Sails, a MVC framework for Node.js, and MongoDB.
- Install dependencies:
npm install
- Start the application:
npm start
ornpm run-script watch
This example uses MongoDB for models and Redis for session and sockets stores.
When Sails is run locally, it will connect to local Redis and MongoDB databases using the default settings. Should you have any issues, you can create and edit config/local.js
to override the default config and use your own. These settings take precedence over all other config files, including those in the config/env
subfolder.
Here is an example file that uses memory for sockets, memory for sessions, and a MongoLab server for the models.
module.exports = {
connections: {
mongodb: {
host: 'ds029541.mongolab.com',
port: 29541,
user: 'foo',
password: 'bar',
database: 'baz'
}
},
session: {
adapter: 'memory'
},
sockets: {
adapter: 'memory'
}
};
For more, see the Sails.js configuration documentation.
Sails is deployed to an instance on Heroku, an application platform. If this is your first time deploying the app, follow steps 1-6. Otherwise, just skip to step 4. For more, see the Heroku Documentation on Node.js applications.
- Create new app:
heroku create app-name
- Add Redis:
heroku addons:add redistogo
- Add MongoDB:
heroku addons:add mongolab
- Deploy the app:
git push heroku master
- Set the environment to production:
heroku config:set NODE_ENV=production
- Insure that at least one instance is running:
heroku ps:scale web=1
Other helpful tasks:
- View application logs:
heroku logs --tail
- Open the application:
heroku open
- View application configuration:
heroku config
- Add existing Heroku Git Origin:
heroku git:remote -a appname
Addons:
The command that Heroku executes to start the application is defined in the procfile. The procfile declares a single process type, web
, and the command needed to run it. For more, see the Heroku documentation.
Node.js (and consequently Sails.js) apps scale horizontally. It's a powerful, efficient approach, but it involves a tiny bit of planning. At scale, you'll want to be able to copy your app onto multiple Sails.js servers and throw them behind a load balancer.
One of the big challenges of scaling an application is that these sorts of clustered deployments cannot share memory, since they are on physically different machines. On top of that, there is no guarantee that a user will "stick" with the same server between requests (whether HTTP or sockets), since the load balancer will route each request to the Sails server with the most available resources. However that means that all room/pubsub/socket processing and shared memory has to be offloaded to a shared, remote messaging queue (usually Redis).Luckily, Socket.io (and consequently Sails.js) apps support Redis for sockets by default.
It is worth mentioning that, when NODE_ENV=production
, config variables are used for the Redis connection values. If these config variables are not available, Sails will try to connect to redis database 0 running on localhost via port 6379. The default password (none) will be used.
The Redis connections are configured in config/sockets.js
and config/session.js
. See options for the Redis adapter if further configuration is needed.
# Backbone Conventions
GET : /:controller => findAll()
GET : /:controller/read/:id => find(id)
POST : /:controller/create => create()
POST : /:controller/create/:id => create(id)
PUT : /:controller/update/:id => update(id)
DELETE: /:controller/destroy/:id => destroy(id)
# You can also explicitly state the action
GET : /:controller/find => findAll()
GET : /:controller/find/:id => find(id)
POST : /:controller/create => create(id)
PUT : /:controller/update/:id => update(id)
DELETE: /:controller/destroy/:id => destroy(id)
Copyright 2015 Lawn Chair Studios. All rights reserved.