Descheduler pods OOM-killed on big clusters

Solution Verified - Updated 2024-05-17T15:11:37+00:00 -

English

• No translations currently exist.

Environment

Red Hat OpenShift Container Platform 4.13 and earlier versions.

Issue

Descheduler pods being killed by Out-of-Memory Killer on clusters and nodes with enough allocatable memory.

Resolution

The issue was fixed for OpenShift Container Platform 4.14.0: OCPBUGS-1995.

The limits were removed.

Root Cause

Limits of Descheduler pods are hardcoded.

The problem is more likely to occur on large clusters because there is a dependency between the memory used by these pods and the size of the cluster.

This solution is part of Red Hat’s fast-track publication program, providing a huge library of solutions that Red Hat engineers have created while supporting our customers. To give you the knowledge you need the instant it becomes available, these articles may be presented in a raw and unedited form.