Nagios Support Forum

Hello everyone,

I am running Nagios XI 2024R1.4.1 on RHEL 8 and currently using Graphios (Python 2.x) to send performance data to InfluxDB. In my graphios.cfg, I have use_service_desc = True, which causes Graphios to create a separate measurement for each service_description. As a result, we now have over 6 000 measurements in InfluxDB, which is becoming unmanageable.

We would like to change this structure (for example, use a single measurement with tags rather than one per service) and it seems like a good time to explore other tools or approaches, since Graphios only supports Python 2 and may break in the future.

Has anyone dealt with a similar issue? What lightweight or more future-proof methods do you recommend for exporting Nagios XI performance data to InfluxDB? Any tips on restructuring the measurements (or alternative exporters) would be greatly appreciated.

Thanks in advance for your help!

Since Graphios uses Python 2 and is outdated, you may consider the following alternatives:

- `nagios-to-influxdb` scripts or community tools
There are several projects or community-run scripts that parse Nagios perfdata and push metrics to InfluxDB using modern libraries and Python 3.

- Search GitHub for updated Nagios-to-InfluxDB exports. Some lightweight tools, use external Nagios scripts or perfdata files and push data via InfluxDB's streaming protocol.

- Write a custom Nagios Event Handler/Notification Script**
Create a lightweight Python 3 or Go script that reads perfdata from Nagios notifications or through periodic parsing of perfdata files, then sends it to InfluxDB as a single measurement with tags for `host`, `service` and other metadata.

- This gives you complete control over measurement structure and tagging. Enables modern authentication, TLS, and bulk injection.