Skip to content

zendesk/ipcluster

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

85 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

IPCluster

repo-checks

IPCluster is a "bare-metal" library for Node.js to manage multi-core concurrency, with sticky sessions and zero-downtime reloading.

Features

  • Sticky sessions (via iptables abuse) - suitable for long-polling
  • Configurable per-process and cluster memory limits
  • Graceful shutdowns and restarts
  • Renice older workers

Dependencies

  • npm modules - toobusy-js, q

IPCluster sticky sessions currently requires iptables for sticky sessions and so is Linux-only

Setup

npm dependencies

npm install

iptables sudo support

sudo iptables support is required in order to support sticky sessions.

Add to sudoers (visudo):

<node> ALL=(ALL) NOPASSWD: /sbin/iptables -t nat -L
<node> ALL=(ALL) NOPASSWD: /sbin/iptables -t nat -[ID] PREROUTING -p tcp -m tcp -s 0.0.0.*/0.0.0.* -d * --dport 80 -j REDIRECT --to-ports ?????
<node> ALL=(ALL) NOPASSWD: /sbin/iptables -t nat -[ID] PREROUTING -p tcp -m tcp -s 0.0.0.*/0.0.0.* -d * --dport 443 -j REDIRECT --to-ports ?????

etc. for other ports

Change <node> to the running user.

Thankfully, iptables does not allow duplicate copies of most flags, so that the * wildcards can't be abused (much). Nonetheless, security feedback is welcome. An alternative for * would be appreciated.

Usage

npm install ipcluster --save

See https://github.com/zendesk/zopim-ipcluster-example for a simple example.

In your application

Deploying new code / Running a new batch of workers

Once you app is ipcluster-enabled, starting it should start a master that will spawn multiple workers. The workers are not child processes (they will outlive the master).

To start a new batch a workers (e.g. when you are deployig new code), you should kill off the old master, and start a new master. The old workers will still be running happily.

Starting a new master will do 2 things:

  • Spawn a new generation of workers, running your latest code for all new incoming connections
  • Become the master for all old workers, and manage their retirement and (eventual) destruction

Note: you can start a new master without first killing the old one. The new master will find and terminate the old master as part of its boot up sequence.

Signals

The master process understands these signals:

  • SIGUSR1: Kills all retired workers.

Contributing

Linting

Run eslint

npm run lint

Testing

ipcluster contains some unit tests made with nodeunit (with many more to come, thanks to you!). At the root of the project run:

npm test

Do run the test suite prior to committing.

Copyright and license

Copyright 2014 Zopim

Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License.

You may obtain a copy of the License at http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.

About

Node.js master/worker clustering module for sticky session load balancing using IPTABLES

Resources

License

Stars

Watchers

Forks

Packages

No packages published