Hello,
We are new to Luminis, and this is actually my first posting to the Luminis Developer Network.
I have been assigned the task to monitor Luminis logs. Does anybody have a mechanism to automatically alert an administrator by email when an error is posted to the logs? I’d like to avoid checking the log file on a daily basis, and have the system notify me when there is an error.
We are running Luminis IV on Linux.
Thanks,
Sam Bootsma
George Brown College
Luminis Logs
Depending on the logging you have setup in cplog4j.properties, (ERROR, DEBUG, etc...) you could get hundreds of emails a day. Really.
Instead, I would "grep ERROR $CP_LOG/cp.log > error.xml", then examine that output file nightly.
Luminis is a huge 500 lbs. gorilla of a program, with MANY intricacies and idiosyncrasies. I hope you realize what you're getting into. Are they asking you to be the Luminis admin? (Ask for a raise and a trip to the SunGard training center in Salt Lake City). Its fun if you like that kind of stuff, I do so my company is paying me nicely for doing what I love anyway. They call it a Platform for a reason, it a big mash-up of programs (LDAP, SunOne calendar, virtual community (group and course studio), iPlanet web servers, and email center, etc….) ;>
Good luck,
Reuben
log monitoring
Here's something I came up with to get an idea of how much we see various errors in our logs.
It's schedule as a cron job. It's pretty rudimentary, but it's better than nothing. We're on Solaris, should work mostly unchanged on Linux:
#!/bin/sh ############################################################################################ # Step 0 - setup ############################################################################################ . /opt/luminis/.cprc DTSTAMP=`date "+%Y-%m-%d %H:%M"` FILENAME=/tmp/error_notification.txt FILEHEAD=/tmp/error_header.txt FILETEMP=/tmp/error_errors.txt echo "Errors logged as of $DTSTAMP" > $FILEHEAD echo "When activity is low, the same errors may be reported several hours in a row." >> $FILEHEAD ############################################################################################ # Step 1 - check for errors and write to $FILETEMP # Note: This won't be perfect, doesn't scan cp.log by date/time, so older # errors might be reported more than once. ############################################################################################ ## Check a: ParallelTimeoutMergingPersonAttributeDaoImpl.java ERRORCOUNT=`grep -c "ParallelTimeoutMergingPersonAttributeDaoImpl.java" $CP_ROOT/logs/cp.log` if [ $ERRORCOUNT -gt 0 ]; then echo "$ERRORCOUNT ParallelTimeoutMergingPersonAttributeDaoImpl errors" >> $FILETEMP fi ## Check b: failed to renderState error ERRORCOUNT=`grep -c "failed to renderState" $CP_ROOT/logs/cp.log` if [ $ERRORCOUNT -gt 0 ]; then echo "$ERRORCOUNT 'failed to renderState' errors" >> $FILETEMP fi ## Check c: Error creating group for course ERRORCOUNT=`grep -c "Error creating group for course" $CP_ROOT/logs/cp.log` if [ $ERRORCOUNT -gt 0 ]; then echo "$ERRORCOUNT 'Error creating group for course' errors" >> $FILETEMP fi ## Check d: Error finding user grep "Error finding user: " $CP_ROOT/logs/cp.log >> /export/home/cpadmin/log/errorfindinguser.log ## Check e: Can't service request to /cp/home/logout ERRORCOUNT=`grep -c "Can't service request to /cp/home/logout" $CP_ROOT/logs/cp.log` if [ $ERRORCOUNT -gt 0 ]; then echo "$ERRORCOUNT 'Cant service request to /cp/home/logout' errors" >> $FILETEMP fi ## Check f: channel rendering errors CALERRORCOUNT=`/usr/xpg4/bin/grep -c -e u12l1n37 -e u13l1n36 -e u14l1n46 $CP_ROOT/logs/cp.log` if [ $CALERRORCOUNT -gt 0 ]; then echo "$CALERRORCOUNT calendar channel errors" >> $FILETEMP fi ## Check g: cpip external user id error ERRORCOUNT=`grep -c "CPIP ERROR sctssb Unable to create exter" $CP_ROOT/logs/cp.log` if [ $ERRORCOUNT -gt 0 ]; then echo "$ERRORCOUNT 'Unable to create external user id' errors" >> $FILETEMP fi ## Check h: Can't service request to /cp/home/next ERRORCOUNT=`grep -c "Can't service request to /cp/home/next" $CP_ROOT/logs/cp.log` if [ $ERRORCOUNT -gt 0 ]; then echo "$ERRORCOUNT 'Cant service request to /cp/home/next' errors" >> $FILETEMP fi ## Check i: "generating cache key" error ERRORCOUNT=`grep -c "generating cache key" $CP_ROOT/logs/cp.log` if [ $ERRORCOUNT -gt 0 ]; then echo "$ERRORCOUNT 'generating cache key' errors" >> $FILETEMP fi ############################################################################################ # Step 2 - mail file ############################################################################################ if test -s "$FILETEMP" then echo "." >> $FILETEMP # triggers end of message cat $FILEHEAD $FILETEMP > $FILENAME mailx -s "Luminis error report" -r "from@school.edu" to@school.edu < $FILENAME fi ############################################################################################ # Step 3 - cleanup ############################################################################################ rm $FILETEMP rm $FILEHEADmeaning?
Do you know what these errors mean?
'Cant service request to /cp/home/logout' errors
'generating cache key' errors
Thanks-
Bill