Werk #14138: Reduce erroneous retrieval and flushing of RRDs

The built-in dashboard Kubernetes and would provide insufficient context information for it's dashlets. This led to error messages in the cmc.log, e.g.,

Error flushing RRD: Usage: rrdtool rrdtool flushcached [--daemon|-d <addr>] <file> [<file> ...]
Error accessing RRD: don't understand 'kube_cpu_usage'

These errors are harmless, but spam the log. With this werk, the erroneous commands are no longer produced if opening the Kubernetes dashboard.

To the list of all Werks