Web nodes are during high user time - PEGA0028

General
, ,
1 
 checkout_enabled: ?
   correlation_id: 0
   interaction_number: -1
   kpi_threshold: 11596411699
   kpi_value: 11677693792
   line: Collection usage threshold exceeded for long-lived object memory pool (G1 Old Gen) INFO:MemoryPool=G1 Old Gen;PoolType=long-lived object;Event=java.management.memory.collection.threshold.exceeded;TimeStamp=2024-03-18 20:45:50,700 GMT;Count=1;NotificationSequenceNo=1;Threshold=11596411699;Used=11677693792;Max=12884901888;Committed=12209618944;Init=8040480768;
   logger: com.pega.pegarules.management.internal.events.JVMNotificationManager
   message_id: PEGA0028
   sequence: 535
   server_id: 089ced43d4963a1a4ec4f2bd03a44216
   timestamp: 2024-03-18 20:45:50,880 GMT
   version: 8

web nodes are going down when user count increases in the system, shows alert PEGA0024.

anybody facing this issue in pega 8.8.2

2

@JohnPaulRaja,C if you need investigation help from our support team log a support issue via the MSP and let us know the INC id here so we can help track it.

3

@JohnPaulRaja,C did you log a ticket with our support team?

The extract provided above is not sufficient - Apart from Pega0028 (Memory collection alerts), you may also be observing Pega0026 alerts, indicating shortage of Database connections.

You would need to work with your team to fix this Pega0026 alerts as this is sufficient to bring down the nodes, being a critical alert.

If you do log a ticket then when you log the INC, please provide the following artifacts

  • PegaRULES and PegaRULES-ALERT logfiles
  • Any heap dump (*.phd) files
  • Any javacore files
  • The verbose GC output file(s)
  • A copy of your JVM settings

Our GCS theam would use these details to gain a picture of what happened at the time of the event and work with you and the team to address root cause(s) for the problem.

If the issue is resolved, please provide a reply here and mark it with accepted solution, thanks.