How can I automatically scale my Droplets based on CPU usage using Terraform and DigitalOcean?

Question

I know that DigitalOcean still does not have auto scaling for Droplets, but I’m using Terraform to manage my infrastructure and want to set up an auto-scaling group for Droplets based on CPU usage.

Ideally, when CPU usage exceeds a certain threshold, new Droplets should be created automatically, and when the usage decreases, they should be removed. What’s the best way to approach this with Terraform and DigitalOcean, and are there any limitations or additional services I need to integrate to achieve this?

Bobby · Answer

Hey! Great question! Indeed, as you mentioned, while DigitalOcean doesn’t have native auto-scaling for Droplets, you can still achieve something similar by combining Terraform with monitoring tools like DigitalOcean’s Monitoring and external automation for scaling. First, set up DigitalOcean Monitoring or an external monitoring tool like Prometheus to keep track of CPU usage on your Droplets. DigitalOcean Monitoring can trigger alerts when CPU thresholds are exceeded, which is key for deciding when to scale up or down: https://docs.digitalocean.com/products/monitoring/ Once you have the monitoring in place, since Terraform isn’t event-driven, you’ll need to use a cron job, CI/CD pipeline (like GitHub Actions or Jenkins), or another automation tool to trigger Terraform runs. The idea is to monitor the CPU usage and, when the usage crosses a defined threshold, trigger Terraform to either scale up by creating more Droplets or scale down by destroying unnecessary ones. Here’s an example of how you can use a variable in Terraform to control the number of Droplets: variable "droplet_count" { default = 1 } resource "digitalocean_droplet" "web" { count = var.droplet_count name = "web-${count.index + 1}" image = "ubuntu-20-04-x64" region = "nyc3" size = "s-1vcpu-1gb" } In this case, the droplet_count variable will control how many Droplets are created. You can modify this value dynamically based on the CPU usage and then trigger a Terraform apply. To distribute traffic across the Droplets, use DigitalOcean’s Load Balancers. Here’s how you can set up a load balancer using Terraform: resource "digitalocean_loadbalancer" "lb" { name = "web-lb" region = "nyc3" forwarding_rule { entry_port = 80 entry_protocol = "http" target_port = 80 target_protocol = "http" } droplet_ids = digitalocean_droplet.web[*].id } This will provide a way to “auto-scale” your Droplets based on demand, even though it’s not built into DigitalOcean’s platform natively. On another note, if you’re just getting started with Terraform and DigitalOcean, this ebook might be helpful: Introduction to Terraform ebook . Let me know if you have any other questions, and good luck with your project! - Bobby

Become a contributor for community

DigitalOcean Documentation

Resources for startups and AI-native businesses

The developer cloud

Start building today

Report this

How can I automatically scale my Droplets based on CPU usage using Terraform and DigitalOcean?

Related questions

How can I use my droplets cloud server as a proxy?

KeepAlived service running but does not executes the check haproxy script

Using HaProxy on Nginx server. Not listening to port 80.

Get domain "referrer" name on instances that behind Load Balancer

High availability infrastructure with Load Balancer, HAProxy and Managed Databases

Cache Doesn't Improve Request Per Second

haproxy - cannot bind socket error - ubuntu 20.04

haproxy - cannot bind socket error - ubuntu 20.04

Proxy backbone/intermediary

Configure subdomain of different host on nginx server which is configured on some other host.

Report this

How can I automatically scale my Droplets based on CPU usage using Terraform and DigitalOcean?

Related questions

How can I use my droplets cloud server as a proxy?

KeepAlived service running but does not executes the check haproxy script

Using HaProxy on Nginx server. Not listening to port 80.

Get domain "referrer" name on instances that behind Load Balancer

High availability infrastructure with Load Balancer, HAProxy and Managed Databases

Cache Doesn't Improve Request Per Second

haproxy - cannot bind socket error - ubuntu 20.04

haproxy - cannot bind socket error - ubuntu 20.04

Proxy backbone/intermediary

Configure subdomain of different host on nginx server which is configured on some other host.