Exadata ILOM memory leak in older ILOM firmware (Doc ID 1310539.1)

Last updated on JUNE 16, 2017

Applies to:

Oracle Exadata Hardware - Version 11.2.0.2 and later
Linux x86-64


Symptoms

Known initial indications include any of the following:

CASE A. A Storage Cell is logging a "System ACPI Power State #0x26" alert in the IPMI log.

CASE B. Receiving e-mails that "A generic sensor has detected an error". The message includes:
Description A generic sensor has detected an error.
Component Name /SYS/ACPI
Trap Additional Info S0/G0: working

Additionally, the ILOM event log indicates several events registered as critical IPMI log events. All have the same description; "System ACPI Power State: ACPI: S0/G0: working"

CASE C. Intermittently (once or twice a week) a Storage Cell server is sending an e-mail message:
Critical: Software Stateful Alert 13_1
Event Time 2011-03-08T17:04:25+00:00
Description Cell configuration check discovered the following problems:
Get Device ID command failed
Get Device ID command failed
Unable to open SDR for reading
Get Device ID command failed
[EXCEPTION] Running /opt/oracle.SupportTools/CheckHWnFWProfile

Additionally, the Alert.log file could contain the following message:
Critical: Software Stateful Alert 12_1

CASE D. Alert log entry:
hostname-ilom issue sunHwTrapSecurityIntrusion: An intrusion sensor has detected that someone may have physically tampered with the system.

CASE E. Alert log entry:
hostname-ilom issue sunHwTrapComponentError: A sensor has detected an error. This generic 'component' trap is generated when the SNMP agent does not recognize the component type.

CASE F. EM GC 12c display:
"Summary Metric evaluation error start - oracle.sysman.emSDK.agent.fetchlet.exception.FetchletException:..."


One additional, common factor between each of these various initial indications is that restarting the ILOM management process does clear up the issue temporarily.

Changes

Current ILOM firmware version is less than the newest version, as summarized in the table below.

/opt/oracle.SupportTools/CheckHWnFWProfile -d
output will show a version less than 3.0.14   For example:

[ILOMVersion]
3.0.9.27.a r58740

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