Re: Controller question

Controller does not keep any important data and can always be replaced.
RAID controllers have cache memory card, which can contain some data not
written to disk, if it was not shut down properly. This data are essential
for recovery, but cache card is separate from controller, has its own
battery backup and can be moved to relpacement controller card without loss
of data. If cache card does contain smth and it is itself destroyed, it
means some data on disk(s) cannot be recovered.
I think that in the reported case system cage was heavily deformed, so it
was not an easy thing to fit replacement card in.
So this was probably the major problem here.

Best regards
    Alexey Pytel



benji@intercomputer.hu on 04/08/99 08:37:28 AM

Please respond to MIDRANGE-L@midrange.com

To:   MIDRANGE-L@midrange.com
cc:    (bcc: Alexei Pytel/Rochester/IBM)
Subject:  Re: Controller question

Matthias!

I don't know how IBM DASD work exactly, but I've read an article somewhere
which stated, that IDE disk controllers (on PCs) store information about
media errors. It's a sort of bitmap I think. The errors may be introduced
during manufacturing because of current technology limitations. Instead of
throwing out every "bad" disk, the bad blocks (?) are recorded in the
controller's memory. The controller will mask these media errors from the
software level (ie. the operating system). Replacing the controller will
change the bitmap mentioned above. At this point you can even throw out
your disk.

I think the same applies to other types of disks/controllers too, including
IBM AS/400 devices. I'd bet IBM has the required hardware to backup the
controllers memory...

                         Benjamin Budai
                            (ICSS Kft. - HUNGARY)






Matthias Oertli <oertlim@s054.aone.net.au> on 99/04/08 17:53:22

Please respond to MIDRANGE-L@midrange.com

To:   MIDRANGE-L@midrange.com
cc:    (bcc: Benjámin Budai/InterComputer)

Subject:  Controller question

Ok, no one answered, I recap the story:

IBM were able to recover data of some disks (which were still good)
ONLY because they were able to get the diskcontroller (which was
broken) working.

I would dearly like to know why it was not possible to just replace
the controller. What data is kept on the disk controller? Can it be
backed up?
Any info (or pointers to some docs) would be appreciated.
Read the story, it's worth it.
Regards,
Matthias

Matthias Oertli wrote:
>
> Supposedly a true story, check it out:
> http://www.as400.ibm.com/australia/daewoo.htm
>
> According to this story, without getting the disk controller
> functional it would not have been possible to recover the data off the
> disks.
> I don't understand this. In case of controller failure, can't you just
> plug in another one?
> Regards,
> Matthias

---------------------------------------------------------------------
Matthias Oertli   <oertlim@s054.aone.net.au>
+---
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to MIDRANGE-L@midrange.com.
| To subscribe to this list send email to MIDRANGE-L-SUB@midrange.com.
| To unsubscribe from this list send email to
MIDRANGE-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator:
david@midrange.com
+---