No monitoring for etcd in the local Rancher cluster with cluster monitoring in Rancher v2.2.x

This document (000020229) is provided subject to the disclaimer at the end of this document.

Situation

Issue

When enabling cluster monitoring in Rancher v2.2.x, upon the local cluster running the Rancher server itself, no monitoring metrics are available for the etcd instances within the cluster.

Pre-requisites

An RKE provisioned Rancher HA cluster
Rancher v2.2.x
Cluster monitoring enabled on the local Rancher cluster

Root cause

Cluster monitoring was introduced in Rancher v2.2.0 and when enabled for a Rancher launched Kubernetes clusters, i.e. a custom cluster or using node templates for an infrastructure provider, will include etcd monitoring metrics. The local cluster running Rancher, whilst provisioned via RKE, is an imported cluster from the Rancher perspective. As a result etcd monitoring metrics are not collected and displayed for this cluster.

Resolution

An enhancement request to include etcd monitoring metric collection, for the local Rancher cluster, within Rancher cluster monitoring is tracked in Rancher Issue #18619.

Disclaimer

This Support Knowledgebase provides a valuable tool for SUSE customers and parties interested in our products and solutions to acquire information, ideas and learn from one another. Materials are provided for informational, personal or non-commercial use within your organization and are presented "AS IS" WITHOUT WARRANTY OF ANY KIND.