H207780: Data corruption exposure with 3.5-inch 300, 450, 600 GB SAS and FC HDDs - IBM Servers



Source

RETAIN tip: H207780

Symptom

A subset of disk drive modules (DDMs) may have an exposure to data loss under a unique set of circumstances during a proximal write, a feature which does a skip operation on the data transfer from Dynamic Random Access Memory (DRAM) to disk to improve performance.

The exposure to data loss requires that the starting Logical Block Address (LBA) is a reassigned LBA.

The affected drives and firmware levels are as follows:

  • Serial Attached SCSI (SAS)
    • VPCA300900EST1 - A3C0
    • VPCA450900EST1 - A3C0
    • VPCA600900EST1 - A3C0
  • Fibre Channel (FC)
    • HUS156030VLF400/HUS1560300FC - JH00, JH01
    • HUS156045VLF400/HUS1560450FC - JH00, JH01
    • HUS156060VLF400/HUS1560600FC - JH00, JH01

This issue has not been reported by any IBM System Storage DS3500 or 4000 or 5000 Storage Controller users.

Affected configurations

The system may be any of the following IBM servers:

  • DS4700 Storage Server, type 1814, any model
  • DS4800 Storage Server, type 1815, any model
  • IBM System Storage DS3512, type 1746, any model
  • IBM System Storage DS3524, type 1746, any model
  • IBM System Storage DS3950 Express, type 1814, any model
  • IBM System Storage DS5020 Disk Controller (1814-20A), any model
  • IBM System Storage DS5100 Storage Controller, type 1818, any model
  • IBM System Storage DS5300 Storage Controller, type 1818, any model

The system is configured with one or more of the following IBM Options:

  • EXP395 Express Expansion Unit (1814-92H), any model
  • EXP810 Storage Enclosure, type 1812, any model
  • IBM System Storage EXP3512 Express, type 1746-A2E
  • IBM System Storage EXP5000 Storage Expansion Unit, type 1818, any model
  • IBM System Storage EXP520 Storage Expansion Unit (1814-52A), any model

This tip is not software specific.

The A3C0, JH00, and JH01 firmware for the hard disk drives is affected.

Solution

The fix for this issue is contained in the following releases:

  • SAS: A3C4 and later
  • FC: JH02 and later

These files are available on the IBM Storage support web site.

The files are available by selecting the appropriate Product Group, type of System, Product name, Product machine type, and Operating system on IBM Support's Fix Central web page, at the following URL:

Additional information

The firmware levels listed in the Fix must be installed in order to remove completely the possibility of encountering this issue.

If the first LBA of a proximal write (caused by two writes that are close to each other sequentially and sent to the drive in close proximity from a time perspective) is a reassigned LBA, an invalid skip mask is used. This creates the possibility that the correct data is not written to the disk.

Applicable countries and regions

 


Document id:  MIGR-5092732
Last modified:  2013-03-14
Copyright © 2014 IBM Corporation