Error when running demo setup

[arnodirlam]

Hi there,

I've been running into some problems running the demo setup.

Steps to reproduce

1. Clone repo
2. cd into demo/setup
3. Generate new SSH key (ssh-keygen -t ed25519) without passphrase and move it to ~/.ssh/nomad
4. Add the corresponding public SSH key in the HCloud web UI for the project, name set to nomad
5. Set env vars TF_VAR_hcloud_token and TF_VAR_ssh_key=~/.ssh/nomad
6. Run terraform init
7. Run terraform apply, confirm with yes for running, and yes for storing the SSH key fingerprint to connect

Outcome

• Server coming up with working Consul and Nomad
• In the Nomad web UI, I can see that the autoscaler and prometheus jobs are healthy
• The redis job is not being placed
• There are no other servers coming up
• In the autoscaler logs, I see the following error:

  2022-12-30T17:15:43.363Z [DEBUG] internal_plugin.nomad-target: triggering run of handler garbage collection
  2022-12-30T17:15:43.473Z [WARN]  policy_manager.policy_handler: failed to get target status: policy_id=57cd4826-25ee-702b-9d36-478b6d7085bf
  error=
  | rpc error: code = Unknown desc = failed to parse HCloud target config: 2 error(s) decoding:
  | 
  | * error decoding 'hcloud_location': failed to get Location with id or name equal to fsn1: unable to authenticate (unauthorized)
  | * error decoding 'hcloud_ssh_keys': failed to get SSHKey with id or name equal to nomad: unable to authenticate (unauthorized)

Other info

• I've enabled the Consul web UI and double-checked the secrets/hcloud/token value

---

Any hints on what else I could try to fix this?

Thanks a lot for your great work with maintaining the plugin and setting up the demo! 🙏

[AndrewChubatiuk]

Hi @arnodirlam I've updated demo in #9 PR. SSH keys will be generated by a module and also updated achubatiuk/hcloud-autoscaler:main docker image

[arnodirlam]

Thanks a lot for the swift and helpful response and PR!

I've checked out the demo-update branch, removed the nomad SSH Key from Hetzner Cloud, and ran Terraform again from scratch. Unfortunately, I'm still getting the same errors in the autoscaler job logs.

Just a guess: It's saying "unable to authenticate" for two things that it needs to query from hcloud, and that's before even provisioning a server, so maybe it's related to the HCloud token?

I don't know if there's anything cached that I need to refresh. But I couldn't find anything. The docker image is forced to be pulled every time, too. 😕

[leoenes]

Hi! I'm facing the same error about decoding "hcloud_location"and 'hcloud_ssh_keys'. Do you know why it happens?

[AndrewChubatiuk]

Sorry for late reply, there was an issue in a plugin, I'll release it soon

[AndrewChubatiuk]

Updated image in demo and prepared a new plugin release