Keepalived ingress scripts timeout and VIP is jumping between nodes frequently on Red Hat OpenShift Container Platform 4

Solution In Progress - Updated 2025-12-15T15:32:22+00:00 -

Issue

On IPI clusters that use VIPs with keepalived static pods it is observed that there are ingress connectivity issues or outages.
VIP moves frequently from one node to the other.
The keepalived pod logs on the ingress nodes have frequent timeouts of the ingress health scripts, causing VRRP election to occur frequently, even when the load is not high and/or the node is mostly idle.
```
Track script chk_ingress is being timed out, expect idle - skipping run
Track script chk_ingress_ready is being timed out, expect idle - skipping run
```

A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.