MAP 0040: Intermittent problem isolation

This MAP provides a structured way of analyzing intermittent problems.

Purpose of this MAP

This MAP provides a structured way of analyzing intermittent problems. It consists of two tables: Hardware Symptoms and Software Symptoms.

Because software or hardware can cause intermittent problems, consider all symptoms relevant to your problem.

How to use this MAP

This MAP contains information about causes of intermittent symptoms. In the following tables, find your symptoms, and read the list of things to check.

When you exchange a FRU, go to Verify a repair to check out the system.

Hardware symptoms

Note: This table spans several pages.

Symptom of hardware problem	Things to check for
Any hardware log entry in the error log.	Use the Hardware Error Report service aid to view the error log and check for: Multiple errors on devices attached to the same SCSI bus. Multiple errors on devices attached to the same async adapter. Multiple errors on internally installed devices only. Contact your service support structure for assistance with error report interpretation.
Hardware-caused system crashes	The connections on the CPU planar or CPU card Memory modules for correct connections Connections to the system planar. Cooling fans operational The environment for a too-high or too-low operating temperature. Vibration: proximity to heavy equipment.
System unit powers off a few seconds after powering On.	Fan speed. Some fans contain a speed-sensing circuit. If one of these fans is slow, the power supply powers the system unit off. Correct voltage at the outlet into which the system unit is plugged. Loose power cables and fan connectors, both internal and external.
System unit powers off after running for more than a few seconds.	Excessive temperature in the power supply area. Loose cable connectors on the power distribution cables. Fans turning at full speed after the system power has been on for more than a few seconds.
Only internally installed devices are failing.	Check the following items that are common to more than one device: Ground connections on all of the disk drives and other types of drives installed. Loose connections on the power cables to the planars, drives, fans, and battery. System unit cooling. Is the input air temperature within limits? Are all the fans running at full speed? Are any of the vent areas blocked? Signal cables to the diskette drives, and the power supply. SCSI device signal cables for loose connectors and terminators. Loose SCSI device address jumpers. Possible contamination of any device that has a cleaning procedure. See the operator guide for cleaning instructions. Excessive static electricity. Correct voltage at the system unit power outlet
Only externally attached devices are failing.	Check the following items that are common to more than one device. Check the SCSI signal cables to the devices for loose connectors and terminators. Check devices that use jumpers to set the SCSI address for loose jumpers. Check any device that has a cleaning procedure for contamination. See the operator guide for cleaning instructions. Check for excessive static electricity. Check the outlet that the device is plugged into for correct voltage. Check the error log for entries for the adapter driving the failing devices. Check the temperature of the devices. Are the cooling vents blocked? Are the fans running? Check for other devices near the failing device that may be radiating noise (displays, printers, and so on).

Software symptoms

Symptom of software problem	Things to check for
Any symptom you suspect is related to software.	Use the software documentation to analyze software problems.
Software-caused system crashes	Check the following software items: Is the problem only with one application program? Is the problem only with one device? Does the problem occur on a recently installed program? Was the program recently patched or modified in any way? Is the problem associated with any communication lines? Check for static discharge occurring at the time of the failure.