cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

[Cloud Integrations] New Enhancement: Cloud Metric Polling OOTB monitoring

KobiB
Dynatrace Contributor
Dynatrace Contributor

Author: Kobi Biton, Principal Product Manager, Cloud Core Integration
Contributors: Dawid Kaszubski, Senior Product Owner | Piotr Kunowski, Principal Software Engineer

 

The Challenge

 

Today when a customer creates/edits a new Connection/Monitoring configuration, upon saving the changes our platform will commence a validation action which will make sure that this monitoring configuration is able to successfully connect to the AWS Account/Azure Tenant for metric polling purposes. We will display and error message if this operation failed.

Many customers asked us to add a feature which will actively monitor the state of each monitoring configuration, especially at runtime (post a successful edit/creation of a monitoring configuration). in simple words: customers expects our platform to raise a problem and receive an alert when their Connection’s monitoring configuration(s) are not able to poll metrics.

 

The Solution

 

Beginning of Dynatrace Platform version 1.296, We have introduced this (much anticipated) capability for our AWS and Azure Integrations. This capability is provided OOTB for new and existing AWS and Azure Integration customers, It requires no configuration from your side.
The following (non exhaustive) prime use-cases are designed to trigger a Problem following an alert:

  • At least one of your AWS/Azure Monitoring Configuration elements is not configured properly or missing: i.e. The configured IAM role does no longer exists or is miss-configured. Your Azure Service Principle is missing or miss-configured, (This is by all means a non exhaustive list).

  • The ActiveGate metric poller is unable to assume the monitoring IAM Role or successfully authenticate using the provided Azure Secret.

 

How will it look like?

 

Our backend process will retry a failed attempt to successfully authenticate with the given monitoring configuration credentials. After few consecutive unsuccessful attempts a platform Problem is to be raised looking similar to the following screenshot:

Problem stateProblem state

 

 

To resolve the error, edit the Connection’s Monitoring Configuration, validate that all the details provided are correct and that all cloud elements (IAM Role, Azure Secret etc…) are intact, once validated save the Monitoring Configuration which will in-turn trigger a detailed validation by our backend which is designed to provide a more granular error message.

But what if there is a Monitoring Configuration which you wish to temporally suspend from metrics polling and yet not get alerted. Just edit the Connection’s Monitoring Configuration and toggle the “Enabled” Radio button, this will Disable (suspend) this monitoring configuration and will not trigger a platform problem until the status is changed back to “Enabled”.

We always look for feedback, enjoy the new capability and stay in touch with us by sharing your experience and suggestion on how can we improve your experience.

0 REPLIES 0

Featured Posts