NO-JIRA: Introduce KMS encryption mode in encryption library

[ardaguclu]

This PR introduces KMS encryption mode into encryption library in library-go. Since there is no keys in KMS, it uses hashsum of KMS Configuration as well as KeyID returned by KMS plugin (by calling an grpc call to Status endpoint).

This PR relies on the building blocks of encryption library that works for other encryption types.

#1876 was a good attempt. This PR adopts some bits from there.

[openshift-ci-robot]

@ardaguclu: This pull request explicitly references no jira issue.

In response to this:

This PR introduces KMS encryption mode into encryption library in library-go. Since there is no keys in KMS, it uses hashsum of KMS Configuration as well as KeyID returned by KMS plugin (by calling an grpc call to Status endpoint).

This PR relies on the building blocks of encryption library that works for other encryption types.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the openshift-eng/jira-lifecycle-plugin repository.

[openshift-ci]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by: ardaguclu
Once this PR has been reviewed and has the lgtm label, please assign deads2k for approval. For more information see the Code Review Process.

The full list of commands accepted by this bot can be found here.

Needs approval from an approver in each of these files:

• OWNERS

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[ardaguclu]

/hold

[ardaguclu]

Opened this openshift/cluster-kube-apiserver-operator#1951 to verify in terms of regression
/cc @flavianmissi @dgrisonnet

[ardaguclu]

/retest

[flavianmissi]

I think depending on the kms plugin to generate the unix socket path creates a fundamental chicken-egg problem where the unix socket path is generated based on a gRPC call to the plugin's Status endpoint, but the plugin requires a unix socket path before it can run. Unless I'm missing something crucial, this simply cannot work... right?

My suggestion is that, for dev preview, we just use the KeyArn from the APIServer config as the key_id. This won't be an issue with the aws kms plugin because the key_id is always matches the KeyArn. This is okay for the aws kms plugin because it does not change the key_id when the key is rotated, so this approach is only good for dev preview. We can refactor it later to something more robust.

What do y'all think?

[ardaguclu]

I think depending on the kms plugin to generate the unix socket path creates a fundamental chicken-egg problem where the unix socket path is generated based on a gRPC call to the plugin's Status endpoint, but the plugin requires a unix socket path before it can run. Unless I'm missing something crucial, this simply cannot work... right?

Unix path is generated based on the hashsum of kmsconfig. It does not include key_id retrieved from Status endpoint. Otherwise, it would indeed be a chicken-egg problem.

[openshift-ci]

@ardaguclu: The following test failed, say /retest to rerun all failed tests or /retest-required to rerun all mandatory failed tests:

Test name	Commit	Details	Required	Rerun command
ci/prow/unit	4a45d50	link	true	/test unit

Full PR test history. Your PR dashboard.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository. I understand the commands that are listed here.