WebLogic Coherence And Kubernetes Ip-tables Configuration, IP Address Masquerading
(Doc ID 2947930.1)
Last updated on MAY 19, 2023
Applies to:Oracle Coherence - Version 220.127.116.11.0 and later
Information in this document applies to any platform.
Using FMW 18.104.22.168 on Kubernetes through OLCNE 1.4 platform. As fmw weblogic uses coherence for managed server communications, there is a requirement to make sure that there are no chains present in ip-tables. Following command was used to check the chain on ip-tables:
After making sure there were no chains in the ip-tables, the weblogic domain was started. Weblogic managed servers coherence cluster formed correctly and all the managed servers were up and running. After a few hours, due to an update on the ip-tables, suddenly the weblogic pods coherence cluster communication was broken and due to that servers availability was impacted. Pods changed to Not Ready state from Ready State. As of now we have put a systemd service to execute every 15 mins to check the ip-tables configuration. This is only a temporary approach. As of now we don't know which process is updating these ip-tables when the weblogic domain is in running state. We would like to know why we need to rely on ip-table entries because multiple softwares may update the ip-tables. Is there a better approach to rely on forming a coherence cluster on Kubernetes without relying on ip-tables? If we need to rely on ip-table entries how to make sure the ip-table updates will not impact the weblogic coherence cluster?
To view full details, sign in with your My Oracle Support account.
Don't have a My Oracle Support account? Click to get started!
In this Document