A simple way to wrap a single-instance module to enable it to work with node cluster.
cluster-rpc
===========
A simple way to wrap node.js modules for use with cluster even though they
were designed to be used in a single non-cluster instance.
Install
=======
``
npm install --save cluster-rpc
Usage
=====
In the master process you will create the real instance of whatever module
you're trying to use (for example express-session/session/memory, sqlite3, level)
and then you will supply the names of the methods you wish to export to worker
processes.
By default each worker will be added when cluster emits a fork event.addOnFork
If needed you can set to false and call addWorker(worker) manually.
`javascript
// You can pick any module with thunk-style callbacks
// For example:
var db = require('level')('./mydb')
// Wrap the instance
var crpc = require('cluster-rpc/master').create({
addOnFork: true // default
, instance: db
, methods: [ 'get', 'put' ]
, name: 'foo-level'
});
// If you set addOnFork to false, You must manually add each worker
// crpc.addWorker(cluster.fork());
crpc.then(function (db) {
// processes are connected and ready
// 'db' is the original instance
});
`
`javascript
// retrieve the instance
var crpc = require('cluster-rpc/worker').create({
name: 'foo-level'
});
// all listed methods will be rpc'd
crpc.then(function (db) {
// db.put, db.get
});
`
Example
=======
`javascript
'use strict';
var cluster = require('cluster');
var crpc;
if (cluster.isMaster) {
crpc = require('cluster-rpc/master').create({
addOnFork: false
, instance: require('level')('./mydb')
, methods: [ 'get', 'put' ]
, name: 'foo-level'
});
crpc.addWorker(cluster.fork());
crpc.then(function () {
db.put('foo', 'bar');
});
}
else {
crpc = require('cluster-rpc/worker').create({
name: 'foo-level'
});
}
crpc.then(function (db) {
setTimeout(function () {
db.get('foo', function (err, result) {
console.log("db.get('foo')", result);
});
}, 250);
});
``