Anyone know when the Metricbeat Prometheus addon will be out of beta? Or at least support basic metrics from Prometheus like CPU/Memory?
Thanks
Hi @gregsh,
We don't have specific dates on that, modules go through a process of feedback until we mark them as stable.
I'm wondering, what do you mean by supporting those basic CPU/Memory metrics? The module should be able to pick anything exported using the Prometheus format
I enabled the Prometheus module in /etc/metricbeat/metricbeat.yml and disabled the Metricbeat module. There are no CPU, memory, network, disk, or other system level metrics; just four prometheus.stats.* metrics. Perhaps I've misconfigured something? I've pointed metricbeat to our main Prometheus server (port 9090). I know this server has good metrics because Grafana shows current data.
The Prometheus module has two metricsets: one is stats and one is collector. The stats metricset gets some metrics from the Prometheus server (only these fields: https://www.elastic.co/guide/en/beats/metricbeat/6.0/exported-fields-prometheus.html). By configuring the collector metricset, you get all the metrics directly from your Prometheus exporters.
1 Like
You're right! Enabling the "collector" metricset did yield many more prometheus metrics. I tried to make some Visualizations for a Dashboard but could not determine what aggregation has the prometheus metrics. I know this is probably out of the scope of the original question. Are there pre-made dashboards for the prometheus collector metricset?
Thanks for your help!
No, unfortunately, we don't have any dashboard for the Prometheus collector metricset. It would be nice to share your dashboard with the OS community once you build one.
This topic was automatically closed after 21 days. New replies are no longer allowed.