Run:AI is installed on GPU clusters. These clusters must have outbound internet connectivity to the Run:AI cloud. Details can be found here: https://support.run.ai/hc/en-us/articles/360010227960-Run-AI-GPU-Cluster-Prerequisites under "Network Requirements"
In some organizations, outbound connectivity is proxied. Traffic originating from servers and browsers within the organizations flows through a gateway that inspects the traffic, calls the destination and returns the contents.
Organizations sometimes employ a further security measure by signing packets with an organizational certificate. The software initiating the HTTP request must acknowledge this certificate, otherwise, it would interpret it as a man-in-the-middle attack.
In-case the certificate is not trusted (or is a self-signed certificate), this certificate must be included in Run:AI configuration for outbound connectivity to work.
The instructions below receive as input a certificate file from the organization and deploy it into the Run:AI cluster so that traffic originating in Run:AI will recognize the organizational proxy server.
A Run:AI cluster is installed by accessing the Administrator User Interface at app.run.ai downloading a yaml file runai-operator.yaml and then applying it to Kubernetes. You must edit the yaml file. Search for httpProxy
Set enabled to true and paste the contents of the certificate under tlsCert