Description of problem: I want automatically-assigned Egress IPs to be highly-available, So that if a node goes down, egress traffic from the namespace keeps working. Version-Release number of selected component (if applicable): 3.10.0 How reproducible: Always Steps to Reproduce: 1. Create an egress IP for a namespace 2. Fail the node holding the IP Actual results: The egress IP is no longer available. Expected results: If a namespace has multiple egress IPs, then it can be made highly-available by having egress IPs on multiple nodes, and using all of them, and ensuring that egress IPs on dead nodes are automatically removed/migrated. Additional info:
PR: https://github.com/openshift/ose/pull/1338
Found the same issue here with my OCP v3.10.14 with netwokrpolicy CNI plugin. I'd tried in the scenario: step1: project with netnamespace set to [192.168.0.71, 192.168.0.70] and set hostsubnet with the two IPs on different nodes --> calling external ap success (with 192.168.0.71) step2: shutdown node with egressIP 192.168.0.71 --> calling external ap fail (timeout)
Moving to MODIFIED to let it be attached to an advisory
I have tested the ha egressIP feature on build v3.10.35 The feature works fine, and no regression issue. Verify the bug.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2018:2376