The Router Pods are flapping. That means they are restarting themselves from time to time because of an error inside the Pod. We'll upgrade to the latest patch level which should solve these issues according to the Changelog provided by Red Hat.
The root cause has been found: There is a memory leak in NetworkManager which leads to out-of-memory situations. This has been identified by Red Hat and is already fixed. The fix is scheduled to be released next week.