ECE PodHealthCheck Defunct Processes Blocked the Worker Nodes
(Doc ID 2854121.1)
Last updated on JANUARY 24, 2024
Applies to:
Oracle Communications BRM - Elastic Charging Engine - Version 12.0.0.4.0 and laterInformation in this document applies to any platform.
Symptoms
Elastic Charging Engine (ECE) PodHealthCheck has no trap in the code. It has no timeout specified in the helm yaml files, so it then uses Kubernetes default of 1 second.
When the code is killed and defunct processes are created, these then use all the Process IDs (PID) on the worker nodes and the servers can no longer be connected to.
Kubernetes default timeout is 1 second which is too low:
ece-customerupdater.yaml: command: ["/bin/sh", "/home/charging/temp/PodHealthCheck.sh"]
ece-diametergateway.yaml: command: ["/bin/sh","/home/charging/temp/PodHealthCheck.sh"]
ece-ecs1.yaml: command: ["/bin/sh","/home/charging/temp/PodHealthCheck.sh"]
ece-ecs.yaml: command: ["/bin/sh","/home/charging/temp/PodHealthCheck.sh"]
ece-emgateway.yaml: command: ["/bin/sh","/home/charging/temp/PodHealthCheck.sh"]
ece-pricingupdater.yaml: command: ["/bin/sh","/home/charging/temp/PodHealthCheck.sh"]
ece-radiusgateway.yaml: command: ["/bin/sh","/home/charging/temp/PodHealthCheck.sh"]
Cause
To view full details, sign in with your My Oracle Support account. |
|
Don't have a My Oracle Support account? Click to get started! |
In this Document
Symptoms |
Cause |
Solution |
References |