[WIP] 📖 Add docs for writing provider integrations #30
Conversation
Adding docs for writing provider integrations Signed-off-by: janiskemper <janis.kemper@syself.com>
jschoone
added
the
Container
There was a problem hiding this comment.
I kept in mind that the state is WIP, so this "review" is more of a reminder of the current state where most details should be added in the ongoing process. I haven't commented on the most important but empty section "Implementing a new Provider Integration" in this review.
docs/PROVIDER.md
Outdated
|
|
||
| In case that there are multiple node images which have to be provided to the user, it makes sense to have separate custom resources so that one custom resource is responsible for only one node image. A pattern where there is a `ProviderClusterStackRelease` that creates different `ProviderNodeImageRelease` custom resources, one for each node image that should be built, is advisable. | ||
|
|
||
| The same pattern that is also used in the objects of `ClusterStackRelease` and `ProviderClusterStackRelease` can be also applied here in a way that `ProviderClusterStackRelease` is set on `ready: true` if and only if all `ProviderNodeImageRelease` custom resources are also `ready: true`. |
There was a problem hiding this comment.
Same here as before. As long as there is no common structure for ProviderNodeImageRelease specs will most likely differ fundamentally resulting in bad readability on provider's code.
There was a problem hiding this comment.
I don't agree here. Why should that be the case? Reliability of code will be mostly about how people write their code, not about how they name their API. And we cannot control how they write the code anyway.
Maybe I misunderstand also what you mean, so you can elaborate maybe?
There was a problem hiding this comment.
It's about using Go's rudimentary support for interfaces to enforce / provide a certain structure for resources expected to be handled the same way for each provider implementation anyway not how the result is achieved using provider specific code that will be fundamentally different.
Take a look at CSI drivers as an example. They all implement common interfaces and work with storage classes. Commonly under parameters the filesystem type was and sometimes is defined as fstype. This meant that is was not always named the same way and moving from cloud to cloud got complicated. These days the external-provisioner handles PVC requests and storage classes. They provided the common name [1] csi.storage.k8s.io/fstype to handle the filesystem type the same way for all CSI drivers [2] and the common interface they share (it's used in e.g. NodeStageVolume under req.VolumeCapability).
[1] https://kubernetes-csi.github.io/docs/external-provisioner.html#storageclass-parameters
[2] https://github.com/kubernetes-csi/external-provisioner/blob/b255ee8f41384b330388c00a5f9252aff343fecf/pkg/controller/controller.go#L612
There was a problem hiding this comment.
We evaluated an interface approach, taking the Cloud Controller Manager as example. However, this is unfortunately not possible, same as with CAPI Provider Integrations, as provider-specific CRDs are needed!
| ## Implementing a new Provider Integration | ||
|
|
||
| // TODO This has to be refined based on https://cluster-api.sigs.k8s.io/developer/providers/implementers-guide/overview. | ||
| // The goal is that users can set up new repositories by themselves and know everything that is needed. |
There was a problem hiding this comment.
It seems that this part will be the most important, so I would recommend addressing it before we merge this PR.
|
@michal-gubricky @NotTheEvilOne @jschoone I have added the documentation that we talked about in yesterday's meeting. I updated the previous section about best practices to make it clearer. @michal-gubricky if there are some open questions to you that I haven't answered yet, please tell me! |
What this PR does / why we need it:
Adding docs for writing provider integrations
Which issue(s) this PR fixes:
Fixes #29
TODOs: