Auto Scaling Group

testgrinder service will be discontinued on January 31, 2025

testgrinder SMPC was discontinued as of February 1, 2024

Auto Scaling group will help you maintain the specified number of running healthy tg-app instances. The recommended number of running tg-app instances is 2. You may run just 1 instance if you can tolerate a short outage while ASG is detecting the outage and replacing the failed tg-app instance. The ASG will be configured to scale with the load automatically up to the limit you specify.

You can manage auto scaling groups on the Auto Scaling Groups page in AWS EC2 console at https://console.aws.amazon.com/ec2

Create Auto Scaling Group

1. Visit the Auto Scaling Group page in AWS EC2 console

2. Click Create Auto Scaling group

3. Step 1: Choose launch template or configuration

   1. Name:

      1. Auto Scaling group name: tg-prod-app

   2. Launch template:

      1. Launch template: choose the launch template created earlier

      2. Version: Latest

   3. Click Next

4. Step 2: Choose instance launch options

   1. Network:

      1. VPC: choose testgrinder VPC

      2. Availability Zones and subnets: choose all

   2. Click Next

5. Step 3: Configure advanced options

   1. Load balancing:

      1. Choose: Attach to an existing load balancer

      2. Choose: Choose from your load balancer target groups

      3. Existing load balancer target groups: choose all

   2. Health checks:

      1. Health check type: ELB

      2. Health check grace period: 120 seconds

   3. Click Next

6. Step 4: Configure group size and scaling policies

   1. Group Size:

      1. Desired Capacity: 2

         1. You may choose to go with just 1 if you can tolerate a short outage while ASG is detecting a failed tg-app instance and replacing it with a healthy one.

      2. Minimum Capacity: 2

      3. Maximum Capacity: 5

         1. If load increases ASG will scale up to this number of tg-app instances. Recommended number is 5, but you may choose a different number.

   2. Scaling Policies: choose Target tracking scaling policy

      1. Scaling Policy Name: Target 50% CPU Utilization

      2. Metric type: Average CPU Utilization

      3. Target value: 50

      4. Instances need: 120 seconds

   3. Click Next

7. Step 5: Add notifications

   1. Configure if you'd like to get notifications from ASG

   2. Click Next

8. Step 6: Tags

   1. Click Next

9. Step 7: Review

   1. Click Create Auto Scaling group

   2. If you chose to receive notifications on Step 7 you should receive a test notification at this point

Once the Auto Scaling group is created, you should see an instance of tg-app launch and accessible at the domain name you configured.