At some point, we’ve all bought something labeled “one size fits all.” But let’s be honest, that commonly means “okay for most,” or “not perfect for anyone.” You wouldn’t buy a suit labeled one size fits all, because you’d want the details tailored to your specific requirements. While the lack of an effective “one size fits all” approach is obvious when we’re talking about clothing, it’s also particularly relevant for performance monitoring solutions.
“One size fits all” monitoring solutions, which lack the flexibility to customize monitored data, rely on grand promises of universal compatibility. However, such platforms often result in less value and more noise post-deployment. The modern production infrastructure requires more customization than the “one size fits all” solution can deliver.
Not only does LogicMonitor give you the best possible starting point when it comes to performance monitoring, but we also provide you with the tools you need to tailor the suit to fit your specific needs. Let’s put this in the context of performance monitoring for Amazon Web Services (AWS):
While LogicMonitor’s long list of CloudWatch integrations and extensive library of LogicModules provide comprehensive performance monitoring for AWS resources out-of-the-box, we also offer the AWS Java SDK as part of our Collector. The built-in AWS SDK acts as a toolbox that allows you to customize the information you see in your LogicMonitor environment, ensuring it aligns with your team’s particular focus and goals.
Here at LogicMonitor, we recently leveraged the built-in AWS SDK to monitor and alert on EC2 scheduled events and CloudWatch Logs.
EC2 scheduled events occur when your instance is scheduled to be stopped, terminated, or rebooted due to underlying hardware issues on AWS’s end. We set up a datasource to help ensure that our team is always prepared for scheduled events when they occur:
The datasource applies to all EC2 instances (devices with the category “AWS/EC2”) and discovers events as instances (to account for multiple scheduled events). LogicMonitor’s built-in AWS SDK lets us get the information elegantly with two simple scripts (one for discovery and one for data collection), as displayed in the following images:
The above datasource automatically detects all EC2 scheduled events and displays the remaining hours until the event will occur. This enables us to set thresholds on the number of hours remaining to determine when to trigger and route alerts to our TechOps team members.
CloudWatch Logs, an AWS log storage service, can be used to collect logs from EC2 instances, as well as other sources. Monitoring CloudWatch Logs enables us to catch all application errors and exceptions, and view them alongside numerical performance data.
We set up an eventsource to pull in all CloudWatch Logs containing “ERROR,” and again leveraging the built-in AWS SDK we were able to easily do this with the following script:
With the above EventSource, all CloudWatch Log entries that contain “ERROR” will trigger event alerts in LogicMonitor. We can potentially use these alerts to explain changes in performance data.
Of course, we always work towards adding out-of-the-box solutions when it makes sense to. In the case of EC2 scheduled events, we’ll add the datasource to our repository, as will also be the case for CloudWatch Logs. However, sometimes providing the tools to support a custom fit is more valuable, as is the case if you’re trying to monitor a custom application.
With LogicMonitor, you can rest assured knowing that you’ll always be able to tailor a custom-fit solution for your environment. So, next time those shiny pre-packaged bells and whistles draw you near, ask how flexible the underlying platform is in case the “one size fits all” approach doesn’t meet your needs.
Interested in learning more about LogicMonitor’s AWS Monitoring capabilities? Come meet
the LogicMonitor team at AWS re:Invent 2016 in Las Vegas! We are giving away prizes and limited edition LogicMonitor swag on-site at our booth #2436. Our Events Team would love to meet you and show how we can help you manage your complex and agile workloads with a product demo. Contact [email protected] with any questions and follow us on Twitter and LinkedIn for more updates.