Alexander Iliev

I just saw the 1.1.1 release and [this](https://github.com/siderolabs/talos/commit/6b7c6110c) commit. I'm about to bump the Talos version to 1.1.1 and see if the issue is still there.

With 1.1.1 the behavior actually got worse, but at least it's no longer intermittent - now it happens every time. No node, except for the one which the bootstrap command...

Thanks, the error was: ``` retrying error: no controlplane endpoints discovered yet ``` and then: ``` retrying error: timeout ``` Turns out that removing the `/cluster/etcd/subnet` and `/machine/kubelet/nodeIP/validSubnets` JSON patches...

Looking at the logs of a node that failed to join the control plane, it looks like the private network was brought up quickly enough. Here's a support bundle from...

I also just hit this and it's kind of blocking. I don't see a way around it short of dropping Terraform and using directly the Hetzner Cloud API.

@phreaker0 thanks for the pointer. I tried to bump the log level of the remote SSH server to `DEBUG3`, but this didn't help much. Something I forgot to mention in...

@phreaker0 right, this seems to be the case. Would you consider a PR fixing this via `Temp::File::tempdir`? Basically this (tested only on Linux): ```diff diff --git a/syncoid b/syncoid index f891099..f323497...

@phreaker0 no worries, thanks for getting back! Indeed using the PID is much simpler. Replacing `time()` with `$$` (and leaving `$remoteuser-$rhost`) would also work, but it wouldn't address #530. I...