_
Ian Smith
SAN/TSM Specialist
_
This email (including any attachments to it) is confidential, legally
privileged, subject to copyright and is sent for the personal attention of the
intended recipient
Hello
I have an old copy storage pool that I want to delete.
I migrated all the primary storage pool data to new primary storage
pools, and backed this up to new copy pools. This was part of a
LTO2-LTO3 migration.
Is there an easy way to tell if all the data in the old copypool resides
in the
Hello
I have a problem with the dsmj client on Solaris 8.
TSM Client version 5.3.3.0, server version 5.2.7.1.
When ever the backup or restore button is pressed on the dsmj interface
an internal server error is detected. There is no obvious errors in the
log and has only stated occurring since
http://www-1.ibm.com/support/docview.wss?uid=swg21197242
You need to touch the error logs that you define in the sysfile. Then
ensure that the oracle user can write to these files.
I usually put them in a user accessible log directory and precreate the
log so the Oracle user can then write to
When I do a select on the events table to get status of scheduled events
into the past, the table suggests nodes missed their schedule even
before they were associated with it!
I need help here, as my statistics for missed backups look terrible!!!
Does it assume a node was associated for ever,
When I do a select on the events table to get status of scheduled
events into the past, the table suggests nodes missed their schedule
even before they were associated with it!
I need help here, as my statistics for missed backups look terrible!!!
Does it assume a node was associated for
it to think the schedule was
active prior to the actual created date.
Adrian Compton
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
Smith, I (Ian)
Sent: 13 November 2006 11:41 AM
To: ADSM-L@vm.marist.edu
Subject: [ADSM-L]
When I do a select
Richard
Does this make sense:
That when a server is associated to a schedule, the events table
populates an entry for the server back to every scheduled time until the
start_time parameter. Hence all the uncertains. I expect that it then
attempts to poulate the returns be cross referencing the
] On Behalf Of
Richard Sims
Sent: 13 November 2006 12:59
To: ADSM-L@VM.MARIST.EDU
Subject: Re: [ADSM-L] Events table
On Nov 13, 2006, at 7:27 AM, Smith, I (Ian) wrote:
Richard
Does this make sense:
That when a server is associated to a schedule, the events table
populates an entry for the server
I have a Windows host that was moved between domains.
It is still trying to backup a directory into an old management class.
There is no statement in the option file or in any global option set.
The problem is the files in the directory are going to the new default
management class- 37 days and
It may be worth turning accounting on, and checking the accounting log
to see where the delay actually resides. Then focusing your attention on
certain components of the infrastructure.
_
Ian Smith
SAN/TSM Specialist
-Original Message-
From: ADSM: Dist
The license file is part of the base code package. If you download a
maintenance level it is never included.
So download the base code and when you actually do the install it will
install the agent.lic file.
_
Ian Smith
SAN/TSM Specialist
-Original
Hello
I'm having a bit of trouble understanding the TSM Central Scheduler API
return codes on standard incremental backup. I just want to clarify when
a schedule is 'failed'. I used to think this was when files changed and
couldn't be backed up for there was issues backing up the registry
objects
Run the tdpoconf passwd utility to create the file.
_
Ian Smith
SAN/TSM Specialist
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
Lawrence Clark
Sent: 23 October 2006 15:56
To: ADSM-L@VM.MARIST.EDU
Subject: Re:
I also see there is a licence error. The agent.lic file is required for
the software to run, or you will get failures during the actual backup
operations.
_
Ian Smith
SAN/TSM Specialist
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL
All options should be placed under the servername stanza in the dsm.sys
file. The opt file should just hold the servername. There are a few
options like the symbolic link etc, but all client options should be in
the sys file.
So take out all the lines from the opt file other than the servername
I have inherited some old Windows 2000 servers to backup and have
installed TSM 5.3.3 client on for standard incremental backup. The TSM
server is 5.2.7.1.
I keep getting the following errors during the backup:
The domain statement is all-local so I thought it would do a
systemobjects whereas
If a node is associated with multiple schedules, will only the first one
(chronologically) be reported in the dsm sched log?
_
Ian Smith
SAN/TSM Specialist
_
This email (including any attachments to it)
Hello
Do people use the CAD to control the scheduler? Is this a good idea,
what happens if the CAD fails?
Is this best practice?
_
Ian Smith
SAN/TSM Specialist
_
This email (including any attachments to
depending on version of tsm and the os.
-Original Message-
From: Smith, I (Ian) [EMAIL PROTECTED]
To: ADSM-L@VM.MARIST.EDU ADSM-L@VM.MARIST.EDU
Sent: 10/13/06 6:31 AM
Subject: CAD Control Client scheduler
Hello
Do people use the CAD to control the scheduler? Is this a good idea,
what
Hi
I am trying to find out all the successful backups for a monthly node -
going back for years.
I am using the following select:
select node_name,hl_name,ll_name,backup_date from backups where
node_name='NODE_MONTHLY' order by backup_date
Is there anyway to get a DISTINCT on the backup date?
,
node_name,hl_name, ll_name from backups
where node_name='NODE_MONTHLY'
order by Backup_date
Leigh
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
Smith, I (Ian)
Sent: 11 October 2006 10:39
To: ADSM-L@VM.MARIST.EDU
Subject: [SPAM: 4.700
60 days for act and summary logs...
Ian Smith
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
David E Ehresman
Sent: 11 October 2006 13:45
To: ADSM-L@VM.MARIST.EDU
Subject: Re: [ADSM-L] SQL Query for monthly backups
I am trying to find out
TSM Client 5.2.4.0 on Windows 2003.
I am doing a point in time restore from the GUI, I looking to the
directory and there are a number of active and inactive directories
displayed.
I select the directory structure and hit restore.
Then in the restore area, a directory appears
Try the following setting:
Set sqldisplaymode wide
_
Ian Smith
SAN/TSM Specialist
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
Joni Moyer
Sent: 04 October 2006 13:27
To: ADSM-L@VM.MARIST.EDU
Subject: [ADSM-L]
By setting the DRMDBBACKUPEXPIREDAYS, allows you to automatically
recycle the TSM DB tapes. This is usually set to something like a 5 day
cycle, based on how realsitic it is that you would want to recover your
TSM environment back to. This means there will always be a critical
volume of tapes
For protection ensure that the ROLL FORWARD mode is on and there is
enough log space to take account for load during the period between TSM
DB Backups. This can be up to 13.2GB. Also set the Database backup
trigger, and ensure there is always scratch tapes. Put the DB and Log
volumes on
Hi
I have a problem on Linux -OS: Red Hat release 3 update 8. The TSM
Client is 5.2.4.0. Any DSMC process uses lots of CPU space.
When we do an strace on the process it is making no system calls.
However it is sitting at 25% CPU- 1 whole CPU.
A manual dsmc has the same issue, and takes ages to
You have migration delay set to 1 day. However migration continue is set
to yes.
It may be that the server is migrating down automatically files that
have not satisfied the migration delay time?
I haven't used this much as I tend to keep migration day at 0 days.
_
Oracle Enterprise Edition can parallelise the backup- i.e there are
multiple channels opened to the SBT_TAPE.
This is defined in the backup script or repository. Usually as a rule of
thumb, limited by the number of CPUs on the server.
MAXNUMMP should match the number of channels allocated by
Hi does anyone know if there is there a 5.3 version of TDP for SQL
planned for release?
It seems on the FTP site the latest version is 5.2.1.4.
_
Ian Smith
SAN/TSM Specialist
Hitachi Data Systems Certified Professional
IBM Tivoli Storage Manager Certified Consultant
We are standardising on 5.3, we have a number of SQL TDP agents on
Windows/Intel- It seems the only product that hasn't been developed on
to 5.3 is this package. Thought I was maybe looking in the wrong place!
_
Ian Smith
SAN/TSM Specialist
Hitachi Data Systems
You need to get Java installed on the machine. You can pull this down
from the Sun website.
_
Ian Smith
SAN/TSM Specialist
Hitachi Data Systems Certified Professional
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
Does this file physically exist:
'/opt/tivoli/tsm/client/oracle/bin/en_AU/tdpo.cat'.
It looks like an Australian English message file??? Were the required language
packs installed?
_
Ian Smith
SAN/TSM Specialist
Hitachi Data Systems Certified Professional
IBM
Hi
Does anyone have experience configuring the TSM clients to backup zones
in Solaris and specifically ZFS.
I am currently deploying but having some issues getting the client to
pick up the zone filesystems. I am running the client from the global
zone. A standard inc does not backup any of the
What was happening at the time? Check earlier in the act log...
i.e. session 23037 and process 659?
_
Ian Smith
SAN/TSM Specialist
Hitachi Data Systems Certified Professional
IBM Tivoli Storage Manager Certified Consultant
-Original Message-
From: ADSM:
]
Smith, I (Ian) [EMAIL PROTECTED] Sent by: ADSM: Dist Stor
Manager ADSM-L@VM.MARIST.EDU
08/29/2006 07:31 AM
Please respond to
ADSM: Dist Stor Manager ADSM-L@VM.MARIST.EDU
To
ADSM-L@VM.MARIST.EDU
cc
Subject
Re: HELP: TSM has recycled itself...
What was happening
TSM 53 has introduced a new option on the checkin command:
WAITTime
Specifies the number of minutes that the server will wait for you to
reply or respond to a request. Specify a value in the range 0-. If
you want to be prompted by the server, specify a wait time greater than
zero. The
This also sounds like the behaviour when the IO station is full
_
Ian Smith
SAN/TSM Specialist
Hitachi Data Systems Certified Professional
IBM Tivoli Storage Manager Certified Consultant
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL
Ensure the PCI slots do not share internal BUS bandwidth. You may end up
with issues there- I moved to dual port PCIe cards for the internal
bandwidth to run many LTO3 drives. Is the server really only PCI, or are
they PCI-X slots?
There is no issues running LTO2 and LTO3 on the same HBA.
I have
It is likely new machine will see the drives as different SCSI
addresses, so be sure to delete and recreate the paths on the TSM
server.
_
Ian Smith
SAN/TSM Specialist
Hitachi Data Systems Certified Professional
IBM Tivoli Storage Manager Certified Consultant
Hello
I am currently running a number of Windows 2003 based TSM 5.2.7.1
servers. They are relatively stable, however I am looking to upgrade to
5.3.
I have had a 5.3.2.0 server running for about 10 months, with minimal
load- I have seen no problems with this instance. I look at the fixes in
Are the drives being used?
What is the mount limit on the device class?
Are the paths online?
_
Ian Smith
SAN/TSM Specialist
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
Bell, Charles (Chip)
Sent: 17 July 2006
In TSM when it labels the tape it gives the volume a media type. See q
libv f=d. This is determined at the hardware level.
This defines to TSM the tape generation, i.e. 394=LTO2, 417=LTO3.
From the q dr f=d, you can see what the drive hardware type can read and
write.
To configure the
You're right, I think it defines the media type from the label...
_
Ian Smith
SAN/TSM Specialist
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
Richard van Denzel
Sent: 10 July 2006 12:22
To: ADSM-L@VM.MARIST.EDU
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
Smith, I (Ian)
Sent: Thursday, July 06, 2006 4:14 PM
To: ADSM-L@VM.MARIST.EDU
Subject: Re: [ADSM-L] ANR8779E: Unable to open drive mt6.0.0.1 error
no=170
The CPF and DPF enable/disable is just a registry change
,
Regards,
Zareyna Salim
Technical Consultant
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
Smith, I (Ian)
Sent: Wednesday, June 28, 2006 6:20 PM
To: ADSM-L@VM.MARIST.EDU
Subject: Re: [ADSM-L] ANR8779E: Unable to open drive mt6.0.0.1 error
Hi
What ultrium driver versions are you using?
We had this problem after upgrading the ultrium device drivers to
6.1.1.2. We had to roll back to 6.0.8.2.
Ensure that the CPF and DPF functions are disabled in the registry by
using the scripts if you are not using this function.
[mailto:[EMAIL PROTECTED] On Behalf Of
Smith, I (Ian)
Sent: Wednesday, June 28, 2006 6:02 PM
To: ADSM-L@VM.MARIST.EDU
Subject: Re: [ADSM-L] ANR8779E: Unable to open drive mt6.0.0.1 error
no=170
Hi
What ultrium driver versions are you using?
We had this problem after upgrading the ultrium device
Versions Data Exists: NOLIMIT
Versions Data Deleted: NOLIMIT
Retain Extra Versions: 45
Retain Only Version: 90
_
Ian Smith
SAN/TSM Specialist
Hitachi Data Systems Certified Professional
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL
LTO2 drives are back compatible with LTO1 media. Ensure that the atape
driver version is at the required level to support gen II.
Put all the tapes physically to read only, to ensure there is no
potential way of writing over the DR data.
_
Ian Smith
SAN/TSM
Can you check the network settings on the HMCH1143 server. If this is
running half duplex, or particularly slow then the sessions will lock
parts of the target disk storage pool and hold them. This may be
directly conflicting with the migration process. As previously stated,
this could also be a
If you set up the snmp alerting from the library- it will send a trap
when the bulk IO loader is full.
_
Ian Smith
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
Bill Kelly
Sent: 04 May 2006 13:34
To:
Show logpinned
This will display any potential sessions pinning the log.
Kick off an incremental database backup asap.
_
Ian Smith
SAN/TSM Specialist
IT Infrastructure
Rabobank International
Thames Court, One Queenhithe
London EC4V 3RL
t: +44 (0)20 7809 3046
f: +44
]
Smith, I (Ian)
[EMAIL PROTECTED]
NK.COM
To
Sent by: ADSM: ADSM-L@VM.MARIST.EDU
Dist Stor
cc
Manager
[EMAIL PROTECTED]
Subject
.EDU Re: HELP
What is the OS?
Could Journaling or some form of, be helpful here?
_
Ian Smith
SAN/TSM Specialist
IT Infrastructure
Rabobank International
Thames Court, One Queenhithe
London EC4V 3RL
t: +44 (0)20 7809 3046
f: +44 (0)20 7809 3599
m: +44 (0)7843 689914
Mailto: [EMAIL
I inherited a TSM environment all running on Windows. So I was stuck
with the OS.
I have deployed multiple HP ML570 servers, twin xeon, 3GB memory. Big
PCIe buses at the back end. Have 7 LTO2 and 7 LTO3 and have seen the
server pretty much pushing them all full bore. Throughput is very good,
I
Check the actlog- there looks to be no scratch tapes! Or a problem mounting
volumes in the library.
_
Ian Smith
SAN/TSM Specialist
IT Infrastructure
Rabobank International
Thames Court, One Queenhithe
London EC4V 3RL
t: +44 (0)20 7809 3046
f: +44 (0)20 7809 3599
m:
Paths go offline due to scsi errors on the line. The drive doesn't
actually report an error so the drive entity stays online.
Paths can go offline for a variety of reasons above and beyond drive
hardware failure.
You can just attempt to put the path back online if it is a temporary
line fault.
Virtual volumes usually exist on another server- so the actual device
class is defined as a device type of server.
Storage pools are then bound to this device class- that exist on the
server defined.
_
Ian Smith
SAN/TSM Specialist
-Original Message-
From:
Hello
Has anyone seen the likes of the following errors on backup storage pool
processing via virtual volumes?
04/12/2006 05:39:15 ANRD bfcputil.c(655): ThreadId40 Error
testing
whether bitfile must be backed up. Callchain:
1050FA49
Hello
I have a situation with two TSM servers which offsite to each other
using virtual; volumes, where the reconcile preview=yes comes up with a
number of volumes that aren't quite right.
What does the actual reconcile do?- is it destructive in any way, and
how long can these processes take?
Did you make any changes to mountlimit or format on the device classes? When
you recreated the drives and paths after the migration did you bind them to the
correct device class/device type?
After the migration did you audit the library?
_
Ian Smith
SAN/TSM
|-+---|
| Smith, I (Ian) | |
| [EMAIL PROTECTED] | |
| Sent by: ADSM: Dist Stor | To|
| Manager | ADSM-L@VM.MARIST.EDU |
| ADSM-L
Q dr f=d
This will tell you read and write format of each drive in TSM.
_
Ian Smith
SAN/TSM Specialist
IT Infrastructure
Rabobank International
Thames Court, One Queenhithe
London EC4V 3RL
t: +44 (0)20 7809 3046
f: +44 (0)20 7809 3599
m: +44 (0)7843 689914
Mailto:
I know that you can use the managedservices option in the opt file and
specify:
managedservices schedule webclient
Meaning the CAD manages the client and the scheduler.
_
Ian Smith
SAN/TSM Specialist
-Original Message-
From: ADSM: Dist Stor Manager
for the BA functionality.
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
Smith, I (Ian)
Sent: Friday, March 03, 2006 4:00 AM
To: ADSM-L@VM.MARIST.EDU
Subject: Re: tdp for mail/EXCH scheduler question !
I know that you can use the managedservices
I see the exact same thing on our TSM servers.
After speaking to IBM they state that the hang is down to a component of
the database not being able to respond during what ever process is
running. During the start of a reclaim process the IO is heavy on the
database and the TSM instance queues
The installable base code is available on the FTP site- but you need the
agent.lic file for the Oracle TDP from the original CD.
I'm looking for a version that runs on Solaris running on x86!
_
Ian Smith
SAN/TSM Specialist
-Original Message-
From: ADSM:
Hello
Other than offsite reclamation, is there any way to easily consolidate
offsite virtual volumes?
Currently I have a server which is electronically vaulting to a remote
TSM server. There are in the region of 3000 volumes in the copy pool,
some 5 MB and smaller in size.
Backup storage pool
Move data?
Regards,
Karel
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
Smith, I (Ian)
Sent: dinsdag 28 februari 2006 13:15
To: ADSM-L@VM.MARIST.EDU
Subject: Offsite Virtual Volumes
Hello
Other than offsite reclamation, is there any way
] On Behalf Of
Smith, I (Ian)
Sent: dinsdag 28 februari 2006 13:15
To: ADSM-L@VM.MARIST.EDU
Subject: Offsite Virtual Volumes
Hello
Other than offsite reclamation, is there any way to easily consolidate
offsite virtual volumes?
Currently I have a server which is electronically vaulting to a remote
TSM
Hello
I have a Windows 2003 TSM server with LTO2 and 3 tape drives with
Ultrium driver 6.0.8.2, everytime it reboots it loses the ultrium device
drivers on some of the LTO2 and 3 tape drives. The actual SCSI ids
change i.e. lb0.1.0.9 then after reboot it becomes lb0.1.0.10. Dioes
anyone know why
Hi
I have been having database performance issues possibly related to
unexpired objects in the database. The problem manifests itself with all
the client sessions during the backup window grinding to a halt. There
are 7 pages of locks on the database and nothing is moving. After
running
What is the format setting on the device class?
If you do a q libv f=d check the media type statement. If you are
rtunning mixed media this may be the cause of the issue.
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
Jim Skinner
Sent: 27 January
Try the /etc/limits file which defines the file size for users etc.
Increase these limits.
_
Ian Smith
SAN/TSM Specialist
IT Infrastructure
Rabobank International
Thames Court, One Queenhithe
London EC4V 3RL
t: +44 (0)20 7809 3046
f: +44 (0)20 7809 3599
m: +44
In AIX 5.x I think the file is in /etc/security/limits
_
Ian Smith
SAN/TSM Specialist
IT Infrastructure
Rabobank International
Thames Court, One Queenhithe
London EC4V 3RL
t: +44 (0)20 7809 3046
f: +44 (0)20 7809 3599
m: +44 (0)7843 689914
Mailto: [EMAIL PROTECTED]
We use a dual core xeon 1.2 GHz HP box. 2 GB of RAM.
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
Richard Mochnaczewski
Sent: 20 January 2006 15:39
To: ADSM-L@VM.MARIST.EDU
Subject: [ADSM-L] Requirements for Administration Center server for
In a cached disk subsystem, the write overhead of RAID 5 can be negated.
Hence the reason if a cache battery dies, write through cache mode is
enabled, as cache may not be able to be destaged in the event of a power
failure- this will have a massive effect on write wait time on the host.
79 matches
Mail list logo
