Move static Egress IP to another node fails with: "Multiple nodes claiming EgressIP" error

Solution Verified - Updated -

Environment

  • Red Hat Openshift Container Platform (OCP) 3.9

Issue

  • When trying to move a static IP address for external project traffic (egress) from one node to another , despite of the configuration has been verified successfully, the application is not working.

Resolution

  • Delete egress IP configuration from node where egress IP is currently allocated.
  • Restart atomic-openshift-node service in the node where the egress IP was allocated, and in the node that will be hosted.
  • Apply the configuration in the desired final node as per commands:
# oc patch hostsubnet node2 -p '{"egressIPs": ["192.168.0.250"]} 
# oc patch netnamespace MyProject -p '{"egressIPs": ["192.168.0.250”]}'
  • Verify the correct behavior of the application.

Root Cause

  • Probably due to the ARP cache

Diagnostic Steps

  • Check atomic-openshift-node log on the node hosting the egress IP:
# journalctl -lu atomic-openshift-node | grep EgressIP
 Feb 26 11:16:39 node2.example.lab.com atomic-openshift-node[98152]: E0123 11:16:39.891333   98191 egressip.go:117] Multiple nodes claiming EgressIP "192.168.0.250" (nodes "192.168.10.125", "192.168.20.120")
  • Component
  • SDN

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.

Comments