Appearance
Ingresses
Create a Kubernetes Ingress to give your web application a public address under apps.remotegpu.ai. The ingress sends requests to a service in your namespace, which passes them to your application's pods.
Before you create an ingress
You need these resources and settings:
- A running application to receive requests.
- A
ClusterIPservice that connects to the application's pods. - A host that matches
<prefix>--<namespace-token>.apps.remotegpu.ai. - The ingress class
traefik, which selects RemoteGPU's ingress controller.
Create an ingress
- Open Kubernetes / Ingresses.
- Select a namespace.
- Click
Create ingress. - Choose the target service.
- Choose the backend port.
- Enter the ingress details.
| Console field | Description |
|---|---|
Ingress name | Kubernetes ingress name in the namespace |
Host prefix | The name to put before your namespace token in the public hostname |
Path | The URL path to send to this service |
Service | Existing service that should receive traffic |
Port | The port to connect to on the service |
The console shows the generated hostname before you create the ingress. For a namespace token such as 123456780000, a host prefix of demo-web becomes demo-web--123456780000.apps.remotegpu.ai.
RemoteGPU sets up HTTPS for the generated apps.remotegpu.ai hostname.
Supported ingress settings
The ingress must use these settings and point to a service in the same namespace.
| Area | Description |
|---|---|
| API version | networking.k8s.io/v1 Ingress |
| Backend type | HTTP rules that send requests to a service |
| Namespace boundary | The ingress must point to services in the same namespace |
| Hostnames | Hosts must match <prefix>--<namespace-token>.apps.remotegpu.ai for the target namespace |
| Ingress class | traefik |
| Path routing | Standard HTTP path routing with pathType |
| TLS | Managed apps.remotegpu.ai hosts use the platform wildcard certificate |
Set spec.ingressClassName: traefik, or omit the field to use traefik by default. The older kubernetes.io/ingress.class annotation is not supported.
Advanced ingress options
Route requests by URL path using Kubernetes Ingress rules. If your app needs URL rewrites or redirects, handle them in the application. RemoteGPU does not support Traefik Middleware custom resources (CRDs) or router middleware annotations in your namespace.
Configure request body size limits in your application. RemoteGPU does not provide a per-ingress setting for this limit.
RemoteGPU handles HTTPS for apps.remotegpu.ai hosts with its wildcard certificate. You do not need to provide a certificate. Leave out cert-manager.io/cluster-issuer and spec.tls for these hosts.
Limits
| Pattern | Support |
|---|---|
NodePort or externalIPs service exposure | Use ingress-backed HTTP routing instead |
| Hostnames that do not match the required format for your namespace | Rejected |
defaultBackend | Not supported |
| Non-service ingress backends | Not supported |
| Per-ingress cert-manager annotations or custom TLS secrets | Not supported for managed apps.remotegpu.ai hosts |
kubernetes.io/ingress.class annotation | Not supported. Use spec.ingressClassName: traefik |
nginx.ingress.kubernetes.io/* annotations | Not supported |
| Creating or editing ingress-controller transport annotations | Not supported |
Traefik Middleware CRDs and router middleware annotations | Not supported |
| Unsupported Traefik annotations or CRDs | Not supported |
| Setting up a custom domain in the Kubernetes console | Not supported |
Troubleshooting
| Symptom | What to check |
|---|---|
| Ingress creation is rejected because of the host | Use the hostname format shown in the error: <prefix>--<namespace-token>.apps.remotegpu.ai |
| The generated URL returns no application response | Check that the ingress points to an existing ClusterIP service with ready pods |
| The console shows an ingress error | Check whether the target service or service port was removed |
| HTTPS configuration is rejected | Do not add spec.tls or cert-manager annotations for managed apps.remotegpu.ai hosts |
| Large requests fail | Check application, upload proxy, or client limits |
Read next
- Read Services to connect the ingress to your application's pods.
- Read Deployments to run the workload first.