IBM BladeCenter S Problem Determination And Service Manual page 45

The OS watchdog timer may be enabled or disabled by default, depending to
the type of blade server. You can enable it or disable it using the advanced
settings of the BIOS configuration utility for the blade server.
I-error messages
I-error messages are displayed in the advanced management module event log.
They occur any time the CPU on the blade server encounters a problem on the
system. Typically, they are the result of a nonresponsive I/O device.
I-errors related to Intel-based blade servers
The root cause of most I-errors is normally one of the following:
v I/O device drivers
v Hard disk drive failure
v PCI device timeout
v Non-terminated SCSI bus
v Real CPU internal failure
v Power supply problems
The following example shows I-error messages:
E BLADE_01 03/10/04 14:13:02 (SN#ZJ1TS1234567) PFA Alert, see preceding error
in system error log.
E BLADE_01 03/10/04 14:13:02 (SN#ZJ1TS1234567) 00151200 MCA: Unrecoverable Error
Detected Proc=1
I BLADE_01 03/10/04 14:13:01 (SN#ZJ1TS1234567) 00151352 MachineCheck Data Status:
A2000000 8C010400
I-errors related to AMD-based blade servers
I-errors result when the CPU encounters a machine check error that cannot be
recovered. Most often, these errors are memory errors, but occasionally they can be
related to the processor. When a machine check error occurs, you will need to
restart the blade server.
Below is an example of a memory error on an AMD blade server reported in the
advanced management module log:
I BLADE_07 08/24/06, 17:40:53 (id14c2n07) Power unit power off event received
E BLADE_07 08/24/06, 17:38:25 (id14c2n07) POSTBIOS: 289 DIMM D02 is Affected.
E BLADE_07 08/24/06, 17:38:25 (id14c2n07) POSTBIOS: 289 DIMM D01 is Affected.
E BLADE_07 08/24/06, 17:38:25 (id14c2n07) Memory disabled: DIMM number 2
E BLADE_07 08/24/06, 17:38:25 (id14c2n07) Memory disabled: DIMM number 1
I2C errors
Error messages related to the BladeCenter S system hardware components and
communications between the components (known as I2C errors) are displayed in
the advanced management module event log.
The following example shows an I2C error message:
Failure reading I2C device. Check devices on bus 7.
Perform these actions to resolve I2C error messages:
1. Restart the advanced management module.
2. Wait two minutes to allow the advanced management module to reset the I2C
bus. If an I2C bus error recovery message does not appear in the event log after
two minutes, reseat the advanced management module.
31
Chapter 2. Diagnostics