Nagios + NDO: Nagios stop writing to log files

[artgopi]

Hi,

last weekend there was a problem with my Nagios. It seems that a problem with the mysql database, that resides in another server, cause Nagios to stop writing logs to the local machine. It look that Nagios just freeze.

Can NDOMOD broker cause this issue? In the past with other installation I never have any problem using NDO.

The NDOMOD is configured do send to a TCP socket in other server that have NDO2DB daemon.

Thanks for any help.

Artur

[mguthrie]

If there's corruption in one of the ndoutils DB tables or if the hard disk was full you could potentially have an issue like this.

[artgopi]

This happen again.

The database disk became full and Nagios stop writing to the logs.

Is there any way to prevent this?

If NDO write to a file, and that file is stored after to the database. Could that help?

thanks,

Artur

[mguthrie]

You can turn off logging of event handlers and passive checks, and you could also run a cron job that deletes entries older than a specified age.

[artgopi]

After some tests, I configure NDOMOD to use a file as output and a command to rotate the log and then store the log in a database (mv and file2sock).

This look more reliable. If the file or file-system where the ndo.dat exist get full, Nagios keeps working, and writing its logs to nagios.log.

The only issue is that the file get rotated where the command as success or not. So I might get some logs lost!