KlusterAlert

Overview

Coming SoonKlusterAlert is not yet available. This documentation is a preview of what's coming.

KlusterAlert monitors your Kubernetes workloads automatically. No rules to write. It detects crashes, restarts, and resource spikes, then notifies your team through Microsoft Teams, Slack, or email with an AI-generated explanation of what happened and why.

How it works

1
Deploy the agent
Install the KlusterAlert Helm chart into each cluster. The lightweight agent runs in its own namespace and uses read-only RBAC. No kubeconfig upload, no inbound firewall rules.
2
Agent polls for issues
Every 60 seconds the agent queries the Kubernetes API for pod and node state. When it detects a problem it pushes the event outbound to the KlusterAlert ingestion API over HTTPS on port 443.
3
KlusterAlert classifies and deduplicates
The event is classified by issue type (CrashLoopBackOff, OOMKilled, ImagePullBackOff, etc.) and deduplicated so the same ongoing issue does not trigger repeated notifications.
4
Notification fires with AI context
A notification is sent to your configured Teams, Slack, or email channel. On Pro and above it includes a plain-English AI explanation and a suggested kubectl command.

Key concepts

Ready to connect your first cluster?

The quickstart walks you through installing the Helm chart and receiving your first alert in under 5 minutes.

Go to Quickstart