GCS versioning without noncurrent cleanup

How to detect GCS versioning without noncurrent cleanup

The strongest signal is a bucket with versioning enabled and no lifecycle rule that actually ages out older generations. A generic lifecycle rule can still leave this detector firing when it never limits older versions.

Check whether the bucket has versioning enabled and whether lifecycle rules would actually clean up older object generations:

gcloud storage buckets describe gs://my-bucket --format=json

In the bucket description, confirm both:

• versioning.enabled is true
• Lifecycle rules either bound retention with age or explicitly target archived or noncurrent versions, such as isLive=false, numNewerVersions, daysSinceNoncurrentTime, or noncurrentTimeBefore

If versioning is enabled but lifecycle only deletes live objects, inspect storage growth over time. Buckets with repeated rewrites, logs, or mutable artifact names are strong candidates.

How this detector confirms GCS versioning cleanup gaps

Cloud Waste Hunter does not treat generic lifecycle presence as enough. The detector only suppresses a finding when it sees a Delete rule that appears to meaningfully limit older version retention. That keeps the page aligned to the implementation boundary: versioning plus missing effective cleanup for versioned buildup.

How to fix GCS versioning without noncurrent cleanup

Define a lifecycle rule that limits how long older generations remain in the bucket.

resource "google_storage_bucket" "versioned" {
  name     = "my-versioned-bucket"
  location = "US"

  versioning {
    enabled = true
  }

  lifecycle_rule {
    condition {
      age                = 60
      with_state         = "ARCHIVED"
      num_newer_versions = 5
    }
    action {
      type = "Delete"
    }
  }
}

If recovery requirements differ across data types, split the data into purpose-specific buckets before applying cleanup.

Caveats and overlap boundaries

This detector is narrower than a generic “bucket has no lifecycle policy” check. A bucket can have lifecycle rules and still be flagged here if those rules never clean up archived generations created by versioning. Conversely, a bucket with an age-based or explicit noncurrent-version delete rule is treated as aligned even if the retention window is conservative.

If the real issue is a bucket with no lifecycle policy at all, that is the broader lifecycle-governance problem and the adjacent gcp-gcs-bucket-missing-lifecycle detector may also fire. This page is specifically about versioning-related storage sprawl.

How Cloud Waste Hunter helps

Cloud Waste Hunter is being built to identify versioned buckets missing effective cleanup controls and point teams toward safer lifecycle policies without assuming generic lifecycle presence is enough. For the broader storage review, continue into the GCP Storage Cost Optimization guide.