LinuxQuestions.org - Looping a bash script

- Linux - Newbie (https://www.linuxquestions.org/questions/linux-newbie-8/)

- - Looping a bash script (https://www.linuxquestions.org/questions/linux-newbie-8/looping-a-bash-script-486097/)

Looping a bash script

Here's the deal - Although I've been around Linux for several years, I'm still a scripting n00b in many ways.

What's going on...

I am running Red Hat Enteprise Linux 4.x on a Dell PowerEdge 1750. Attached to it is a Dell PowerVault 220S configured as an 800 GB RAID 5. I am currently transferring 700+ GB of data from a TerraVault via a mounted SMB share to the 800 GB RAID 5.

I am currently monitoring the progress manually via the following command shell command:

[rhe4 ~]$ df -lh | grep "/dev/sdb1"
/dev/sdb1 808G 194G 573G 26% /vol1

What I would ultimately like to do is the following:

I'd like to append the output to file every 3 minutes until 730G has been reached. A time and date stamp would be nice too. Something along the lines of:

Fri Sep 22 14:07:14 PDT 2006
/dev/sdb1 808G 194G 573G 26% /vol1

Fri Sep 22 14:10:34 PDT 2006
/dev/sdb1 808G 194G 573G 26% /vol1

Obviously I need to create some sort of while-do-loop but not really sure how to go about doing this and I'm not versed in sed/awk, ergo - the reason I need a little help here.

Any takers?

Thanks in advance!

Put this in a file filename.sh

Code:

#!/bin/sh

while true

do

  date

  df -lh | grep "/dev/sdb1"

  sleep 180

done

Then chmod +x filename.sh.

The sleep command is in seconds, so sleep 180 is 3 minutes. To kill it you'll have to CTRL+C. Otherwise, you can add a variable before the while loops starts and then increment it within the loop and exit when it has run a certain number of times.

Stepping through...

Quote:

Originally Posted by pljvaldez

Put this in a file filename.sh

Code:

#!/bin/sh

while true

do

  date

  df -lh | grep "/dev/sdb1"

  sleep 180

done

Then chmod +x filename.sh.

The sleep command is in seconds, so sleep 180 is 3 minutes.

jvaldez - thanks for your attempt - I think something more needs to be done however...

I got the filename, setting the environment...

while true... hmmm... but what are we comparing it to? While -WHAT- is true?

From what I can see...

it's going to do the following...

while true... (while the following is true...)
do (do the following...)
date - outputs date to stdout
df -lh | grep "/dev/sdb1" - outputs drive info to stdout
sleep 180 - pauses for 180 seconds (3 minutes)
done - end loop

Once the script reaches the done marker, there's nothing left to be "true" as we haven't compared the current data size x against the maximum (730G). Also, since the next itteration would be incremented by 3 minutes, it would no longer be true, but false. Also, we haven't appended anything to a log file...

We're missing 2 key components.

So in my seudo/quazi code here's the logic I am seeing...

Start loop
Append date to copystat.log
Append df -lh | grep "/dev/sdb1" to copystat.log
check df -lh | grep "/dev/sdb1" for current drive capacity
if current drive capacity is less than 730G loop after 3 minutes
else if current drive capacity is 730G or more, end loop

Hope this helps.

I'm not sure as to how to write a script to accomodate this logic.

Thanks agian,

I stand corrected

Quote:

Originally Posted by pljvaldez

Put this in a file filename.sh

Code:

#!/bin/sh

while true

do

  date

  df -lh | grep "/dev/sdb1"

  sleep 180

done

I stand corrected - You script is working beautifully - other than output to file and testing against the destination size.

So this is better than nothing.

Thanks again

Code:

#!/bin/bash



echo -n Enter limit:

read LIMIT

while true

do

  REST=`df -lh | grep "/dev/hda2" | awk '{print $4;}'`



  if [[ $REST =~ "[0-9]*M" ]]

  then

    # cut M from size

    REST=${REST%M}

  

    if [[ "$REST" -lt "$LIMIT" ]]

    then

      # exit loop

      break

    fi



    # add stuff to file

    echo -n `date` >> copystat.log

    echo `df -lh | grep "/dev/hda2"` >> copystat.log

  fi



  sleep 3

done

If I understand you, you can try this script.

Here's another crack (probably not the most elegant, but it worked for me assuming there's no way to jump from 729G to 731G in 3 minutes time)

Code:

#!/bin/sh



#global size limit

LIMIT="730G"



while true

do

  date > date.txt

  df -lh | grep "/dev/sdb1" > df-out.txt

  

  #create single line with date and df output

  paste date.txt df-out.txt >> logfile.log

  

  i=`awk '{ print $3 }' df-out.txt`

  

  #exit loop if size from df -lh = LIMIT

  if [`echo $i` = `echo $LIMIT`]; then

      exit

  fi



  #wait 3 minutes (180 seconds) then run again

  sleep 180



done