Monitoring Swift AMH Node

The eG agent periodically executes tests on the SWIFT AMH node using the JBoss AS/EAP server and JVM, collects the necessary statistics, and reports them to the eG manager. These tests are mapped to specific layers of the SWIFT AMH node layer model (see Figure 1). 

Figure 1 : The layer model of the Swift AMH Node

Using the metrics reported, administrators can find quick and accurate answers for the following performance questions:

• Are the AMH services available and responsive on the node?

• Are the total number of JMS connections optimal?

• Are there too many connections inactive at any given time?

• Are there too many sessions created in each connection that might impact connection performance?

• Are the queues in started state?

• Are there too many queues in on-operation state?

• Is the logical node active and running?

• Is the number of degraded or stopped services too high?

• Is the physical running?

• Are there too many logical nodes created on given physical node?

• Is it too long since the last heartbeat was received from physical node?

• Are the key processes running on the node?

• Is the time take by queue to process the message withing tolerance limit?

• Are there too many high processing time recorded by the queue?

• Is there threadpool able to generate enough threads to process the messages optimally?