Before you begin
You need:- An Obsy account (sign up free)
- A running Kubernetes cluster with
kubectlaccess - A Datadog or Grafana Cloud account (for metrics and traces)
Step 1 — Create your account
Go to app.obsy.ai/signup, fill in your name, email, and a password, then verify your email.Step 2 — Complete onboarding
After your first login, the onboarding wizard walks you through:- Name your organization — this becomes the slug for your public status page.
- Connect an observability platform — pick Datadog or Grafana Cloud and paste your API key.
- Add your first cluster — give it a name and paste your
kubeconfigor use the in-cluster agent.
Step 3 — Install the OTel collector
Once your cluster is registered, open it in OTel Collector, then click Install Collector. Obsy deploys an OpenTelemetry gateway + node collector via Helm into theobsy-system namespace. The process takes about 2 minutes. You’ll see a green Healthy badge when done.
Your cluster needs outbound internet access on port 443 to reach the Helm chart registry and your observability platform’s ingestion endpoint.
Step 4 — Send your first trace
Add the OTel SDK to one of your services and point it at the gateway:Step 5 — View data in your platform
Head to Datadog or Grafana — your traces, metrics, and logs should appear within a minute. Back in Obsy, open Telemetry Review and select your service to see a quality and cost breakdown.What’s next?
Set up sampling
Reduce cost without losing signal by configuring smart sampling rules.
Route alerts to Obsy
Send Datadog or Grafana alerts to Obsy for automatic RCA and incident creation.
Configure incidents
Set up Slack channels, auto-create rules, and notification handles.
Publish your status page
Give your customers a real-time view of your system health.