DL360 Disk Failures

[Craig Smith_4]

I have a DL360 G1, with the following configuration.

2x 18.2G drives as a mirrored pair, running off the integrated ROC. OS is W2K Advanced Server. No external devices attached.

After approx 1 hour the first disk fails (goes red, and an alert shows up in the IM log saying disk X has failed), approx an hour after that the second disk fails, and the server blue screens (Obviously!), again an error is written to the IM log saying the server has stopped.

Both disks have been replaced to no avail, as has the ROC and the drive back plane.

Firmware levels have been upgraded to the following:

Disks: B014
ROC: 1.50b
System Bios: 07/04/02

When you reboot the server after failure, it boots fine, oddly with no POST errors.

Any suggestions as to the cause of this problem anybody?

[Ron Amos]

Hi Craig,
BF01865222
18.2 GB Wide Ultra3 SCSI Hot-Plug
189395-001
A new firmware upgrade (B014) has been released for the hard drive model 18.2 GB Wide Ultra3 SCSI Hot-Plug that corrects a Self-Monitoring Analysis and Reporting Technology (S.M.A.R.T.) calculation problem if excessive read errors, seek errors, or reallocated sectors occur.

The firmware upgrade also contains an enhancement to the S.M.A.R.T. technology by adding detection of excessive write errors if they occur.

It would seem that you must have 18.2 GB Wide Ultra3 SCSI Hot-Plug ??? Model #BFO1865222 ??? Part #189395-001 with latest Firmware.

You also have the latest Firmware on Integrated Smart Controller (ROC) and System Board. The Integrated Smart Controller should be using Low Voltage Differential (LVD) and has a Peak Transfer Rate of 80 MB/s (internal). Intended for Protocol Wide Ultra2 SCSI

Wide Ultra3 SCSI is backward compatible but will only operate at 80 MB/s using the Integrated Smart Controller. If you are using ???Wide Ultra3 SCSI Hard Drives??? be sure that you are also using ???Low Voltage Differential (LVD) SCSI Cable???. But you mention no External Devices and this Server doesn???t have SCSI Cabling so I would recommend replacing ???Backplane Board, Hot Plug SCSI Disk Drive Cage Part #173829-001??? even though it may have been replaced prior. Also suggest running ARRAY Diagnostics Utility (ADU) by Booting on a Smart Start and save the Report Text (About 20 Pages). Then E-Mail this Report Text to ENTERPRISE.ECS@COMPAQ.COM with ???ITRC Response??? in Subject. After this call (800) 652-6672 and ask the Server Support Engineer to examine this Report Text with our ADU Reader to get additional insight as to what is happening here.

PS: I wouldn???t discount the System Board Part #173837-001 for 550, 800, 866, & 933Mjz Models OR Part #224928-001 specific to 1.0Ghz Model (Both Part #???s these include Thermal Pads required).

I will be interested in your resolution.

Ron

[Craig Smith_4]

Problem solved by replacing System Board.

ADU and System Diags showed no faults!