So, You Want to Start Monitoring Your AWS Account?
Before implementing an AWS monitoring solution, address “What” and “Why”.
Before we jump into our recommendations for best practices for monitoring AWS accounts, we need a 30,000′ view of why we need to monitor activity and what we are trying to achieve.
What Do We Need to Monitor, and Why?
First off, many teams make monitoring a priority to head off small problems before they become big ones. Creating and implementing a monitoring plan that collects data across your entire AWS infrastructure helps you to detect and troubleshoot complicated failures when they occur.
Being proactive in cloud data collection assists your speed to recovery, and removes a ton of frustration in the event of failure. Through a broad but structured monitoring program, you can predictive potential outages, rather than waiting for something bad to happen. In some cases you may find yourself even preventing failures completely.
Yet many of us still do not make monitoring a priority for managing AWS accounts, due to lack of time and resources. We cannot stress enough how important monitoring is to the health of your cloud environment. With the cloud management platforms available in the marketplace today, which easily connect to AWS instances or cloud environments, not having enough resources is no longer a legitimate excuse. A good plan, paired with an appropriate management or automation solution, can fill this void. This is where prioritizing a list of monitoring tasks and automating those routine tasks can yield significant dividends.
Your monitoring plan should address, at minimum, the following questions:
- What are your goals for monitoring?
Your goals should directly express what you are trying to achieve, for example: catch problems before an outage results, prevent overspending and track usage levels, etc.
- What resources will you monitor?
You want to monitor as much as possible. Start small and expand into as many areas as possible across your cloud environment.
- How often you monitor will these resources?
Hourly, daily, weekly, monthly? This answer depends on your environment, available resources, and where the resources are located within the infrastructure (ie. Production or development/testing)
- What monitoring tools will you use?
AWS provides a number of monitoring tools to provide deep visibility into many aspects of the AWS environment, and those tools form the basis of your monitoring approach. But you’ll also want to make sure you are able to enforce the best practices and address issues found by the monitoring platforms. Thus, a cloud management platform can (and should) be used in tandem with the AWS monitoring tools.
- Who will look at the monitoring data and determine when issues occur?
Humans or automated management solution, or both?
- Who should be notified when something goes wrong?
Having a clear escalation process is critical to the success of any monitoring initiative. We recommend a flexible notifications capability that alerts team members about issues where and how they want to be notified. Maybe it’s via email or text. Maybe they want a Slack message or something to be put in their ServiceNow queue. Regardless, making sure the right team members have the information they need to handle issues is critically important to the success of your monitoring strategy.
Remember, a plan without action is clearly a plan to fail. Be sure to subscribe to our blog and follow along in our next post as we outline the next phase of implementing a monitoring plan with our recommended 3-Step Process to Start Monitoring Your AWS Cloud Environments.