For a highly scalable FmW environment, whether it’s WebCenter Content, Identity Management, or any other application, clustering is necessary. This makes the entire system give high throughput, higher user transaction output, and above all gives a High-Availability environment.
Post-cluster set up, sometimes when you’re starting the additional nodes, start up fails. One of the many reasons for it could be that the nodes (host machines) are set up similar to the parent machine. Continue reading for the easy solution that will get your system back up-and-running in no time.
A cluster is a collection of multiple server instances running behind a web tier. For end users, it is entirely transparent and acts as a single point of entry to application.
When setting up the cluster and extending it to multiple nodes, the scale out or scale up mechanism can be used, as detailed by Oracle here.
The issue listed here relates to the scale out operation where in post set up, starting the additional node for the managed server shows the error that password authentication is not successful and connection to the Admin server is failing.
The root cause for this error is that the additional node host machine clock (time) is not set exactly as the host machine, which is running the Admin server.
All the node hosts in a cluster should have the same time and date set.
To figure out if this is the error, review the Admin server log file. If you see the error illustrated in the screenshot below (the messages seen after “Running” status is all coming from the additional node when during start up, user authorization is checking with the Admin server), you’ve found your issue and now you know how to correct it.