Installing Component Pack for Connections

The installation package for Component Pack for HCL Connections includes sets of Docker images for each offering and the Helm charts for installing the images.

General requirements

  • Helm v2
  • Kubernetes 1.16 or later, with sufficent resources
  • Docker Registry up and running
  • NFS set up and accessible from Kubernetes worker
Note: If you're using a single-node Kubernetes installation, you need to ensure that it can schedule non-system pods, and that it is tagged as the infrastructure host so that pods that need an infrastructure tag can be scheduled.

All steps in the installation tasks that follow must be run on the node that has Helm v2 installed and configured to communicate with the Kubernetes cluster where Component Pack will be started. The non-root user should handle Helm, and the same user should have kubectl configured against the same Kubernetes cluster.

Environment example

For the sake of example when you do the install, let's assume an environment that has Connections installed and both Customizer and Orient Me enabled (where Orient Me uses ElasticSearch as the backend database):

  • User is accessing Connections from www.mysite.com
  • When you hit www.example.com, you hit the Nginx running on proxy01.internal.mysite.com listening on ports 80 and 443. This proxy is used to decide which requests to send to Customizer and which goes directly to the IBM HTTP Server.
  • IBM HTTP Server is running on port 443 on ihs01.internal.mysite.com
  • The Kubernetes master is k8smaster01.internal.mysite.com. If you're using multiple Kubernetes masters, you need a load balancer in front of them. In that case, k8smaster01.internal.mysite.com is pointing to the load balancer, behind which are one or more Kubernetes masters.
  • Your Docker Registry is running on mydockerregistry01.internal.mysite.com:5000. The user name is admin and the password is password. From each Kubernetes worker you can run docker login –u admin –p password mydockerregistry01.internal.mysite.com:5000
  • Your NFS server is accessible on 192.168.168.192 from all the Kubernetes workers.
  • On the NFS server, you have a folder called /pv-connections, and inside it you have the folder structure described in Setting up persistent volumes with NFS.
  • Your user name for the WasND web console is wasadmin, and your password is waspassword

The following topic steps you through the tasks to get all Component Pack applications up and running. If you aren't using a given application in your environment, you can skip that task.