Skip to main content

Mock Services Operations

This guide covers operational aspects of running Microcks mock services in Kubernetes environments.

Architecture Overview​

                           ┌──────────────────────┐
                           │   Staqr API Pod     │
                           │                      │
                           │  Uses connection:    │
                           │  mock-optus-wsg     │
                           └──────────┬───────────┘
                                      │
                                      ▼
┌─────────────────────────────────────────────────────────────┐
│                    Kubernetes Cluster                        │
│  ┌─────────────────┐    ┌─────────────────────────────────┐ │
│  │   ConfigMap     │    │        Microcks Pod             │ │
│  │                 │    │                                  │ │
│  │ connections:    │    │  ┌─────────────────────────┐    │ │
│  │   optus:        │───▶│  │ /soap/WsgService/9.2   │    │ │
│  │     endpoint:   │    │  └─────────────────────────┘    │ │
│  │     microcks:   │    │                                  │ │
│  │     8080/...    │    │  Port: 8080 (ClusterIP)         │ │
│  └─────────────────┘    └─────────────────────────────────┘ │
└─────────────────────────────────────────────────────────────┘

Deployment​

Mock services are deployed via Helm subchart and are only enabled in development:

Environmentmicrocks.enabledBehaviour
DevelopmenttrueMicrocks deployed, mock connections active
StagingfalseMicrocks not deployed, UAT connections
ProductionfalseMicrocks not deployed, real carrier connections

Verify Deployment​

# Check Microcks pod
kubectl -n staqr-dev get pods -l app=microcks

# Check service
kubectl -n staqr-dev get svc -l app=microcks

# Check ingress
kubectl -n staqr-dev get ingress -l app=microcks

View Logs​

# Microcks application logs
kubectl -n staqr-dev logs -l app=microcks -f

# Seed job logs (mock imports)
kubectl -n staqr-dev logs -l job-name=staqr-dev-seed-mocks

Health Monitoring​

Health Endpoint​

# Internal (from within cluster)
kubectl -n staqr-dev exec -it deploy/staqr-api -- \
  curl http://staqr-dev-microcks:8080/api/health

# External (via ingress)
curl https://mocks.go.prvidr.io/api/health

Expected response:

{"status": "UP"}

Liveness/Readiness Probes​

The Helm chart configures probes automatically:

livenessProbe:
  httpGet:
    path: /api/health
    port: 8080
  initialDelaySeconds: 60
  periodSeconds: 10

readinessProbe:
  httpGet:
    path: /api/health
    port: 8080
  initialDelaySeconds: 30
  periodSeconds: 5

Common Operations​

Import Mock Definition​

# Copy WSDL into cluster
kubectl -n staqr-dev cp mocks/microcks/import/optus-wsg-9.2.wsdl \
  staqr-dev-microcks-0:/tmp/

# Import via API
kubectl -n staqr-dev exec -it staqr-dev-microcks-0 -- \
  curl -X POST http://localhost:8080/api/artifact/upload \
    -F "file=@/tmp/optus-wsg-9.2.wsdl"

List Imported Services​

kubectl -n staqr-dev exec -it deploy/staqr-api -- \
  curl -s http://staqr-dev-microcks:8080/api/services | jq '.[].name'

View Invocations (Request Logs)​

kubectl -n staqr-dev exec -it deploy/staqr-api -- \
  curl -s "http://staqr-dev-microcks:8080/api/invocations/WsgServiceOrderService/9.2?page=0&size=10" | jq

Clear Mock Data​

# Delete and recreate pod (loses imported mocks)
kubectl -n staqr-dev delete pod -l app=microcks

# Seed job will re-import mocks on pod startup

Troubleshooting​

Microcks Pod Not Starting​

# Check pod events
kubectl -n staqr-dev describe pod -l app=microcks

# Common issues:
# - PVC not bound (storage class unavailable)
# - Image pull failure (registry access)
# - Resource limits too low

Mock Returns 404​

# Check if service is imported
kubectl -n staqr-dev exec -it deploy/staqr-api -- \
  curl http://staqr-dev-microcks:8080/api/services | jq '.[].name'

# Check exact endpoint path
kubectl -n staqr-dev exec -it deploy/staqr-api -- \
  curl http://staqr-dev-microcks:8080/api/services | jq '.[].operations[].resourcePaths'

# Re-run seed job
kubectl -n staqr-dev delete job staqr-dev-seed-mocks
# ArgoCD will recreate it

Connection Timeout to Microcks​

# Verify service is reachable
kubectl -n staqr-dev exec -it deploy/staqr-api -- \
  nc -zv staqr-dev-microcks 8080

# Check network policy
kubectl -n staqr-dev get networkpolicy

# Check service endpoints
kubectl -n staqr-dev get endpoints staqr-dev-microcks

SOAP Request Not Matching​

Ensure:

  1. SOAPAction header matches WSDL operation:

    curl -H "SOAPAction: SubmitServiceOrder" ...

  2. XML namespace matches WSDL targetNamespace:

    <soapenv:Envelope xmlns:wsg="http://optus.com.au/wsg/9.2">

  3. Request validates against XSD schema (Microcks validates by default)

Slow Startup​

Microcks with embedded MongoDB takes 60-90 seconds to start. The Helm chart accounts for this:

startupProbe:
  httpGet:
    path: /api/health
    port: 8080
  failureThreshold: 30
  periodSeconds: 5  # 150 seconds total

Scaling Considerations​

Single Replica Only​

Microcks with embedded MongoDB (uber profile) does not support horizontal scaling:

replicaCount: 1  # Do not increase

For high-availability requirements, use Microcks with external MongoDB (not recommended for mock services).

Resource Limits​

Recommended resources for development:

resources:
  limits:
    cpu: "1"
    memory: 1Gi
  requests:
    cpu: 500m
    memory: 512Mi

Backup and Recovery​

Export Mock Configurations​

# Export all services
kubectl -n staqr-dev exec -it staqr-dev-microcks-0 -- \
  curl http://localhost:8080/api/services/export > microcks-backup.json

Restore After Pod Replacement​

The Helm chart's seed job (job-seed-mocks.yaml) handles this automatically by:

  1. Running as post-install/post-upgrade hook
  2. Importing all WSDL/OpenAPI files from /mocks/

If manual restore is needed:

# Re-import backup
kubectl -n staqr-dev cp microcks-backup.json staqr-dev-microcks-0:/tmp/
kubectl -n staqr-dev exec -it staqr-dev-microcks-0 -- \
  curl -X POST http://localhost:8080/api/artifact/upload \
    -F "file=@/tmp/microcks-backup.json"

Metrics and Alerting​

Prometheus Metrics​

Microcks exposes metrics at /actuator/prometheus:

# ServiceMonitor (if using Prometheus Operator)
apiVersion: monitoring.coreos.com/v1
kind: ServiceMonitor
metadata:
  name: microcks
spec:
  selector:
    matchLabels:
      app: microcks
  endpoints:
    - port: http
      path: /actuator/prometheus

Suggested Alerts​

# Microcks down
- alert: MicrocksDown
  expr: up{job="microcks"} == 0
  for: 5m
  labels:
    severity: warning
  annotations:
    summary: "Microcks mock service is down"

# High error rate
- alert: MicrocksHighErrorRate
  expr: rate(http_server_requests_seconds_count{status=~"5..",app="microcks"}[5m]) > 0.1
  for: 10m
  labels:
    severity: warning

Security Considerations​

  1. No authentication in dev: Microcks runs with KEYCLOAK_ENABLED=false for convenience
  2. Internal access only: Service type is ClusterIP (not exposed externally except via ingress)
  3. Ingress protection: Consider IP allowlisting for admin UI access
  4. No sensitive data: Mock responses should not contain real customer data