Hi James,
If I understand well, every week, you backup 160 GB of data to new tapes
that are taken offsite ?
Have you tried to use backup STG. This command only put offsite the new
files. Say you have
1% of new files, you only have to offsite some GB and not some hundred GBs.
You do not need
to purchase DRM to do this, there are plenty of scripts that can helps you.
On your backup site, do you have a TSM server w/ a 3575 lib ? If yes, that's
is the solution,
you only move "week" data. Else, sometime, you will need to reclaim your
offsite tapes and
regenerates a full set of unfragmented-copy-tapes. I did this for a 8 TB
site and only a few tapes
(and DLT) are moved offsite every day.
As mentionned in the list, set the retention to 0. or scan IDLE drives an
manually dismount tapes
from a admin sched.
Hope this helps
Rv
----- Original Message -----
From: "Poehlman, James" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, April 02, 2001 6:51 PM
Subject: Export Node Mount Waits
> TO *SMers - All,
> New to this so please bear with.
>
> ENVIRONMENT:
> TSM Server IBM RS6000 R50 2 Procs 1GB ram
> AIX 4.3.3 MNT_lev 004
> TSM Server Version 3.7.2
> Library IBM3575 - L12 2 drives - NOT XL
> No disk pools.
> All pools sequential to 3575.
>
> APPLICATION:
> Export node every weekend - active files for client node.
> (WHY? - Company Policy requires weekly full backups of this
> client to go off site.)
> This is in addition to daily incremental to the primary storage pool.
> Exports:
> 395,963 items
> 160,980,891,020 bytes.
> Copies active files from 71 sources tapes (2 storage pools) to 15 output
> tapes.
> Starts 0300 Saturday morning and runs until Sunday Afternoon.
>
> PROBLEM:
> Export node waits for retention timeout on source tape before
> unmounting it and mounting the next source tape. With a 5 minute
> mount retention this wastes almost 3.5 hours. Even with 1 minute
> retention wastes way over 1 hour. I have other thing that need to
> happen over the weekend that also need these drives.
>
> QUESTION:
> Is there a way around this? Is there a fix for it?
> Is there a better way to get a set of offsite tapes for
> the active files for the node without doing a selective
> backup to new storage pool?
> Please don't mention backup sets. Also, please don't tell
> me I need more drives. Upgrade request for them has been in the
> works, for more than 6 months along with upgrade to a L18.
> These systems are orphans, so to speak. Upper management does
> not like to spend money on them. Another story.
>
> Thanks to all,
>
> James D. Poehlman (Denny)
> AIX / Unix Systems Administrator
> Senior Technical Specialist
> Mid-Range Server Technology Team
> Black and Decker
> North American Power Tools - Information Systems
> [EMAIL PROTECTED]
> Voice 410-716-3039
> Cell 410-375-5974
>
