Exalogic Compute Nodes Hang When More Than 32 Virtual Machines Are Running (Doc ID 2098113.1)

Last updated on FEBRUARY 11, 2016

Applies to:

Oracle Exalogic Elastic Cloud Software - Version 2.0.6.2.0 to 2.0.6.2.3
Oracle Exalogic Elastic Cloud Software - Version 2.0.6.1.1 to 2.0.6.1.2
Linux x86-64
This MOS Note Applies To Following Releases:

- Exalogic Virtual Releases 2.0.6.1.x on X4-2 racks
- Exalogic Virtual Releases 2.0.6.2.x on X4-2 and X5-2 racks

Symptoms

In Exalogic X4-2 and X5-2 Virtual racks when running over 32 VM's on a compute node, the following symptoms are observed.

  1. Guest VM's are all hung or inaccessible on given node

  2. OCFS timeouts in all compute node logs

  3. Hung node cannot access NFS or other hosts via infiniband network.

The following errors can also be observed in the compute node /var/log/messages log.

Changes

 Adding additional virtual machines can trigger this problem.

Cause

Sign In with your My Oracle Support account

Don't have a My Oracle Support account? Click to get started

My Oracle Support provides customers with access to over a
Million Knowledge Articles and hundreds of Community platforms