When debugging include-exclude specs, don’t look first at your client options
file: do ‘dsmc q inclexcl’, which will both evaluate the specs and look for
server-side contributors and intrinsic OS contributors - which may have
interactions with what you coded on the client. And always issue that
From my experience, this likely has nothing to do with the TSM server, but
rather either configuration issues with the client or permissions on the
dsmerror.log, or its location.
Start with a simple command like ‘dsmc q fi’, as an ordinary user and then
root, to see if there are issues with dsme
The simple answer is that you have to provide scratch tapes, else your TSM
system will be crippled. Look around your system. Examine the Activity Log for
the past week. Find out what has been happening to your scratch tapes. Look
particularly for lots of tape volumes in Filling status and in Rea
The common services/daemons management arrangement in Linux is a
start/stop/reset/status script in /etc/init.d/ for each service, kicked off by
trigger specs in /etc/rc.d/, which in turn is managed via the chkconfig
command, conventionally via corresponding chkconfig numbers coded near the top
Ruth - Go into the Mac’s Utilities folder and run the Console utility, which
likely will reveal logged indications of problems encountered. Having that
active when you try to start the process should show immediate messages.
Richard Sims, Boston University
SELECT * FROM SYSCAT.COLUMNS
at least in TSM 5.
Richard Sims
A bad mount can cause a ‘df’ command to hang, and general file systems list
processing via system calls to have problems.
Make sure that your computer systems have monitoring to catch these things, to
prevent general functionality problems.
On Sep 12, 2014, at 3:11 PM, Jeanne Bruno wrote:
> Do
There can be other causes of this message, such as preemption by higher
priority tasks (particularly restores, retrieves, and recalls). See what the
server Activity Log contains at the time that the client encountered its error,
and examine the log prior to that for resource consumption (drives,
On Jul 2, 2014, at 3:34 AM, Robert Ouzen wrote:
> INCLUDE “Volumes/Drobo/Movies/…/*” mgsign
Robert -
One thing I can see is that the above line should instead be:
INCLUDE “/Volumes/Drobo/Movies/…/*” mgsign
You may be able to use command ‘dsmc query inclexcl’ to check the contents.
Ric
On Apr 16, 2014, at 12:15 PM, Nick Laflamme wrote:
> I'll bet today's lunch money that what's going on is that these aren't
> scratch volumes; they're volumes that were assigned to the pool with a DEF
> VOLUME command.
But then the volumes would not show as Scratch Volume?: Yes
Leonard needs
By extension, a logical question: Has Expiration been running? (In a neglected
TSM server, it might not be, and getting that going could free a bunch of
space.)
If Expiration has been running, a recourse is to go looking for obsolete nodes
and filespaces to dispose of.
If your system has USB, y
Brian -
The most common way to achieve your goal is to do an UPDate STGpool, with
ACCess=READOnly.
Richard Sims
Standard procedure in a TSM implementation is to have a copy storage pool as a
safety, which allows a defective primary storage pool tape’s contents to be
recovered, as via Restore Volume.
Ricnard Sims
The purpose of sending tapes (or, more generally, data) offsite is as part of a
disater recovery regimen. IBM Redbooks provide a wealth of practical guidance,
where the TSM DR redbook at http://www.redbooks.ibm.com/abstracts/sg246844.html
is and older one, but worth reviewing. (Search on “TSM” a
Standard procedure is to repeat the Move Data on the possibly bad tape such
that the next attempt will use a different tape drive, which may be able to get
a few more entries off the tape. Repeat further within reasonable bounds.
Thereafter you can perform a Restore Volume (assuming you have a C
Commonly, HSM archives are backed up as well. If that’s the case, restoral
would be the much less painful method.
Richard Sims, still at Boston University
Having abundant physical memory in your system doesn't mean that your server
process will get to use it. Operating systems can impose per-process limits,
according to configuration values. See
http://pic.dhe.ibm.com/infocenter/tsminfo/v6r3/index.jsp?topic=%2Fcom.ibm.itsm.srv.install.doc%2Ft_sr
Do SELECT * FROM SYSCAT.TABLES WHERE TABNAME='___'
on a given table name to see what columns it has.
(Table schema can change over time.)
Richard Sims
On Dec 18, 2012, at 11:32 AM, Jeannie Bruno wrote:
> Is there a 'current_day' column on the status table? just curious, it's
> giving me
It seems that some process in your shop managed to give the tape an internal
label which differs from its barcode label, possibly by a mis-labeling episode,
or some kind of physical tape copying.
I would do a Move Data to get the contents off that tape and onto a legitimate
cartridge, to get it
First, make sure expiration is running regularly, and run an Expire Inventory
to try to clear some more space now.
Look for any filespaces not receiving new data recently, and check with the
owner to see if obsolete, where junkers can be deleted.
Given your limited capacity, you should turn collo
20 matches
Mail list logo
