Quickstart example DO not working

[BoKKeR]

I have tried deploying 2 times, the steps finish successfully, the droplets are created. When trying to access the IP I get:

default backend - 404

When accessing the host I get an SSL error.

Could you verify that the config is still valid for DO?

Also, does the server droplet have to be the same size as the worker node? Isn't that wasteful?

[bashofmann]

You have to use the hostname. The ingress controller inside the cluster routes incoming traffic to the Rancher pods via the hostname. If you just access it with the IP address, it does not know where to route to and returns a 404.

The SSL error/warning in the browser is expected, these quickstarts use self-signed certificates. You can just skip the error and trust the certificate. If you configure your own DNS names, instead of a xip.io one, you can also use Let'sEncrypt or provide your own certificates. See https://rancher.com/docs/rancher/v2.x/en/installation/install-rancher-on-k8s/#4-choose-your-ssl-configuration.

[Canuckaholic]

I'm trying the DO QuickStart and while running terraform apply --auto-approve it errors with:

Error: RKE version is not supported [v1.19.7-rancher1-1 v1.19.6-rancher1-1 v1.19.4-rancher1-1 v1.19.3-rancher1-1 v1.19.2-rancher1-1 v1.18.9-rancher1-1 v1.18.6-rancher1-2 v1.18.3-rancher2-1 v1.17.17-rancher1-1 v1.17.16-rancher1-1 v1.17.14-rancher1-1 v1.17.13-rancher1-2 v1.17.13-rancher1-1 v1.17.12-rancher1-1 v1.17.11-rancher1-1 v1.17.6-rancher2-2 v1.17.5-rancher1-1 v1.17.4-rancher1-2 v1.17.4-rancher1-1 v1.17.3-rancher1-1 v1.17.2-rancher1-2 v1.17.2-rancher1-1 v1.17.0-rancher1-2 v1.17.0-rancher1-1 v1.16.15-rancher1-4 v1.16.15-rancher1-2 v1.16.14-rancher1-1 v1.16.10-rancher2-2 v1.16.10-rancher2-1 v1.16.9-rancher1-1 v1.16.8-rancher1-1 v1.16.6-rancher1-1 v1.16.3-rancher1-1 v1.16.2-rancher2-1 v1.16.2-rancher1-1 v1.16.1-rancher1-1 v1.15.12-rancher2-7 v1.15.12-rancher2-5 v1.15.12-rancher2-4 v1.15.12-rancher2-2 v1.15.12-rancher1-1 v1.15.11-rancher1-3 v1.15.11-rancher1-1 v1.15.6-rancher1-2 v1.15.5-rancher1-2 v1.15.4-rancher1-2 v1.15.3-rancher1-1 v1.14.10-rancher1-1 v1.14.10-rancher1-0 v1.14.9-rancher1-2 v1.14.9-rancher1-1 v1.14.8-rancher1-1 v1.14.7-rancher1-1 v1.14.6-rancher1-1 v1.14.5-rancher1-1 v1.14.1-rancher1-2 v1.13.12-rancher2-1 v1.13.12-rancher1-1 v1.13.11-rancher1-1 v1.13.9-rancher1-1 v1.13.7-rancher1-1 v1.13.5-rancher1-2 v1.13.5-rancher1-1 v1.13.4-rancher1-2 v1.13.4-rancher1-1 v1.13.1-rancher1-2 v1.13.1-rancher1-1 v1.12.10-rancher1-2 v1.12.9-rancher1-1 v1.12.7-rancher1-4 v1.12.7-rancher1-3 v1.12.7-rancher1-2 v1.12.6-rancher1-1 v1.12.5-rancher1-2 v1.12.3-rancher1-1 v1.12.1-rancher1-1 v1.11.9-rancher1-2 v1.11.9-rancher1-1 v1.11.8-rancher1-1 v1.11.3-rancher1-1 v1.11.1-rancher1-1 v1.10.11-rancher1-1 v1.10.5-rancher1-2 v1.10.5-rancher1-1 v1.10.1-rancher2-1 v1.9.7-rancher2-2 v1.9.5-rancher1-1 v1.8.11-rancher2-1] got v1.18.12-rancher1-1

  on ../rancher-common/rancher.tf line 16, in resource "rancher2_cluster" "quickstart_workload":
  16: resource "rancher2_cluster" "quickstart_workload" {

[bashofmann]

@Canuckaholic Just tried it out and works fine for me. Did you modify rke_kubernetes_version or workload_kubernetes_version ?

[Canuckaholic]

Hello. So I figured it out. You will get this error if you try to create a droplet on DO that is anything less than 4GB.

Adrian (AJ) 778-772-7787

On Feb 25, 2021, at 12:22 PM, Bastian Hofmann notifications@github.com wrote:

@Canuckaholic Just tried it out and works fine for me. Did you modify rke_kubernetes_version or workload_kubernetes_version ?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.

[bashofmann]

Closing this issue. The default droplet size is s-2vcpu-4gb.

[senica]

EDIT: Hmm, I just ran the same command again terraform apply auto-approve and it went through on the second go round without changing anything.

Would you happen to have your example terraform tfvars file handy? I'm using the default and only modified the do_token and password and am getting this error with Digital Ocean. It creates a 4GB, but still errors out with the same error:

module.rancher_common.rancher2_cluster.quickstart_workload: Creating...
╷
│ Error: RKE version is not supported [v1.21.5-rancher1-1 v1.21.4-rancher1-1 v1.20.11-rancher1-2 v1.20.10-rancher1-1 v1.20.9-rancher1-1 v1.20.8-rancher1-1 v1.20.6-rancher1-1 v1.20.5-rancher1-1 v1.20.4-rancher1-1 v1.19.15-rancher1-1 v1.19.14-rancher1-1 v1.19.13-rancher1-1 v1.19.12-rancher1-1 v1.19.10-rancher1-1 v1.19.9-rancher1-1 v1.19.8-rancher1-1 v1.19.7-rancher1-1 v1.19.6-rancher1-1 v1.19.4-rancher1-1 v1.19.3-rancher1-2 v1.19.3-rancher1-1 v1.19.2-rancher1-1 v1.18.18-rancher1-2 v1.18.18-rancher1-1 v1.18.17-rancher1-3 v1.18.17-rancher1-2 v1.18.17-rancher1-1 v1.18.16-rancher1-3 v1.18.16-rancher1-2 v1.18.16-rancher1-1 v1.18.15-rancher1-3 v1.18.15-rancher1-2 v1.18.15-rancher1-1 v1.18.14-rancher1-3 v1.18.14-rancher1-2 v1.18.14-rancher1-1 v1.18.10-rancher1-2 v1.18.9-rancher1-1 v1.18.8-rancher1-1 v1.18.6-rancher1-2 v1.18.6-rancher1-1 v1.18.3-rancher2-2 v1.18.3-rancher2-1 v1.17.17-rancher2-3 v1.17.17-rancher2-2 v1.17.17-rancher2-1 v1.17.17-rancher1-3 v1.17.17-rancher1-2 v1.17.17-rancher1-1 v1.17.16-rancher1-3 v1.17.16-rancher1-2 v1.17.16-rancher1-1 v1.17.14-rancher1-1 v1.17.13-rancher1-1 v1.17.12-rancher1-1 v1.17.9-rancher1-2 v1.17.9-rancher1-1 v1.17.6-rancher2-2 v1.17.6-rancher2-1 v1.17.5-rancher1-1 v1.17.4-rancher1-2 v1.17.4-rancher1-1 v1.17.3-rancher1-1 v1.17.2-rancher1-2 v1.17.0-rancher1-2 v1.17.0-rancher1-1 v1.16.15-rancher1-4 v1.16.15-rancher1-3 v1.16.15-rancher1-2 v1.16.15-rancher1-1 v1.16.14-rancher1-1 v1.16.13-rancher1-1 v1.16.10-rancher2-2 v1.16.10-rancher2-1 v1.16.9-rancher1-1 v1.16.8-rancher1-2 v1.16.8-rancher1-1 v1.16.7-rancher1-1 v1.16.6-rancher1-2 v1.16.6-rancher1-1 v1.16.4-rancher1-1 v1.16.3-rancher1-1 v1.16.2-rancher2-1 v1.16.2-rancher1-1 v1.16.1-rancher1-1 v1.15.12-rancher2-8 v1.15.12-rancher2-7 v1.15.12-rancher2-6 v1.15.12-rancher2-5 v1.15.12-rancher2-4 v1.15.12-rancher2-3 v1.15.12-rancher2-2 v1.15.12-rancher1-1 v1.15.11-rancher1-3 v1.15.11-rancher1-1 v1.15.11-rancher1-0 v1.15.10-rancher1-1 v1.15.9-rancher1-1 v1.15.7-rancher1-1 v1.15.6-rancher1-2 v1.15.5-rancher2-2 v1.15.5-rancher1-2 v1.15.5-rancher1-1 v1.15.4-rancher1-1 v1.15.3-rancher1-1 v1.15.2-rancher1-1 v1.15.0-rancher1-1 v1.14.10-rancher1-1 v1.14.10-rancher1-0 v1.14.9-rancher1-1 v1.14.8-rancher2-1 v1.14.8-rancher1-1 v1.14.7-rancher1-1 v1.14.6-rancher1-1 v1.14.5-rancher1-1 v1.14.3-rancher1-1 v1.14.1-rancher1-2 v1.14.1-rancher1-1 v1.13.12-rancher2-1 v1.13.12-rancher1-1 v1.13.11-rancher1-1 v1.13.10-rancher1-2 v1.13.10-rancher1-1 v1.13.9-rancher1-2 v1.13.9-rancher1-1 v1.13.7-rancher1-1 v1.13.5-rancher1-3 v1.13.5-rancher1-2 v1.13.5-rancher1-1 v1.13.4-rancher1-2 v1.13.4-rancher1-1 v1.13.1-rancher1-2 v1.13.1-rancher1-1 v1.12.10-rancher1-2 v1.12.10-rancher1-1 v1.12.9-rancher1-1 v1.12.7-rancher1-4 v1.12.7-rancher1-3 v1.12.7-rancher1-2 v1.12.7-rancher1-1 v1.12.6-rancher1-2 v1.12.6-rancher1-1 v1.12.5-rancher1-2 v1.12.5-rancher1-1 v1.12.4-rancher1-1 v1.12.3-rancher1-1 v1.12.1-rancher1-1 v1.11.9-rancher1-3 v1.11.9-rancher1-1 v1.11.8-rancher1-1 v1.11.6-rancher1-1 v1.11.5-rancher1-1 v1.11.3-rancher1-1 v1.11.2-rancher1-2 v1.11.2-rancher1-1 v1.11.1-rancher1-1 v1.10.12-rancher1-1 v1.10.11-rancher1-1 v1.10.5-rancher1-2 v1.10.5-rancher1-1 v1.10.3-rancher2-1 v1.10.1-rancher2-1 v1.10.0-rancher1-1 v1.9.7-rancher2-2 v1.9.7-rancher2-1 v1.9.5-rancher1-1 v1.8.11-rancher2-1] got v1.20.10-rancher1-2
│ 
│   with module.rancher_common.rancher2_cluster.quickstart_workload,
│   on ../rancher-common/rancher.tf line 25, in resource "rancher2_cluster" "quickstart_workload":
│   25: resource "rancher2_cluster" "quickstart_workload" {

[bashofmann]

@senica Should be fixed with https://github.com/rancher/quickstart/commit/8e9a85467cb9e4dc6ed86c3a06438d5f9782dd84. It seems the specific RKE version has been removed from the list of supported versions wich is provided to Rancher by https://github.com/rancher/kontainer-driver-metadata.