Bashscripts.org

Hi, First post to this forum.
Need some help to gather the following information from Lotus Notes log files.
I need to search maillogs and grep out anything which took more than 2 minutes and a total of all mail

Can anyone provide any 1 liners which may capture :

anything above 00:02:00
and total of lines/emails in the logfile.

,Size:,1K,Time:,00:00:01,Hop,Count:,2

My script is mostly echoing and then crunching the lines
Starting,to,Analyse,logs,at,Tue,May,,1,11:41:45,NZST,2012
Start,Date,=,,20/03/2012,09:10:45,Router:
End,Date,=,,20/03/2012,09:11:26,Router:
Crunching,,210,,lines


So it shows how many lines are crunched which is 1 step done.
Just need to grep out the +2Mins

Any help on a 1 liner would be much appreciated.
Last request was too much

This sounds like a fun project for bash or Perl! What have you tried so far?
You could almost certainly use awk to gather your numbers. You'd probably want to do your math separately. I can think of a few ways to approach this.

Thanks for the prompt reply.

Your ideas would be much appreciated as I have only managed some re-direction.

Trying to apply some awk or sed for output but to do the sums I am trying to work out.

I'm trying some Perl 1 liner scripts which just strip the data out. As these logs have a lot of info

perl -wne'while(/[\w\.]+@[\w\.]+/g){print "$&\n"}' test.txt

perl -wne'while(/[\w\.]+@[\w\.]+\w+/g){print "$&\n"}' test.txt

perl -wne'while(/[\w\.\-]+@[\w\.\-]+\w+/g){print "$&\n"}' emails.txt | sort -u > output.txt

These can strip out the emails, but I need to get the time columns and size, so awk is preferable for that I'm thinking.

If someone can provide a script to awk out the 5 relevant columns into another file for safe keeping that would be a good start

This is what I have so far.

[size=90]#!/bin/sh
MAILLOG=/root/logs/notes.log
OUTPUTFILE=mylog.txt

echo "Starting to Analyse logs at `date`" > $OUTPUTFILE
echo "Start Date = " `head -1 $MAILLOG | awk '{ print $1, $2, $3 }'` >> $OUTPUTFILE
echo "End Date = " `tail -1 $MAILLOG | awk '{ print $1, $2, $3 }'` >> $OUTPUTFILE
echo "Crunching " `wc -l $MAILLOG | awk '{ print $1 }'` " lines" >> $OUTPUTFILE
echo >> $OUTPUTFILE
echo "==================" >> $OUTPUTFILE
echo >> $OUTPUTFILE
echo "======= Host not found ========" >> $OUTPUTFILE

echo >> $OUTPUTFILE
echo "=> Domains associated with Host Not found" >> $OUTPUTFILE
grep "Host not found" $MAILLOG | sed -e 's/.*for name=//' | sed -e 's/ type=.*//' | sort | uniq -c >> $OUTPUTFILE
echo "==================" >> $OUTPUTFILE


so nothing yet on +2 MINS.

any advice?

Hi jeo, could you post you ideas so i could have a look. In need of some inspiration.
Thankyou for your reply