Sharing Lock Not Removed When Node Crashed
(Doc ID 2743369.1)
Last updated on DECEMBER 01, 2023
Applies to:
Oracle Communications BRM - Elastic Charging Engine - Version 11.3.0.6.0 and laterInformation in this document applies to any platform.
Goal
On Oracle Communications Elastic Charging Engine (ECE) 11.3.0.6.0 version, High Availability:
It is observed that the sharing lock is not removed when the node is crashed.
Below are observations based on the node crash due to side effect of parallel incremental loading and billing run:
- Long Garbage Collection (GC) happened due to memory constraint on couple of nodes.
- Nodes were evicted from the cluster.
- Locks existing on the SharingCustomerLock remained and not cleared.
- Locks were existing on CustomerInBrmTransactionLock.
- Reloading accounts cleared the lock on CustomerInBrmTransactionLock but not from SharingCustomerLock.
- That caused failures in multiple orders/billing.
- Manual clearing of sharing lock has helped to fix the order/billing.
- Here it is observed that lock from sharing cache has not been removed and also incremental loading has not cleared the sharing lock.
- It cancelled the start due to timer.
Solution
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
| Goal |
| Solution |
| References |