Shell Programming and Scripting

hi,

I've some log files in a path /SYSTEM/custom/data/MNP/xmlerror as:

MNP_PORT_IN_P200902191347563283_20090220181630_err_1.xml
MNP_PORT_IN_P200902171717114365_20090220211814_err_2.xml
MNP_PORT_IN_P200902191349023233_20090220181923_err_1.xml
MNP_PORT_IN_P200902191613464167_20090220121403.xml
MNP_PORT_IN_P200902172005315413_20090220125338_err_2.xml
MNP_PORT_IN_P200902121649264151_20090220160804.xml
MNP_PORT_IN_P200902172005315413_20090220125338_err_3.xml

Now i want the count of only those log files who are not having _err_1, _err_2, etc in the file name ie in the above case its 2 (count of .xml files only without having _err_number) in a day ie in the above case date is 20th Feb 2009 and that to of every half a hr.

ie. in my script i've given a sleep time of 5 mins so i want the count of files from 19:00 to 19:25 and from 19:30 to 20:00 of 20th Feb 2009.

And if the count of files is greater than 25 then an sms should be sent and an script "stop_bulk_engine.sh" should be executed.

Plz help me in this regard.

Thanks.

Considering the sample input you have given...

The below command will give you the xml's which have been generated in 1 day

Code:

$ find *.xml -mtime -1 -print >>out.xml
$cat out.xml
MNP_PORT_IN_P200902121649264151_20090220160804.xml
MNP_PORT_IN_P200902171717114365_20090220211814_err_2.xml
MNP_PORT_IN_P200902172005315413_20090220125338_err_2.xml
MNP_PORT_IN_P200902172005315413_20090220125338_err_3.xml
MNP_PORT_IN_P200902191347563283_20090220181630_err_1.xml
MNP_PORT_IN_P200902191349023233_20090220181923_err_1.xml
MNP_PORT_IN_P200902191613464167_20090220121403.xml

The below will give you the count of files not having _err_1,_err_2 etc...

Code:

$ awk -F "_" '{if ($6!="err") print }' out.xml | wc -l
2

Hi,

Hope this also can do that..

Code:

sed -n '/^[A-Z].*[a-z]_[0-9].xml$/!p' inputfile|wc -l

Code:

sed -n '/\([A-Z]\{3\}\)_\([A-Z]\{4\}\)_\([A-Z]\{2\}\)_P\([0-9]\{18\}\)_\([0-9]\{14\}\).xml/p' inputfile|wc -l

Thanks
Sha

I want to extract the timestamp from the file name and then take the count of every half an hr giving a sleep time of 5 mins between every half an hr for a day and taking that count in a variable and then performing some actions.

for example:
for files of 23rd Feb 2009
from 10:00 am to 10:30 am
and from 10:35 am to 11:00 am
and so on till 12:00 am of 24th Feb 2009

and staring from the night 12 of 23rd Feb 2009.

Thanks

hi

below is my script:

#!/bin/sh

smsto1=0132777094
timeout=300

while :
do

TIME=`date '+%H%M'`
START=0700
END=2300


if [ $TIME -ge $START ] && [ $TIME -le $END ]
then

rcvd_file_cnt=`ls -l /SYSTEM/custom/data/MNP/xmlerror | sed -n '/\([A-Z]\{3\}\)_\([A-Z]\{4\}\)_\([A-Z]\{2\}\)_P\([0-9]\{18\}\)_\([0-9]\{14\}\).xml/p' |wc -l`

echo "Count is" $rcvd_file_cnt

[ $rcvd_file_cnt -eq 0 ] && { sendsms_all.exe $smsto1 "(ALERT) Greater Than 25 files received today.";}

[ $rcvd_file_cnt -gt 0 -a $rcvd_file_cnt -le 2 ] && { sendsms_all.exe $smsto1 "(ALERT) Greater Than 25 files received.";}
fi

sleep $timeout

done

In this i want to I want to extract the timestamp from the file name and then take the count of every half an hr giving a sleep time of 5 mins between every half an hr for a day.

But with the above one im getting the count of all files including different days but i need it wrt 1 day and of every half an hr.

please help in the same regard.