sshd: no hostkeys available -- exiting
denver-s opened this issue · comments
In variables.tf I did not set any signer_target_host_key as I did not understand what it is and how to generate, furthermore it seemed optional. My pod tezos-remote-signer-forwarder-mybaker-0 gives me this error:
It's explained here, but yes, the README should be more clear
https://tezos-docs.midl.dev/deploy-remote-signer.html#ssh-endpoint-host-key
For now I commented this line of tezos-remote-signer-forwarder.sh to make it work.
ok, but I am not sure if it will work without the host key. You should generate one, so if your cluster breaks for any reason, and you recreate it, your signer will connect to it as if nothing happened... in the absence of this, it will refuse to connect.
By the way, this is the source code for the docs. Feel free to open PRs, you can do it directly on github.
Yes, I will generate a host key. I was wondering if there is a guide to it or is it just a ssh-keygen command to execute on my terminal.
By the way, this is the source code for the docs. Feel free to open PRs, you can do it directly on github.
Thank you, I will.
I didn't write it down, I'd appreciate if you add it to the guide. Something like:
Thanks. I did this way, deleted the pod and it worked!
variable "signer_target_host_key" {
type = string
default = <<-EOK
-----BEGIN OPENSSH PRIVATE KEY-----
<the key>
-----END OPENSSH PRIVATE KEY-----
EOK
}
@nicolasochem As a side note, it's probably needed to perform a check for non-optional values to be provided before Terraform execution.
Mmm I see. Still, I think some kind of checks need to be implemented.