Skip to content

Conversation

@ethnt
Copy link

@ethnt ethnt commented Jan 15, 2022

I re-implemented the code in #1157 after bringing it up to date with master. I think this would solve the issue mentioned at #904. If I need to add typing, please let me know 🙏

To add a bit of color: I had an initial deployment that failed (mostly related to usage of sops-nix, but that feels unrelated). Subsequent deploys weren't able to be performed because it would never use the generated SSH key. This seems like it could be a semi-common condition — however, I'm not sure about the reasoning behind the original conditional.

@roberth
Copy link
Member

roberth commented Mar 1, 2023

Hi @ethnt,

According to the git blame, we switched to setting the authorized_keys via the NixOS configuration at some point 8ad1138

The original issue describes a situation where an activation did happen, in which case this change doesn't seem useful, unless NixOps is unaware that the activation succeeded. Maybe it isn't quite sure and the cur_toplevel is only set when it's absolutely sure that the activation succeeded? In that case --force-reboot does indeed explain it.

It seems like this is something we could write a regression test for, based on, or in: https://github.com/NixOS/nixops/blob/master/integration-tests/default.nix

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants