Well, maybe I missed something, too. Keep in mind that when discussing
technical issues, the details are critical. It would be helpful to always
provide hard data to go along with the verbal description. As a rule, one
can only provide an answer that is as specific as the question. :-)
With that in mind, it would help to know the *exact* command you are
using, and the *exact* output you are seeing. Options file is helpful, as
is output from "dsmc query inclexcl".
Now the info I gave you will cause the client to traverse only those
directories that you specified. So how about trying it to confirm the
behavior? Open up a Windows OS prompt and do the following:
cd /d c:\
md "rh test1\dira"
md rhtest2
md rhtest3\datafiles
cd rh test1\dira
echo this is a test > testfile.out
copy testfile.out \rhtest2
copy testfile.out \rhtest3
copy testfile.out \rhtest3\datafiles
cd program files\tivoli\tsm\baclient
dsmc i "c:\rh test1\*" c:\rhtest2\ c:\rhtest3\datafiles\ -su=y
In the second-to-last command above you may need to switch to a different
drive and/or directory, depending on where the TSM client is installed on
your system. If you client *is* installed in C:\Program Files\Tivoli\TSM,
then you can just copy & paste this into the command line. Oh, I should
also mention that if your options file (dsm.opt) already has
includes/excludes in it, then depending on what is excluded, you may see
only some or no files backed up. For the purposes of this test, I
recommend a minimal options file, something like this:
COMMMETHOD TCPIP
TCPSERVERADDRESS your.tsm.server.address
NODENAME yournodename
PASSWORDACCESS GENERATE
You, of course, have to plug in the appropriate TSM server address and
node name.
Below is the output when I do this on my system. Note that the file in
c:\rhtest3 is not backed up because we said to back up
c:\rhtest3\datafiles. Also note that *only* the directories I specified
are traversed (I have no explicit include/exclude statements in my options
file). And just to be clear, my TSM client is installed in C:\TSM; your
install directory may differ.
If the principles I show here vary from what you are looking for, then it
would be helpful if you could provide a more detailed description of what
you are trying to accomplish, and what exactly you are seeing (including
actual sample output, like I show here).
C:\>c:
C:\>cd \
C:\>md "rh test1\dira"
C:\>md rhtest2
C:\>md rhtest3\datafiles
C:\>cd rh test1\dira
C:\rh test1\dira>echo this is a test > testfile.out
C:\rh test1\dira>copy testfile.out \rhtest2
1 file(s) copied.
C:\rh test1\dira>copy testfile.out \rhtest3
1 file(s) copied.
C:\rh test1\dira>copy testfile.out \rhtest3\datafiles
1 file(s) copied.
C:\rh test1\dira>cd \tsm\baclient
C:\TSM\baclient>dsmc i "c:\rh test1\*" c:\rhtest2\ c:\rhtest3\datafiles\
-su=y
IBM Tivoli Storage Manager
Command Line Backup/Archive Client Interface - Version 5, Release 2, Level
2.10
(c) Copyright by IBM Corporation and other(s) 1990, 2003. All Rights
Reserved.
Node Name: STORMAN
Session established with server STORMAN: Windows
Server Version 5, Release 2, Level 2.1
Server date/time: 05/13/2004 16:49:45 Last access: 05/13/2004 16:47:26
Incremental backup of volume 'c:\rh test1\*'
Incremental backup of volume 'c:\rhtest2\'
Incremental backup of volume 'c:\rhtest3\datafiles\'
Directory--> 0 \\storman\c$\rh test1 [Sent]
Directory--> 0 \\storman\c$\rh test1\dira [Sent]
Normal File--> 17 \\storman\c$\rh test1\dira\testfile.out
[Sent]
Successful incremental backup of '\\storman\c$\rh test1\*'
Directory--> 0 \\storman\c$\rhtest2 [Sent]
Normal File--> 17 \\storman\c$\rhtest2\testfile.out [Sent]
Successful incremental backup of '\\storman\c$\rhtest2\*'
Directory--> 0 \\storman\c$\rhtest3 [Sent]
Directory--> 0 \\storman\c$\rhtest3\datafiles [Sent]
Normal File--> 17
\\storman\c$\rhtest3\datafiles\testfile.out [Sent]
Successful incremental backup of '\\storman\c$\rhtest3\datafiles\*'
Total number of objects inspected: 8
Total number of objects backed up: 8
Total number of objects updated: 0
Total number of objects rebound: 0
Total number of objects deleted: 0
Total number of objects expired: 0
Total number of objects failed: 0
Total number of bytes transferred: 1.57 KB
Data transfer time: 0.00 sec
Network data transfer rate: 0.00 KB/sec
Aggregate data transfer rate: 0.78 KB/sec
Objects compressed by: 0%
Elapsed processing time: 00:00:02
C:\TSM\baclient>
Andy Raibeck
IBM Software Group
Tivoli Storage Manager Client Development
Internal Notes e-mail: Andrew Raibeck/Tucson/IBM@IBMUS
Internet e-mail: storman AT us.ibm DOT com
The only dumb question is the one that goes unasked.
The command line is your friend.
"Good enough" is the enemy of excellence.
"ADSM: Dist Stor Manager" <ADSM-L AT VM.MARIST DOT EDU> wrote on 05/13/2004
14:48:23:
> Andy,
> I must be confused (what else is new). I believe I
> tried running a backup as you suggested. I think that
> I still ran through all the directories/folders in the
> backup domain, this still took too long. Maybe I
> missed something.
>
> --- Andrew Raibeck <storman AT US.IBM DOT COM> wrote:
> > If I understand this correctly, include/exclude
> > isn't really the issue
> > since from the GUI you were picking the specific
> > subdirectories that you
> > wanted to process. Why not simply do the same thing
> > from the CLI?
> >
> > For example, if from the GUI you were selecting
> > directories C:\Some Dir,
> > D:\MyDir, and D:\work\datafiles, then why not just
> > do this?
> >
> > dsmc i "c:\some dir\*" d:\mydir\
> > d:\work\datafiles\ -subdir=yes
> >
> > Note that the asterisk is required in the first
> > example because the
> > embedded space in the file specification
> > necessitates using quotation
> > marks. If you don't care why, then you can skip the
> > rest of this. For
> > those who want to know, it has to do with how the
> > Windows command
> > processor parses quotes, backslashes, and escape
> > characters.
> >
> > If I had used a command like this, without the
> > asterisk:
> >
> > dsmc i "c:\some dir\" d:\mydir\
> > d:\work\datafiles\ -subdir=yes
> >
> > Then the Windows command processor would have
> > treated the ending
> > slash-quote combination in "c:\some dir\" as an
> > embedded quotation mark
> > rather than a delimiter, passing the following
> > arguments to the TSM
> > client:
> >
> > arg1: i
> > arg2: c:\some dir" d:\mydir\ d:\work\datafiles\
> > -subdir=yes
> >
> > which would make the client think that arg2 was the
> > entire file
> > specifications, rather than three file specs and an
> > option. What we really
> > want to pass to the client is:
> >
> > arg1: i
> > arg2: c:\some dir
> > arg3: d:\mydir\
> > arg4: d:\work\datafiles\
> > arg5: -subdir=yes
> >
> > If this doesn't make sense, don't worry about it....
> > just use the asterisk
> > as I showed above and it should work fine.
> >
> > Regards,
> >
> > Andy
> >
> > Andy Raibeck
> > IBM Software Group
> > Tivoli Storage Manager Client Development
> > Internal Notes e-mail: Andrew
> > Raibeck/Tucson/IBM@IBMUS
> > Internet e-mail: storman AT us.ibm DOT com
> >
> > The only dumb question is the one that goes unasked.
> > The command line is your friend.
> > "Good enough" is the enemy of excellence.
> >
> > > > > Hi,
> > > > >
> > > > > I'm backing up a NAS device with 1M+ files. As
> > > > other
> > > > > posts indicate, the only way to get any
> > > > performance in
> > > > > this setup is to split the file backups across
> > > > > multiple sessions. In my case I have two
> > backup
> > > > hosts,
> > > > > each running two backup streams. I can start
> > each
> > > > of
> > > > > the four backup streams using the GUI and
> > > > selecting
> > > > > appropriate folders. A kind of load balancing
> > is
> > > > > achieved by picking an appropriate mix of
> > folders
> > > > for
> > > > > each backup stream. Each backup runs in a few
> > > > hours
> > > > > each stream processing a subset of the 1M
> > files.
> > > > >
> > > > > My problem is I can't figure out how to
> > accomplish
> > > > the
> > > > > same thing using the command line and
> > associated
> > > > > include/excludes. I'd like to schedule the
> > four
> > > > > streams to run the same way, each with an
> > > > appropriate
> > > > > set of folders/files to process. It would be
> > > > wonderful
> > > > > if I could somehow see how this is
> > accomplished
> > > > via
> > > > > the GUI, as I can't seem to get the same
> > behavior
> > > > > using commands. Each combination I have tried
> > > > always
> > > > > results in each backup stream
> > processing/scanning
> > > > all
> > > > > of the 1M files! This takes way too many hours
> > and
> > > > > doesn't provide the same parallel behavior I
> > get
> > > > when
> > > > > running the GUI. How can I limit each stream
> > to
> > > > > processes only the files/folders I desire
> > without
> > > > > scanning everything; the same way as the GUI
> > seems
> > > > to
> > > > > function?
> > > > >
> > > > > Thanks for your help,
> > > > > Rodney Hroblak
> > > > > ADP
> > > > >
> > > > >
> > > > >
> > > > >
> > > > > __________________________________
> > > > > Do you Yahoo!?
> > > > > Yahoo! Movies - Buy advance tickets for 'Shrek
> > 2'
> > > > >
> > > >
> > >
> >
> http://movies.yahoo.com/showtimes/movie?mid=1808405861
> > > > >
> > >
> > >
> > >
> > >
> > >
> > > __________________________________
> > > Do you Yahoo!?
> > > Yahoo! Movies - Buy advance tickets for 'Shrek 2'
> > >
> http://movies.yahoo.com/showtimes/movie?mid=1808405861
>
>
>
>
>
> __________________________________
> Do you Yahoo!?
> Yahoo! Movies - Buy advance tickets for 'Shrek 2'
> http://movies.yahoo.com/showtimes/movie?mid=1808405861
|
