Runtime Monitoring

Every exported runtime agent includes a built-in web dashboard and reports live status to Studio or Nodlyn Cloud. You can operate, inspect, and control agents from the machine itself or remotely — including from a phone browser.

Two monitoring layers

Layer	Where	What you see
Agents tab	Studio sidebar or Nodlyn Cloud	Fleet view: online/offline, running job, version drift, remote commands, activity feed.
Runtime dashboard	On the agent machine (default port `9090`)	Jobs, schedule, logs, variables, approvals, active run controls, Swagger API.

Heartbeats (status to Studio/Cloud) are separate from event ingress (ERP/scripts → agent). See Event Ingress.

On-prem vs cloud monitoring

Choose at export time under Agent monitoring:

This on-prem studio

Agent appears in the local Studio Agents tab after install.
Heartbeats every ~30 seconds; marked Offline after ~90 seconds without check-in.
Auto-refreshes every 15 seconds in the UI.

Nodlyn Cloud (remote)

Sign in with your Nodlyn Cloud account during export (on-prem Studio login is not enough).
Install the ZIP on the target machine the same way as on-prem monitoring.
View agents at https://nodlyn.com/app/ → Agents tab — works from any browser, including mobile.
Hybrid pattern: design in cloud or desktop → export with cloud monitoring → run on edge → monitor from phone or browser anywhere.

Agent status values

Status	Meaning
Starting	Agent just launched; setting up.
Idle	Online, waiting for triggers or events.
Running	Executing a workflow run.
Errors (Degraded)	Online but recent runs failed — check runtime logs.
Paused	Service up; workflows paused — triggers ignored until Resume.
Offline	No heartbeat — crashed, network loss, or shut down.
Pending	Registered but no first heartbeat yet — start the agent on the target machine.

Remote commands (Agents tab)

From Studio or Cloud, send commands delivered on the next heartbeat (≤30 s):

Pause / Resume — stop or accept triggers.
Run / Dry run — start a manual execution.
Cancel run — stop the active job.
Stop agent — shut down the runtime service.

Each agent row includes Dashboard ↗ — opens the local runtime dashboard URL reported by the agent.

Integrated runtime dashboard

On the agent host, open http://localhost:9090 (or the port set at export):

Route	Purpose
`/` or `/dashboard`	Overview, schedule, recent runs, active run controls.
`/activity`	Step-by-step results for a run.
`/logs`	Rolling log tail.
`/variables`	Workflow and global variables.
`/approvals`	Pending business approvals.
`/compact`	Mobile-friendly minimal view — status and controls on small screens.
`/swagger`	Interactive API docs for ingress and dashboard JSON endpoints.

View status on mobile

Cloud Agents tab — open https://nodlyn.com/app/ on your phone when cloud monitoring was selected at export.
On-prem Agents tab — open Studio on the same network (or VPN) from a mobile browser.
Runtime /compact — bookmark or share the dashboard URL on the agent’s LAN (e.g. http://192.168.1.50:9090/compact).
Share / QR Code — generate a QR link to view, trigger, or run the workflow from your phone; use Start Tunnel for access outside the local network. See Share Workflow & QR Code.

After export

The Agents tab shows a Waiting for agent to connect… banner for ~10 minutes after export. Install and start the agent on the target machine; status updates automatically when the first heartbeat arrives.