Skip to main content

Infrastructure Incident Alerts

Infrastructure incident alerts notify administrators and operations teams when Kadeck's monitoring system detects issues with Kafka clusters and associated components. These alerts enable proactive response to infrastructure problems, helping maintain system availability and data integrity.

Prerequisites​

Configuring alert integrations for infrastructure incidents requires Kadeck administrator privileges.

Configuration​

You can configure alert integrations for infrastructure incidents in two ways:

  • Default integration - A single integration that receives notifications for all infrastructure incident types
  • Per-incident type - Specific integrations configured for individual incident types, allowing targeted routing to different teams or systems

Configuration is managed through the Kadeck Administration interface at Administration → Alerts → Notifications.

Supported Incident Types​

Kadeck monitors and can alert on the following infrastructure incident types:

  • Broker Is Down - Reports when a broker is down or unreachable
  • Schema Registry Is Down - Reports when the Schema Registry is down or unreachable
  • Connect Worker Is Down - Reports when a Kafka Connect worker is down or unreachable
  • Under Replicated Partitions - Reports when a partition has fewer replicas than the configured replication factor
  • Offline Partition - Reports when a partition has no leader and cannot be read from or written to
  • No Controller - Reports when the cluster has no controller assigned

Future Enhancements​

Additional monitoring checks and incident types will be added in upcoming releases. Custom incident definitions will also be supported, enabling organizations to create alerts for specific infrastructure conditions relevant to their environment.