how to Deploy LiveKit to Kubernetes. (self hosting) and is it necessarily needs a domain for the deployment? I need to be on-prem fully

Thanks for reaching out, Yasser. I'm an AI assistant and I'm here to help. Give me a moment while I research this topic.

is the domain necessary? what if i put internal ip instead

Great news! You can definitely deploy LiveKit using an internal IP instead of a domain, though it does need a few specific tweaks to make sure everything routes smoothly within your private network. Here’s a friendly guide to help you get it set up: Steps to Use an Internal IP: 1. Use the

<http://external-dns.alpha.kubernetes.io/internal-hostname|external-dns.alpha.kubernetes.io/internal-hostname>

annotation in your Service or Ingress resource to point to your internal DNS records. This will link to the internal IP. 2. If you’re working with a Service of type

LoadBalancer

or

ClusterIP

, configure it to use the internal IPs of your nodes. 3. For Ingress, just specify the internal hostname right in the annotations to make sure traffic goes to the internal IP. 4. Here’s a quick YAML example: `metadata: annotations: external-dns.alpha.kubernetes.io/internal-hostname: internal.example.com`This sets up an internal DNS record using your internal IP. 5. Don’t forget to tweak your Helm chart values to include those internal hostnames. You can check out the

server-sample.yaml

file for some handy configuration ideas. 6. Make sure your Service type (

LoadBalancer

,

NodePort

, or

ClusterIP

) is set up just right. If you’re using

NodePort

, you might need to specify whether to use public or private IPs. Follow these steps, and you’ll have LiveKit up and running on an internal IP in no time. If you hit any bumps or want a hand with anything else, just let me know—I’m here to help! Sources: docs/sources/service.md | gclb-sample.yaml

what about the api key and secret is it will be auto generated in this case

Just a heads-up: the API key and secret aren’t automatically created when you deploy LiveKit to Kubernetes, even if you’re using an internal IP. You’ll need to generate them yourself and add them to your deployment setup. Here’s how you can do it: 1. Run this command to generate your API key and secret: `livekit-server generate-keys`Make sure to save the key and secret somewhere safe! 2. Add those credentials to your Kubernetes config. For example, if you’re using the Helm chart, put them in the

keys

section of your

values.yaml

file like this:

livekit:
  keys:
    your-api-key: your-api-secret

3. Or, if you prefer, you can set these up using environment variables or Kubernetes secrets for a more secure approach. If you want, I’m here to help you through any of these steps—just let me know!