Node API
Inspect the node serving the request: an unauthenticated health check, runtime statistics and uptime, the active main configuration, and a configuration reload trigger.
The node endpoints report on the single node that handles the request, normally the Master. They cover a lightweight health probe, live system statistics, and the node’s own configuration. The health check is the only endpoint in the whole API, alongside metrics, that needs no token.
Health check
GET /api/pingReturns basic status and identity for the node. This endpoint is unauthenticated, making it suitable for liveness and readiness probes.
curl http://<master>:7000/api/ping{
"status": "ok",
"nodeId": "node-1",
"clusterName": "universe-cluster",
"master": true
}| Field | Type | Description |
|---|---|---|
status | string | Always ok when the node is serving. |
nodeId | string | Identifier of the responding node. |
clusterName | string | Hazelcast cluster name the node belongs to. |
master | boolean | Whether the node runs as the Master. |
Point container or load-balancer health checks at /api/ping. It does not require a token and never touches the cluster scheduler.
Node information
GET /api/nodeReturns the node ID, a human-readable uptime string, and a block of system statistics.
curl http://<master>:7000/api/node \
-H "Authorization: Bearer YOUR_TOKEN"{
"nodeId": "node-1",
"uptime": "2h 14m 06s",
"systemStats": {
"cpuLoad": 0.32,
"usedMemoryMB": 4096,
"totalMemoryMB": 16384
}
}| Field | Type | Description |
|---|---|---|
nodeId | string | Identifier of the responding node. |
uptime | string | How long the node process has been running. |
systemStats | object | Runtime statistics for the node, such as CPU load and memory usage. |
Node configuration
GET /api/node/configReturns the node’s active main configuration, the same fields written to config.json on disk. Use this to confirm cluster membership, role, and port settings without shell access to the host.
curl http://<master>:7000/api/node/config \
-H "Authorization: Bearer YOUR_TOKEN"{
"address": "127.0.0.1",
"port": 6000,
"apiPort": 7000,
"nodeId": "node-1",
"clusterName": "universe-cluster",
"isMasterNode": true,
"masterAddress": "127.0.0.1",
"masterPort": 6000,
"masterApiPort": 7000,
"debug": false,
"maxRamMB": 16384,
"maxCpu": 800
}| Field | Description |
|---|---|
address | Address this node advertises to the cluster. |
port | Hazelcast cluster port (default 6000). |
apiPort | REST API port (default 7000). |
nodeId | Unique node identifier. |
clusterName | Hazelcast cluster name shared by every member. |
isMasterNode | Whether the node runs the API and owns state. |
masterAddress / masterPort / masterApiPort | Coordinates of the Master used to join the cluster. |
debug | Whether verbose logging is enabled. |
maxRamMB / maxCpu | Capacity limits the scheduler honors when placing instances. |
Reload configuration
POST /api/node/reloadTriggers a reload of the node configuration from disk. This endpoint is defined but not yet fully implemented; it currently returns 501 Not Implemented.
curl -X POST http://<master>:7000/api/node/reload \
-H "Authorization: Bearer YOUR_TOKEN"{
"error": "Not Implemented"
}To reload instance configurations at runtime today, use the config reload console command via the Commands API.