IBM ServeRAID, User Reference

Page: 179 / 226

v Improper SCSI bus termination
Isolating hardware problems
You can use the following techniques to isolate most hardware problems:
v Check error codes within the ServeRAID Manager when a physical drive fails to
respond to a command. Research these codes in the
Hardware Maintenance
Manual and Troubleshooting Guide for your server.
v
In non-hot swap systems, make sure the physical drives are properly attached to
the cable connectors. Start with the connector closest to the SCSI terminator and
work your way forward to the connector closest to the controller. Examine each
SCSI device and ensure that it has the proper jumper settings.
v While the server is turned off, reseat the ServeRAID controller in its PCI slot and
all cables and disk devices on the SCSI bus.
v Examine the cables for bent or missing pins, nicks, crimps, pinches, or
over-stretching.
v Temporarily attach the physical drives to an integrated SCSI controller, if
available. Start the BIOS. As the BIOS POST runs, review the status of the
physical drives and the negotiated data rates. Determine if it is correct.
From the BIOS, choose an option which will list all the devices attached to the
controller. Select one of the physical drives and initiate a media test. This will test
the device and the entire SCSI bus. If you see errors on the integrated SCSI
controller, try to determine if it is the physical drive or the cable by initiating a
media test on other physical drives. Test both online and defunct physical drives
to determine if the test results are consistent with the drive states on the
ServeRAID controller. You can also move hot-swap physical drives to a different
position on the backplane and retest to see if the results change.
If the problem persists, swap out the SCSI cable and run a media test again on
the physical drives. If the physical drives pass the test, the previous cable is bad.
This is a valuable technique for isolating a failing component in the SCSI path.
Note: Depending on the rate at which it negotiates data with low-voltage
differential (LVD) SCSI devices, an integrated controller might produce
different results than the ServeRAID controller.
v
Use the system diagnostics to test the ServeRAID subsystem. Press F2 to start
diagnostics. If the subsystem fails the test, disconnect the physical drives from
the ServeRAID controller; then, reset the controller to the factory-default settings.
Run the diagnostic tests again. If the subsystem passes the diagnostics test,
then attach the disks to the ServeRAID controller, one channel at a time, and run
the tests again to isolate the channel of the failing component. If the controller
continues to fail diagnostic tests, call your IBM service representative for further
assistance.
Note: Be sure to use the most recent diagnostic tests available for the server.
v
Disconnect the first physical drive marked defunct from the cable or backplane.
Restore the ServeRAID controller to the factory-default setting. Attempt to import
the RAID configuration from the physical drives. Depending on how the failure
occurred, this technique might have mixed results. There is a reasonable chance
that all physical drives will return to an online state, except for the physical drive
that is disconnected.
v Move the ServeRAID controller into a different PCI slot or PCI bus and retest.
Notes:
1. When attaching an LVD ServeRAID controller to storage enclosures, set the
data rate for the channel to the data rate supported by the enclosure.
Chapter 18. Solving ServeRAID problems
163