nagflux

Nagflux

A connector which transforms performancedata from Nagios/Icinga(2)/Naemon to InfluxDB/Elasticsearch

Nagflux collects data from the NagiosSpoolfileFolder and adds informations from Livestatus. This data is sent to an InfluxDB, to get displayed by Grafana. Therefor is the tool Histou gives you the possibility to add Templates to Grafana.

Nagflux can be seen as the process_perfdata.pl script from PNP4Nagios.

Dependencies

Golang 1.5+

Install

go get -u github.com/griesbacher/nagflux
go build github.com/griesbacher/nagflux

A x86-64 Linux binary will be added to the releases. Here the link to the latest Release.

Configure

Here are some of the important config-options:

Start

If the configfile is in the same folder as the executable:

./nagflux

else:

./nagflux -configPath=/path/to/config.gcfg

Debugging

• If the InfluxDB is not available Nagflux will stop and an log entry will be written.
• If the Livestatus is not available Nagflux will just write an log entry, but additional informations can't be gathered.
• If any part of the Tablename is not valid for the InfluxDB an log entry will written and the data is writen to a file which has the same name as the logfile just with the ending '.dump-errors'. You could fix the errors by hand and copy the lines in the NagfluxSpoolfileFolder
• If the Data can't be send to the InfluxDB, Nagflux will also write them in the '.dump-errors' file, you can handle them the same way.
• If the logs are showing files are being read (in DEBUG mode) but nothing is going into InfluxDB, check the perfdata template to ensure it matches OMD format. See Perfdata Template for more details.

Dataflow

There are basically two ways for Nagflux to receive data:

• Spoolfiles: They are for useful if Nagflux is running at the same machine as Nagios
• Gearman: If you have a distributed setup, that's the way to go

With both ways you could enrich your performance data with additional informations from livestatus. Like downtimes, notifications and so.

Targets can be:

• InfluxDB, that's the main target and the reason for this project.
• Elasticsearch, more a prove of concept but it worked some time ago ;)
• JSON, to parse the data by an third tool.

OMD

Nagflux is fully integrated in OMD-Labs, as well as Histou is. Therefor if you wanna try it out, it's maybe easier to install OMD-Labs.

Perfdata Template

Nagflux supports a couple of Perfdata templates (see main_test.go for some supported formats). By default it assumes you have the OMD formattemplate. If you are setting this up manually (not using OMD) please ensure your perfdata template is as follows:

Host

DATATYPE::HOSTPERFDATA\tTIMET::$TIMET$\tHOSTNAME::$HOSTNAME$\tHOSTPERFDATA::$HOSTPERFDATA$\tHOSTCHECKCOMMAND::$HOSTCHECKCOMMAND$

Service

DATATYPE::SERVICEPERFDATA\tTIMET::$TIMET$\tHOSTNAME::$HOSTNAME$\tSERVICEDESC::$SERVICEDESC$\tSERVICEPERFDATA::$SERVICEPERFDATA$\tSERVICECHECKCOMMAND::$SERVICECHECKCOMMAND$

If you are using Nagios the default templates will not work. Use the above templates with config host_perfdata_file_template and service_perfdata_file_template, respectively.

DEMO

This Dockercontainer contains OMD and everything is preconfigured to use Nagflux/Histou/Grafana/InfluxDB: https://github.com/Griesbacher/docker-omd-grafana

Presentations

• Here is a presentation I held about Nagflux and Histou in 2016, only in German, sorry: Slides
• That's the first one from 2015, also only in German. Slides - Video