this client goes down, we would like to also scale down its consumers, up to the point of removing all of them. The intention is to then have all those messages being routed to the GenericExchange, where there's a pool of generic consumers taking care of them.
The problem is that if I delete data.queue.2 (in order to remove its binding which will lead to new messages being routed to the GenericExchange) all its pending messages will be lost.
Here's a simplified architecture view:
It would be an acceptable solution to let the messages expire with a TTL in the client queue, and then dead letter them to the generic exchange, but then I also need to stop the topic exchange from routing new messages to this "dying" queue.
So what options do I have to stop the topic exchange from routing messages to the client queue where now there's no consumer connected to it?
Or to explore another path - how to dead letter messages in a deleted/expired queue?
If the client queue is the only one with a matching binding as your explanation seems to suggest, you can just remove the binding between the exchange and the queue.
From then on, all new messages for the client will go through the alternate exchange, your "generic exchange", to be processed by your generic consumers.
As for the messages left over in the client queue, you could use a shovel to send them back to the topic exchange, for them to be routed to the generic exchange.
This based on the assumption the alternate exchange is internal. If it's not internal, you can target it directly with the shovel.
As discussed with Bogdan, another option to resolve this while ensuring no message loss is occuring is to perform multiple steps:
remove the binding between the specific queue and the exchange
have some logic to have the remaining messages be either consumed or rerouted to the generic queue
if the binding removal occurs prior to the consumer(s) disconnect, have the last consumer disconnect only once the queue is empty
if the binding removal occurs after the last consumer disconnect, then have a TTL on messages with alternate exchange as the generic exchange
depending on the options selected before, have some cleanup mecanism to remove the lingering empty queues
I wonder if the following scenario is possible:
Create an exchange of type direct
Publish a message to that exchange with routing key rk1
After that:
Create a queue which accepts messages with routing key rk1
Consume message published to exchange
It seems like if there is no queue present, the message is dropped and there is no way to receive it.
So basically I want to be able to produce messages when there are no consumers present. And consume them some time later.
It seems like if there is no queue present, the message is dropped and there is no way to receive it.
Yes, this is correct, but it's only part of the story.
A message queue is the storage location for messages published to the server. A consumer is a designated connection set to receive messages put into a queue. The exchange is simply a location to push messages. It contains the routing semantics to determine which messages wind up in the queues on the server. When a message cannot be routed to a queue and/or consumer, there are various semantics that can apply, but the default is that the message is dropped.
Options for dealing with unroutable messages:
Alternate exchange - designates a different exchange where messages can be dumped if they cannot be routed to a queue on the current exchange. This can be thought of similar to how TCP/IP works when a destination host is not reachable on the current subnet, and traffic is forwarded to the gateway. Note that a queue must be bound to the alternate exchange for the message to be dumped into. A typical case might be to have it configured as fanout exchange with one queue to trap all messages sent into the alternate exchange.
Mandatory or Immediate - return a message back to the sender if it can't be delivered. The server does not store the message.
Mandatory designates that the message must be deliverable to a queue at the time it is published. If the message is not routable, the publisher will receive a basic.return.
Immediate designates that, in addition to being deliverable, must be immediately routed to a consumer on a particular queue (e.g. it's not good enough that it be dumped in a queue for pickup later - it has to be delivered to the end consumer right now.
In every case, if there is no queue, the server cannot store the message.
The entity queue is the one that is supposed to hold the messages , so without a queue the messages will be lost.
However in case you do not create any exchange with appropriate routing key you may leverage dead lettering feature in rabbitmq.
Another solution could be to declare the queue with the binding after the exchange and before publishing the message; this way the message will be routed and stored, but you may have to add some TTLs ( https://www.rabbitmq.com/ttl.html ).
I am integrating several .Net modules using pub/sub messaging using RabbitMQ and MassTransit. Most of the message subscription shall be durable. But some shall be transient. When a consumer dies the messages shall not be stored and already queued messages shall be discarded.
In each module I create 1 bus with 2 receive endpoints. One is configured as durable and non-auto-delete. The other one is configured as non-durable and auto-delete. Each gets its own set of consumers. This works as expected.
Now I am trying to implement request/response messages. Here comes the problem because now the sender has to decide to which exchange to route to. And that is wrong as I want receiver to decide whether to use durable or transient queue.
My questions:
Is there a better way how to support durable and transient subscription at the same time?
Why is MassTransit binding message exchange to an endpoint exchange that is bound to an endpoint queue? Why cannot the message exchange be directly bound to the endpoint queue?
Lets assume that all request consumers in one module are either durable or transient. Is it possible to declare one "module"-exchange which is then bound to either durable or transient queue? So the sender addresses the module exchange and module decides to which queue to bind. How to convince MassTransit to do so?
A module is using durable subscriptions that survive through restarts of module and also broker. After some time, admin (so in run-time of the system) decides to disconnect this module from the system. Can the module somehow unsubscribe everything and let MassTransit to remove the durable exchanges and queues?
Your question starts with request/response sent to an unknown endpoint, and ends with removing exchanges. These are different things, I suppose.
I cannot answer point-by-point, just will try to clear up things.
Request/response by definition requires you to know where you send stuff. As per MassTransit convention, the endpoint address is always an exchange/queue pair address. Therefore, you cannot let receiver decide who will handle this message, it will be delivered to the exchange/queue of the endpoint where you send it to.
About the "unsubscribe" - MassTransit deletes nothing. You have to clean up the binding that is not being used manually or by using the management API.
I'm trying to setup RabbitMQ in a model where there is only one producer and one consumer, and where messages sent by the producer are delivered to the consumer only if the consumer is connected, but dropped if the consumer is not present.
Basically I want the queue to drop all the messages it receives when no consumer is connected to it.
An additional constraint is that the queue must be declared on the RabbitMQ server side, and must not be explicitly created by the consumer or the producer.
Is that possible?
I've looked at a few things, but I can't seem to make it work:
durable vs non-durable does not work, because it is only useful when the broker restarts. I need the same effect but on a connection.
setting auto_delete to true on the queue means that my client can never connect to this queue again.
x-message-ttl and max-length make it possible to lose message even when there is a consumer connected.
I've looked at topic exchanges, but as far as I can tell, these only affect the routing of messages between the exchange and the queue based on the message content, and can't take into account whether or not a queue has connected consumers.
The effect that I'm looking for would be something like auto_delete on disconnect, and auto_create on connect. Is there a mechanism in rabbitmq that lets me do that?
After a bit more research, I discovered that one of the assumptions in my question regarding x-message-ttl was wrong. I overlooked a single sentence from the RabbitMQ documentation:
Setting the TTL to 0 causes messages to be expired upon reaching a queue unless they can be delivered to a consumer immediately
https://www.rabbitmq.com/ttl.html
It turns out that the simplest solution is to set x-message-ttl to 0 on my queue.
You can not doing it directly, but there is a mechanism not dificult to implement.
You have to enable the Event Exchange Plugin. This is a exchange at which your server app can connect and will receive internal events of RabbitMQ. You would be interested in the consumer.created and consumer.deleted events.
When these events are received you can trigger an action (create or delete the queue you need). More information here: https://www.rabbitmq.com/event-exchange.html
Hope this helps.
If your consumer is allowed to dynamically bind / unbind a queue during start/stop on the broker it should be possible by that way (e.g. queue is pre setup and the consumer binds the queue during startup to an exchange it wants to receive messages from)
Is is possible to configure a RabbitMQ exchange or a queue in such a way that at most one message with a given routing key is pending at any time? If new message arrives, the old one would be dropped and the new one enqueued.
If such option is not available, what would be the best way to implement this at the application level? I.e. when application receives a message how can it check if there any more pending messages?
You need to install Last Value Cache and enable it. Your exchange will be type "x-lvc", which inherits from the direct exchange type.
each time you connect to MQ, create a queue and bind to this exchange. It will deliver the most recent message to the queue. It is perfect for making sure you get only the most uptodate message. All other messages sent to this exchange are discarded unless there is a queue connected. So once connected you will continue to receive updates.
here are installation instructions:
https://github.com/simonmacmullen/rabbitmq-lvc-plugin
here is a similar question:
RabbitMQ messaging - initializing consumer