Update documentation around self-managed Dashboards#37055
Conversation
This is automatically downloaded in materialize-terraform-self-managed
Not much extra steps are required if just using the terraform modules
| - source_labels: | ||
| - __meta_kubernetes_pod_name | ||
| target_label: pod | ||
| - job_name: podMonitor/materialize-sql/0 |
There was a problem hiding this comment.
One thing I wanna callout is the materialize-sql endpoints require authentication if the auth mode isn't None https://materialize.com/docs/security/self-managed/authentication/. This means passing in a SQL username/password in the scrape config.
| {{< note >}} | ||
| This guide uses the [prometheus-community](https://github.com/prometheus-community/helm-charts) Helm chart to install Prometheus. | ||
| {{< /note >}} | ||
| - A self-managed instance of Materialize running with materialize-terraform-self-managed Terraform module with |
There was a problem hiding this comment.
I feel strongly that we need more instructions here! Tell me a little bit more information about:
- When do I need to enable observability? eg: I don't have an o11y stack installed already
- What does enabling observability do? Eg: installs prometheus, grafana etc.
- What if I already have an o11y stack and a prometheus store installed? What do I do next?
| ```bash | ||
| curl -O https://raw.githubusercontent.com/prometheus-community/helm-charts/refs/heads/main/charts/prometheus/values.yaml | ||
| ``` | ||
| {{< tip >}} |
There was a problem hiding this comment.
Let's leave forward looking statements about roadmap / future updates out of docs the please :)
| You can find them in the [Metrics Scraping](https://materializeinc.github.io/materialize-monitoring/metrics/scraping/) page. | ||
|
|
||
|  | ||
| Classic ScrapeConfigs are also available in [Metrics Scraping](https://materializeinc.github.io/materialize-monitoring/metrics/scraping/). |
There was a problem hiding this comment.
What are classic scrapeconfigs? Older scrape configs?
| {{< note >}} | ||
| This guide uses the [prometheus-community](https://github.com/prometheus-community/helm-charts) Helm chart to install Prometheus. | ||
| {{< /note >}} | ||
| - A self-managed instance of Materialize running with materialize-terraform-self-managed Terraform module with |
There was a problem hiding this comment.
So .. there's only one bullet point. So, not sure if we need the "one of"
| ```bash | ||
| curl -O https://raw.githubusercontent.com/prometheus-community/helm-charts/refs/heads/main/charts/prometheus/values.yaml | ||
| ``` | ||
| {{< tip >}} |
There was a problem hiding this comment.
Let's remove the tip. While appreciate the spirit of this, in reality, let's not incorporate timeframe as things like product priorities could change.
| {{< note >}} | ||
| This guide uses the [prometheus-community](https://github.com/prometheus-community/helm-charts) Helm chart to install Prometheus. | ||
| {{< /note >}} | ||
| - A self-managed instance of Materialize running with materialize-terraform-self-managed Terraform module with |
There was a problem hiding this comment.
Per meeting discussion: enable_observability=true bundles both the infra + the flag(s) to enable getting the metrics, yes?
For orgs with observability stacks, we need to show just flipping the flag to enable/load scrape configs.
| If you are using materialize-terraform-self-managed, you already have Grafana available. | ||
|
|
||
| ## 3. Optional. Visualize through Grafana | ||
| You can use the following to enable port-forwarding to access the Grafana UI: |
There was a problem hiding this comment.
We could make this into a steps:
- the port forwarding and the open Grafana UI the port forwarding
- the open Grafana UI
And move the warning up.
|
This was heading in the wrong direction. Going to restart. |
4 participants
Motivation
The current documentation assumes a Helm install with no Grafana or Prometheus.
While that is possible, we want to not be so in the business of managing Grafana and
Prometheus best practices in this repository.
Most of materialize-terraform-self-managed already supports standing up Grafana and Prometheus
leaving a lot of this to not be super interesting.
Description
This updates prometheus.yml to a better scrapeconfig since this is downloaded by default by materialize-terraform-self-managed implicitly.
Provide updated documentation for managing in materialize-terraform-self-managed.
Establish up-front baseline assumptions.
Link to more advanced documentation for now.
Verification
(cd docs/user; hugo server)