Hi all,
Anyone could help me with a simple statement to look for the exact
successful elapsed time for 23/2/2005. I tried out with the cast fn but it
returns no matching result.
SCHEDULE_NAME: KULBPM01_SYSOBJ
START_TIME: 2006-02-23 04:34:49.00
END_TIME: 2006-02-23 04:37:20.00
EL
Hi,
yes there are two lines for that specific filespace. One for primary and
one for the copy stg pool.
The strange thing is, that q occ shows me that there should be 441.000
files in the primary stgpool and q backup (-inact -subdir=yes) shows
only 190.000 files.
Am Mittwoch, den 22.02.2006
I have seen a 1 gigabit NIC run at speeds as fast as 450 megabits, for a
sustained period of time, on an AIX TSM server. 450 megaBITs/sec is 162
gigaBYTEs/hour. In our 14-hour backup window, we typically back up 1TB
of data.
So your NIC is probably not the problem. Look first at your network -
all
Is this client writing to disk or to tape? If tape, what kind?
Orville L. Lantto
Glasshouse Technologies, Inc.
From: ADSM: Dist Stor Manager on behalf of Troy Frank
Sent: Wed 2/22/2006 4:08 PM
To: ADSM-L@VM.MARIST.EDU
Subject: Re: [ADSM-L] tsm
I can't speak
>> On Tue, 21 Feb 2006 09:28:30 -1000, Aaron Becar <[EMAIL PROTECTED]> said:
> Just a quick question. Has anyone every had TSM disk volumes just
> disappear on them? For example if you do a listing of the files where
> the volumes should be there is nothing there? Thanks!
Hmm. When an admin
I can't speak to all brands/models since I haven't seen them all.
However, all the brands/models of 1GB NIC that I have seen only work if
set to auto. I've never seen an instance where forcing 1gb/full duplex
was even an option. Of course, this also means that the switchport has
to be set to auto
Chances are it's not a TSM issue at this point but a network issue.
Usually speeds this slow on a GB infrastructure indicate that
"auto-negotiation" is not working. ( never does ).
Make sure all NIC's and switch ports are locked down to 1GB at full duplex.
Gerald Michalak
TSM - Certified V5 Admi
>From my testing dsmcad doesn't complain about the '-server=' option, but
doesn't use it
Matthew Glanville
You are asking a hardware question. X86 will be different from RISC 6000
and I suspect OS/390 is really fast.
What hardware?
Are the HBAs sharing a bus?
Are the NICs sharing a bus?
What kind of bus are the cards in?
What CPU? How fast?
Is TSM running solo on the server?
Are you going to disk? Tape?
Hello,
First of all, I know almost nothing about Tivoli so bear with me :)
Now, do you have any idea what's the maximum throughput between a TSM client
and server, during a backup? Let's assume that we have a GB infrastructure
connecting them. I'm experiencing really low speeds (3-4MBps) and
unfo
Query OCCupancy reports primary and copy storage pool populations.
The client is obviously unaware of copy storage pools, which are
exclusively the server's interest.
The output of the Query OCCupancy *should* include one line for each
storage pool instance. Something may be awry if that is not t
Hi,
I have a strange phenonemon on our TSM Server (5.3.2.2, z/Linux).
I have a node ("oldnode") with old data I have imported some time ago
from our ADSM Server (3.x z/VM).
"q occ oldnode /filespace1" shows me the following:
number of files: 441.029
phyical space occupied: 1.089.444 MB
I generat
Adam,
This is only a guess, but maybe the move nodedata command is transferring
compressed data. If the data isn't being decompressed upon read and
re-compressed upon write then that would explain what appears to be lower
tape utilization. See what the query nodedata command shows before and
afte
If you have the luxury of re-creating the problem, try again and put
TESTFLAG DISABLENQR in the dsm.opt file (search on DISABLENQR in the
archives to see what people have said about this before).
That turns off NQR and uses CLASSIC restore. It is known that sometimes
CLASSIC will outperform NQR.
What version TSM?
What OS?
Met vriendelijke groet, With kind regards,
Richard van Denzel.
-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED] On Behalf Of
Thomas Rupp
Sent: woensdag 22 februari 2006 16:26
To: ADSM-L@VM.MARIST.EDU
Subject: AW: [ADSM-L] (Too) long
Some additional information:
Server: IBM eserver xSeries 346, Intel Xeon CPU, 3.2GHz
Database: 20GB on 4 Volumes (EMC AX100)
Storagepool: 1317GB on 13 Volumes (EMC AX100)
No tape activity is involved as all data (40MB) is restored from disk.
So TSM seems to spend most of the time scanning throug
Tomas,
This can be normal behaviour. It all depends on which media the files
are stored.
For instance I've done a restore of an 800MB Unix file system which was
scattered over 12 LTO-1 tapes and it took 45 min.
The way to improve performance for these kind of restores is using
colocation.
Met v
Hi *SMers,
I had to restore a subdirectory (2000 files) for a fileserver with a lot
of files
(9.5 million files according to QU OCC). According to the docu it was a
no-query-restore (NQR) so all the work was done at the server.
I know that a NQR can take a lot of time, but I think 4:43 hours is to
The dsmcad process can take a -se option, but sure about the
-virtual node option though. I prefer to start it with the -optfile
option so that if I have different domains (or filesystems) for each
"instance" running on the client, each of them will pick up only the
right filesystems.
My
We have set up a new primary storage pool, using lto2 in a 3584 library,
with collocation by group. TSM5.3.2.1 on w2k3.
We are moving nodedata by collocation group from the original pool to the
new one in order to collocate our existing data.
In the original pool we see estimated tape capacities of
On Feb 22, 2006, at 7:14 AM, goc wrote:
i'm aware that dsmc sched can work with -se or -virtualnodename
parameter
but does dsmcad works this way also ?
Goran - The method for having multiple independent schedules with
dsmcad is documented in Appendix A of the Unix client manual.
(This
Goran,
Unless they've changed things recently, dsmcad *only* works with the
parameter '-optfile='.
Don't believe the older documentation either (see apar IC32030).
To make life interesting, on Solaris at least, dsmcad will accept
*anything* supplied on the command line, ignore it, and then sit t
hi all,
just a quick one ...
i'm aware that dsmc sched can work with -se or -virtualnodename parameter
but does dsmcad works this way also ?
i belive not coz it invoces the dsmc shed command, right ?
that would be the anwser to my problems, i'm having big trouble setting
mulitple schedules on se
23 matches
Mail list logo
