Hide Forgot
Description of problem: Customer needs to have application (inside OpenShift) access an application (external to Openshift) that is running on an IP address that falls within the servicesSubnet range for their OpenShift Cluster. The only way to correct this would be to change the servicesSubnet and serviceNetworkCIDR values, which appears to be unsupported as well as breaking various services if you change it on a currently running cluster. Steps to Reproduce: 1. Change the servicesSubnet and serviceNetworkCIDR values in the master-config file 2. Restart atomic-openshift-master services Actual results: Cluster does not respond properly (oc only works ~50% of the time) Expected results: Cluster adjusts properly and continues to work.
This is currently showing up on the blocker report but should not be since it is an RFE
Docs PR for 3.4: https://github.com/openshift/openshift-docs/pull/3112
Eric and Dan -- If you think this issue should go into the 3.11 Release Notes, please enter Doc Text above. I am not clear on what the resolution is. "Because changing the `servicesSubnet` and `serviceNetworkCIDR` values is not supported in openshift-sdn, changing those values in the master configuration file and restarting will result in OpenShift failing. Changing the IPs addresses is supported by OVN." ?? Thank you Michael
There does not need to be anything in the 3.11 release notes; the feature was not implemented, so nothing changed in 3.11.