Server Removed From Cluster View Due To PeerGone When JMS SAF Is Configured (Doc ID 1086357.1)

Last updated on AUGUST 04, 2023

Applies to:

Symptoms

The servers are periodically dropping out of the cluster.

Log messages are shown below:

This issue only occurs with the specific configuration below:
      DomainA configured with a SAFAgent that is configured to forward messages to a DistributedDestination on DomainB.
      DomainB configured with a SAFAgent that is configured to forward messages to a DistributedDestination on DomainA.
      Both DomainA and DomainB are configured with a cluster containing 2 managed servers.
      DistributedDestination on DomainA is configured with the same JNDI name as DistributedDestination on DomainB. (This is the key requirement to cause this issue)
      SAFRemoteContext is configured with t3s protocol in the destination URL. (Using t3 will not reproduce the issue)
      The server names and resource names of DomainA are different from the server names and resource names on DomainB.

If you use t3 in the SAFRemoteContext destination URL instead of t3s, the problem will not occur.

Cause

My Oracle Support provides customers with access to over a million knowledge articles and a vibrant support community of peers and Oracle experts.