ADSM-L

Re: Increase of tape usage after activation of new release

1998-05-11 15:40:04
Subject: Re: Increase of tape usage after activation of new release
From: Ruth Mitchell <mitchell AT TC.CORNELL DOT EDU>
Date: Mon, 11 May 1998 15:40:04 -0400
Having the same problem on an AIX (4.2) platform, server V3.1.1.1, we
followed up with IBM. We just received word that APAR IX78152 has been
issued on this. Since there's no PTF available yet, the temporary fix is:

  Until the fix becomes available, customers can identify the
  volumes that appear to have incorrect percent reclaimable
  values.  These are volumes whose percent reclaimable is very
  high (e.g., 100%) even though the percent utilization is high.
  Customers can then do either or both of the following:
  .
  o Use the Update Volume command to mark these volumes as
    unavailable so they are not selected for reclamation.
  .
  o Use Move Data or reclamation to transfer files from the
    affected volumes to a new volume.

--Ruth Mitchell
  Cornell Theory Center/NCS
  Cornell Theory Center/NCS

On May 7, 12:17pm, "James R (Jim) Owen" wrote:
}
} We have the same problem after upgrading AIX ADSM Server from v2.1.0.12
} to v3.1.1.2.  IBM ADSM Level 2 and developer are currently investigating
} our problem (and possibly yours, as well -- they just told me yesterday
} that someonelse had the same problem on another platform).  If you have
} not already called ADSM Support, please call them now.  The problem, as we see
} it, is that before the upgrade, "Query Volume Format=Detailed" showed
} %utilization and %reclaimable values where
}         %utilization + %reclaimable <= 100%
} After the upgrade and some other seemingly extraneous processing, we have >100
} tapes that show expected %utilization *and* 100% reclaimable!
}
} Updating a tape stgpool w/ REClaim=99 causes these 100+ tapes all to be
} scheduled for reclamation including at least one tape that is really 98%
} utilized and full of data!  We are hoping and expecting IBM to resolve
} our problem (pmr06818.180) sometime today or tomorrow so that we can
} resume tape reclamation this weekend.  For now, I've disabled our reclamation
} scheduling.  You probably should do that, too!
}
} Again, the problem seems caused by an error in calculating the reclaimability
} of tapes after the upgrade.  AUDit Volume for the 98% utilized tape shows 
there
} are 0 damaged files on the tape. AUDIT RECLAIM LAST=NO that shows that we do
} not have any V3 aggregate reclamation problems (0 definate and 0 possible), so
} that is not a contributiing factor in the miscalculation of the reclaimability
} for these tapes.
}
} IBM level 2 support just called asking for some additional disgnostic
} information from our server database.  When I mentioned your problem,
} She recommended that you call IBM's World Trade Support immediately, and
} "DO NOT ALLOW ANY MORE RECLAMATION PROCESSING" until IBM can fix the
} bugus 100% reclaimability of those tapes (assuming that we both have
} the same problem.)  Good luck to us.
}
} Jim.Owen AT Yale DOT Edu (USA: 203.432.6693)
}
} Stefan Cina wrote:
} >
} > Two weeks ago we have updated ADSM MVS Server V2 to ADSM MVS Server
} > V3.1.0.2. Since then ADSM acts differently:
} > - On Sonday and Monday space reclamation on 100 to 400 tapes (5 to 10
} > before the update)
} > - The last weekend ADSM has mounted 700 tapes (100 before the update)
} > Is this normal?
} > Thanks for any suggestions.
} > Regards.
} >
} > Stefan
}
}-- End of excerpt from "James R (Jim) Owen"
<Prev in Thread] Current Thread [Next in Thread>