Health Checks
The Intelligence Hub exposes HTTP endpoints that can be used by load balancers, orchestration platforms, or monitoring tools to check the health and availability of the hub.
Liveness
The /v2/livez endpoint returns an HTTP 200 response when the hub application is live. Use this endpoint to determine whether the hub process is running and responsive.
Readiness
The /v2/readyz endpoint returns an HTTP 200 response when the hub is ready to receive network traffic. Use this endpoint to determine when the hub is fully initialized and available to handle requests.
Endpoint Behavior by Deployment Mode
The response from each endpoint depends on how the hub is deployed.
| Deployment Mode | Node Role | Standby Mode | /v2/livez | /v2/readyz |
|---|---|---|---|---|
| Standalone | — | — | 200 | 200 when processing data, 503 until then |
| High Availability | Primary | Any | 200 | 200 when processing data, 503 until then |
| High Availability | Secondary | passive | 200 | 503 |
| High Availability | Secondary | active | 200 | 200 when processing data, 503 until then |
readyz reports whether the node is processing data, not whether it is the primary. A passive secondary processes none, so it always returns 503, which lets a load balancer route only to the primary. An active secondary returns 200 once its connections and pipelines are running, and a primary that is starting or reloading returns 503 until its own are. A node keeps the primary role while it is not ready, so a sustained 503 from the primary is worth alerting on. See Standby Modes.
active cluster, route traffic only to a node that returns 200 and reports "role": "primary". The status code alone can select a ready secondary, and the role alone can select a primary that is not yet serving data.In an active cluster, readiness also gates promotion: a node only takes the primary role once it reports ready.
Response Body
Both endpoints return a JSON response body containing a deployment object that describes the current node’s deployment context.
| Field | Type | Description |
|---|---|---|
status | string | Whether the node is ready. Either OK or NOT_READY. |
reason | string | Why the node is not ready. Only present when status is NOT_READY. |
deployment.mode | string | The deployment mode. Either standalone or ha. |
deployment.role | string | The role of this node. Either primary or secondary. Only present in HA mode. |
deployment.standbyMode | string | How the nodes of this cluster behave while standing by. Either passive or active. Only present in HA mode. See Standby Modes. |
deployment.nodeId | string | The unique identifier for this node. Only present in HA mode. |
deployment.primaryNodeId | string | The unique identifier for the primary node. Only present in HA mode. |
Standalone example:
{
"status": "OK",
"deployment": {
"mode": "standalone"
}
}
High Availability example:
{
"status": "OK",
"deployment": {
"mode": "ha",
"role": "primary",
"standbyMode": "active",
"nodeId": "node-1",
"primaryNodeId": "node-1"
}
}
High Availability example, a node that is not ready:
{
"status": "NOT_READY",
"reason": "Runtime is not ready.",
"deployment": {
"mode": "ha",
"role": "secondary",
"standbyMode": "passive",
"nodeId": "node-2",
"primaryNodeId": "node-1"
}
}