Monitoring

Monitoring ensures that workflows and integrations operate reliably across connected systems.

The Demiton platform records detailed execution data so operators can observe system behavior and diagnose failures.

Monitoring focuses on three primary areas:

• workflow execution
• integration health
• infrastructure performance

Workflow Observability

Every workflow execution produces a BlueprintRun record.

Each run includes:

• run identifier
• execution start time
• execution status
• step execution results

Individual steps are recorded as StepRun records.

Each StepRun includes:

• step name
• execution status
• error details
• execution duration

These records provide a complete audit trail for workflow execution.

Execution Status

Workflow runs progress through several states:

PENDING
RUNNING
SUCCEEDED
FAILED
SUSPENDED
CANCELLED

These states allow operators to quickly determine system behavior.

Integration Monitoring

Adapters communicate with external systems such as ERP platforms or field systems.

Monitoring should track:

• API failures
• authentication errors
• network connectivity issues

Repeated adapter failures may indicate:

• credential expiration
• API changes
• service outages

These issues should be investigated immediately.

Queue Monitoring

The worker runtime processes jobs from a queue.

Queue monitoring should track:

• queue length
• worker throughput
• processing latency

High queue latency may indicate that additional worker capacity is required.

Failure Handling

When workflows fail, the runtime records error information.

Operators should review:

• failed step logs
• adapter error messages
• system responses

This information allows teams to diagnose the cause of failure.

Operational Dashboards

Monitoring dashboards typically display:

• workflow success rates
• workflow execution duration
• adapter error rates
• queue backlog

These dashboards provide real‑time visibility into platform health.

Alerting

Alerting systems should notify operators when:

• workflows repeatedly fail
• adapter connections break
• queue backlog grows unexpectedly

Timely alerts allow operators to respond quickly to operational issues.

Summary

Monitoring ensures that workflows remain observable and reliable.

By tracking workflow runs, adapter behavior, and queue performance, organizations can maintain operational stability across complex system environments.

---