SAP Commerce (Hybris)- Clustering Issue, Solved - Process engine on specific node!!

Communication of nodes in SAP commerce(Hybris) environment was failing in cluster with following error:

INFO   | jvm 1    | main    | 2020/06/17 22:59:45.448 | WARN  [TransferQueueBundler,hybris-broadcast,hybrisnode-0] [UDP] JGRP000032: hybrisnode-0: no phy

sical address for 5eac3c39-17dc-2661-2d3c-f53d97c5433a, dropping message

In the cluster environment task of process engine communicate using udp or tcp as configured using JGROUP. 

Probable Solution: 

Sometimes this communication disrupt and task skip the jobs to do. We might need to stop backend task processing on particular node or nodes which can be achieved with simple configuration : 

just mention following in local property  of the specific node the process engine will not start: 

Before 5.5 :

task.excluded.cluster.ids=1,2

After 5.5: 

task.engine.loadonstartup=false

Also, groovy script can help if need to stop on the fly or to know the status of process engine.

go t to hac and run groovy on script tab :

to stop : 

import de.hybris.platform.task.TaskEngine

import de.hybris.platform.task.TaskService

 

 TaskEngine engine = taskService.getEngine()

 boolean running = engine.isRunning()

 if(running)

 engine.stop()

to start : 

import de.hybris.platform.task.TaskEngine

import de.hybris.platform.task.TaskService

 

 TaskEngine engine = taskService.getEngine()

 boolean running = engine.isRunning()

 if(running)

 engine.start()

Cheers !!