HP3000-L Archives

May 1995, Week 4

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
"Rudderow, Evan" <[log in to unmask]>
Reply To:
Rudderow, Evan
Date:
Thu, 25 May 1995 18:26:00 EDT
Content-Type:
text/plain
Parts/Attachments:
text/plain (50 lines)
Stan Sieler <[log in to unmask]> wrote:
 
>One of our clients has a 4.0 system with several dozen ldevs
>in mpexl_system_volume_set (yes, we've told them that isn't a good
>idea).  8 drives are on one HPFL string (the rest are non-HPFL).
>One of those went "bad" (red fault light), but now appears to be working.
>
>However, when they boot up, and do a DSTAT ALL, they see
>most of the disks are ok, but all 8 drives on the HPFL string are
>in the LONER state (and can't be accessed via file system).
 
<snip>
 
>We've power cycled ... no success.
>We've replaced the HPFL controller card ... no sucess.
>We've searched HPSL ... no success.
>We've booted up with drive 34 turned off ... no sucess.
 
This is a *really* outside chance --
 
Once upon a time before I started using user volumes I added a brokered 7933
to my S/950; I didn't need the space right away, so in volutil I set
permanent space to 0%.   Later I upped permanent space to 100%.  Things went
fine -- until the next START NORECOVERY.  (I've forgotten what state the
volume was in after that re-boot disc.)  HP came in and, somehow, got it
working.  At the next re-boot the same thing happened -- the drive got lost,
once again HP did magic.  Third time, same story; I got rid of the drive.  I
don't know if it was a hardware problem or if, by initially setting
permanent space 0% the volume label got screwed up (although I seem to
recall that the RC center also said there were irregularities with volume
related tables on the volume master.)
 
So, did your customer ever do something creative like what I described
above?
 
Some other outside chances:
 - You said you looked at the volume labels, try looking at the
volume-related OS tables on Ldev 1 (sorry, I don't know where they are or
where they live)
 - Could there be something funky going on with generation numbers that might
make the system think that these MPEXL_SYSTEM_VOLUME_SET members belong to
some other MPE...SET?
 
From what you describe it doesn't appear to be a hardware problem --
although it might have been caused by H/W.
 
Good luck!
 
 -- Evan

ATOM RSS1 RSS2