Whenever the heap utlization goes over 90%, a red alert is issued by the production monitoring system. This time, however its over the admin server, so little less panic.
Analysis
It’s about heap utilization, so the most sensible logs to look at are the gc-logs and heap dumps. Transferring heap dumps takes time, hence I started analyzing the gc-log of the admin server.