Skip to content

Events and Alerts

Edge Builder uses events to provide information on issues arising on an edge node and alerts to monitor aspects of edge nodes.

Events

Edge Builder uses a system of Events to provide information about issues that arise on an edge node. Events are stored in the Edge Builder Controller, and are one of the types described in the following table:

Type Description
Internal Any internal events raised when the Edge Builder Controller is managing the edge node
Resource Events that are triggered when a monitoring alert is triggered on the edge node; for example when the CPU usage exceeds the threshold
App Events referring to any issues with an app running on an edge node
Node Events relating to any other issues associated with actions performed on an edge node

All events are associated with an edge node. When an event is registered, it is marked as active.
When an edge node has active events associated with it, its state changes to Warning, unless the edge node is or has been in the Down state, in which case the event state will be Error.

You must review all active events, using one of the following:

  • The REST API, using the following command where controller-ip must be replaced with the IP address of the Edge Builder controller:

    curl -ski -X GET "http://controller-ip:8085/api/events?node=Node1" \
    -H "Content-type: application/json" \
    -H "Authorization: eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJFeHBpcmVzQXQiOjE2MTEzMTQyNjEsIlBhc3N3b3JkIjoiRWRnZUJ1aWxkZXIxMjMiLCJVc2VybmFtZSI6ImlvdGVjaCJ9.c-VgQx1Iq184Vi_j9FdsnQcN6zWjTnPfKnm2C692Rbw"
    
  • The Edge Builder CLI, using the following command:

    edgebuilder-cli event view -n node1
    

Alerts

Edge Builder can assign and monitor configurable alerts on nodes. These alerts utilise Kapacitor and are customised using the Chronograf user interface. In Chronograf, alert rules are created, where custom configurations for the alerting can be set up. Examples of possible alerts include CPU usage above/below certain threshold, disk usage alerts, and many more.

Alert "rules" are created first, and these are described using TICKScripts. These are created automatically using Chronograf, but can also be manually edited. Once these rules are created, this can be added to a node by using the rule to describe an alertDefinition in Edge Builder, before then deploying this as an alert on a specific node.

Whenever Kapacitor detects that an alert has been triggered on the node, it will send this information to Edge Builder and an event will be created describing the details of the alert.

For a detailed tutorial on configurable alerts, see the configurable alerts tutorial