Rhino Compute Setup

The headless Rhino that solves definitions, on its own Windows machine separate from the app server.

Prerequisites

  • Windows Server or VM (Azure, AWS, or on-premises)
  • A Rhino license: Core Hour Billing, standalone, or Cloud Zoo

Steps

  1. Create a VM. Azure · AWS · or on-premises Windows Server.

  2. Install Rhino Compute. Follow the Deploy to IIS guide.

  3. Install the VektorNode fork. It carries changes Selva relies on: block instances, and the definition_not_cached reply that makes pointer reuse safe. In PowerShell as Administrator on the compute VM:

    irm https://raw.githubusercontent.com/VektorNode/compute.rhino3d/8.x.selva/script/update_compute_server/update_compute_selva.ps1 | iex

    The script takes no arguments. It downloads the latest 8.x.selva build, stops the RhinoComputeAppPool app pool and the Rhino.Compute site, backs up the current deployment to C:\RhinoComputeBackups\ (keeping the last five), swaps in the new rhino.compute and compute.geometry folders under C:\inetpub\wwwroot\aspnet_client\system_web\4_0_30319\, restarts IIS, and hits http://localhost/healthcheck to confirm. On failure it rolls back to the previous backup. Logs land in C:\Logs\RhinoCompute\.

  4. Enable block instances. Set RHINO_COMPUTE_CREATE_HEADLESS_DOC=true and restart the service.

  5. Install the Selva plugin. Under the rhino.compute user account, open the Grasshopper Package Manager and install Selva plus any other plugins your definitions need. Restart afterwards. Plugins installed as a different user aren’t found at solve time.

Updating

Re-run the same script as Administrator; it always pulls the latest 8.x.selva build and keeps a rollback backup.

Connecting

Register the server at /admin/compute with its URL and an optional API key. The data provider persists this config; there is no env var for it.