Figure out checking if given hardware ID is available to use

[invidian]

Right now we ask user to select hardware IDs which should be used for nodes, but we could as well pick any available hardware from Tinkerbell server. In both cases, we need to check if given hardware is ready to run workflows. The easiest way of doing this is just to see if there are any workflows associated with a given hardware.

[gianarb]

I think @mmlb and @splaspood @markyjackson-taulia can give us an idea about how to do this moving forward not for v0.1.0 (tech preview)

we need to check if the given hardware is ready to run workflows.

In practice, we have to identify if give hardware can be a good candidate for provisioning. I think using the previous workflow is an easy way to do it and it will work for v0.1.0 but moving forward we will have to use a combination of factors.

For not it works mainly because deprovisioning and machine update won't be supported in v0.1.0. For v0.1.0 the lifecycle is the simples one as possible I presume:

1. Create a Cluster via kubectl apply
2. Identify nodes that didn't run a workflow yet because it means they are available
3. Run workflow

That's it.

[invidian]

Something like this works fine: https://github.com/tinkerbell/cluster-api-provider-tink/blob/b3153920195a2dbdbfe70271f958aaac508b178e/controllers/tinkerbell.go#L128-L216.

[invidian]

Note, that code above is vulnerable to TOCTOU. In the future we should have some sort of locking mechanism, which will allow us to reserve hardware for a given cluster.

[gianarb]

Note, that code above is vulnerable to TOCTOU. In the future we should have some sort of locking mechanism, which will allow us to reserve hardware for a given cluster.

I still think we should specify a list of hardware ID because I don't want to use all my 10 NUCs for ClusterAPI :P but yes! We also need something smarter than a list! xD

[invidian]

I'm reworking #17 right now to make it based on #8, so we can use most likely labels to figure out if hardware is available.