Re: Remote clients transferring GBs over WAN

[John Wright]

>From 4.1 onwards there is Adaptive sub-file backup for Mobile Clients.
I don't use it or know much about it but I think it allows you to backup
the changed bytes/blocks of a file on a laptop rather than the whole
changed file.
This may be of some use to you depending on the data they are backing up.

Cheers
John



"Richard Sims"
<[EMAIL PROTECTED]>

Sent by: To: ADSM-L
"ADSM: Dist
Stor Manager"cc:
<[EMAIL PROTECTED]   Subject: Re: Remote clients 
transferring GBs over WAN
IST.EDU>


26-Mar-2003
13:17
Please respond
to "ADSM: Dist
Stor Manager"
<[EMAIL PROTECTED]
IST.EDU>





>The organization I work for has four main establishments. Typically,
someone
>will go "on mission" to another establishment and perform a backup of his
>laptop. Since his dsm.opt points to a server that that is hundreds, if
not,
>thousands of kilometers away, this backup will be done over the WAN and
not the
>usual LAN. Occasionally, this user might even attempt to backup 1 or 2 GBs
over
>the WAN. This obviously causes some problems.
>
>What to do about this? Blocking the port in the firewall comes to mind but
that
>is not a real answer.  Has anybody encountered this problem and what, if
>anything, did you do? Thank you...

Such backups have to be painful for the laptop user as well.  I should
think
that following the procedures outlined in the Windows client manual under
"Performing a backup with limited bandwidth" should yield relief.
There is a whitepaper on the IBM site describing it in fuller detail.

  Richard Sims, BU

Re: Why so many Filling tapes?

[John Wright]

Depends how many tape drives you have and how many migration is set to use
I would have thought.
If there are four drives and migration is allowed to use them all it will.
Therefore four tapes will be mounted but its unlikely all will be written
to fully leaving you with several filling tapes as opposed to one full.

Also it must be quicker for it to stream to multiple drives as opposed to
one.

Cheers
John





"Roger Deschner" <[EMAIL PROTECTED]>

Sent by: "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]>
04-Feb-2003 01:03
Please respond to "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]>




To: ADSM-L

cc:
Subject:Why so many Filling tapes?

I have a non-collocated sequential (tape) storage pool. Why should there
ever be more than one tape in Filling status? I have four. Why can't it
fill one up and then start another? These are Super-DLT tapes, so each
one means something - 200gb compressed.

I understand why I have lots of Filling tapes in a collocated tape pool
- that's the whole idea of collocation. But this is happening in a
non-collocated tape pool.

Roger Deschner  University of Illinois at Chicago [EMAIL PROTECTED]
== "Sometimes it pays to stay in bed on Monday, rather than spending ===
= the rest of the week debugging Monday's code." -Dan Salomon ==

Re: Error Message : 'number of mount points available for removab le media.'

[John Wright]

Try a q path to see if the paths to the drives are ok.




"PINNI, BALANAND (SBCSI)" <[EMAIL PROTECTED]>

Sent by: "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]>
27-Jan-2003 16:55
Please respond to "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]>




To: ADSM-L

cc:
Subject:Re: Error Message : 'number of mount points available for 
removab le
media.'

Hi,

Drive may show as available .But did your library come up ok after
./dsmserv
?I feel its library problem.
Balanand Pinni

-Original Message-
From: Nick Rutherford [mailto:[EMAIL PROTECTED]]
Sent: Monday, January 27, 2003 10:44 AM
To: [EMAIL PROTECTED]
Subject: Error Message : 'number of mount points available for removable
media.'

Hello all,

We've started getting the following messages in the TSM log whenever any
tape
processing attempts to start

ANR4571E Database backup/restore terminated - insufficient number of mount
points available for removable media.

A  similar message is issued when we run migration or reclamation.

On Saturday, our 3494 tape library lost connection with our TSM server.
Both
the
TSM server and the 3494 have been re-booted and the Token ring connection
has
now been restored to the library manager.

We're running a TSM V5.1.1.0 server on a AIX/F50 platform.
IBM 3494 tape library shared between the TSM server and MVS/OS390
partition.

All 4 drives have been physicaly reset.
IBM 3494 says the drives are all available.
All 4 drives can be seen by the AIX & TSM.
The device class says to use a mount limit of 'DRIVES'.

All LAN definitions on the ATL look o.k also.

Can anybody confirm whether this message is drive or library manager
related
and
any potential reasons why this should be displayed.

many thanks,
Nick Rutherford


*
This e-mail is confidential and intended solely for the use of the
individual to whom it is addressed.  Any views or opinions presented
are solely those of the author and do not necessarily represent
those of Honda of the UK Manufacturing Ltd.

If you are not the intended recipient please notify the sender
immediately by return e-mail and then delete this message from your
system.  Also be advised that any use, disclosure, forwarding,
printing or copying of this e-mail if sent in error is strictly
prohibited.  Thank you for your co-operation.
*

Re: DELETE FILESPACE Question

[John Wright]

Make sure you make a copy of the current volhist file somewhere safe
before you restore the old db though just in case!.




"David E Ehresman" <[EMAIL PROTECTED]>

Sent by: "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]>
10-Jan-2003 13:19
Please respond to "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]>




To: ADSM-L

cc:
Subject:Re: DELETE FILESPACE Question

>Is there a chance to reconstruct the data ??

It won't be easy.  But you can backup your current data base, restore
the data base to an earlier backup that has the archive files.  Export
the archive files.  Restore your current data base from backup.  Import
the exported archive files.

Re: Non IBM Libarary for TSM

[John Wright]

This HP library sounds very similar to the SUN L20/40/60 Library
(Re-badged HP KIT) We have one using DLT and it functions ok. It has also
remote management card also so you can administer the library, for the
most part, from a web  type interface.
ADIC Scalars are also supported I think.




"Lloyd Dieter" <[EMAIL PROTECTED]>

Sent by: "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]>
18-Dec-2002 14:35
Please respond to "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]>




To: ADSM-L

cc:
Subject:Re: Non IBM Libarary for TSM

IBM OEM's an HP library that IBM calls the 3600, that will use either DLT
or LTO.  They start at 2 drives/20 slots, and go up to 6 drives/60 slots.

They are supported by TSM, but I have no personal experience with them.

Regards,

-Lloyd

On Wed, 18 Dec 2002 17:14:47 +0530
Raghu S <[EMAIL PROTECTED]> wrote:

> Hi
>
> I work on TSM soluion design. Most of the time i see customer unhappy
> with the price of 3583 LTO.
> ( I stopped suggesting single drive libraries for TSM ). I found Magstar
> 3570 MP is a suitable one for medium sized corporates.I have no idea on
> Non IBM libraries.Can anybody suggest me Non IBM libraries and
> corresponding URLs ( Should work with TSM as efficient as IBM libraries
> and cheaper).
>
> Thanks in advance
>
> Regards
>
> Raghu.
>
>
>
> Richard Foster
>  HYDRO.COM>  cc:
> Sent by: "ADSM: Subject: Re: timeout
> value for processes Dist Stor
> Manager"
> <[EMAIL PROTECTED]
> T.EDU>
>
>
> 12/18/2002 04:20
> PM
> Please respond
> to "ADSM: Dist
> Stor Manager"
>
>
>
>
>
> The Mountwait parameter only specifies how long the process will wait
> for a tape to be mounted after the process has been allocated the
> necessary drive(s).
>
> If I understand Henrik's question, there doesn't appear to be a
> parameter which governs how long a process will wait for the tape drives
> to become available.
>
> We just run a script which cancels all Reclaim processes, whether they
> are active, waiting for tape mounts or waiting for drives.
>
> Richard Foster
> Norsk Hydro
>
> 
> mountwait parameter in devclass definition
>
> regds
> Raghu
>
> 
> Hi all!
>
> Is there a server option or a another why to specify for how long a
> process will be active?
>
> Example, If all drives are busy I might not want reclamation to be
> pending for hours until two drive becomes available.
>
> Henrik.Wahlstedt
>


--
-
Lloyd Dieter-   Senior Technology Consultant
 Registered Linux User 285528
   Synergy, Inc.   http://www.synergyinc.cc   [EMAIL PROTECTED]
 Main:585-389-1260fax:585-389-1267
-

Re: ?? Cost on a storage frame on 3494

[John Wright]

Last retail price I had on a frame was about £21k though discounts are 
available.

3590H. I looked into these, they can be SCSI or Fibre channel. If you 
already have SCSI the default upgrade is to 3590H SCSI.
For a few hundred £ more you can have them converted to FC at the same 
time you have the upgrade done.
These can be directly fibre attached to the Unix host or to the host via a 
switched network but don't forget to include fibre HBAs for the Host in 
your budget.

I hear IBM are doing a seasonal special on this very upgrade right now.

Regards
John




"Lisa Cabanas" <[EMAIL PROTECTED]>

Sent by: "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]>
11-Dec-2002 15:47
Please respond to "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]>

 
 

To: ADSM-L

cc: 
Subject:?? Cost on a storage frame on 3494

Howdy all,

any one have a quick and dirty estimate of what a storage frame runs on a
3494 (retail, gov't)?  My IBM contacts are out of the office today, and 
one
of my bosses wants to talk storage planning today.  I have prices for the
L12 and D12.

While I'm at it-- any of you guys with the new H drives-- what do they run
(the FC ones?)

thanks!!

lisa

Re: Relocating TSM DB (storage)

[John Wright]

Any idea if moving database data between volumes like this will free any
fragmented space as it moves the data?

Cheers
John



"Zlatko
Krastev/ACIT"
   To: ADSM-L

Sent by: cc:
"ADSM: Dist  Subject: Re: Relocating TSM DB (storage)
Stor Manager"
<[EMAIL PROTECTED]
IST.EDU>


22-Nov-2002
15:17
Please respond
to "ADSM: Dist
Stor Manager"
<[EMAIL PROTECTED]
IST.EDU>





Think for "available" as total space and for "assigned" as maximum used.
This is like hot-spare disk you may or may not put in your RAID array.
TSM allows a volume to be deleted only if it is smaller than Available
minus Assigned, i.e. there is available but unassigned space to move data
from volume to be deleted.

Zlatko Krastev
IT Consultant






"Robert L. Rippy" <[EMAIL PROTECTED]>
Sent by: "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]>
22.11.2002 16:49
Please respond to "ADSM: Dist Stor Manager"


To: [EMAIL PROTECTED]
cc:
Subject:Re: Relocating TSM DB (storage)


If you don't extend and create more 'available' space, then how is there
space available to move the DB's to when you start deleting volumes?
Before TSM can use DB space, you must extend and make the space available
for use.

Thanks,
Robert Rippy




From: Andy Carlson <[EMAIL PROTECTED]> on 11/22/2002 09:42 AM

Please respond to "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]>

To:   [EMAIL PROTECTED]
cc:
Subject:  Re: Relocating TSM DB (storage)

Do not extend the DB before doing the delete dbvol.  The delete command
uses the empty, enextended space to move the data from the volumes to be
deleted.


Andy Carlson|\  _,,,---,,_
Senior Technical Specialist   ZZZzz /,`.-'`'-.  ;-;;,_
BJC Health Care|,4-  ) )-,_. ,\ (  `'-'
St. Louis, Missouri   '---''(_/--'  `-'\_)
Cat Pics: http://andyc.dyndns.org/animal.html


On Fri, 22 Nov 2002, Robert L. Rippy wrote:

> Yes, you are correct. Create the new DB's and extend the DB and then
delete
> the old DB's. The data will migrate over to the new extended space. Any
> reason you had 37 1GB volumes for your DB's. IBM told me not to really
go
> over 10 volumes on your DB and to be sure they aren't on the same drive
if
> you can help it.
>
> I would suggest that if you could, create 5 8GB new volumes for a total
of
> 40GB and move the DB over to the new 8GB partitions.
>
> Thanks,
> Robert Rippy
>
>
>
> From: "Thach, Kevin" <[EMAIL PROTECTED]> on 11/22/2002 09:30 AM
>
> Please respond to "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]>
>
> To:   [EMAIL PROTECTED]
> cc:
> Subject:  Relocating TSM DB (storage)
>
> I am wanting to move our TSM DB which is currently on ESS, and has 37
1GB
> dbvolumes, to another set of ESS disk.  If I define the new dbvolumes on
> the
> new disks, do I then just extend the db, and then do a del dbvol on the
old
> ones?  Does the delete make the data migrate over?
>
> Do I have to do a reduce db for any reason before I do the delete dbvol?
> Can I do the delete with the server up and running?
>
> Thanks
>
> This E-mail contains PRIVILEGED AND CONFIDENTIAL INFORMATION intended
only
> for the use of the Individual(s) named above.  If you are not the
intended
> recipient of this E-mail, or the employee or agent responsible for
> delivering it to the intended recipient, you are hereby notified that
any
> dissemination or copying of this E-mail is strictly prohibited.  If you
> have
> received this E-mail in error, please immediately notify us at (865)
> 374-4900 or notify us by E-mail at [EMAIL PROTECTED]
>

Re: Backup Question

[John Wright]

It may just be that the file had been updated between the end of the
previous incremental and the start of this one. Consequently it would be
flagged for backup when TSM builds its list.
By the time TSM gets around to backing an application or a person could
have amended the file again thus giving a time stamp later than the
initial TSM scan.

As long as the file is not changing at the point it gets backed up, i'm
guessing that you would see no error messages. I'm not sure if TSM would
know or bother to check if the files size/attributes had changed between
the initial scan and backup unless it found it open.
It knows its supposed to back it up and just does it.

Regards
John




"Thomas Denier" <[EMAIL PROTECTED]>

Sent by: "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]>
08-Nov-2002 17:34
Please respond to "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]>




To: ADSM-L

cc:
Subject:Re: Backup Question

> Zoltan,
> Thanks for your reply, but here's what I thought:
> After it builds a list of files to backup, it then traverses/sweeps the
> tree(File System) to do the actual backup. At this time if it finds a
> new file in its path, it will back it up. If the file is placed in that
> path after TSM has gone past that File System Structure it will not go
> back to back it up since it has no knowledge of that file.
> The reason for my explaination is that we have seen many files whose
> modified time is much later than the Backup Start time and it is still
> backed up.
> Let me know your thoughts.

TSM clients from 3.7 onward have been designed to overlap the later
part of the file scan with transfer of new and changed files found in
the earlier part of the scan. On a lot of our clients the time stamps
in the log indicate that the scan stretches out over most of the duration
of the backup. With that kind of timing behavior the scan might pick up
files created toward the end of the backup, depending on the location
of the files.

Re: restore to diff. client

[John Wright]

On unix if you want to restore data from client A onto client B logon to
client B and type dsmc -virtualnode=A.
As long as you know the password for client A you can then restore all the
stuff you like.

Alternatively logon to client B, amend the dsm.opt file os that it looks
the same as that for client A, restart the scheduler and you can restore.
John




"Michelle Wiedeman" <[EMAIL PROTECTED]>

Sent by: "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]>
07-Nov-2002 12:25
Please respond to "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]>




To: ADSM-L

cc:
Subject:restore to diff. client

Hi!

Does anyone have a clue how to do a restore from one client to another on
the adsm commandline interface?

Thnx!
michelle

Re: Safe to backup DB while c

[John Wright]

DB backups seems to work ok and I have run them alongside clients
sessions. This how I understand it works, similar to and Oracle backup
really.

The DB backup seems to 'freeze' database activity even if client sessions
are active and generates its image to backup. While the database activity
is frozen and the database backup running the transactions generated by
the client are written to the Logvolumes instead of Database volumes. When
the backup complete these Log volume transactions are sync'd back binto
the database.

Some considerations to here are that your DB backup  is out of date as
soon as it completes if it backed up while clients were active.
Should you need to recover your DB you will be losing data straight away.
Roll forward recovery mode may be an option here plus the various ways of
running incrementals etc so you could look into that option.

You must have sufficient recovery log space available to take all the
client transactions whilst the backup is running. You can monitor how much
log is used between dbbackups by reset log consumption statistics.



John







"Sam Sheppard" <[EMAIL PROTECTED]>

Sent by: "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]>
06-Nov-2002 17:08
Please respond to "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]>




To: ADSM-L

cc:
Subject:Safe to backup DB while c

 Top of message 
>>--> 11-06-02  09:06  S.SHEPPARD (SHS)Safe to backup DB while c

I have never heard of or seen any restrictions or problems with backing
up the DB while backups are running.

Sam Sheppard
San Diego Data Processing Corp.
(858)-581-9668
---`


 Top of message 
>>--> 11-06-02  07:48  ..NETMAIL (001) Safe to backup DB while c
Date: Wed, 6 Nov 2002 10:40:17 -0500
From: "Matt Simpson" <[EMAIL PROTECTED]>
Subject: Safe to backup DB while client backups running?
To: [EMAIL PROTECTED]
_Top_of_Message_

We're trying, usually in vain, to get our TSM environment under
control. We have daily scripts that perform backups of our database,
one onsite and one offsite.  The scripts check to see if any backup
sessions are active, and if they are, they reschedule themselves to
try again in 10 minutes.

It's getting more and more difficult to find any time during the day
when backups aren't running. We keep spreading backup schedules out
to try to get stuff to run successfully.  Our offsite tapes are
supposed to be sent offsite by 8 AM, but sometimes they're not
getting completed until mid-afternoon.

Ideally, I know it would be safer to backup the database when it's
not being updated by backup processes. But it seems like there's no
way we're going to be able to reach Nirvana.  If I want my DB backups
to run at any kind of predictable time, it looks like we're going to
have to allow them to run while client backups are running.  Am I
setting myself up for major trouble if I do that?

Actually, I'm not 100% sure it's not already happening.  I know we
don't start the database backup if any client backup sessions are
active.  But I'm not sure if there's anything in place that prevents
a client backup from starting after the DB backup starts.  If that's
true, are we just wasting time spinning our wheels waiting for a
window when no backups are running to start the DB backup?
--


Matt Simpson --  OS/390 Support
219 McVey Hall  -- (859) 257-2900 x300
University Of Kentucky, Lexington, KY 40506

mainframe --   An obsolete device still used by thousands of obsolete
companies serving billions of obsolete customers and making huge obsolete
profits for their obsolete shareholders.  And this year's run twice as
fast
as last year's.

---`