Log File with latest filename monitoring

[RIDS_I2MP]

Hi Team,

We are looking to monitor an application log file but the log file name changes as per the application configuration settings for max size.
It created multiple log files with XXX_date_01.log, XXX_date_02.log so on.. and latest logs are present in newest file (XXX_date_02.log).
Could you please let us know if we can achieve this latest log file monitoring. it should through an alert if we found any "ERROR" keyword.

Also, could you please share which agent needs to be installed in order to achieve this requirement from Nagios XI.
Kindly share any link or documentation if this can be achieved from Nagios end.


Thanks,
Nitin Parate

[gsmith]

Hi

Want you would want to do is create a script that does the following:
1. determine the most current file (XXX_date_01.log)
2. find any instances of the word ERROR in the file
3. get the latest ERROR message based on timestamp in log (assuming there are timestamps in log)
4. If current time - timestamp is < 5 minutes set errorlevel to 2, else errorlevel = 0
5. if errorlevel =2 echo "ERROR: (message from step 3) & exit 2
else if errorlevel =0 echo "OK" & exit 0
else echo "unknown issue" & exit 3

Note: You would have to accomodate the "edge case" where the log file closes and a new one is created during
the "look bacK' period of 5 minutes

The script returns a message and an exit status to Nagios.

The return codes Nagios recognizes are:
0 for OK
1 for Warning
2 for Critical
3 for Unknown

Install NCPA on the machine to be monitored:

Image5.jpg

Put your script in the /usr/local/ncpa/plugins directory on the machine to be monitored.

Run the Configuration Wizard: NCPA on the Nagios server.
Keep or remove the default checks as you desire. Near the bottom of the page use the
"file picker" (circled) to select the script on the remote machine:

Image6.jpg

Continue with the Wizard. Since the script checks the log entries for the previous
5 minutes you would need to set the service to check every 5 minutes.

Note: There may be away to tell the script to look back X minutes based
on the frequency of the service check - this would be the ideal implementation.

The following documentation describes using NCPA in depth:
https://www.nagios.org/ncpa/help.php#ac ... the-plugin

Thanks