My Oracle Support Banner

Exalogic Compute Nodes Intermittently Crash and Reboot Due to a Memory Leak in Size-192 Allocations Within The Kernel Slab Area (Doc ID 1985070.1)

Last updated on JANUARY 03, 2024

Applies to:

Oracle Exalogic Elastic Cloud Software - Version 2.0.0.0.0 and later
Linux x86-64

Symptoms

You have an Exalogic system configured in either a physical or virtual configuration where you have applications running that utilize the Sockets Direct Protocol (SDP) for interprocess communication or utilizes Java Database Connectivity (JDBC) to access a database instance running on an interconnected Exadata system. Intermittently, you are experiencing unexpected crashes and automatic restarts for the compute nodes, with call traces that indicated the kernel has identified an Out Of Memory situation and attempted to recover from this by launching an Out Of Memory Killer process to reclaim memory, such as the following:

 

 

Changes

 

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
Changes
Cause
Solution
 Ensure you have already reconfigured OTD to used TCP rather than SDP then first ensure the directions provided in the following note have been followed to achieve this
 Remove any usage of -Dcom.sun.sdp.conf to avoid additional, implicit, sources of unexpected SDP usage that could trigger the leak
References


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