IBM admits disk problems -- MIDRANGE-L

Here's a Q&A statement from IBM:


Contingency Press Statement/Q&A

IBM is aware that a limited number of customers have reported problems with
certain IBM 10KRPM hard disk drives.  IBM is working with these customers
on an individual basis to resolve their issues in accordance with IBM's
standard warranty practices by repairing or replacing the drives, as
appropriate.

Q.   Is it true that large numbers of iSeries customers are experiencing
system failures and losing critical data from their computer files?

A.   No. A limited number of customers have experienced higher than normal
fallout rates with their direct access storage devices (DASD).  The disk
drives in question are our 8.5 GB and 17.5 GB 10K rpm drives.  IBM is
replacing the failed drives under warranty and working with customers on an
individual basis to help them avoid any data loss outages.

     Although we have not yet discovered the root of the failures, we have
identified a number of issues and have released a DASD fix-pack that
contains program temporary fixes (PDFs) that will immediately correct those
problems.  Although they are limited to a few customers, we are working
aggressively to identify and provide a final fix to the problem.  IBM
remains committed to providing its customers high-quality products and
services.

Q.   Since this is a potential problem for other customers using this disk,
what steps are you taking to alert them of the problem?

A.   We are using IBM's established services practice for high impact
pervasive fixes.  The PTFs for the DASD availability fixpack have been
posted as high impact pervasive PTFs (HIPERS). In addition, the HIPERs have
been provided as a package on the ww.as400service.com fixes and updates
Website.

Q.   What are you encouraging customers do to protect themselves in case of
failure?

A.   As is standard practice, we are encouraging our customers to implement
good systems management procedures and use available iSeries tools to
minimize outages due to system device or application failures.  Best
practices for limiting exposures due to disk failure include:

   Regular Systems backups
   Implementing RAID or Mirroring to protect from data loss due to a single
   drive failure.
   Apply PTFs across all partitions, particulrly the DASD Availability
   Fixpack.
   Limit exposures to data loss failures by replacing any failed drive
   promptly.  Customers can significantly reduce replacement times by using
   Service Director as a means to automatically notifying IBM Service of a
   drive failure.  In addition, customers can automatically send a break
   message to designated users when the system detects a critical condition
   such as DASD failure.

Q.   What is the extent of the problem? Precisely what drives if any are
affected? Is this due to a bad batch of drives?

A.   The iSeries 8.5 GB and 17.5 GB 10Krpm drives have been affected,
however, the failures do not appear to be related to a specific vintage of
drives.