Receiving mails with "msg: Event Listeners are taking too long to respond or they are not alive."

Resolving The Problem

If the mail is received during startup of the system and stops after some minutes:

The Event Listener is still in the startup time. The time the event listener needs for startup is different for each system. If you receiving the mail only the first ten minutes you do not need any action.

Event Listener appear to start, but do not respond:

You are getting the mail since start of the system and the occurrence didn't end after some minutes. Please verify that you are on version 5006 or later.

For details see: NFX2799 and NFX8018

Mail starts after some time after start of the system and no listener is running:
Verify that you have a bilisten.pid in the installation directory of Sterling B2B Integrator and that the process is running (Unix/Linux: Copy the process id from bilisten.pid. If ps -ef | grep <bilisten pid> return nothing the listener is not running, Windows: Under Services proof that the Sterling B2B EventListener is started ). Verify that a listen happens on the event listener port (Linux/Unix: netstat -a | grep <EVENT_PORT from sandbox.cfg>, Windows: netstat -an | find "<EVENT_PORT from sandbox.cfg>). If one of the verifications fails, please do a <si installation directory>/bin/stopListener.sh/cmd and afterwards a startListener.sh/cmd. Verify that the listener starts without any errors.

You receive the mail after an out of memory of ActiveMQ:

Restart of the listener will solve the out of memory error. But you will still receive the mail notifications. You need to restart the whole system to stop the mail notifications. For solving the OOM of ActiveMQ search for articles about ActiveMQ. If you are not able to find a solution for your situation open a new case with support, please.

You receive the mail and ActiveMQ failed or did not start:

The load of the events can no longer be handled from the system. Solutions are one or all of below points, Best practice is to create or modify the values in the customer_overrides.properties.

1. Reducing the creation of visibility events: Verify the setting in the visibility.properties or customer_overrides.properties for visibility_coverage. If you need the information from screens like communication session please set it to CommBase (entry in customer_overrides.properties: dmivisibility.visibility_coverage=CommBase) . If you do not need such information reduce it further.
2. Increase delays for the monitors:Change settings for Monitor:  
   resourceMonitor.DBResourceMonitor.delay=240000
   resourceMonitor.BIResourceMonitor.delay=360000
3. Increase the queue for the event listener: dmivisibility.event_input_queue_capacity=3072. If you go beyond this values you can run into performance issues.
4. Reduce the time between persistence of events: dmivisibility.persistent_batching_interval=1000