moving files from a dir in one machine to a dir in another machines

Shell Programming and Scripting
1

Hi, I am a unix newbie.I need to write a shell script to move my oracle READ WRITE datafiles from one serevr to another. I need to move it from /u01/oradata/W1KK/.. to /u01/oradata/W2KK,
/u02/oradata/W1KK/.. to /u02/oradata/W2KK.

That is, I actaully am moving my datafiles from one database to another.. database W1KK to W2KK .. both of these databases are on different machines. This is what I have now. I get the list of datafiles into a file called
W1KK_dbfiles.txt. After that, I donot know how to use sed or awk to move the files using scp.Can someone please help with this?

#!/bin/ksh
sqlplus -s "/ as sysdba" <<EOF>W1KK_dbfiles.txt
       set heading off feedback off verify off
        SELECT name FROM v$datafile WHERE ENABLED='READ WRITE';
       exit;
EOF
2

lets just say you are moving files to "/u02/oradata/W2KK" - directory. this system has

username -  W2KK
hostname - u02
then your sftp will work like this

sftp W2KK@u02
(here ask for password)
cd /u02/oradata/W2KK/ 
put W1KK_dbfiles.txt

this will put W1KK_dbfiles.txt into the directory /u02/oradata/W2KK

3

Thanks.But that is not what I am looking for. I will be getting the list of READ WRITE files that need to be moved by -

#!/bin/ksh
sqlplus -s "/ as sysdba" <<EOF>W1KK_dbfiles.txt
set heading off feedback off verify off
SELECT name FROM v$datafile WHERE ENABLED='READ WRITE';
exit;
EOF

the database w1kk is on a machine, say m1 and the database w2kk is on another machine, say m2. The directory structures where the datafiles are located are almost the same. On machine m1 it is-
/u01/oradata/w1kk
/u02/oradata/w1kk
/u03/oradata/w1kk
..
..
..

and on machine m2, it is
/u01/oradata/w2kk
/u02/oradata/w2kk
/u03/oradata/w2kk
..
..
..

Once I identify the files that need to be scp-ed, I need to move the file
/u01/oradata/w1kk to /u01/oradata/w2kk
/u02/oradata/w1kk to /u02/oradata/w2kk
/u03/oradata/w1kk to /u03/oradata/w2kk

I tried to use sed to replace the w1kk to w2kk bu it didnt work. I just didn't know how to do it. This is what I tried with sed -

#!/bin/ksh
sqlplus -s "/ as sysdba" <<EOF>W1KK_dbfiles.txt
set heading off feedback off verify off
SELECT name FROM v$datafile WHERE ENABLED='READ WRITE';
exit;
EOF

for i in `cat W1KK_dbfiles.txt`
do
echo $i
echo `sed 's/W1KK/W2KK/g' $i `
done

This is wrong because - it tells me it doesn't identify $i as a file and second, it loops through the W1KK_dbfiles.txt for all the records, instead of using sed to act on each line of the file. I am just lost. Can someone please help?