'Pool timed out' whilst using daemon

Help
1

Hi,

I’m having a strange error with Atuin even when the daemon is enabled (and listening).

The error is:

Error: pool timed out while waiting for an open connection

Location:
    atuin/src/command/client.rs:108:18

My config is:

auto_sync = false
dialect = "uk"
enter_accept = false
key_path = "$XDG_RUNTIME_DIR/agenix/atuin_key"
secrets_filter = true
sync_address = "<REDACTED>"
workspaces = true

[daemon]
enabled = true
sync_frequency = "120"
systemd_socket = true

[sync]
records = true

I am using ZFS. Can’t recall if that was fixed in Atuin?

The thing is, the daemon should be listening… so I guess the pool error means the client isn’t configured correctly?

Any ideas?

Thanks.

2

Could you share your atuin doctor as well please?

3

No need!

As it turned out, Nix hadn’t activated the daemon. I believe the hook was falling back to direct write, which ran into ZFS on $HOME.

Thanks for the quick reply! :smiley:

1 Like
4

No worries! Glad you got it sorted

5

And fwiw for anyone finding this via google or something;

The daemon is not a 100% cure-all for any ZFS related issues. It is a workaround that does as much as we can to avoid the issue

The upstream ZFS problem:

It is possible that it could happen still, though it would show up in the daemon logs and not in your shell :smiley:

6

Good shout. For those interested in my setup, I created a ZFS zvol for Atuin, on a EXT4 filesystem to mitigate this issue. That way I can be certain my Atuin is syncing fine, and not running into any delays.

My Nix declaration for that mount is here: nixfigs/hosts/nixos/MORPHEUS-LINUX/disks.nix at main · shymega/nixfigs · GitHub