Skip to main content

nodejs push java redis

Disclaimer: The code below is just a general guidelines and I am not pasting full code so you would have to fill in the blanks if you want to use it.

Continuing http://neopatel.blogspot.com/2014/02/redis-publish-subscribe-to-nodejs.html  finally I was able to write the nodejs app that would listen to events from redis and publish message to browser.  The architecture looks like



1) Your browser will use SockJS javascript library to connect to the nodejs server and it would then listen to events. The normal html code looks like

<script src="http://cdn.sockjs.org/sockjs-0.3.min.js"></script>
<script type="text/javascript">
    isOpen = false;
    var newConn = function(){
        var sockjs_url = 'https://xxx.com/push';
        var sockjs = new SockJS(sockjs_url);
        sockjs.onopen = function(){
            sockjs.send(JSON.stringify({"userName": "kpatel@acme.com", "sessionId": "e1930358-87d2-4d2b-86b6-2b2373acfaf1"}));
            isOpen = true;           
            console.log('Connected.');
        };
        sockjs.onmessage = function(e){
            console.log("Got message:" + e);           
            if(e.data == 'fileSystemEvent'){
                si.call = si.fn();
            }
        };
        sockjs.onclose = function(){
       
            var recInterval = setInterval(function(){
                    newConn();
                    if(isOpen) clearInterval(recInterval);
                }
                , 15000
            );
            console.log('Closed.');
            console.log('Connecting...');
        };
    };
    newConn();
</script>

2) Now on server first thing you need is a c10K server like nginx or haproxy.
I had to install nginx1.4.3 because the default that came with sudo apt-get didnt had the websocket support.
I just used this to install nginx 1.4.3
wget http://nginx.org/download/nginx-1.4.3.tar.gz
tar xvzf nginx-1.4.3.tar.gz
cd nginx-1.4.3

./configure \
--user=nginx                          \
--group=nginx                         \
--prefix=/etc/nginx                   \
--sbin-path=/usr/sbin/nginx           \
--conf-path=/etc/nginx/nginx.conf     \
--pid-path=/var/run/nginx.pid         \
--lock-path=/var/run/nginx.lock       \
--error-log-path=/var/log/nginx/error.log \
--http-log-path=/var/log/nginx/access.log \
--with-http_gzip_static_module        \
--with-http_stub_status_module        \
--with-http_ssl_module                \
--with-pcre                           \
--with-file-aio                       \
--with-http_realip_module             \
--without-http_scgi_module            \
--without-http_uwsgi_module           \
--without-http_fastcgi_module        
make
sudo make install

3) Then I had to add the below upgrade headers to make nginx upgrade the http socket to websocket

    location /push {
            proxy_pass              http://localhost:8090;
        proxy_set_header        X-Real-IP $remote_addr;
        proxy_set_header        X-Forwarded-For $proxy_add_x_forwarded_for;
        proxy_set_header X-Forwarded-Proto https;
        proxy_set_header        Host $http_host;
        # WebSocket support (nginx 1.4)
        proxy_http_version 1.1;
        proxy_set_header Upgrade $http_upgrade;
        proxy_set_header Connection "upgrade";

    } 

4) I started the nodejs server on 8090 port and tomcat was on 8080. Now bare bones nodejs wont give you everything you need so you need several packages. I installed all of them using

sudo add-apt-repository ppa:chris-lea/node.js
sudo apt-get update
sudo apt-get install nodejs
sudo npm install -g redis sockjs socket.io winston



a) I used winston for logging like log4j is in java
b) redis package to talk to redis
c) sockjs for sockjs connections code

I just installed the modules globally as this is a scratch app

5) Inside nodejs app when a connection comes the client will pass username and sessionid. so nodejs needs to validate that sessionid with tomcat and then add it to a local map. Then when it receives the redis event it needs to just write it to the socket associated with the username or reject it if there is no open connection to the user.

var sockjs  = require('sockjs');
var http    = require('http');
var redis   = require('redis');

//    Store all connected sessions
var sessionMap = {};

//    Create redis client
var redisClient = redis.createClient(6379, '127.0.0.1');

//    Create sockjs server
var sockjsServer = sockjs.createServer();

// Sockjs server

sockjsServer.on('connection', function(conn) {
    conn.on('data', function(message){
        var data = JSON.parse(message || "{}");
        //    call function to check if current user is exist in db
        result = validateSession(data.userName, data.sessionId, function(response){
            logger.info("Got response" + response);
            if(response.success === true){
                addConnectionToSessionMap(conn,data.sessionId,data.userName);
                conn.write(JSON.stringify({success:true, statusCode:response.statusCode}));
            } else {
                logger.info("Got error: " + response.message);
                conn.write(JSON.stringify({success:false, statusCode:response.statusCode}));
                conn.end();
            }
        });
    });
    conn.on('close', function(){
        //    remove connection from Maps if connection is closed
        if(connMap[conn.id]){
            removeConnection(conn, '');
            conn.end();
        }
        logger.info("Closing...");
    });
});

redisClient.subscribe(config.redis_push_channel);

//    Push incoming message to all connected users
redisClient.on("message", function(channel, message){
    var data = JSON.parse(message || "{}");
    //figure out the connections to write message based on users and write message to them
    //conn.write(data)
});


//    Create http server
var server = http.createServer();
//    Hook sockjs in to http server
sockjsServer.installHandlers(server, {prefix:'/push'});   
server.listen(8090);

6) But coming from java world nodejs seemed fragile as it would just kill itself on every exception
so that when I added logging and in winston I added "exitOnError: false"

var winston = require('winston');

var logger = new (winston.Logger)({
  transports: [
    new winston.transports.File({ filename: __dirname + '/logs/push.log', json: false })
  ],
  exceptionHandlers: [
    new winston.transports.File({ filename: __dirname + '/logs/push.log', json: false })
  ],
  exitOnError: false
});

7) also nodejs wont have usual startup/stop things like tomcat has so you need to cook up your own something like

export NODE_PATH=/usr/lib/node_modules
nohup node push_server.js 1>>"nohup.out" 2>&1 &
echo $! > push.pid"

In short I am still excited for nodejs because if your startup has JS skills then JS developers can quickly start pitching into nodejs code and write server code.

Also being event model it scales pretty fine similar to the way c10k servers like nginx and haproxy scales. To me bigger learning was event based programming where you just write what happens on what event and let the framework take care of things. Good thing was that the browser has a live connection and I can publish message to redis when a file is added and in browser console logs for all users I can see the event instantly :).


Comments

Popular posts from this blog

Haproxy and tomcat JSESSIONID

One of the biggest problems I have been trying to solve at our startup is to put our tomcat nodes in HA mode. Right now if a customer comes, he lands on to a node and remains there forever. This has two major issues: 1) We have to overprovision each node with ability to handle worse case capacity. 2) If two or three high profile customers lands on to same node then we need to move them manually. 3) We need to cut over new nodes and we already have over 100+ nodes.  Its a pain managing these nodes and I waste lot of my time in chasing node specific issues. I loath when I know I have to chase this env issue. I really hate human intervention as if it were up to me I would just automate thing and just enjoy the fruits of automation and spend quality time on major issues rather than mundane task,call me lazy but thats a good quality. So Finally now I am at a stage where I can put nodes behing HAProxy in QA env. today we were testing the HA config and first problem I immediat...

Adding Jitter to cache layer

Thundering herd is an issue common to webapp that rely on heavy caching where if lots of items expire at the same time due to a server restart or temporal event, then suddenly lots of calls will go to database at same time. This can even bring down the database in extreme cases. I wont go into much detail but the app need to do two things solve this issue. 1) Add consistent hashing to cache layer : This way when a memcache server is added/removed from the pool, entire cache is not invalidated.  We use memcahe from both python and Java layer and I still have to find a consistent caching solution that is portable across both languages. hash_ring and spymemcached both use different points for server so need to read/test more. 2) Add a jitter to cache or randomise the expiry time: We expire long term cache  records every 8 hours after that key was added and short term cache expiry is 2 hours. As our customers usually comes to work in morning and access the cloud file server it ...

Spring 3.2 quartz 2.1 Jobs added with no trigger must be durable.

I am trying to enable HA on nodes and in that process I found that in a two test node setup a job that has a frequency of 10 sec was running into deadlock. So I tried upgrading from Quartz 1.8 to 2.1 by following the migration guide but I ran into an exception that says "Jobs added with no trigger must be durable.". After looking into spring and Quartz code I figured out that now Quartz is more strict and earlier the scheduler.addJob had a replace parameter which if passed to true would skip the durable check, in latest quartz this is fixed but spring hasnt caught up to this. So what do you do, well I jsut inherited the factory and set durability to true and use that public class DurableJobDetailFactoryBean extends JobDetailFactoryBean {     public DurableJobDetailFactoryBean() {         setDurability(true);     } } and used this instead of JobDetailFactoryBean in the spring bean definition     <bean i...