Creating a CloudWatch Dashboard for Monitoring EC2 Instances in AWS

Module 3: CloudWatch Dashboard Assignment

Problem Statement: You work for XYZ Corporation. To maintain the security of the AWS account and the resources you have been asked to implement a solution that can help easily recognize and monitor the different users. Also, you will be monitoring the machines created by these users for any errors or misconfigurations.

Tasks To Be Performed:

1. Create a dashboard which lets you check the CPU utilization and networking for a particular EC2 instance.

1. Preparation:

   • I first ensured that I had two EC2 instances running: instance1 and instance2.
     

2. Building My Dashboard:

   • I logged into AWS and went straight to CloudWatch.
   • On the left side, I clicked on Dashboards and then hit Create Dashboard.
   • I gave my dashboard the name EC2s and clicked Create dashboard

3. Populating the Dashboard with Widgets:

   • Inside my new dashboard I get prompted to Add widget.
   • I chose the Line widget because it visually made the most sense for what I wanted.
   • Under Metrics, I navigated to EC2 -> Per-Instance Metrics.
   • I selected CPUUtilization and then filtered by my EC2 instance’s ID or name for accuracy.
   • For networking, I picked metrics like NetworkPacketsIn and NetworkPacketsOut. Again, I made sure to filter by my EC2 instance.
     Here I picked Instance1 will repeat steps for Instance2
   • Once I added all my metrics, I clicked Create widget.
   • The result is a dashboard with 2 widgets for Instance1 and Instance2
     

4. Saving My Work:

   • I was happy with how everything looked, so I clicked Save dashboard.

5. Simulating Load on Instance1: ^877754

   • To compare the metrics side-by-side, I decided to simulate some load on instance1 while leaving instance2 idle.
   • I SSH-ed into instance1.
   • To increase CPU usage, I ran: while true; do openssl dgst -sha256 /dev/zero; done &
   • To generate some network traffic, I used the curl command to download a large file multiple times: while true; do curl -O https://dumps.wikimedia.org/enwiki/latest/enwiki-latest-pages-articles.xml.bz2; done &
   • This step ensured that instance1 was now actively using CPU and network resources, while instance2 remained idle.

6. Results from CloudWatch Metrics: After initiating the CPU-intensive task on instance1

   NetworkPacketsIn:
   Instance1 showcased heightened incoming packets due to the network tasks.

   CPUUtilization:
   High CPU usage was evident on Instance1 from the hashing operation.