Cloud Pak for Integration

Component configuration

This page contains guidance on how to configure the release of all the components of the Integration Cloud Pak. This includes:

  1. Navigator
  2. Asset Repo
  3. App Connect Enterprise
  4. MQ
  5. EventStreams
  6. API Connect
  7. Aspera
  8. Operations Dashboard
  9. Deleting components

App Connect Enterprise guidance

values.yaml example

Configure the following values on your helm chart when you install App Connect Enterprise.

image:
  pullPolicy: Always
  pullSecret: `<deployer-dockercfg pull secret>` #Get this secret by running oc get secrets on the namespace you want to deploy in
license: accept
replicaCount: 1
security:
  initVolumeAsRoot: false
selectedCluster:
- ip: icp-proxy.`<cluster-hostname>` #For example, icp-proxy.cluster-openshift-0819-management.fyre.ibm.com
  label: local-cluster
  namespace: local-cluster
  value: local-cluster
tls:
  hostname: `<cluster-hostname>`

API Connect

Releasing APIC Cheatsheet

Here are some typical values you can use to release a small APIC instance. proxy_ip will be the IP of your master node if you co-located proxy and master.

  • registry: docker-registry.default.svc:5000/apic/
  • image pull secret: <deployer-dockercfg-pull-secret> #Get this secret by running oc get secrets on the namespace you want to deploy in
  • storage class: <block-storage-class-name> #Get storage classes available by running oc get sc
  • helm tls secret: apic-ent-helm-tls
  • FQDNs:
    • mgmt FQDNs: management.<proxy-ip>.nip.io (x4)
    • portal FQDNs: portaldirector.<proxy-ip>.nip.io portalweb.<proxy-ip>.nip.io
    • analytics FQDNs: analyticsingestion.<proxy-ip>.nip.io analyticsclient.<proxy-ip>.nip.io
    • gateway FQDNs: apigateway.<proxy-ip>.nip.io gatewayendpoint.<proxy-ip>.nip.io
  • mode: dev
  • cassandra replicas: 1
  • cassandra volume size: 20GB
  • analytics volume size: 20GB

Deleting a component

If things dont go as planned or you want to remove the component release completely, you can remove by:

  1. Deleting the helm release with:
    helm del --purge <release-name> --tls
    
  2. Removing all components in a specific namespace with:
    kubectl delete pod,sts,ds,sts,job,pvc -n <namespace-name> --all
    
close