E1: KER: After Short Time the E1 Kernels Are Restarted on an Enterprise Server, the Overall Performance Is Degraded, the Transactions Are Delayed and Most All of the Call Object Kernels Are Accumulating Outstanding Requests and Generate Call Stack Dumps (Doc ID 2820968.1)

Last updated on SEPTEMBER 01, 2024

Applies to:

Symptoms

After the Enterprise Server is started, the performance is good, but is degraded abruptly after short time (e.g. 15 minutes) the enterprise server was restarted.
When the performance is degraded, the applications are running very slow and the UBEs can barely be submitted.
Also the kernels running on the Enterprise Server are accumulating a high number of outstanding requests and the Call Object Kernels are generating call stack dumps while releasing the user session resources by calling JDB_FreeEnv API.
Checking for possible causes as mentioned in Doc ID 1902552.1 - JDB4100007 JDB_FreeEnv Warning, all the queries are running fine with no delays and there are no records locked or deadlocks reported on database server.

The information captured into the Call Object Kernel logs are showing that there are delays when running JDB_FreeEnv API, as noted below:

Changes

 No changes have been made to the existing configuration or infrastructure

Cause

My Oracle Support provides customers with access to over a million knowledge articles and a vibrant support community of peers and Oracle experts.