oracle_gg_monitoring_nagios [My wiki]

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.
=====Overview=====
This way partially uses the script method but this time, instead of receiving spam A LOT. I will integrate the script with nagios so I can be notified ONLY when there is a problem.


=====Nagios Basics=====
You can use ANY script with nagios as soon as it provides the correct exit status. Nagios script should exit with:

  * 0 - If all is OK
  * 1 - If it is Warning
  * 2 - If it is Critical
  * 3 - If it is unknown

Knowing that, I have modified the previous script to exit accordingly:

<Code:bash|Script to be integrated with Nagios>
#!/bin/bash
OIFS=$IFS
IFS="
"
NIFS=$IFS
function status {
OUTPUT=`$GG_HOME/ggsci << EOF
info all
exit
EOF`
}
function alert {
for line in $OUTPUT
do
GNAME=$(echo "${line}" | awk -F" " '{print $3}')
GSTAT=$(echo "${line}" | awk -F" " '{print $2}')
GTYPE=$(echo "${line}" | awk -F" " '{print $1}')
if [[ ${GTYPE} == "MANAGER" ]] || [[ ${GTYPE} == "REPLICAT" ]] || [[ ${GTYPE} == "EXTRACT" ]]
then
  case ${GSTAT} in
    STOPPED)
     STAT="STOP"
    ;;
    ABENDED)
     STAT="ABEND"
    ;;
    RUNNING)
     STAT="OK"
    ;;
    *)
     STAT="UNKNOWN"
    ;;
  esac
fi
done
echo ${STAT}
case ${STAT} in
  ABEND)
   exit 2
   ;;
  STOP)
   exit 1
   ;;
  OK)
   exit 0
   ;;
  *)
   exit 3
   ;;
esac
}
export GG_HOME=$(find /u01/app -name "ggsci"  2>&1 | grep -v "Permission denied");
export GG_HOME=${GG_HOME%??????}
export ORACLE_HOME=$(echo `tail -1 /etc/oratab` | awk -F":" '{print $2}')
export LD_LIBRARY_PATH=$ORACLE_HOME/lib
export ORACLE_SID=$(echo `tail -1 /etc/oratab` | awk -F":" '{print $1}')
status
alert
</Code>

The following scripts, exports the ORACLE_HOME, ORACLE_SID (as long as it is the last one) and the Golden Gate Home location (as long as it is somewhere under /u01/app). The scripts also exits appropriately as listed above.

Now to make that script works we have to make coupel things:
=====Configure The monitoring=====
We have to configure two thigns here:

  * Nagios on Client Side
  * Nagios on Server Side

The client configuration will involve re-configuring the NRPE cfg and sudoers (as we need nagios to execute the script as oracle owner)
The Server Configuration Will include adding this config on the nagios server.

====Configure Nagios Client====
Nagios usually comes with plugin called NRPE. Locate where is the nrpe.cfg and edit it as follows by appending this line at the end:

<Code:bash|Edit NRPE cfg>
command[check_gg_oracle]=sudo -u oracle /usr/local/nagios/libexec/check_gg_oracle.sh
</Code>

It is import the script to be executed with Oracle user as the golden gate doesn't have a client way of connecting to the console.

Once that is done, we have to update the sudoers as well. I know there is way more secure way, but I have done as follows:

<Code:bash|Update Sudoers>
## Same thing without a password
# %wheel        ALL=(ALL)       NOPASSWD: ALL
nagios ALL=(ALL) NOPASSWD: ALL
</Code>

That will allow the nagios to execute any command as any user. I know quite unsecure, feel free to change it with ALL=(oracle)

Once this is done, it is ready to add that in the nagios itself :) 

====Configure the Nagios Server====
Once this is all done, we are ready to add this feature to our server:

Go to:

  - Configure -> Core Config
  - Services
  - Add New -> Fill the fields as follows: 
  * Common Settings
    * Config Name: check_gg_oracle
    * Description: Checking Oracle Golden Gate (Or whatever you want)
    * Check Command: check_nrpe
    * $ARG$: check_gg_oracle
    * Mangage Hosts: Add the needed hosts which you want to monitor.
  * Check Settings and the rest of tabs:
    * As desired according your needs

After all this is done, you should be able to see the monitoring status for that service in your dashboard.