mjhjmtu

file.txt

servername1:DBNAME1:/u01/app
servername2:DBNAME2:/u01/dbs
servername3:DBNAME3:/u01/app1

I want to use above file contents line by line and execute it in shell script

command1 hostname=servername1 db=dbanme1 location=/u01/app 
command2 hostname=servername1 db=dbanme1 location=/u01/app

after the above commands it has to take the second line values and execute

while IFS=':' read -r hostname db location
do
 set -- hostname="$hostname" db="$db" location="$location"
 command1 "$@"
 command2 "$@"
done <file.txt

Or, if you need to preserve the positional parameters,

while IFS=':' read -r hostname db location
do
 command1 hostname="$hostname" db="$db" location="$location"
 command2 hostname="$hostname" db="$db" location="$location"
done <file.txt

Both of these would read file.txt line by line, and for each line read the three :-delimited fields into the three variables hostname, db and location. If there are more :-delimited fields on any row, these would be added onto the value of location.

Then the commands are executed.

Using gnu parallel:

$ parallel --colsep ":" 'command1 hostname=1 db=2 location=3; command2 hostname=1 db=2 location=3' :::: file.txt 

For each line in file.txt the commands are excecuted. The lines get splitted by :, so that 1,2,3 get replaced by the values in the columns.

Note that parallel start the processes in parallel. By default the number of jobs running in parallel is the number of cores. You can specify how many jobs are allowed to run in parallel with -j. If you need to keep the output in the order of the lines executed, use the -k parameter.