Production monitoring

Goal

Deploy Application production-monitoring on site-a. You can change agent.name.

Three microservices start: sensor-simulator, smart-anomaly-detector, and diagnostic-service. User rules and the export rule are already in place.

The file is deploy/steps/03-application-production-monitoring.yaml.

YAML

production-monitoring

apiVersion: datasance.com/v3
kind: Application
metadata:
  name: production-monitoring
spec:
  natsConfig:
    natsAccess: true
    natsRule: production-monitoring-export
  microservices:
    - name: sensor-simulator
      agent:
        name: site-a # override for your ECN agent name
      images:
        registry: 1
        amd64: ghcr.io/datasance/pot-edge-patterns/sensor-simulator:1.2.0
        arm64: ghcr.io/datasance/pot-edge-patterns/sensor-simulator:1.2.0
        riscv64: ghcr.io/datasance/pot-edge-patterns/sensor-simulator:1.2.0
        arm: ghcr.io/datasance/pot-edge-patterns/sensor-simulator:1.2.0
      natsConfig:
        natsAccess: true
        natsRule: sensor-publisher
      container:
        env:
          - key: TELEMETRY_MODE
            value: replay
          - key: SITE_ID
            value: plant-01
          - key: REPLAY_CSV_PATH
            value: /assets/manufacturing/replay.csv
          - key: REPLAY_SEGMENTS_PATH
            value: /assets/manufacturing/segments.json
          - key: REPLAY_SPEED
            value: "1"
          - key: REPLAY_LOOP
            value: "true"
      schedule: 10

    - name: smart-anomaly-detector
      agent:
        name: site-a # override for your ECN agent name
      images:
        registry: 1
        amd64: ghcr.io/datasance/pot-edge-patterns/smart-anomaly-detector:1.2.0
        arm64: ghcr.io/datasance/pot-edge-patterns/smart-anomaly-detector:1.2.0
        riscv64: ghcr.io/datasance/pot-edge-patterns/smart-anomaly-detector:1.2.0
        arm: ghcr.io/datasance/pot-edge-patterns/smart-anomaly-detector:1.2.0
      natsConfig:
        natsAccess: true
        natsRule: anomaly-processor
      container:
        env:
          - key: SUBSCRIBE_SUBJECT
            value: telemetry.machine.>
          - key: TEMP_THRESHOLD
            value: "90"
          - key: VIBRATION_THRESHOLD
            value: "10"
          - key: ERROR_RATE_THRESHOLD
            value: "15"
          - key: JOURNAL_STREAM
            value: JOURNAL_PRODUCTION
          - key: DEBOUNCE_SEC
            value: "30"
      schedule: 20

    - name: diagnostic-service
      agent:
        name: site-a # override for your ECN agent name
      images:
        registry: 1
        amd64: ghcr.io/datasance/pot-edge-patterns/diagnostic-service:1.2.0
        arm64: ghcr.io/datasance/pot-edge-patterns/diagnostic-service:1.2.0
        riscv64: ghcr.io/datasance/pot-edge-patterns/diagnostic-service:1.2.0
        arm: ghcr.io/datasance/pot-edge-patterns/diagnostic-service:1.2.0
      natsConfig:
        natsAccess: false
      container:
        ports:
          - internal: 8080
            protocol: tcp
            external: 8081
      schedule: 30

This creates the production-monitoring NATS account tied to the export rule.

sensor-simulator

Image ghcr.io/datasance/pot-edge-patterns/sensor-simulator:1.2.0. User rule sensor-publisher.

Variable	Default	Purpose
TELEMETRY_MODE	replay	Replay the manufacturing CSV
SITE_ID	plant-01	Site label in the JSON
REPLAY_LOOP	true	Repeat after the last row

You can set REPLAY_SEGMENT to fault-high-vibration-m002, fault-overheating-m003, or fault-quality-m004.

smart-anomaly-detector

Image ghcr.io/datasance/pot-edge-patterns/smart-anomaly-detector:1.2.0. User rule anomaly-processor.

It subscribes to telemetry.machine.>, publishes notify.alerts.> and journal.events.>, and writes stream JOURNAL_PRODUCTION.

diagnostic-service

Image ghcr.io/datasance/pot-edge-patterns/diagnostic-service:1.2.0. natsAccess is false. Container port 8080 is published as external port 8081. The dashboard reaches it later through a Router Service.

Deploy

potctl deploy -f deploy/steps/03-application-production-monitoring.yaml

What you should see

potctl get microservices
potctl describe microservice sensor-simulator

Logs show machineId M001 through M004 on telemetry.machine..

potctl describe microservice smart-anomaly-detector

natsAccess is true and natsRule is anomaly-processor.

The Controller has created the plant NATS account. Copy its id on the next page before you deploy the import rule.

Common mistakes

An image tag other than :1.2.0.

agent.name is site-a, but that Edgelet node is not provisioned.

The export rule is missing, so the Application cannot bind production-monitoring-export.

Next

Account public key

See anything wrong with the document? Help us improve it!

Edit this page on GitHub!