Here is my use case:
I have two endpoints: one with MQ and the second with TCP/IP
I have to replace a legacy server which accepts queries from remote TCP/IP clients. Once the socket is open with the client, data is exchanged in both sides. the server sends asynchronously MQ data through TCP/IP and receive data from clients asynchronously also. Each data message sent has to be acknowledged. The constraint here is that I have to use the same socket.
I created two routes
from("netty4:tcp://ipAddress:port?sync=true").to("wmq:queue:toQueue")
from("wmq:queue:fromQueue").to("netty4:tcp://ipAddress:port?sync=true")
I start the first queue to receive session open request from clients and then I start the second route to start sending data but I cannot use the same channel.
I tried to get the remote port of the first route and used it in the second route but I have a ConnectException because netty4 tries to open a new socket which is already open.
I found that netty4 can be used asynchronously using the AsyncProcessor but I didn't find any example dealing with my use case.
The only idea I found is that I have to create a standalone server which open the sockets with the clients and make it communicate with the two endpoints.
Is there any way to implement this situation using camel only?
any help on this subject is really appreciated.
Your code won't be able to run as it is for your use case. I also suspect you are trying to use Camel as IP server framework and not an integration in this case.
Lets review Apache Camel's concept of producers and consumers. In the integration world we talk about client and servers as consumers and producers. This might seem like a language difference until you realise a consumer(typically a client) can also be a producer(server).
Some helpful definitions:
1. Producer: A producer is an entity capable of creating and sending a message to an endpoint. A typical example would be code like .to("file:data/outbox") as this produces a file.
2. Consumer: A consumer is an entity that receives messages produced by a producer, it wraps these messages in an exchange and sends them to be processed. A typical example would be code like from(jms:topic:xmlOrders)
A rule of thumb is that typically consumers are the source of the messages being routed.
BIG NOTE:
These two definitions are not set in stone a producer can also be an endpoint using the from and a consumer can be an endpoint using the to.
So in your case let's break up the route:
from("netty4:tcp://ipAddress:port?sync=true").to("wmq:queue:toQueue")
In this route you are creating a Netty server that sends a message to a queue. Here your netty endpoint acts as a consumer(yes it is in the from clause) however this creates a Netty4 Server at the IP address and endpoint you specified. This then send a message to another consumer which is the MQ client which act as a consumer again. So two consumers? Where is the producer? The client connecting to the netty server will act as producer.
Let's look at the second piece of the route:
from("wmq:queue:fromQueue").to("netty4:tcp://ipAddress:port?sync=true")
Here you are creating a client/consumer for the MQ services and then creating a client/producer to the netty server. Essentially you are creating a NEW client here that connects to the SERVER you created in the first route.
So in short your route creates a Netty server that send a message to MQ then creates a MQ client that sends a message to a Netty client which connects to the server you have created. It wont work like this.
Go read about message exchange patterns for further reading, but I would suggest that if you are just using Netty and MQ then maybe Camel is a bit overkill as it is a integration platform and not a IP server platform.
Related
I'm building a message broker which communicates with clients over ZeroMQ PUSH/PULL sockets and has the ability to exclude clients from messages they're not subscribed to from the server side (unlike ZeroMQ pub/sub which excludes messages on the client side).
Currently, I implement it in the following way:
Server: Binds ZeroMQ PULL socket on a fixed port
Client: Binds a ZeroMQ PULL socket on a random or fixed port
Client: Connects to the server's PULL socket and sends a handshake message containing the new client's address and port.
Server: Recieves handshake from client and connects a PUSH socket to the client's PULL server. Sends handshake response to the client's socket.
Client: Recieves handshake. Connected!
Now the client and server can communicate bidirectionally and the server can send messages to only a certain subset of clients. It works great!
However, this model doesn't work if the clients binding PULL sockets are unable to open a port in their firewall so the server can connect to them. How can I resolve this with minimal re-architecting (as the current model works very well when the firewall can be configured correctly)
I've considered the following:
Router/dealer pattern? I'm fairly ignorant on this and documentation I found was sparse.
Some sort of transport bridging? The linked example provides an example for PUB/SUB.
I was hoping to get some advice from someone who knows more about ZeroMQ than me.
tl;dr: I implemented a message broker that communicates with clients via bidirectional push/pull sockets. Each client binds a PULL socket and the server keeps a map of PUSH sockets so that it can address specific subscribers. How do I deal with a firewall blocking the client ports?
You can use the router/dealer to do this like you say. By default the ROUTER socket tracks every connection it has. The way it does this is by having the caller stick the connection identity information in front of each message it recieves. This makes things like pub/sub fairly trivial as all you need to do is handle a few messages server side that the DEALER socket sends it. In the past I have done something like
1.) Server side is a ROUTER socket. The ROUTER handles 2 messages from DEALER sockets SUB/UNSUB. This alongside the identity info sent as the first part of a frame allows the router to know the messages that a client is interested in.
2.) The server checks the mapping to see which clients should be sent a particular type of data using the map and then forwards the message to the correct client by appending the identity again to the start of the message.
This is nice in that it allows a single port to be exposed on the server. Client side we do not need to expose ports, simply just connect to the server ROUTER socket.
See https://zguide.zeromq.org/docs/chapter3/ for more info.
Previously I have setup a device to connect to Google Cloud IoT Core and when the device would connect it would get a callback from the server to a topic with the device configuration file.
I am currently moving off of that to a local MQTT broker on a Raspberry Pi with Mosquitto running. What I am trying to do is figure out how to replicate the sending of the configuration file when connected.
Is there a way for other clients to know when a new client connects? If so I could then just have a client running on the Pi that would is responsible for sending messages.
The thought was that the Pi would hold the configuration file of the connected device and once it connects it would be sent back over to it via a topic for that device
Or is there another solution that I dont even know about yet that accomplishes this? I have not set up my own end to end MQTT communication before so I dont exactly have a path forward here
Is there a way for other clients to know when a new client connects?
Not as part of the protocol; but it's simple enough to publish a message upon connection. I can see a few ways of accomplishing your aim:
Publish all of the configurations as retained messages. You would require a 'config' topic per device for this (say config/uniquedevicename). When the client starts up it will subscribe to the relevant topic and receive its configuration (the 'server' would need to publish a message to each of the config channels with the retain flag set).
Upon starting the device publishes a message requesting its configuration; the server is subscribed to the relevant topic and responds with the configuration. You can use preset topics for the response or pass the topic in as part of the request.
For one (or a small number of devices) option 1 would be very simple to implement; however option 2 provides more flexibility.
I want to implement a chat application which uses "MQTT protocol" in order to send messages from the device(Android phone). I need a "Kafka broker" which would run on the server and listen to these messages.
For this, I need an MQTT proxy, but even after googling a lot I could not find any open source MQTT proxy. Please suggest if there is an open source MQTT proxy. And if not, then, is it possible to implement one of my own?
There is an MQTT Proxy from Confluent, but it is not open source. You can use it free against a single broker, or 30-day trial for a cluster of more than one broker.
I am aware that the Tibco EMS provides Fault Tolerance in a hot backup configuration on the server side as detailed in the User's Guide, this answer and here.
But on the client side does Tibco EMS provide out of the box solution for fault-tolerant clients?
An example: on the topic Sports.F1.PitStop two clients (server1, server2) register as publishers. The idea being that should something go wrong on server1 (i.e. publisher on server1 goes down), server2 would seamlessly continue to publish on the topic. So the question is, does Tibco EMS provide such client-side fault-tolerance capability?
No.
EMS (or JMS) does not support a client-side fault tolerance feature. The reason is simple : typically, publishers processes don't know each other.
To elaborate:
Topics usually accepts many publishers (more than two).
In a pub/sub scenario, publishers don't know subscribers, and to an extent they don't typically know the other publishers.
Your solution:
My first question regarding your solution: Why can't both servers publish messages at the same time ?
I assume you have a good reason (like messages from server1 and server2 being redundant). In that case, then you will have to have some kind of communication between your "active" and "passive" server.
Possibility 1 : Server2 is connected to a simple service/rmi/other heartbeat mechanism, and can tell if/when Server1 as stopped publishing.
Possibility 2 : Server2 is itself subscribed to the topic, and can tell when messages have stopped.
Last note:
In case you meant shared "subscriptions" (as in, one durable subscription being shared between two servers): the new JMS 2.0 API is supporting this feature. EMS 8 is the only version of EMS supporting JMS 2.0.
I'm not sure if I fully understand your question. EMS is a message broker. It brokers messages. Thats all it does. If you have multiple servers publishing to the same topic, then that is fine for EMS. If your publishers themselves are in a fail-over configuration, so that only one is actively sending and the other takes over when the first one fails, then EMS doesn't care.
Managing the fail-over mechanism from one publisher to the other, that is something you have to develop yourself. EMS doesn't offer anything to support that. You need some kind of mechanism for server2 to know when server1 is down. There are several ways you can do that:
Use a heartbeat mechanism where the active server sends periodical heartbeat messages (possibly across EMS as well) and the passive server listens to them and when they stop, assume server1 is down.
Use an active open connection between server1 and server2 (open a simple tcp port for example and stop it from closing automatically by periodically sending a dummy message), and have an open read on that port. You'll get a connection error the moment server1 goes down and you don't have to wait for timeouts on heartbeats.
Use a third party monitoring tool, such as openview, tivoli or nagios to detect server1 is down and inform server2.
There is a feature on EMS you can use. EMS has system admin topics that you can subscribe to and that log basically anything that goes on internally. You can subscribe to connect and disconnect events to monitor any component connecting to or disconnecting from EMS. If server1 fails, this will be visible as a disconnect event.
heartbeat mecanism:
https://support.tibco.com/s/article/Tibco-KnowledgeArticle-Article-33918
For example:
client_heartbeat_server=10
client_timeout_server_connection=35
server_heartbeat_client=10
server_timeout_client_connection=35
When deploying a web application running on a traditional web server, you usually restart the web server after the code updates. Due to the nature of HTTP, this is not a problem for the users. On the next request they will get the latest updates.
But what about a WebSocket server? If I restart or kill the old process all connected users will get disconnected. So my question is, what kind of strategy have you used to deploy a WebSocket server smoothly?
You're right, every connected user will be disconnected if the server restarts.
I think the less bad solution is to tell to the client to reconnect in the onClose method of the client.
WebSockets is just a transport mechanism. Libraries like socket.io exist to build on that transport -- and provide heartbeats, browser fallbacks, graceful reconnects and handle other edge-cases found in real-time applications.
In our WebSocket-enabled application, socket.io is central to ensuring our continuous deployment setup doesn't break users' active socket connections.
If clients are connected directly to sever that does all sockets networking and application logic, then yes - they will be disconnected, due to TCP layer that holds connection.
If you have gateway that clients will be connecting to, and that gateway application is running on another server, but will communicate and forward messages to logical server, then logical server will send them back and gateway will send back to client responses. With such infrastructure, you have to implement stacking of packets on gateway until it will re-establish connection with logical server. Logical server might notify gateway server before restart. That way client will have connection, it will just wont receive any responses.
Or you can implement on client side reconnection.
With HTTP, every time you navigate away, browser actually is creating socket connection to server, transmits all data and closes it (in most cases). And then all website data is local, until you navigate away.
With WebSockets it is continuous connection, and there is no reconnection on requests. Thats why you have to implement simple mechanics when WebSockets getting closing event, you will try to reconnect periodically on client side.
It is more based on your specific needs.