Looking for volunteers to test a new feature - observability for your Mimir cluster made easier!

[09jvilla]

Hi Mimir community -

We’re looking for volunteers to test out a new feature we’ve built to make it easier to observe your Grafana Mimir cluster.

Using our most recent Helm chart (v3.0), users can deploy a Grafana Agent to scrape metrics and logs from their Grafana Mimir cluster by adding a few extra lines to their Helm values file.
Users must provide their own Prometheus (or Prometheus-compatible metrics database) and Loki to store the scraped metrics and logs collected by the Agent.

For those who don’t want the overhead of managing their own Prometheus and Loki, we’ve made it easy to remote-write those metrics and logs to Grafana Cloud, which provides a hosted Prometheus and Loki. Besides giving you a zero-maintenance way to store the metrics and logs data you’ve collected, Grafana Cloud also provides a hosted Grafana instance with a one-click experience to install all the dashboards and recording rules necessary to make sense of it. Besides decreasing your operational overhead for observing Mimir, using Grafana Cloud to observe your Mimir also gives you the peace of mind that the observability for your metrics database is in a separate environment than the database itself.

We’re looking for users to test out this experience of observing their Mimir cluster with Grafana Cloud, described in more detail here on our ‘Self-hosted Grafana Mimir integration for Grafana Cloud’ page.

For users with small Mimir installations (~1 tenant, ~10 pods), the metrics and logs required to monitor Mimir should fit within Grafana Cloud’s free tier, meaning that you can monitor your Mimir install in Grafana Cloud at no charge. Medium and larger size installations may generate enough logs and metrics to exceed free tier limits, and we would be happy to waive any costs for users interested in trying out this new experience for 3 months as a thanks for your feedback.

If you’re interested in helping us make Mimir better, let us know by commenting on this discussion or reaching out on the community slack!