ECA034: POTENTIAL FOR DATA CORRUPTION IN CLUSTER ENVIRONMENT


Subject: ECA034: POTENTIAL FOR DATA CORRUPTION IN CLUSTER ENVIRONMENT
Updated Netfinity server RETAIN ECA034: 

    Record number:       H167847
    Device:              D/T8680         Service code:      33
    ECA number:          ECA034          Parts source:      LOC
    EC number:           EC000000        Publish flag:      R
    Mandatory:           Y               Hit count:         UHC00004
    Automatic shipment:  N               Success count:     USC0000
    Availability date:   99/11           Publication code:  PC50
    Plant of control:    23              Date created:      O99/11/02
    Tip key:                             Date last altered: A99/11/24
                                         Owning B.U.:       USA


Abstract: ECA034: POTENTIAL FOR DATA CORRUPTION IN CLUSTER ENVIRONMENT

PURPOSE:

This ECA is being released to cover a rare, but potentially serious issue resulting from the manual movement of cluster resources from one node to another under MSCS.

This ECA is "MANDATORY".

To order this ECA, contact your IBM Branch Office, Central Response Office or Designated Country ECA Coordinator.


This ECA is scheduled to be withdrawn December 31, 2000.


FEATURES:

  Type,                  Machines Affected
  Model,   With          and/or Feature/Device     B/M to be    Service  System
  Stage    B/M           Description               Installed    Hours    Hours
  X        B/M0000000    SERVERAID-3L              B/M0000000   00.5     00.5
  X        B/M0000000    SERVERAID-3H              B/M0000000   00.5     00.5
  X        B/M0000000    SERVERAID-3HB             B/M0000000   00.5     00.5

PHYSICAL CHECK:



PREREQUISITES:

A ServeRAID 3H, 3L, or 3HB adapter in a cluster, or failover pair configuration.


COMPANION: None.

CONCURRENT: None.


DETAIL:

There is a rare chance of data corruption during rapid manual failover of a cluster environment.

The Prerequisite:

A ServeRAID 3H, 3L, or 3HB adapter in a cluster, or failover pair configuration.

Recently, in IBM's lab stress testing, a data corruption problem was found
in a Cluster Environment with a ServeRAID 3x adapter.

This problem will not occur in a production environment.
It can, under very rare circumstances, occur if a customer were to perform rapid,
manual failovers from one node of the cluster to the other and back.
Even under these circumstances it would be very unlikely that you would experience this problem.

Note: Due to the potential seriousness for this problem, IBM recommends that you update the ServeRAID Firmware as soon as possible.


FIX:

Apply ServeRAID 3.50C.

This file is available at the following URL:

http://www.pc.ibm.com/searchfiles.html
Search on "ServeRAID 3.50C firmware"

Firmware image files (e.g. 09n7686a.exe) are available from the following URL:

http://www.pc.ibm.com/searchfiles.html
search on "09n7686a"

Any corresponding Text file (e.g. 09n7686a.txt) is viewable on-line and may be printed.

Directions for creating a Firmware update diskette:

The downloaded file (e.g. 09n7686a.exe) is a self-extracting executable file that will prompt for a diskette.
Follow instructions on the display to create the diskette.


NOTES:

This ECA is "MANDATORY".

This ECA is scheduled to be withdrawn December 31. 2000.


USA:
IBM PSS/TSS CE's should record all time to Service Code 33, ECA034, Other Office 990.
Travel Time = 0.8Hrs.

USA Business Partners: Warranty claims for ECA Reimbursement
should be submitted via ECLAIM using Type Service 0E.
Refer to the Service Support Guide (SSG) for details.


EMEA:

IBM PSS CE's should record all time to Service Code 33, ECA034, (System M/T) Other Office 990.
Travel Time = 0.8Hrs.

EMEA Business Partners: Refer to Warranty Claim System;
Use Emergency Claim 5 (ECA) and enter into CPPS:
ECA Number 034
Machine Type (system affected).


Back to  Jump to TOP-of-PAGE

Please see the LEGAL  -  Trademark notice.
Feel free - send a Email-NOTE  for any BUG on this page found - Thank you.