3592 standalone drives at DR Site

[T. M. Ioia]

Hi all,

We just recently started using a 3494 ATL with 3592 drives in our data
center. We are also creating our DR tapes using the tape library and ejecting
the tapes to send to off-site storage.

At our DR site, we do not have a 3494 ATL tape library but we have stand-
alone 3592 drives.

I'm having trouble deciding how to best set up our DR recovery process.

Can I maintain the same addresses, for example 600-603 and just change the
LIBRARY YES to LIBRARY NO and MTL NO to MTL YES in the device definitions
using HCD?

I have a feeling I'm going to run into trouble with SMS by doing this. Do I
have to create a new MTL definition, storage classes, storage groups, etc.

Initially, we just want to restore the system at the DR site using the 3592
tape drives. Longer term, as we utilize the 3494 more and more, we will want
to be able to run jobs that think they have a tape library available.

Ideally and conceptually, what I would like to do, is operate in ATL mode at
our data center and in MTL mode at the DR site, with the least amount of
reconfiguration as possible.

I have no way of testing this before going to the DR site. All of our 3592
drives are in the 3494, no standalone drives. I thought I could define one of
the 3592 as part of an MTL, but it wouldn't let me bring it on-line because the
system knows that it is inside the ATL

Any ideas?

Thanks.

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to list...@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[Lorne Dudley]

We had this problem at one of our DR exercises.

Our solution was to define a separate set of non-library drives that
were always offline at our home installation.

For example

IEE457I 21.05.31 UNIT STATUS 407
UNIT TYPE STATUS VOLSER VOLSTATE
A000 3590 OFFLINE /REMOV
A001 3590 OFFLINE /REMOV
A002 3590 OFFLINE /REMOV
A003 3590 OFFLINE /REMOV

IEE457I 21.06.26 UNIT STATUS 483
UNIT TYPE STATUS VOLSER VOLSTATE
0A00 359L O-NRD-R /REMOV
0A01 359L O-NRD-R /REMOV
0A02 359L O-NRD-R /REMOV
0A03 359L O-NRD-R /REMOV

Regards

Lorne Dudley
Queen's University
Kingston, Ontario

[Clark, Kevin]

T. M.

If the DR site is under VM, that just attached the devices (6xx-6xx) as
your own define3d standalone devices. Toy will have to define new 6XX
addresses with the LIBRARY NO.

Kevin

Hi all,

reconfiguration as possible.

Any ideas?

Thanks.

This e-mail message and any attachments transmitted with it are confidential and are intended solely for the use of its authorized recipient(s). If you are not an intended or authorized recipient, you are hereby notified that any disclosure, copying, distribution or taking any action in reliance on the information contained in this e-mail is prohibited. If you have received this message in error or are not authorized to receive it, please immediately notify the sender and delete the original message and all copies of it from your computer.

[Lizette Koehler]

At DR you may need to do a DISCONNECT on your VCAT catalog. We ran into a
problem where at DR (same as your setup) we kept getting a message about a
Library and we did not have one. The DISCONNECT of the VCAT corrected that
issue.

Lizette

----------------------------------------------------------------------

[Lizette Koehler]

I should be more clear - VOLCAT VGENERAL catalog (my term VCAT for short)

Lizette

> Subject: Re: 3592 standalone drives at DR Site

>
> At DR you may need to do a DISCONNECT on your VCAT catalog. We ran
> into a
> problem where at DR (same as your setup) we kept getting a message
> about a
> Library and we did not have one. The DISCONNECT of the VCAT corrected
> that
> issue.
>
> Lizette
>

----------------------------------------------------------------------

[Neubert, Kevin]

Look here as well: http://www-01.ibm.com/support/docview.wss?uid=isg3S1001830.

Regards,

Kevin

-----Original Message-----
From: IBM Mainframe Discussion List [mailto:IBM-...@bama.ua.edu] On Behalf Of T. M. Ioia
Sent: Thursday, January 07, 2010 6:00 PM
To: IBM-...@bama.ua.edu

[John Kelly]

<snip>

Our solution was to define a separate set of non-library drives that were
always offline at our home installation.

</snip>

Wouldn't you have a problem with SMS assigning to the non-library drives
at the DR or do you use different ACS' or unitnames at DR?

Jack Kelly
202-502-2390 (Office)

[Darth Keller]

We have both manual 3590 & 3592 tape drives at DR. We had issues with the
wrong tapes being mounted on the wrong tape drives because they both have
the same device type. We don't have any stand-alone 3592's at 'home' -
they're managed as part of the native library we have defined.

We got around it temporarily by varying the different drives offline
depending on which drives we needed. Our final solution for the next DR
was to create an MTL for our 3592 tape drives and allowing SMS to manage
them. The Operating Systems group had to define another library and, yes,
we had to make SMS changes for the MTL. We had to delete the old library
definitions and define new definitions for the DR VTS and the MTL. It's
worked like a charm since then.
ddk

This e-mail message and all attachments transmitted with it may
contain legally privileged and/or confidential information intended
solely for the use of the addressee(s). If the reader of this
message is not the intended recipient, you are hereby notified that
any reading, dissemination, distribution, copying, forwarding or
other use of this message or its attachments is strictly
prohibited. If you have received this message in error, please
notify the sender immediately and delete this message and all
copies and backups thereof. Thank you.

[Lorne Dudley]

Hello John

We assign to 3590 units, whether tape library or not.

Also EXPORT VOLCAT DISCONNECT as others have pointed out.

Regards

Lorne

John Kelly wrote:
> <snip>
> Our solution was to define a separate set of non-library drives that were
> always offline at our home installation.
> </snip>
>
> Wouldn't you have a problem with SMS assigning to the non-library drives
> at the DR or do you use different ACS' or unitnames at DR?

----------------------------------------------------------------------

[T. M. Ioia]

Hi Lorne, can you elaborate more on what you mean by "We assign to 3590
units, whether tape library or not." ?

Doen't your DATACLAS routine have to resolve the unit to a STORAGE GROUP
that specifies a single tape library, and that tape library will be either an ATL
or MTL, but can't be both.

Thanks.

[Lorne Dudley]

My colleague who does the SMS routines will be away until next week.

I'm sending direct to your email a copy of the SMS DATACLAS routines we
are using.

At the DR site we do not have a tape library but non tape library drives
accept tapes from our robot. I'm not aware of any special SMS activity
that we had to do. I'll check with my colleague on his return.

Regards

Lorne

[Lorne Dudley]

Hello ti...@YAHOO.COM .

Your email address bounced when I tried to send you the SMS DATACLAS we
are using.

Send me a good email address direct to dud...@queensu.ca and I'll try
again.

[SUBSCRIBE IBM-MAIN M. Ioia]

Thanks Lorne (a fellow Canadian, I'm in Toronto). Please try again to
ti...@yahoo.com. It was de-activated because I hadn't used it in awhile.
Should be ok now.

I appreciate the info, it sounds like a clean solution and similar to what I had
in mind, I'm anxious to see it.

[John Kelly]

<snip>

Your email address bounced when I tried to send you the SMS DATACLAS we
are using.

</snip>

I'm not too sure what good a DATACLASS ACS routine will be, unless it's
for the media or to tell the SC and SG routine what to do. If it's non
SMS, the SC routine will have to make the decision, not the DClass.

Jack Kelly
202-502-2390 (Office)

[SUBSCRIBE IBM-MAIN M. Ioia]

Thanks for the email Lorne. I had a look at your DATACLAS routine and it
really doesn't help to understand how you are accomplishing the transparency
between your datacenter and DR site. Your routine sets DATACLAS=DC3494
for any Library Tape devices. I would have to see the STORCLAS and
STORGRP routines to see how this might eventually be mapped to a STORAGE
GROUP name, and what those storage groups look like.

I just noticed that I can specify multiple LIBRARY names for the STORAGE
GROUP. Does anyone know if SMS will cycle through multiple LIBRARY names
until it finds one that is on-line? This way, I can define an ATL and an MTL
and SMS will always find the ATL at the datacenter and only find the MTL at
the DR site. Does this sound plausible? Anybody ?

[SUBSCRIBE IBM-MAIN M. Ioia]

Hi Darth, I find you post interesting because it is along the lines of what I
would like to accomplish.

Are you saying that you are in fact using the same library name at 'home' and
at the DR site? This approach seems like its really the only way to keep it
somewhat simplified, otherwise wouldn't the CA-1 (entry exit routine) have to
be changed as well, mine has the library name hard-coded?

Do you have to go through any ISMF changes at the DR site, or is the DR
IODF with MTL runed on for the devices sufficient for SMS to come up with an
MTL instead of an ATL ? What about the TCDB, will it be ok with the switch
from ATL to MTL, or will it have to be re-built somehow?

The more I think about this, the more complex it becomes. Sounds like a good
excercise for an IBM Redbook.

thanks

[Brian Fraser]

>Does this sound plausible?

Yes.

I have both PROD & DR Libraries defined in my STORAGE GROUP.

Only the PROD Libraries come online and available while running in PROD, and
only the DR libraries come online and available when running in DR.

[Lorne Dudley]

ti...@YAHOO.COM

I'll describe our DR exercise a bit and get you more information next
week when the SMS guy returns, including a sample of all of the SMS code
if you wish.

When we go to the DR site we take our backup tapes from a week back
(leaving the most current at our offsite home location).

We normally run with DFrmm, which we turn off at the DR site. Other
than restore of the system to a state of a week old +, we do not run a
lot of tape jobs in DR exercise mode. Any tape jobs that do run use the
catalog at that point in time for input. Any output tapes required are
mounted using a manual recording system.

Of course, in a real DR mode things would be different. We would have
DFrmm up and go through the process of restoring updated data on return
to the restored home production environment, not an easy task.

We use the IBM DR centre in Toronto. How about you ?

Regards

Lorne

[Mike Wood]

When your ATL and MTL have different library names this is ok for creating
new data in production/DR. But for referencing existing data you will have
problems because the TCDB library location for each private volume will be
atl, but in DR you want it mounted in mtl.

So, you would need to alter the volume entries accordingly or look at having
just 1 library defined, and using the same name, but at DR site alter the
library entry to be MTL.

The volcat only needs to be disconnected if you use stand-alone drives which
are not system managed MTL at DR.

Mike Wood RMM Development

On Sat, 9 Jan 2010 13:33:18 +0800, Brian Fraser <brianm...@GMAIL.COM> wrote:

>>Does this sound plausible?
>
>Yes.
>
>I have both PROD & DR Libraries defined in my STORAGE GROUP.
>
>Only the PROD Libraries come online and available while running in PROD, and
>only the DR libraries come online and available when running in DR.
>

[SUBSCRIBE IBM-MAIN M. Ioia]

Hi Mike, I think we are on the same wavelength....you describe exactly what
is going through my mind.

>>So, you would need to alter the volume entries accordingly or look at having
just 1 library defined, and using the same name, but at DR site alter the
library entry to be MTL.<<

To re-cap:

Option 1 -
- two Libraries defined, one ATL, one MTL.
- ATL comes on-line at home site, MTL comes on-line at DR site
- STORAGE GROUP will list both library names
- no changes to ACS routines, all allocations point to same STORAGE GROUP
- at DR modify TCDB entries to point to MTL library name (how?)
- CA-1 library exit needs to know about both libraries (any other implications?)

Option 2 -
- one library defined
- use a different DR IODF with ATL devices defined as MTL devices
- modify ACDS to re-define the library definition as a Manual library
- any CA-1 implications ?

I think Option 2 might be easier to do, but need to investigate tape
management implications.

Do you know if any of the RMM docs address this problem?

thanks.

[SUBSCRIBE IBM-MAIN M. Ioia]

Hi Lorne, we now use IBM DR as well, we just moved from another vendor.

At this stage, our DR process is similar to yours, although I'd like to move it
futher along in terms of getting closer to our real production environment.

We have CA-1 up and running for stand-alone tapes, but want to replicate our
tape library to make production runs as transparent as possible. See my
other post at alternative ways of getting tape library transparency at DR site.

[Lorne Dudley]

Hello ti...@YAHOO.COM !

I talked with my SMS guy.

It turns out that we have never written to 3590-1 tapes during our DR
exercise. We read from them to restore the system with separately
defined non-library drives, but never write, so our SMS routines do not
come into play. (We do a little writing to 3490 tapes.)

As a result of this discussion this will be on our agenda for the next
DR exercise.

Here is part of the code in our current SMS definition.

FILTLIST LIB_TAPE INCLUDE('3590-1') /*Library Tape */

WHEN (&UNIT EQ &LIB_TAPE /* 3590 TAPE */
AND &STORCLAS = 'SCLIBDUP') /* Tape Library SC */
DO
EXIT /* Exit with STORCLAS=SCLIBDUP*/
END
/* */
WHEN (&UNIT EQ &LIB_TAPE) /* 3590 TAPE */
DO
SET &STORCLAS = 'SCLIBDUP' /*Force SCLIBDUP */
EXIT
END

The initial thought from our SMS guy was to disable the allocation of
3590-1 to the ATL at the DR site.

We will be talking about this over the next week or so and we will
definitely be writing to 3590-1 tapes at our next exercise.

Thanks to all who contributed to this discussion.

Lorne Dudley
Queen's University
Kingston, Ontario

----------------------------------------------------------------------

[Scott Rowe]

But what about a real disaster? Would you need to write 3592 tapes? I would think so, and in that case you should certainly have it worked out (and tested) now, not wait until the real thing

>>> Lorne Dudley <dud...@QUEENSU.CA> 1/13/2010 3:21 PM >>>
Hello ti...@YAHOO.COM !

CONFIDENTIALITY/EMAIL NOTICE: The material in this transmission contains confidential and privileged information intended only for the addressee. If you are not the intended recipient, please be advised that you have received this material in error and that any forwarding, copying, printing, distribution, use or disclosure of the material is strictly prohibited. If you have received this material in error, please (i) do not read it, (ii) reply to the sender that you received the message in error, and (iii) erase or destroy the material. Emails are not secure and can be intercepted, amended, lost or destroyed, or contain viruses. You are deemed to have accepted these risks if you communicate with us by email. Thank you.