Installing the WebHMI Helm Chart Using Rancher

After successfully installing the K3s Kubernetes cluster, you can use Helm to deploy WebHMI. Before installing the WebHMI Helm chart, it is recommended to create the Docker Hub and TLS secrets for secure access. Learn more

To install the WebHMI Helm chart:

  1. Add the Helm repository.

    The repository contains the WebHMI application chart and all its required components—CoreDNS, MetalLB, and the HAProxy Ingress Controller.

    1. In Rancher, open your cluster view, go to Apps > Repositories , and click Create .

    2. Enter the following details, and then click Create to add the repository.

      • In Name , enter a descriptive chart name, such as webhmi .

      • In Index URL , enter https://download.example.com/helm-charts/ .

  2. Install the Helm chart.

    1. In Rancher, navigate to Apps > Charts .

    2. In Repository , select the repository that you added in step 1.

    3. Locate the WebHMI chart and select it. When the detail page opens, click Install to start the guided two-step installation process.

  3. In the first dialog, configure the following basic settings, and then click Next.

    • Namespace: Leave this as default . If you use a different namespace, it must match the namespace that you used when creating your secrets. Learn more

    • Release Name: Enter a unique name, such as ico . This name will be used as a prefix to all Kubernetes resources related to this deployment.

  4. On the Edit Options tab in the next dialog, configure the following General Settings that are required for a functional WebHMI deployment.

    • WebHMI Replica Count —The number of WebHMI pods to run. Adjust based on your cluster’s CPU and memory capacity. You can start with the default value and scale later.

    • FrameWorX Server Connections—The hostname(s) or IP address(s) of the Genesis machine(s) with a FrameWorX server that WebHMI connects to.

      Click Add Row and enter at least one server. You can optionally include a port; for example, hostname:8443. If you add multiple servers, WebHMI load-balances between them.

    • Use HTTPS for FrameWorX Connections —Initially, leave this disabled. To set up secure communication between WebHMI and the FWX server, refer to Securing WebHMI–FrameWorX Communication with HTTPS .

    • WebHMI Hostname (FQDN) —A fully qualified domain name (FQDN) that will be used to access WebHMI through a browser, such as webhmi.example.com . It must resolve to the cluster’s load balancer IP.

    • Enable HTTPS / TLS Secret Name —Enable HTTPS (recommended) and specify the name of the Kubernetes TLS secret that stores your certificate.

  5. In Dependencies, configure the load balancer and ingress controller, using the following settings:

    • Install HAProxy Ingress Controller —Installs the HAProxy ingress controller for routing external traffic into the cluster. Enabled by default; disable only if you are providing your own ingress controller.

    • Enable On-Prem Load Balancer—Select to install MetalLB with the deployment.

      Unlike cloud environments, on-premises clusters do not include a built-in load balancer service. If you plan to use a different load balancer, disable this option.

    • External Load Balancer IP —An unused IP address in your network that MetalLB assigns to the ingress controller, creating the entry point to your cluster.

    • Install CoreDNS —Leave this disabled unless your cluster does not already have a DNS service. When using Rancher with K3s, CoreDNS is included by default.

  6. Configure the optional settings as desired.

    • If you want to use your own display files and choose a storage backend, set up the Storage for Displays section for mounting custom displays via SMB or via NFS .

    • In Autoscaling , enable automatic scaling of WebHMI pods based on CPU and memory usage. Set the minimum and maximum replica counts, the target utilization thresholds, and optionally enable overprovisioning to keep spare capacity ready.

    • In Advanced Settings, you can adjust the logging level, override the container image repository or tag, or customize CPU and memory resource limits.

  7. When finished, click Install to start the deployment. Rancher begins provisioning the WebHMI components. This process may take a few minutes while container images are downloaded and pods start up.

What's Next?

Verifying the Helm Chart Installation in Rancher