Post April 2023 ACLs no longer allowed by default so wait for changes

[kumichou]

what

• Wait a little while after configuring bucket ownership controls on a new S3 bucket before configuring the bucket ACL

why

• Post April 2023, ACLs are disabled by default for all new S3 buckets which can cause a race condition when trying to apply a canned ACL to a new S3 bucket - https://aws.amazon.com/about-aws/whats-new/2022/12/amazon-s3-automatically-enable-block-public-access-disable-access-control-lists-buckets-april-2023/
• If you try to use the resourceaws_s3_bucket_acl too quickly after the resource aws_s3_bucket_ownership_controls then ACLs may still be disabled on your new S3 bucket and your terraform apply will fail. If you re-run your apply then it works just fine the second time around as the propagation of the change in bucket ownership has finished.

references

• None

[danfsd]

From my personal experience fixing this issue locally doesn't require the use of a time_sleep resource, instead, we can make it work by adding a depends_on like on #175, #176 and #177.

[kumichou]

That's fine with me if the depends_on works. I thought I had tried that locally to fix the issue but apparently not.

[rubyisbeautiful]

mentioning issue #174 to assist finding

[kumichou]

I'm going to close this PR then since there's another that uses depends_on to correct the issue.