Last updated on NOVEMBER 05, 2016
Applies to:Oracle WebLogic Server - Version 9.2 to 18.104.22.168.0
Information in this document applies to any platform.
A WLS JMS Message Bridge deployed to a cluster and this bridge consumes messages from a topic on another remote WL instance. The Message Bridge is targeted to a cluster of 2 nodes. The connection to the topic is durable. As such, it is expected one node of the cluster to have a successful Message Bridge connection and the other node to not be active but continue to attempt to connect until the first node fails and thus releases the durable subscriber lock.
However, the node, that is retrying to connect, does not close its connections on the remote WLS instance. Each time, it attempts to connect and verify the durable subscriber lock, it leaks a connection. Currently, the Message Bridge is configured to retry every 60 seconds. Therefore, it is leaking about 1 connection per minute. It was once noticed more than 1000 connections from the backup/standby Message Bridge node to the remote server. Bouncing the backup/standby node removed these 1000 orphaned connections. Netstat showed only one socket from the bridge server to the remote server for those 1000+ connections. After the bounce the connection count is steadily climbing.
ENVIRONMENT RESOURCE CONFIGURATION
+Cluster1 that contains Server1 and Server2
+JMS module targeted to a cluster
+Uniform distributed queue (targeted to JMSServer-1, JMSServer-2)
+Source Bridge Destination (Topic in WLS 12.1.1)
+Target Bridge Destination (UniformDQ in WLS 12.1.1)
Sign In with your My Oracle Support account
Don't have a My Oracle Support account? Click to get started
Million Knowledge Articles and hundreds of Community platforms