ec2-metric-alarm.md

Define dynamic scaling policy

In this lab you will define a dynamic scaling policy for your autoscaling group based on CPU level.

Helpful commands

Get instance IPs

aws ec2 describe-instances --instance-ids <instance_id> --query 'Reservations[0].Instances[0].PrivateIpAddress'

Lab

Provision a group

ansible-playbook provision-elb-ec2-metric-alarm-playbook.yml

Verify instance states

aws autoscaling describe-auto-scaling-groups --auto-scaling-group-names workshop-elb-ec2-metric-alarm-asg --query 'AutoScalingGroups[0].Instances'

Verify ELB state

aws elb describe-instance-health --load-balancer-name workshop-elb-ec2-metric-alarm-lb

Verify alarms

aws cloudwatch describe-alarms --alarm-names cpuDown_workshop-elb-ec2-metric-alarm-asg cpuUP_workshop-elb-ec2-metric-alarm-asg

Copy stress rpm onto any of the instances

aws autoscaling describe-auto-scaling-groups --auto-scaling-group-names workshop-elb-ec2-metric-alarm-asg --query 'AutoScalingGroups[0].Instances'
aws ec2 describe-instances --instance-ids <instance_id> --query 'Reservations[0].Instances[0].PrivateIpAddress'
scp -i <ssh_key> rpm/stress.rpm <user>@<instance_ip>:/tmp/

SSH to the machine, install 'stress' and run it

ssh <user>@<instance_ip> -i <path_to_key>
sudo su
rpm -i /tmp/stress.rpm
stress -c 50

Watch the current CPU level and verify that ASG scales up

watch uptime

Stop 'watch uptime' and verify that ASG scales down

Clean up

ansible-playbook provision-elb-ec2-metric-alarm-playbook.yml --extra-vars "state=absent" --tags "asg"
ansible-playbook provision-elb-ec2-metric-alarm-playbook.yml --extra-vars "state=absent" --tags "lc"