Shell Programming and Scripting

Dear All,

I am not getting the expected output :::

df -h | awk -v limit=70 'NR>0 && int($5) >=limit'

df -h
Filesystem Size Used Avail Use% Mounted on
/dev/cciss/c0d0p10 2.0G 236M 1.6G 13% /
/dev/cciss/c0d0p1 985M 33M 903M 4% /boot
/dev/cciss/c0d0p3 9.7G 363M 8.8G 4% /files/BACKUP
/dev/cciss/c0d0p11 985M 18M 917M 2% /home
/dev/cciss/c0d0p5 9.7G 3.5G 5.8G 38% /opt
none 3.0G 0 3.0G 0% /dev/shm
/dev/cciss/c0d0p2 15G 4.5G 9.3G 33% /product
/dev/cciss/c0d0p9 3.9G 33M 3.7G 1% /tmp
/dev/cciss/c0d0p7 5.8G 2.7G 2.9G 49% /usr
/dev/cciss/c0d0p8 4.9G 458M 4.2G 10% /var
/dev/vg_database/lv_database
345G 255G 73G 78% /database
/dev/vg_database/lv_oradata6
40G 35M 38G 1% /database/oradata6
/dev/vg_database/lv_oradata7
9.7G 33M 9.2G 1% /database/oradata7
/dev/vg_database/lv_backup
177G 2.1G 166G 2% /database/backup


It should return

/dev/vg_database/lv_database
345G 255G 73G 78% /database


Please help me out

Thanks

unless you copy/pasted it wrong, it looks like there TWO lines of what you're looking for.
If it is on ONE line - it works as expected.

its basically the one line... but when i run this it does not return the required result...

df -h | awk -v limit=70 'NR>0 && int($5) >=limit'

please note that the partition mounted on /database is not being returned for results which are true i.e.

/dev/vg_database/lv_database 345G 255G 73G 78% /database

BR/Asad

$ cat asa.txt

Code:

Filesystem Size Used Avail Use% Mounted on
/dev/cciss/c0d0p10 2.0G 236M 1.6G 13% /
/dev/cciss/c0d0p1 985M 33M 903M 4% /boot
/dev/cciss/c0d0p3 9.7G 363M 8.8G 4% /files/BACKUP
/dev/cciss/c0d0p11 985M 18M 917M 2% /home
/dev/cciss/c0d0p5 9.7G 3.5G 5.8G 38% /opt none 3.0G 0 3.0G 0% /dev/shm
/dev/cciss/c0d0p2 15G 4.5G 9.3G 33% /product
/dev/cciss/c0d0p9 3.9G 33M 3.7G 1% /tmp
/dev/cciss/c0d0p7 5.8G 2.7G 2.9G 49% /usr
/dev/cciss/c0d0p8 4.9G 458M 4.2G 10% /var
/dev/vg_database/lv_database 345G 255G 73G 78% /database
/dev/vg_database/lv_oradata6 40G 35M 38G 1% /database/oradata6
/dev/vg_database/lv_oradata7 9.7G 33M 9.2G 1% /database/oradata7
/dev/vg_database/lv_backup 177G 2.1G 166G 2% /database/backup

$ nawk -v limit=70 'NR>0 && int($5) >=limit' asa.txt
/dev/vg_database/lv_database 345G 255G 73G 78% /database

nawk command isnt found

nawk -v limit=70 'NR>0 && int($5) >=limit' asad.txt
-bash: nawk: command not found

I had your output in the file gAshvin as follows

/dev/cciss/c0d0p10 2.0G 236M 1.6G 13% /
/dev/cciss/c0d0p1 985M 33M 903M 4% /boot
/dev/cciss/c0d0p3 9.7G 363M 8.8G 4% /files/BACKUP
/dev/cciss/c0d0p11 985M 18M 917M 2% /home
/dev/cciss/c0d0p5 9.7G 3.5G 5.8G 38% /opt none 3.0G 0 3.0G 0% /dev/shm
/dev/cciss/c0d0p2 15G 4.5G 9.3G 33% /product
/dev/cciss/c0d0p9 3.9G 33M 3.7G 1% /tmp
/dev/cciss/c0d0p7 5.8G 2.7G 2.9G 49% /usr
/dev/cciss/c0d0p8 4.9G 458M 4.2G 10% /var
/dev/vg_database/lv_database 345G 255G 73G 78% /database
/dev/vg_database/lv_oradata6 40G 35M 38G 1% /database/oradata6
/dev/vg_database/lv_oradata7 9.7G 33M 9.2G 1% /database/oradata7
/dev/vg_database/lv_backup 177G 2.1G 166G 2% /database/backup


$ awk -v limit=70 'NR>0 && int($5) >=limit' gAshvin
/dev/vg_database/lv_database 345G 255G 73G 78% /database

I assume that you might have broken the line and the awk encountered the line as 2 seperate lines and you missed the correct output.

I tested this scenario on my test machine as follows

$ df -h
Filesystem Size Used Avail Use% Mounted on
/dev/mapper/VolGroup00-LogVol00
71G 3.8G 63G 6% /
/dev/sda1 99M 16M 78M 18% /boot
tmpfs 500M 0 500M 0% /dev/shm
/dev/hda 681M 681M 0 100% /media/ashvin


$ df -h | awk -v limit=18 'NR>0 && int($5) >=limit'
/dev/sda1 99M 16M 78M 18% /boot
/dev/hda 681M 681M 0 100% /media/ashvin

Please do double check the df -h output in your case

On my machine result, If I change the clause "limit=18" by " limit=5", The output does not display the line
/dev/mapper/VolGroup00-LogVol00 71G 3.8G 63G 6% /
in output. This is because the new line character by df -h is treated as EOR by awk and hence the test condition 'NR>0 && int($5) >=limit' fails for this line ( which is infact treated 2 seperate records by awk)
Hope this explains your failes output.
Try to redirect the df -h output in a file, verify the file and then use awk over the file.