Hi All,
we use Helm 2.x to create new deployment configs on OKD 3.10 cluster and we
configure the image stream tag to deploy a new application pointing to a
specific tag. The template used for a deploy is reported below as example:

# Source: ocp-generic/templates/04_service.yaml
apiVersion: v1
kind: Service
metadata:
  name: "test-api-dev"
  namespace: "ocp-dev"
  labels:
    application: "test-api-dev"
spec:
  ports:
   - name: "http"
     port: 8080
     targetPort: 8080
     protocol: "TCP"
   - name: "https"
     port: 443
     targetPort: 443
     protocol: "TCP"
  selector:
    deploymentconfig: "test-api-dev"
  sessionAffinity: None
  type: ClusterIP
---
# Source: ocp-generic/templates/03_deploymentconfig.yaml
apiVersion: apps.openshift.io/v1
kind: DeploymentConfig
metadata:
  name: "test-api-dev"
  namespace: "ocp-dev"
  labels:
    application: "test-api-dev"
spec:
  replicas: 1
  selector:
    deploymentConfig: "test-api-dev"
  strategy:
    type: Rolling
    rollingParams:
      updatePeriodSeconds: 1
      intervalSeconds: 1
      timeoutSeconds: 180
  triggers:
   - type: ConfigChange
   - type: ImageChange
     imageChangeParams:
       automatic: false
       containerNames:
        - "test-api-dev"
       from:
         kind: ImageStreamTag
         name: "test-api:latest"
         namespace: "ocp-dev"
  template:
    metadata:
      labels:
        deploymentConfig: "test-api-dev"
        application: "test-api-dev"
        date: "1591894524"
    spec:
      dnsPolicy: ClusterFirst
      restartPolicy: Always
      hostAliases:
      securityContext: {}
      terminationGracePeriodSeconds: 60
      terminationMessagePath: /dev/termination-log
      containers:
       - name: "test-api-dev"
         image: "docker-registry.default.svc:5000/ocp-dev/test-api:latest"
         imagePullPolicy: Always
         ports:
          - containerPort: 8080
            protocol: "TCP"
          - containerPort: 443
            protocol: "TCP"
         env:
          - name: "SYSLOG_LOGLEVEL"
            value: "DEBUG"
         readinessProbe:
          initialDelaySeconds: 20
          timeoutSeconds: 5
          periodSeconds: 30
          httpGet:
             path: /actuator/health
             port: 8080
             scheme: HTTP
---
# Source: ocp-generic/templates/02_imagestream.yaml
kind: ImageStream
apiVersion: image.openshift.io/v1
metadata:
  name: "test-api"
  namespace: "ocp-dev"
  labels:
    application: "test-api-dev"
spec:
  dockerImageRepository: "docker-registry.default.svc:5000/ocp-dev/test-api"
  tags:
    - name: "latest"
---
# Source: ocp-generic/templates/05_route.yaml
apiVersion: route.openshift.io/v1
kind: Route
metadata:
  name: "test-api-dev"
  namespace: "ocp-dev"
  labels:
    application: "test-api-dev"
spec:
  host: "test-api-dev.test.local"
  port:
    targetPort: "http"
  path: "/"
  tls:
    termination:  edge
    insecureEdgeTerminationPolicy: Redirect
    certificate: |-

    key: |-

    caCertificate: |-

    destinationCACertificate: |-

  to:
    kind: Service
    name: "test-api-dev"
    weight: 100
  wildcardPolicy: None

After the first deploy we noticed that the configuration of deployments
config reports the image parameter under container section with the sha256
value and not the image tag value. We also noticed that the configuration
reports the lastTriggerImage parameter with the same value. If we tried to
release a new DC the configuration points to the first shs256 and not the
new one.
If we remove manually from the DC YAML config the lastTriggerImage param
and we override the image sha256 tag with the image stream tag the
deployment work fine and doesn't change the config.

Is it a normal behavior? Is it possible to avoid this config with the
sha256?

Thanks a lot in advance,
Marcello
_______________________________________________
users mailing list
users@lists.openshift.redhat.com
http://lists.openshift.redhat.com/openshiftmm/listinfo/users

Reply via email to