"loopyWriter.run returning. connection error" messages in the kube-apiserver pod logs

Problem

• Why do we have below mentioned messages continuously getting logged in kube-apiserver logs?

Environment

• Platform9 Managed Kubernetes – All Versions

• Upstream Kubernetes Releases - till v1.21

Cause

• Those messages are generated by the gRPC layer whenever a connection to etcd is closed, and they are not a symptom of an issue.

Resolution

There is no workaround that will reduce or prevent these messages from being logged as of now, however they have been suppressed by a patch to upstream kubernetes and are expected to not be logged in future Platform9 Managed Kubernetes releases that are based on kubernetes v1.22 or greater.