There's already clear documentation on how to manually install the metricbeat agent to various operating systems, including Windows, here. It can also be deployed using the Helm charts shown here, making it very easy to deploy to multiple Kubernetes nodes (say as a DaemonSet).
The problem is that the Docker image that the Helm chart is deploying to the containers is a Linux-based one (based on CentOS, as stated here). Various things in the Helm chart are also Linux-specific, such as the liveness probe code here.
Are there any instructions or even pointers on how to deploy to Windows nodes in an automated way? Creating the Docker image so its base layer is a Windows one would be great.
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.