What question do you have?:
Is there a way to configure the contour and envoy pods to use private images when using the Contour Gateway Provisioner? When installing it following the instructions listed here: https://projectcontour.io/getting-started/#:~:text=release%2Dcontour%2Denvoy-,Option%203%3A%20Contour%20Gateway%20Provisioner,-The%20Gateway%20provisioner, I have the ability to modify the image location of the projectcontour namespace deployment by manually editing the spec.containers.image field for the deployment resource.
However, since this is a dynamic setup, when I create a gateway resource in my application namespace, that then creates the new contour and envoy pods in that application namespace by design. The problem is, those pods all point to the public images. (ie. ghcr.io/projectcontour/contour:v1.33.2, docker.io/envoyproxy/envoy:distroless-v1.35.8). I don't see anything in the https://projectcontour.io/quickstart/contour-gateway-provisioner.yaml file which allows me to modify the image locations for those resources either. If I manually update the image location in the deployment and daemonset resources in the application namespace, they just get reverted after a few minutes.
I need a way to use the dynamic provisioner and have it support an air-gapped environment. I've tried using the helm install (https://projectcontour.io/getting-started/#:~:text=2/2%20Ready-,Option%202%3A%20Helm,-This%20option%20requires), but that does not seem to work for my specific use case so I need the dynamic provisioner method to work.
Anything else you would like to add:
[Miscellaneous information that will assist in solving the issue.]
Environment:
- Contour version: 1.33.2
- Kubernetes version: (use
kubectl version): 1.34.2
- Kubernetes installer & version: Tanzu TKGS
- Cloud provider or hardware configuration: Tanzu TKGS
- OS (e.g. from
/etc/os-release):
What question do you have?:
Is there a way to configure the contour and envoy pods to use private images when using the Contour Gateway Provisioner? When installing it following the instructions listed here: https://projectcontour.io/getting-started/#:~:text=release%2Dcontour%2Denvoy-,Option%203%3A%20Contour%20Gateway%20Provisioner,-The%20Gateway%20provisioner, I have the ability to modify the image location of the projectcontour namespace deployment by manually editing the spec.containers.image field for the deployment resource.
However, since this is a dynamic setup, when I create a gateway resource in my application namespace, that then creates the new contour and envoy pods in that application namespace by design. The problem is, those pods all point to the public images. (ie. ghcr.io/projectcontour/contour:v1.33.2, docker.io/envoyproxy/envoy:distroless-v1.35.8). I don't see anything in the https://projectcontour.io/quickstart/contour-gateway-provisioner.yaml file which allows me to modify the image locations for those resources either. If I manually update the image location in the deployment and daemonset resources in the application namespace, they just get reverted after a few minutes.
I need a way to use the dynamic provisioner and have it support an air-gapped environment. I've tried using the helm install (https://projectcontour.io/getting-started/#:~:text=2/2%20Ready-,Option%202%3A%20Helm,-This%20option%20requires), but that does not seem to work for my specific use case so I need the dynamic provisioner method to work.
Anything else you would like to add:
[Miscellaneous information that will assist in solving the issue.]
Environment:
kubectl version): 1.34.2/etc/os-release):