Dashboard API

The coordinator's embedded dashboard exposes a JSON API alongside the web UI. All endpoints are served on the same address as /metrics (set via --metrics-addr).

Base URL

Base URL

http://<coordinator>:9090

Endpoints

GET /api/v1/pipeline

Returns the pipeline summary.

Response:

{
  "pipeline": "orders-demo",
  "run_id": "a1b2c3d4",
  "source_kind": "mysql",
  "sink_type": "iceberg+rest",
  "tables": 1,
  "workers": 2,
  "started_at": "2026-09-17T10:00:00Z",
  "uptime_s": 3600,
  "status": "healthy",
  "snapshot_active": false,
  "maintenance_enabled": true
}

Field	Type	Description
pipeline	string	Pipeline name from the spec
run_id	string	Unique run identifier
source_kind	string	Source driver (mysql, postgres, kafka)
sink_type	string	Sink type (iceberg+rest, clickhouse, couchbase)
tables	int	Number of configured tables
workers	int	Number of connected workers
started_at	string	RFC3339 timestamp of pipeline start
uptime_s	int64	Seconds since start
status	string	healthy, degraded, or failed
snapshot_active	bool	True during initial snapshot phase
maintenance_enabled	bool	True if any table has maintenance enabled

---

GET /api/v1/tables

Returns all streams (source → target).

Response:

[
  {
    "source": "shop.orders",
    "target": "bronze.orders",
    "write_mode": "upsert",
    "position": "mysql-bin.000003:12345",
    "lag_s": 2.5,
    "rows_total": 150000,
    "rows_rate": 1200.5,
    "commits": 450,
    "commit_failures": 0,
    "commit_latency_ms": 45.2,
    "equality_deletes": 1200,
    "deletes_dropped": 0,
    "snapshot_progress": 1.0,
    "maintenance": {
      "compaction": {
        "runs": 12,
        "last_run": "2026-09-17T11:00:00Z",
        "files_removed": 48,
        "files_added": 6,
        "bytes_before": 104857600,
        "bytes_after": 98566144
      },
      "expiry": {
        "runs": 6,
        "last_run": "2026-09-17T11:05:00Z",
        "snapshots_removed": 30
      },
      "orphan": {
        "runs": 1,
        "last_run": "2026-09-17T11:00:00Z",
        "files_deleted": 42,
        "bytes_freed": 10485760
      }
    }
  }
]

---

GET /api/v1/tables/{name}

Returns one stream by source or target name.

Parameters:

Name	In	Description
name	path	Source or target table name (e.g. shop.orders or bronze.orders)

Response: Same as a single element from GET /api/v1/tables.

Status codes:

• 200 OK — stream found
• 404 Not Found — no stream matches the name

---

GET /api/v1/workers

Returns all connected workers.

Response:

[
  {
    "name": "orders-demo-bronze.orders-0",
    "status": "streaming",
    "epoch": 0,
    "tables": ["bronze.orders"],
    "last_ack_s": 1.2,
    "inflight_bytes": 524288,
    "committed": {
      "bronze.orders": "mysql-bin.000003:12345"
    }
  }
]

Field	Type	Description
name	string	Worker group name (<pipeline>-<target>-<index>)
status	string	starting, snapshotting, or streaming
epoch	uint64	Reset count (bumped on supervisor reset)
tables	[]string	Tables this worker is assigned to
last_ack_s	float64	Seconds since last acknowledgement
inflight_bytes	int64	Unacked batch bytes
committed	map	Per-table last committed CDC position

---

GET /api/v1/workers/{name}

Returns one worker by name.

Parameters:

Name	In	Description
name	path	Worker group name

Response: Same as a single element from GET /api/v1/workers.

Status codes:

• 200 OK — worker found
• 404 Not Found — no worker matches the name

---

GET /api/v1/events

Returns the operational event log, newest first.

Query parameters:

Name	Default	Description
type	(all)	Filter by event type (e.g. commit_failure, worker_reset)
worker	(all)	Filter by worker name
limit	200	Max events to return (cap: 1000)

Response:

[
  {
    "ts": "2026-09-17T11:05:32.123456789Z",
    "type": "compaction",
    "worker": "orders-demo-bronze.orders-0",
    "table": "bronze.orders",
    "message": "compaction",
    "fields": {
      "files_removed": 48,
      "files_added": 6
    }
  }
]

---

GET /api/v1/logs

Returns the coordinator's structured log tail, newest first.

Query parameters:

Name	Default	Description
level	debug	Minimum level: debug, info, warn, error
limit	500	Max log lines to return

Response:

[
  {
    "ts": "2026-09-17T11:05:32.123456789Z",
    "level": "INFO",
    "msg": "coordinator: worker acked",
    "attrs": {
      "worker": "orders-demo-bronze.orders-0",
      "position": "mysql-bin.000003:12345"
    }
  }
]

---

GET /api/v1/stream

Server-Sent Events (SSE) endpoint for real-time updates.

On connect, the server pushes a full snapshot event containing the complete state (pipeline, tables, workers, events, logs). After that, only delta events are pushed:

Event type	Payload
snapshot	Full state on connect
state	Updated pipeline/tables/workers
event	New operational event
log	New coordinator log line

A periodic : ping comment keeps idle proxies from timing out.

Example (browser):

SSE browser example

const es = new EventSource('/api/v1/stream');
es.addEventListener('snapshot', (e) => {
  const state = JSON.parse(e.data);
  renderDashboard(state);
});
es.addEventListener('state', (e) => {
  const delta = JSON.parse(e.data);
  updateDashboard(delta);
});

Example (curl):

SSE curl example

curl -N http://coordinator:9090/api/v1/stream

---

POST /api/v1/actions/cancel

Terminates the pipeline gracefully — sends Shutdown to all workers.

Response:

{"ok": true}

Status codes:

• 200 OK — shutdown initiated
• 500 Internal Server Error — shutdown failed

---

POST /api/v1/actions/restart/{worker}

Resets one worker's session: bumps its epoch, cancels its gRPC session, and triggers a supervisor reset. The worker reconnects with a fresh epoch.

Parameters:

Name	In	Description
worker	path	Worker group name

Response:

{"ok": true}

Status codes:

• 200 OK — restart initiated
• 500 Internal Server Error — restart failed

---

GET /healthz

Kubernetes liveness probe. Returns 200 OK with body ok when the coordinator is running.

GET /readyz

Kubernetes readiness probe. Returns 200 OK with body ok when the coordinator is ready to serve traffic.

---

Error responses

All error responses return plain text in the body:

Error response example

no such stream

HTTP status codes follow standard REST conventions:

Code	Meaning
200	Success
400	Bad request (invalid query parameter)
404	Resource not found
500	Internal server error