Automate a slow scaledown of an AWS ASG in Asgard

With this CLI tool, you can automate a slow scaledown of an AWS ASG in Asgard. If your server has long lived connections, you might want to terminate the instance one a time to allow time to redistribute the load across the cluster.

How to use

By default, the tool will scale down an existing ASG to 0 instance with a 5 minute pause between ASG resize requests

npm install
node index.js --asgardHost <your-asgard-hostname> realtime-ASG-CARFGI70UCAL-v524

To stop multiple instances on each scaledown request, a rate option can be used:

node index.js --asgardHost <your-asgard-hostname> --rate 2 realtime-ASG-CARFGI70UCAL-v524

To request a scaledown every 3 minutes, a delay option can be used (whole numbers only):

node index.js --asgardHost <your-asgard-hostname> --delay 3 realtime-ASG-CARFGI70UCAL-v524

Not yet implemented

I haven't implemented some things, because I don't yet need them. I'm happy to take pull requests if you'd like to see something get added

1. Allow HTTPS communication to Asgard
2. Allow authentication with Asgard
3. Reading a default configuration file for settings that don't change often like asgardHost