d involved in the procedure. For the procedure itself
you should be following the Admin Guide instructions, which guide you
through the multiple commands involved in the procedure. The section
on Mirroring the Database and Recovery Log goes through the steps,
including dsmfmt.
Richard Sims, BU
, else you need to take appropriate action there as well.
Richard Sims, BU
started, and consider making them Unlimited if not already, or do
whatever can be done under your HP operating system to boost the
memory that your server is allowed, and restart it.
Perhaps someone else with an HP server has more perspective on this.
Richard Sims, BU
If the tsm deletes the data and keeps the object in the DB why?
Have you done an Expire Inventory since deleting the object?
Richard Sims, BU
Is there a list of supported Tape devices someplace that I could reference.
Go to http://www.tivoli.com/support/storage_mgr/adsercli.htm
click on Requirements and Supported Devices
Richard Sims, BU
what happens.
Richard Sims, BU
Hm, not for me... WinZip complains about 524 bytes missing.
Now for the classic question: Did you FTP it in Binary form?
(That's the standard problem in situations like these. And
no one reporting the problem has yet specified how they
performed the download.)
Richard Sims, BU
were able to backup these non-owned files. Now they get ANS1136E errors =
I'd first do a 'dsmc Query Access' from the owner of the files to see if
the SET Access is still in effect. This will give you more info if you
have to proceed, then, to contact Support.
Richard Sims, BU
. The
one prevailing condition is that once you advance your client level,
you are committed to it because of the advanced content of the new
backups.
Richard Sims, BU
advisable
to back up databases using an API-based utility which participates in the
database environment to back it up from the inside, and thus get a
consistent and restorable image. TSM provides Data Protection agents for
some databases, but for others you'll have to seek another source.
Richard Sims
to the proclamations. If you've
dealt with other vendors you'll probably agree that IBM is far and
away better than most other vendors in responsiveness and action. It's
just when they make such a display of intending to do something and don't
follow through that their credibility suffers with the customer base.
Richard Sims, BU
' or similar command? What do various
'mtlib' queries show?
And just what is evidenced in TSM "not seeing" the drives?
Richard Sims, BU
Shekhar - You're thrashing. What you need to do is sit down with the
Admin Guide and Concepts manual and get a handle on setting
up policies. If you don't get a firm grasp on that you'll just have
endless problems as you continue with TSM administration.
Richard Sims, BU
for your
future viewing.
Richard Sims, BU
ill govern
how much data moves in each server transaction. The performance of
your tape technology is also a factor.
Richard Sims, BU
Initially I put my include/exclude statements in the dsm.opt file
but when I run dsmc it complains that it's an invalid option.
From the Backup/Archive manual for TSM on AIX:
"Place this option in the include-exclude options file
or the client system options file (dsm.sys)."
Ri
thresholds.
Richard Sims, BU
ve a valid
backup copy group. This file will not be backed up.
The server policy definitions need to be checked out, according to this
message.
Richard Sims, BU
prevents interoperation.
Richard Sims, BU
ENCES AND OTHER RESOURCES" toward the bottom of the file,
and "SQL" therein. There are also various sample SELECT statements
sprinkled throughout the categorized file.
Richard Sims (just another ADSMer, and compulsive documenter), BU
if that instance of the file is already in the
backup storage pool - which would be a good thing, allowing you to recover it
to the primary pool. A RESTORE STGpool Preview=Yes would probably tell you
if this were so.
Richard Sims, BU
around in the file system looking for
something new to back up.
Richard Sims, BU
Redbook "Guide to Sharing and
Partitioning IBM Tape Library Data"
(SG24-4409)
Richard Sims, BU
or filesizes,
whichever is larger).
If this stuff were easy, anyone could do it, and then where would we be as
overpaid technicians?
Richard Sims, BU
be concerned about from the numbers: You say this was
a disk restoral; but some 9000 files were restored whereas the backup indicates
there were about 10x as many before.
Richard Sims, BU
/system02_ORASHSDA.dbf
Richard Sims, BU
er installation, you could cd \tcpip\bin and enter the
OS/2 command 'tcpipcfg' and click in the Network tab, which I
think will have that setting. I last did this years ago, and things
may have changed...
Richard Sims, BU
IBM_scsi_tape_IUG.ps and/or IBM_scsi_tape_IUG.pdf
IBM_scsi_tape_PROGREF.ps and/or IBM_scsi_tape_PROGREF.pdf
If you have Magstar tape drives or libraries, this is a must-have.
Richard Sims, BU
anywhere. Check the Messages manual appropriate to your
TSM level.
Richard Sims, BU
problems.
Richard Sims, BU
that it will still traverse that area looking for directories to back up.
Follow the manual recommendation and code EXCLUDE.DIR instead, to wholly
exclude the directory.
Richard Sims, BU
'lsof' to see what processes are listening on those RPC ports.
Richard Sims, BU
not think I had
changed anything, but went ahead let it save. That appeared to fix
the problem. Otherwise, run an AUDITDB to reconcile the database with
reality.
Richard Sims, BU
to
issues and will take action when documentation needs improvement.
Richard Sims, BU
on specific systems
experiencing the problem.
Richard Sims, BU
--- Response from "Ask Symantec service" ---
Perhaps this is a permissions issue. There are two things I would try:
- Have the NAV services log in under a local Admin account instead of
LocalSys
% is
used, it of course means that there is no problem occurring and so the
value at this time doesn't mean much. If instead it can be determined from
the abend dump what the memory utilization level was, that would be a
real indication if memory exhaustion was playing a role.
Richard Sims, BU
backup storage pool Unavailable
for a moment, attempt a restoral, unmark, and look in the Server Activity
Log for what volume it could not get.
Richard Sims, BU
it interactively, from
the command line, versus a schedule.
Richard Sims, BU
/etc/security/limits
definitions and make sure that root's memory utilization is not artificially
constrained.
Richard Sims, BU
ic of http://people.bu.edu/rbs/ADSM.funcdir .
Richard Sims, BU
tivoli.tsm.server.rte42.config[215]: 15732 Trace/BPT trap(coredump)
Gerald - You indicate that your /usr should not be filling, so that should
not be the cause. Another possibility is that swap space is not
active on your AIX system.
Richard Sims, BU
Products Family: A Practical Guide (SG24-4632).
Richard Sims, BU
into this situation.
If this is the case, recreate the old management class definition.
In your backup, you are using ADSM to back itself up. But if you lost the
file system that it's in, what would you use to restore it?
Richard Sims, BU
unto itself in this regard.
Richard Sims, BU
Richard, what is the difference between using "dsmc retrieve -fromnode"
and "dsmc retrieve -virtualnode -pa". Why do you prefer -virtualnode over
-fromnode?
Eliza - -FROMNode and -FROMOwner are part of the facility for users sharing
server-stored files, defined by filename via Set
If I deleted a tape of the adsm data base, I can restore it ?
You would have to restore your *SM database to a point where that
tape volume existed. Deleting filespace and volumes is thorough.
Richard Sims, BU
When all devices are in use all the other libary clients get
"ANR1401W Mount request denied" and the scheduled job fail.
Message 1401 is usually a summary message. Were there any detail messages
preceding it indicating the actual problem?
Is anybody else getting connection refused when trying to get to
www.ibm.redbooks.com ?
You mean www.redbooks.ibm.com. But, yes, it is denying access, still.
Richard Sims, BU
www.redbooks.ibm.com is finally back in service, now.
ould require up to 25% in additional
space". (page 590)
Most of us use the empirical approach of allocating a reasonable amount of
space to start. Then we monitor usage (which has to be done anyway) and
are prepared to add more db vols as necessary.
Richard Sims, BU
than the permissions settings of the original files.
Your instructions should say to use the -p option of the tar
command, to restore fields to their original modes, ignoring
the present umask. The tar command should thus look like:
tar -xpf IP22087.tar
thanks, Richard Sims, BU
anged over time. The workspace
usually needs to be contiguous, as well, and when things get tight, untoward
events can transpire. How TSM is actually servicing some of these queries
also plays a part. Your mileage may vary, things change, and all that.
Richard Sims, BU
manuals: http://www.tivoli.com/support/storage_mgr/pubs/admanual.htm
ADSM manuals: http://www.tivoli.com/support/storage_mgt/adsm/pubs/admanual.htm
Richard Sims, BU
Does anyone know where there is a more concise explanation of the fields in
the accounting record.
One place is http://people.bu.edu/rbs/ADSM.QuickFacts .
See section near bottom "ACCOUNTING RECORD FORMAT" and the
Communications Wait definition.
Richard Sims, BU
,
and in any case you have the AIX doc online at
www.rs6000.ibm.com/doc_link/en_US/a_doc_lib/aixgen/topnav/topnav.htm .
Errno 46 means Device not ready. Have a look at that drive.
Richard Sims, BU
to be at a current one.
Richard Sims, BU
I'm currently running TSM 3.7on Solaris on an E10k. We're trying to used Commethod
Sharedmem.
...
Can anyone tell me how much sharedmem TSM tries to get by default and how I can
modify this setting?
I haven't seen documentation describing TSM's utilization of shared memory.
But you may want to
I need to restore last month's backup of our *SM database temporarily
Krish - If you have sufficient disk space (unlikely) it would be ideal to
leave your current server config untouched and restore the temp
server config to other space.
Otherwise, you have to take over the existing
implies, you need to log on to it from the server console.
Had you ever before logged onto it from an admin client?
Richard Sims, BU
like to learn of any platforms where this is not
the case. I should think it would have to be due to some peculiarity in the
operating system. Conversely, one would not expect TSM servers to be
developed for platforms which were not fully versatile.
Richard Sims, BU
So, if you instance is installed outside of the default, try adding a
"cd" to the directory where your dsmserv.dsk file is located.
Good general advice. Keep in mind also that the rc.adsmserv itself does
a 'cd' to what it thinks is the server directory, so an alternate,
modified version should
. See the latter
at http://www.tivoli.com/support/storage_mgr/clients.html#linux,
referring to Migration Information in the Readme.
Richard Sims, BU
09/12/2000 20:58:40 ANS1228E Sending of object 'C:' failed
09/12/2000 20:58:45 ANS1435E An Error Occurred saving the Key.
09/12/2000 20:58:45 ANS1428E Registry Backup function failed.
Bill - This could be due to lack of space on the system drive to
hold the staged Registry files.
Richard
, whereas
the NT server manual defines it as taking the operand as "FILE(LicenseFileName)".
thanks, Richard Sims, BU
I am attempting to connect an RS/6000 SP system to a 3494 tape library.
I think I have followed the manuals correclty, but cannot access the
library using the mtlib commands. I keep getting the message that says
the library is offline to the host.
Um, is the library offline to the host? Go to
s hampered. Just another thing to look at.
Richard Sims, BU
, and would cost far less than your time in
doing API work. Alternately, is there any way to share the disk with
another system (NFS, etc.) such that it could be backed up from there?
Richard Sims, BU
Above and beyond that, is there information out in the big wide world as to what
factors to consider and what parameters to check (end-to-end i.e. ADSM server,
network, and ADSM client) to maximize restore performance, especially in the
case of a large amount of data (10-100 Gb) all at once to a
.
Richard Sims, BU
ent sprayd
because it invites heavy, mystery network loads where the 'spray'
command is mis-used. At least it requires no files on either end.
Richard Sims, BU
1 to TAPE2 simply by
updating the definition of DISK1 to specify TAPE2 as its next storage pool.
You could cause a storage pool to be skipped by setting its MAXSize value
to a silly, small number.
Richard Sims, BU
may not exceed the TCP/IP limits currently in effect
for your operating system. The TSM client is programmed to check.
Code a value smaller in TSM, or boost your opsys limits, and the message
*should* not appear.
Richard Sims, BU
start-up. Note that each
# dismount may take a minute or so. RBS
/usr/bin/mtlib -l /dev/lmcp0 -d -f /dev/rmt1
/usr/bin/mtlib -l /dev/lmcp0 -d -f /dev/rmt2
/usr/bin/mtlib -l /dev/lmcp0 -d -f /dev/rmt3
/usr/bin/mtlib -l /dev/lmcp0 -d -f /dev/rmt4
Richard Sims, BU
There was a rumour that adsmpipe would be supplied and supported with V4.
I'd be surprised if it were: IBM has consistently said that ADSMPIPE is
obsolete and part of the past. What we really need, of course, is
long overdue enhancement of the capabilities of the command line clients.
Richard
documented is in the back of
"IBM SCSI Tape Drive, Medium Changer, and Library Device Drivers:
Installation and User's Guide" (GC35-0154)...
FTP to ftp.storsys.ibm.com, in /devdrvr directory.
You can also do 'mtlib -\?', but that doesn't show allowed combinations
of operands.
Richard Sims, BU
How do I find out the total storage used by a node
Do 'Audit License', then 'Query Auditoccupancy'.
Break down with 'Query Occupancy'.
Richard Sims, BU
it, and
for the spot restoral of individual files.
In summary, get those baseline numbers and use them to help isolate the
problem areas. Identifying problems is 90% of their solution. And stay a
huge fan of the product. :-)
Richard Sims, BU
and what can happen. If
nothing else, you'll have a lot more real confidence.
Richard Sims, BU
version of the product in transitioning from
IBM ADSM to Tivoli TSM development. I'm avoiding 3.7 - and very much hope that
4.x exhibits stabilized development.
Richard Sims, BU
reveal what's really happening.
Richard Sims, BU
. That is something they need to address, not
leave you to deal with. Solaris 2.5.1 is very old stuff now, and your site
needs to make decisions on either upgrading that system or isolating it (give
it a tape drive and let it do its own primitive backups).
Richard Sims, BU
Restore or Retrieve Files to
Another Workstation", which avoids all the authorization complexities.
Richard Sims, BU
body wants to be in the hellish situation of having to do a
dsmserv auditdb. In Unix systems that support it, create an /etc/rc.shutdown
that does a dsmadmc to perform a Halt, to shut TSM down nicely when the unix
system is shut down cleanly (which is not always the case, unfortunately).
Richar
et it that way?
I'd like to see Tivoli publish a study on the value of this procedure,
showing a typical database's performance before compaction, immediately
after, and then periodically thereafter as a normal amount of activity
causes refragmentation. Otherwise we're just speculating on its value.
Richard Sims, BU
in "interesting" situations,
Hopefully this won't occur, as we have a policy of testing all changes in
the lab before applying them to production.
Wow, what a terrific philosophy! If Tivoli did that, customers would be
able to perform an Import and it would work!
Richard Sims, BU
are AIX, at least.
Richard Sims, BU
We didn't set the EXPInterval, I think it is 0, am I right?
The Force is not that strong in this advisor, Phillip: you'll have
to do a Query OPTions in your server and tell me. According to the
server Admin Ref, the default is 24 (hours) these days.
Richard
Does someone know of a quick query to find out how much data was backed up
last nite?
as i am trying to forcast how much tapes are needed on a daily basis
Dwight's response about getting info from accounting records is the best,
detailed info source for backup, archive, and HSM usage. You
://www.storage.ibm.com/hardsoft/tape/3590/prod_data/3590perform.pdf
It's not overly exciting, but provides some relative values.
Richard Sims, BU
the last backup was done.
Another approach is to delve into the Archives and Backups tables,
though that's more voluminous.
Richard Sims, BU
When executing the sentence Query FIlespace F=D indeed sample the field
Last Backup Start Dates / Time .
But it shows it in white , without information this way for all the nodes .
It doesn't visualize anything that field .
Juan - The emptiness means that there is nothing to report; that is,
Being slightly pedantic, it actually means that there has been no
incremental backup performed. There could be backup data stored
for the node if a selective backup was used.
Thanks, Trevor - quite so. Thanks also to Wanda, who spotted the
API being used, which I missed.
Richard
hope things will improve.
Recall the recent postings of a 3.7 customer who found that Import does not
work in 3.7.
Richard Sims, BU
a client attempts to retrieve a file and
the server detects an error in the file copy in the primary storage pool,
the server marks the file as damaged. At the next attempt to access the
file, the server obtains the file from a copy storage pool.
Richard Sims, BU
Unavailable state to Readwrite, and do an Update Drive to
make it Online again. (I've been there.)
You should never simply say "Oh, well" and stop looking for a tape that has no
reason to be elsewhere. Bad idea, because it does not solve the problem.
Richard Sims, BU
are probably
too small for that user. Confer with your Unix people about this:
they should be able to advise.
Richard Sims, BU
. Your original library
definition had SCRATCHCATegory=401, with 402 implied for 3590s. Your
new definition conflicts with that.
Richard Sims, BU
of upgrading our 3590 drives within this
current server level, and I'd like to be sure.
thanks, Richard Sims, BU
s cleaning abilities.
Richard Sims, BU
. The current Exclude capability is rather limited.
Richard Sims, BU
1 - 100 of 4098 matches
Mail list logo
