ASPI and SCSI communication errors as well as certain hardware error conditions can be difficult to isolate and to correct. This difficulty is compounded when the problem is intermittent.
The nature of these errors, include components on the system bus used for connecting the server to the storage device. This includes:
· Server system and motherboard
· SCSI card
· SCSI cable
· ASPI modules
· Storage library and its drives
Trouble-shooting these types of problems must take into account the whole sub-system.
Many times the tests used by a hardware technician trouble-shooting a problem call, may not be equivalent to how the system is actually used during "normal" production operations when the error occurs. That is, using the actual server sub-system with all of its hardware and software and under the same conditions current with the error.
Some trouble-shooting may consist of running front panel tests on the storage library itself or attaching a laptop to the device and executing various operations directly on the storage device. Utility programs may also be used. While these methods are valuable, they often fall short of a comprehensive approach to understanding and resolving the problem.
This is because these types of tests do not fully emulate what is actually occurring in the SCSI/ASPI communication when the system is running in the production environment. These types of tests may simply miss an intermittent error.
If using these standard trouble-shooting procedures to isolate a persistent problem component has not provided a resolution to the error condition, then using a SCSI Analyzer is the next logical step to take.
A SCSI analyzer is a device that traps a "Bus trace" of all communication between the ASPI module and the target SCSI device. The data in this "trace" can analyzed by a hardware engineer, providing them with key communication as well as the error sequence leading up to, and including the actual failed operation.
The benefit of using a SCSI analyzer is that it operates independently on the actual server system, connected to the storage library being used when the error condition occurs. The SCSI analyzer can be used along with all of the software programs and hardware components normally utilized in the production environment. This provides a clear and independent picture of what is occurring in the error sequence.
Another benefit is that the information provided in a Bus trace will end any discussion over whether the nature of the problem is software or hardware which is helpful to both hardware and software vendor alike.