You are here

How to Monitor ECS Containers on AWS

IT Modernization and Containers

Containers are increasingly becoming a core technology in modern IT infrastructure. They provide for resource isolation without the overview of the virtual hardware in server virtualization. If you aren’t already familiar with containers, contact us at Opsview to help you understand their value and importance in your modern IT infrastructure.

Docker, the pioneer in container images, can construct a system image that contains your application code, all its necessary dependencies, the precise required version of the programming language you use, and its host operating system. Docker allows developers to create exact configurations that aren’t affected by the host servers on which they run, or any other processes running around them. Testing, development with peers, and production deployment is consistent, dependable, and reliable. Docker is the standard container format on Amazon AWS.

Monitoring Docker Containers on AWS

The isolation that Docker provides is a tremendous advantage for DevOps teams. That said, this isolation can come at a price if you don’t have a solution to monitor the health of the container. CloudWatch metrics of EC2 performance are helpful, but lack the granularity and scope that an in-process, fit-for-purpose solution like Opsview can provide.

Monitoring Docker containers in the Amazon cloud is not substantially different from monitoring Docker containers with other environments. Our software agent allows you to examine and monitor the following details about your Docker container:

  • CPU usage
  • RAM usage
  • Disk usage
  • A list of the top processes
  • Total uptime

As with all aspects of Opsview monitoring, this data is collected and collated into our easy-to-use graphical interface, giving you both the convenience of a bird’s eye view and a detailed snapshot for more intense troubleshooting sessions. Opsview also stores your monitoring data in its database so you can review historical performance, etc.

Note, if you’re looking to further automate Opsview in your container infrastructure, and you use Puppet or Ansible for automation and configuration management, you can use our Puppet or Ansible modules to automatically register new containers with the Opsview server. How cool is that?

Innovative Service Flows with Opspacks and Docker

The Docker monitoring service is performed via our extensible Opspacks system. This plugin architecture allows for any service to be monitored, and the vast majority of applications, at a deeper level.

Combining Opsview Docker monitoring with Opspacks, you can monitor current EC2 spot instance pricing for your Docker containers. If at least some of your operations have worker processes that do not have to be running all the time, using this pairing you can monitor the AWS Price List via API to coordinate the most economical time to run your spot instances.

Since the Opspacks technology can be adapted to nearly any accessible API, you can coordinate Docker monitoring and operations based on any other metric. Applying container methodology like triggering Docker containers based on stock pricing, sports scores, or any other available data, opens nearly limitless possibilities.

If you’re using containers, you need to monitor them. If you aren’t using containers, but you want to learn more about them or monitoring in general, please contact us. We’re here to help you get the most out of your IT investment.

Get unified insight into your IT operations with Opsview Cloud

More like this

Extending Puppet into your Monitoring Environment
Blog

Opsview has introduced the Opsview Puppet Module, which automatically registers compute instances for monitoring within Opsview Monitor. 

The DevOps Manifesto
Blog

DevOps practices are more popular today than ever before. Read this to discover how to further the emergence of DevOps culture in your...

Simplify memory and disk utilization monitoring in AWS
Blog

How to improve your AWS EC2 memory monitoring. Since AWS CloudWatch does not have access to your OS, some monitoring metrics may be missing,...