audit: update as of 2021-06-02

[cncf-ci]

Audit Updates wg-k8s-infra

[k8s-ci-robot]

Hi @cncf-ci. Thanks for your PR.

I'm waiting for a kubernetes member to verify that this patch is reasonable to test. If it is, they should reply with /ok-to-test on its own line. Until that is done, I will not automatically test new commits in this PR, but the usual testing commands by org members will still work. Regular contributors should join the org to skip this step.

Once the patch is verified, the new status will be reflected by the ok-to-test label.

I understand the commands that are listed here.

Details

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/test-infra repository.

[spiffxp]

/ok-to-test

[spiffxp]

I'm confused. #2089 landed 13h ago. It updated the audit script to dump these files, and preemptively included them from a manual run.

This is behaving like either:

• the audit script hasn't been updated
• the service account running this doesn't have sufficient bigquery read access to see these resources

Going to check log files

[spiffxp]

#2100 should have fixed, but it broke the script's ability to run in CI

#2104 should fix those problems

[spiffxp]

#2120 - reviewed using the script here

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: cncf-ci, spiffxp

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

The pull request process is described here

Details

Needs approval from an approver in each of these files:

• OWNERS [spiffxp]

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

[spiffxp]

This was a huge pain to review, my browser kept hanging on the "files changed" page. I used the script from #2120 to categorize changes, see that PR for everything checked specifically. But I'll try to sum up the followup issues:

• buckets should not have ACLs enabled (they should be using IAM at the bucket level instead)
  • the vast majority of this is buckets created by kube-up.sh -> need to find out if kube-up.sh requires per-object ACLs
  • the next largest pattern is GCR buckets that our scripts didn't pre-provision
  • the kubernetes_public_billing bucket has them but I don't think it needs them
• non-prod buckets shouldn't have retention enabled
• there are some k8s-staging buckets that do
• storage class should always be STANDARD
  • kubernetes-public has bucket(s) that are not
• audit logs should always be present
  • there are some projects that lack system_event or had it removed by this audit PR, unclear if this is a visibility issue, or the log is truly missing
• logs created by e2e tests should not be present
  • I thought we had disabled logging for these, but the resources are showing up. Would rather not allowlist logs resources, because we want to catch unexpected new logs; OTOH this would be an annoying denylist to maintain.

The manual changes leftover after running that script that I was ok with:

• k8s-infra-ii-sandbox changes: I'm mostly ignoring these, they are usually non-k8s-infra-code-driven or manual changes which is fine for a sandbox
• k8s-staging-cluster-api-gcp changes due to image-builder/capg: add more permissions to the service account #2061