ADSM-L

Re: TSM annoyances

2001-11-12 09:13:06
Subject: Re: TSM annoyances
From: "Kauffman, Tom" <KauffmanT AT NIBCO DOT COM>
Date: Mon, 12 Nov 2001 09:00:00 -0500
Mark -

Right now we run tapes off-site once per day. And only on weekdays. In the
worst-case scenario, we end up at the D/R hotsite with tapes suitable to
recover to 70+ hours prior to the actual outage. Our Executive management
has decreed that this is not good enough. Because of the increasing tie-in
of electronic order processing and shop-floor control, they want no more
than a three-hour data loss.

We can't get to this by shuttling tapes; we're going to need to move the
off-line redo logs electronically.

As for how often we need them -- we haven't - yet. They do get used in our
twice yearly D/R recovery tests (as I write this the away team should have
five systems running through mksysb restores; it's that time of year again
:-).

Current process - the logs get processed by brarchive/backint on a
half-hourly cycle; two copies, to two different management classes and disk
pools. The storage pools get copied once per day. This will be replaced by a
triggered copy at the completion of the brarchive script, with both pools
copying to tape to move off-site and one pool also copying to the off-site
tape library.

In general, *SM at our site is for D/R; any benefits we get in terms of
local file recovery are a plus, but not the reason the product is installed.

Tom Kauffman
NIBCO, Inc

> -----Original Message-----
> From: Mark Stapleton [mailto:stapleto AT BERBEE DOT COM]
> Sent: Monday, November 12, 2001 1:31 AM
> To: ADSM-L AT VM.MARIST DOT EDU
> Subject: Re: TSM annoyances
>
>
> On Wed, 7 Nov 2001 09:01:03 -0500, it was written:
> >It can take two weeks to fill an
> >LTO tape with our SAP/Oracle redo logs - but you need them
> to recover to
> >point of failure. These tapes tend to go off-site at about
> 10% (of 95 GB).
> >The reclaim starts just after they check out of the library.
> We're trying to
> >find an inexpensive way of moving these off-site
> electronically, but we run
> >from a low of 12 GB per day to an observed high of 45 GB per day.
>
> Think about the need for this (off-siting tape backups of log files).
>
> How often do you need the logs? How often during the day do you back
> them up? How often have you had media failure that requires using the
> copy pool copies? After thinking these through, how important *is* it
> to offsite log files?
>
> I understand the administrator's desire to offsite 'em, but is there a
> viable business case for electronic offsiting? If there's a tape drive
> or tape volume bottleneck in conjunction with the redo log backups,
> there is probably a need to review the resources you currently have,
> with a thought toward increasing those resources.
>
> --
> Mark Stapleton (stapleton AT berbee DOT com)
>
<Prev in Thread] Current Thread [Next in Thread>