"Image not found" from nowhere #681
Description
Hello,
I was working on testing new configurations for NAT GW on my cluster when I got suddenly the following error:
$ ./hetzner-k3s create --config ./cluster_config.yaml
_ _ _ _____
| |__ ___| |_ _____ __ ___ _ __ | | _|___ / ___
| '_ \ / _ \ __|_ / '_ \ / _ \ '__|____| |/ / |_ \/ __|
| | | | __/ |_ / /| | | | __/ | |_____| < ___) \__ \
|_| |_|\___|\__/___|_| |_|\___|_| |_|\_\____/|___/
Version: 2.4.2
[Configuration] Validating configuration...
[Configuration] ...configuration seems valid.
[Instance hacksessible-master1] Creating instance hacksessible-master1 (attempt 1)...
[Instance hacksessible-master1] Creating instance hacksessible-master1 failed: {
"error": {
"code": "invalid_input",
"message": "image not found",
"details": {}
}
}
[Instance hacksessible-master1] Waiting 1 seconds before retry...
[Instance hacksessible-master1] Creating instance hacksessible-master1 (attempt 2)...
[Instance hacksessible-master1] Creating instance hacksessible-master1 failed: {
"error": {
"code": "invalid_input",
"message": "image not found",
"details": {}
}
}
[Instance hacksessible-master1] Waiting 2 seconds before retry...I've tried to change the type of instances and the image without success
DEBUG=true does not provide the HTTP outputs, so I cannot check if it comes from the CLI or from Hetzner.
PS: If you also have a solution for a solid NatGW configuration, I am up to it, I cant use the one from Cilium (because of resource requirements) and the one described in the documentation seems broken (the CLI cannot connect to the master instance to configure it because of the Cloud init script closing its access), I am searching for a "hybrid" configuration, where my node is able to have a public IP, but all my kubernetes related pods etc. share the same NatGW