ADSM-L

Re: Disaster recovery

1997-04-23 16:32:38
Subject: Re: Disaster recovery
From: "Pittson, Timothy ,HiServ/NA" <tpittson AT HIMAIL.HCC DOT COM>
Date: Wed, 23 Apr 1997 16:32:38 -0400
Would setting the reclamation threshold to 0 for the existing offsite
copypool  have the same end result as what Bill describes below ??  For
the IBMer's listening in - would ADSM process the tapes in nodename
order which, in effect, would collocate the offsite copypool if your
onsite storagepools are collocated. ??

Tim Pittson
tpittson AT himail.hcc DOT com
>----------
>From:  Bill Colwell[SMTP:bcolwell AT DRAPER DOT COM]
>Sent:  Thursday, April 17, 1997 4:51 PM
>To:    ADSM-L AT VM.MARIST DOT EDU
>Subject:       Re: Disaster recovery
>
>Why don't you make another copypool which is collocated?
>
>The thread of your messages is that you are worried about the number of
>tapes which would be called in from a non-collocated copypool during a
>client restore when the primary volumes are marked destroyed.  You can have
>more than one copypool.  So, some weekend, created another stgpool
>type=copy, backup your primary pools to it using 'numpr=a_large_number'.
>Whether you define this pool as collocating or not it will be more or less
>collocated because the input pool is collocated.  Send the new copypool
>volumes offsite and set up your DR plans to use them.  Then update it to be
>non-collocating and continue to do your backups to it.  After enough time
>that the tail end is big enough to be worrisome, repeat the process.
>
>You will use more volumes and your database will grow, but this may get you
>what you are really after which is a mostly collocated set of offsite
>volumes.
>
>Bill Colwell
>The Charles Stark Draper Laboratory
>Cambridge Ma.
>
>
>_________________________Reply Header_________________________
>Author: ADSM-L AT vm.marist DOT edu
>Subject: Disaster recovery
>04-17-1997 04:17 PM
>
>Hi all, yes I'm still fretting over Disaster Recovery.  If you lose
>the site that has your MVS server, and about 50 servers that are in
>that computer room... you have to take ALL your offsite tapes (which
>belong to ALL your market offices and other buildings) to your hotsite
>location, because I dont have a daily report showing which offsite
>tapes belong to which servers?  Does anyone know if IBM is working on
>a way to remedy this?  SHOW VOLUMEUSAGE doesn't work for copy pool
>tapes.  I can run a (loooong running) job today that does a Q CON on
>each offsite tape, then SAS's the list, and gives me a list.  But it's
>such a pig I can't run that every day.  And if we lose the one location
>with the mainframe and all the servers in that location, I can't run
>the job.  It'd be too late then.  Does anyone know if they're going
>to fix a way to report offsite tapes by node?    Does the DRM component
>do this?
>THANKS!
>Julie Phinney
>JULPHINN @ EMPHESYS.E-MAIL.COM
>
<Prev in Thread] Current Thread [Next in Thread>