checking errpt

[icu812]

Does anyone have a script that checks erros in the errpt(AIX box) on a daily basis. I have approx. 32 AIX boxes and have been tasked with monitoring errpt on a weekly basis. Help! Thanks

 

[bjverzal]

E-mail me at WVerzal@komint.com and I'll send you my error log monitoring script.

Bill.

 

[cspilman]

Here is a script that might work for you:

#!/bin/ksh

TOTALERRS=`errpt | grep -v "IDENTIFIER" | wc -l`
OLDERRS=`cat /usr/local/bin/errpt.count`
((NEWERRS=TOTALERRS-OLDERRS))

if [ ${NEWERRS} -gt 0 ]
then
errpt | grep -v "IDENTIFIER" | head -${NEWERRS} | cut -c 42- |
while read ERRMSG
do
echo "errpt:${ERRMSG}" | /usr/bin/mailx user@domain.com
done
fi

echo ${TOTALERRS} > /usr/local/bin/errpt.count
Regards,
Chuck

chuck.spilman@nokia.com

 

[bi]

icu812:

I haven't tried Chuck's script, but I do use Bill's script. I start it up at boot time and have set in the script for it to run every five minutes. I started using it just as we were starting to have massive disk problems (cause by an overheated computer room) and it saved me and my users.

 

[cat400]

Here is a simple script that can be cron'd to run each day at 23:59. It mails the current days error log entries to the admin.

#!/bin/ksh
##########################################################################
#
# ckerrpt: This script will look for all errors in the errpt
# for the current day. It should be run at 23:59 to catch all
# errors that occurred that day.
#
##########################################################################
#set -x
tempfile=/tmp/cherrpt.log

#-------------------------------------------------------------------------
# Create header for mail
#-------------------------------------------------------------------------
print "host: "$(hostname) "has experienced the following errors ..." > $tempfile
print >> $tempfile

#-------------------------------------------------------------------------
# Create a date string that will match errpt date format MMDDHHMMYY
#-------------------------------------------------------------------------
today=$(date +%m%d)'[0-9][0-9][0-9][0-9]'$(date +%y)

#-------------------------------------------------------------------------
# Place each matching errpt line into the temp file
# Igonore tty errors because there may be too many of them and they are benign
# for the most part.
#-------------------------------------------------------------------------
errpt | egrep $today | egrep -v "tty" >> $tempfile

#-------------------------------------------------------------------------
# If there are more then two lines in the temp file then assume there were
# matches in the errpt and send mail.
#-------------------------------------------------------------------------
set $(wc -l $tempfile)
if [[ $1 != "2" ]] then
mail -s &quot;$(hostname): daily error report&quot; $ADMIN_MAIL < $tempfile
fi

rm $tempfile

 

[MoshiachNow]

The following is a script that gives a statistcis of &quot;Application terminated Abnormaly&quot; (core) errors in errpt.
Cut&paste the following into the terminal window:

#============================================
ksh

function core_source {
CORE_PATH=&quot;/tmp/core_path&quot;
CORE_SOURCE_LOG=&quot;/tmp/core_source.log&quot;
CORE_STATISTICS=&quot;/tmp/core_stat&quot;
CORE_SORT=&quot;/tmp/core_sort&quot;

#create statistics for all application cores for this host and add it to the main log:
echo &quot;-----------&quot;
echo &quot;Application cores statistics:&quot;
errpt -a -j C60BB505 |awk '/PROGRAM NAME/,/ADDITIONAL INFORMATION/' | grep -vE &quot;PROG|ADD&quot; | sort > $CORE_STATISTICS
cat $CORE_STATISTICS | sort -u > $CORE_SORT
for CORE_REASON in `cat $CORE_SORT |cut -d&quot; &quot; -f1`
do
CORE_COUNT=`grep $CORE_REASON $CORE_STATISTICS |wc -l`
printf &quot;%-20s %6i%s %8s \n&quot; $CORE_REASON $CORE_COUNT
done
#Now collect the sysdump and SYSINTR info:
echo &quot;-----------&quot;
}
core_source