Following Telegraf official README.md, I could compile and run Telegraf on OpenBSD. Due to my lack of knowledge on Go, this turned to be a quite instable method. Some Telegraf release would compile, some not. Anyway, thanks to Landry@, it is now possible to use Telegraf from OpenBSD ports.Continue reading “A better way to run Telegraf on OpenBSD”
While experimenting with Telegraf, SNMP and InfluxDB, I ended up filling some measurements with tags and fields that have the same name. InfluxDB works nicely with such user-case by adding numbers (in format “_###”) to those keys. But that’s a mess to deal with when using Grafana. And a mess in general.
I tried to run InfluxData Telegraf on OpenBSD 6.2 but it wasn’t available in the Ports nor was I able to compile a binary from sources. But the latter has changed since I have an OpenBSD 6.3 instance running. Here’s how to compile and run Telegraf on OpenBSD.
I’ve enabled an OpenBSD unbound(8) daemon that is used as a central DNS cache resolver. Now I needed to know what it was doing and how it performed. The question was answered grabbing statistics from unbound and render them using Grafana.
The whole monitoring stack is composed of Net-SNMP, Telegraf and InfluxDB for the metrics part ; and syslogd(8), Logstash and Elasticsearch for the logs part. Of course, most of those run on OpenBSD (6.3) ; except Telegraf, which is not available (yet).
I regularly check my pf(4) activity using pfctl(8) and pflogd(8). I already monitored pf using collectd(1) and rrdtool. This time, I wanted to use my already configured InfluxDB/Grafana system. The thing is, Telegraf is not (yet) available in OpenBSD 6.3. So I used a Telegraf container to remotely poll OpenBSD using the native SNMP OPENBSD-PF-MIB.