IBM Support

Possible corruption on ServeRAID 7e SCSI (Adaptec HostRAID) - IBM xSeries Servers

Troubleshooting


Problem

RETAIN tip H184689: A data corruption problem has been identified in the current ServeRAID 7e SCSI code. The problem has not been reported from internal or external customers, nor has it been observed at any time during the many IBM test cycles.

Resolving The Problem

Source

RETAIN tip: H184689  

Symptom

A data corruption problem has been identified in the current ServeRAID 7e SCSI code. The problem has not been reported from internal or external customers, nor has it been observed at any time during the many IBM test cycles. It occurs rarely while the code recovers from a media (Hard Drive) error. Because of unexpected timing conditions, the code reads data from an incorrect memory buffer and writes it back to the drive as good data. This results in corrupted data written to the drives without any error notification.  

Affected configurations

The system may be any of the following IBM eServer xSeries servers:  

  • eServer xSeries 346, Type 8840, any model
  • eServer xSeries 306, Type 8836, any model
  • eServer xSeries 236, Type 8841, any model
  • eServer xSeries 226, Type 8648, any model
  • eServer xSeries 206, Type 8482, any model

The storage option may be configured with one of the following embedded SCSI options:  

  • Adaptec AIC-7902 Ultra320 SCSI controller

The ServeRAID 7e (Adaptec HostRAID) SCSI_version 3.00.036 and earlier Firmware for the Microsoft Windows is affected.
 
The ServeRAID 7e (Adaptec HostRAID) SCSI_version 3.00.056 and earlier Firmware for the Linux is affected.
 
The ServeRAID 7e (Adaptec HostRAID) SCSI_version 2 and earlier Firmware for the NetWare is affected.  

Solution

ServeRAID 7e SCSI driver version 3.00.63 and higher for the Microsoft Windows
ServeRAID 7e SCSI driver version 3.00.63 and higher for Linux
ServeRAID 7e SCSI driver version 3.00.65 and higher for Netware 6.5

These files can be found at the following URL:

SERV-RAID

Workaround

The workaround is to set either of the drives in the RAID 1 array to the defunct state. Essentially, this is accomplished by failing one of the drives in the mirrored pair manually. The manual failing can be done by using the ServeRAID Manager or the Ctrl-A BIOS tool.
 
Note: Setting the drive to the defunct state eliminates the redundancy of the RAID 1 array and increases the chances for data loss. Data loss means information stored on the device that is no longer located on the device or is no longer accessible due to some type of error. You should implement additional backup measures to compensate for the lack of redundancy resulting from this workaround.  

Additional information

Systems configured with ServeRAID 7e SATA or SAS, or any RAID level on the ServeRAID 7e SCSI other than RAID 1, are not exposed to the risk.

Document Location

Worldwide

Operating System

Older System x:Operating system independent / None

[{"Type":"HW","Business Unit":{"code":"BU016","label":"Multiple Vendor Support"},"Product":{"code":"HW20J","label":"Older System x->xSeries 206"},"Platform":[{"code":"PF025","label":"Platform Independent"}],"Line of Business":{"code":"","label":""}},{"Type":"HW","Business Unit":{"code":"BU016","label":"Multiple Vendor Support"},"Product":{"code":"HW20N","label":"Older System x->xSeries 226"},"Platform":[{"code":"PF025","label":"Platform Independent"}],"Line of Business":{"code":"","label":""}},{"Type":"HW","Business Unit":{"code":"BU016","label":"Multiple Vendor Support"},"Product":{"code":"HW20Q","label":"xSeries 346"},"Platform":[{"code":"PF025","label":"Platform Independent"}],"Line of Business":{"code":"","label":""}},{"Type":"HW","Business Unit":{"code":"BU016","label":"Multiple Vendor Support"},"Product":{"code":"HW20R","label":"Older System x->xSeries 236"},"Platform":[{"code":"PF025","label":"Platform Independent"}],"Line of Business":{"code":"","label":""}}]

Document Information

Modified date:
29 January 2019

UID

ibm1MIGR-61617

Page Feedback