AWS Infrastructure monitoring

[Gopi.Krishnan]

Hi,

I have a Kubernetes cluster deployed in AWS and my application is running in that. Now I am able to monitor my application via AppDynamics java agent. I want to monitor the infrastructure now and I have used the autoscaling feature as well.

So what is the best possible way to monitor the infra of my environment with a machine agent? Or is there any other way to do the same?

Regards,
Gopikrishnan 

[Philip.Rogers]

Hi Gopi-

Thanks for your inquiry. This sounds like a great use case for our new Kubernetes Cluster Agent. I would encourage you to take a quick look at our Cluster Agent docs to familiarize yourself with the agent and reply back with any specific questions.

https://docs.appdynamics.com/display/PRO45/Monitoring+Kubernetes+with+the+Cluster+Agent

The "Cluster Agent Requirements" page within this link mentions compatibility with "Amazon EKS version 1.11, 1.13, 1.14 with the Kubernetes metrics-server deployed."

One Cluster Agent consumes a single Server Visibility license.

[Gopi.Krishnan]

Hi Philip,

Got your point and for kubernetes deployed on AWS instances its better to have cluster agent for monitoring the infra. One more case what I have is there are some java applications running on AWS VMs in that case is it better to have machine agent deployed on AWS or is it better to use AWS extension for monitoring?

Regards,

Gopikrishnan 

[Philip.Rogers]

@Gopi.Krishnan 

If deploying machine agents inside your VMs is an option, such as EC2 instances, that will provide the best result for infra monitoring, with the added benefit of correlating to application performance where applicable. 

That being said, the AWS extensions do provide their own added value as well and can be installed as a supplement. 

[Gopi.Krishnan]

Thanks Philip.

Is there any differentiator like what extra details we get from extension monitoring as compared to machine agent monitoring?

Regards,

Gopikrishnan

[Philip.Rogers]

@Gopi.Krishnan 

The machine agent will be more targeted to that specific machine's hardware metrics versus an AWS extension.

https://docs.appdynamics.com/display/PRO45/Hardware+Resources+Metrics 

The above link lists the metrics the machine agent will gather (with server visibility license enabled for best results). The machine agent will have pre-built functionality and 

https://www.appdynamics.com/community/exchange/extension/aws-ec2-monitoring-extension/

And here's a link to our EC2 extension as an example. Notice the metrics section of this extension will take you to AWS' list of metrics here: https://docs.aws.amazon.com/AmazonCloudWatch/latest/monitoring/aws-services-cloudwatch-metrics.html

Also important to note, server visibility is a built-in part of the AppDynamics GUI, whereas an extension requires you to create a dashboard manually with the new metrics you gather.