Empty and delete a bucket - CoreWeave Docs

CoreWeave AI Object Storage only deletes a bucket once it is completely empty. The bucket must contain no current objects, no noncurrent object versions, no delete markers, and no incomplete multipart uploads. If any of these remain, s3:DeleteBucket returns BucketNotEmpty. This guide walks through emptying a bucket of every kind of content it can hold, then deleting the empty bucket. The four cleanup steps are independent and can run in parallel: skip any that do not apply to your bucket.

Prerequisites

Before you start:

An Access Key for the organization that owns the bucket.
A configured S3-compatible client (AWS CLI, s3cmd, or Boto3).
Permission to call s3:DeleteObject, s3:DeleteObjects, s3:AbortMultipartUpload, and s3:DeleteBucket on the bucket.

Deleting a bucket is permanent. Once a bucket and its contents are deleted, the data cannot be recovered. Confirm the bucket name and contents before you start.

Step 1: Delete current objects

Delete all current-version objects in the bucket.

AWS CLI
s3cmd
Boto3

aws s3 rm with --recursive removes every current object in the bucket. Replace [BUCKET-NAME] with the name of your bucket.

Delete all current objects with AWS CLI

aws s3 rm s3://[BUCKET-NAME] --recursive

For very large buckets, see Bulk deletion performance for tuning options.

s3cmd rm with --recursive removes every current object in the bucket. Replace [BUCKET-NAME] with the name of your bucket.

Delete all current objects with s3cmd

s3cmd rm --recursive s3://[BUCKET-NAME]

Set environment variables for your CoreWeave credentials:
```
export ACCESS_KEY_ID="[ACCESS-KEY-ID]"
export SECRET_ACCESS_KEY="[SECRET-ACCESS-KEY]"
```
Alternatively, configure your CoreWeave credentials to work with the AWS CLI. We recommend using a separate profile for CoreWeave AI Object Storage to avoid conflicts with your other AWS profiles and S3-compatible services. If you don’t set up this configuration, you might encounter errors when using AI Object Storage.
Configure CoreWeave credentials
1. Create a new credentials file and profile in your CoreWeave configuration directory.
  Create a new credentials file and profile
  AWS_SHARED_CREDENTIALS_FILE=~/.coreweave/cw.credentials aws configure --profile cw
2. When prompted, provide the following values:
  - AWS Access Key ID: The Access Key ID of your CoreWeave AI Object Storage Access Key.
  - AWS Secret Access Key: The Secret Key of your CoreWeave AI Object Storage Access Key.
  - Default region name (Optional): To set a default region, see CoreWeave Availability Zones.
  - Default output format: Use json for JSON output.
3. Set the default endpoint URL to the appropriate endpoint for your use case:
  - The primary endpoint, https://cwobject.com, for use outside a CoreWeave cluster.
  - The LOTA endpoint, http://cwlota.com, for use inside a CoreWeave cluster. The LOTA endpoint routes to the LOTA path for best performance.
  Set the primary endpoint for local development
  AWS_CONFIG_FILE=~/.coreweave/cw.config aws configure set endpoint_url https://cwobject.com --profile cw
4. Set the S3 addressing_style to virtual:
  Set virtual addressing style
  AWS_CONFIG_FILE=~/.coreweave/cw.config aws configure set s3.addressing_style virtual --profile cw

This script pages through s3:ListObjectsV2 and deletes objects in batches of 1000 using s3:DeleteObjects. Replace [BUCKET-NAME] with the name of your bucket.

Delete all current objects with Boto3

import os
import boto3
from botocore.client import Config

BUCKET = '[BUCKET-NAME]'

boto_config = Config(
    region_name='US-EAST-04A',
    s3={'addressing_style': 'virtual'}
)

s3 = boto3.client(
    's3',
    endpoint_url='https://cwobject.com',
    aws_access_key_id=os.environ['ACCESS_KEY_ID'],
    aws_secret_access_key=os.environ['SECRET_ACCESS_KEY'],
    config=boto_config
)

deleted = 0
paginator = s3.get_paginator('list_objects_v2')
for page in paginator.paginate(Bucket=BUCKET):
    contents = page.get('Contents', [])
    if not contents:
        continue
    s3.delete_objects(
        Bucket=BUCKET,
        Delete={'Objects': [{'Key': o['Key']} for o in contents]},
    )
    deleted += len(contents)

print(f"Deleted {deleted} current objects from {BUCKET}.")

Bulk deletion performance

For large buckets, choose the deletion approach that matches the scale you need:

s3:DeleteObjects batches delete up to 1000 objects per call. The Boto3 example above uses this approach. It is the most efficient way to delete millions of objects from a single client.
Parallel aws s3 rm workers: Split deletion across prefixes and run multiple aws s3 rm s3://[BUCKET-NAME]/[PREFIX] --recursive commands in parallel. Reasonable for hundreds of thousands of objects when the keys partition cleanly by prefix.
rclone delete: Run rclone delete [REMOTE]:[BUCKET-NAME] for incremental deletion with progress reporting. Combine with --transfers to control concurrency. See Versioned buckets for rclone configuration.

Mass deletion is a write-heavy workload. Running thousands of concurrent delete operations against a single bucket can affect other clients reading from or writing to the same bucket. For very large deletions, schedule the work outside peak hours, or stage it across prefixes.

If you do not need immediate deletion, an Expiration lifecycle rule deletes objects on a schedule without consuming client capacity.

Step 2: Delete noncurrent versions and delete markers

Skip this step if the bucket does not have versioning enabled or suspended. To check:

Check the versioning status of a bucket

aws s3api get-bucket-versioning --bucket [BUCKET-NAME]

If the response includes "Status": "Enabled" or "Status": "Suspended", the bucket can hold noncurrent versions and delete markers, both of which must be removed before deletion. The simplest way to remove every noncurrent version and delete marker is to list them and delete them by VersionId.

AWS CLI
s3cmd
Boto3

This shell loop pages through list-object-versions and deletes every version and delete marker it finds. Replace [BUCKET-NAME] with the name of your bucket.

Delete all versions and delete markers with AWS CLI

BUCKET="[BUCKET-NAME]"

aws s3api list-object-versions --bucket "$BUCKET" \
    --query '{Objects: Versions[].{Key: Key, VersionId: VersionId}}' \
    --output json > versions.json

aws s3api delete-objects --bucket "$BUCKET" --delete file://versions.json

aws s3api list-object-versions --bucket "$BUCKET" \
    --query '{Objects: DeleteMarkers[].{Key: Key, VersionId: VersionId}}' \
    --output json > markers.json

aws s3api delete-objects --bucket "$BUCKET" --delete file://markers.json

If the bucket has more than 1000 versions or delete markers, repeat both blocks until both listings return no entries.

Delete all versions and delete markers with Boto3

import os
import boto3
from botocore.client import Config

BUCKET = '[BUCKET-NAME]'

boto_config = Config(
    region_name='US-EAST-04A',
    s3={'addressing_style': 'virtual'}
)

s3 = boto3.client(
    's3',
    endpoint_url='https://cwobject.com',
    aws_access_key_id=os.environ['ACCESS_KEY_ID'],
    aws_secret_access_key=os.environ['SECRET_ACCESS_KEY'],
    config=boto_config
)

deleted = 0
paginator = s3.get_paginator('list_object_versions')
for page in paginator.paginate(Bucket=BUCKET):
    items = []
    for v in page.get('Versions', []):
        items.append({'Key': v['Key'], 'VersionId': v['VersionId']})
    for m in page.get('DeleteMarkers', []):
        items.append({'Key': m['Key'], 'VersionId': m['VersionId']})
    if not items:
        continue
    for i in range(0, len(items), 1000):
        s3.delete_objects(
            Bucket=BUCKET,
            Delete={'Objects': items[i:i + 1000]},
        )
        deleted += len(items[i:i + 1000])

print(f"Deleted {deleted} versions and delete markers from {BUCKET}.")

For more on versioned bucket workflows, see Versioned buckets.

Step 3: Abort incomplete multipart uploads

Incomplete multipart uploads do not appear in normal object listings, but they consume storage and prevent bucket deletion. They are the single most common reason DeleteBucket returns BucketNotEmpty when the bucket appears empty in the Cloud Console or in aws s3 ls. To check whether the bucket has incomplete multipart uploads:

List incomplete multipart uploads

aws s3api list-multipart-uploads --bucket [BUCKET-NAME]

If the response is empty, skip to Step 4. If it lists any uploads, follow Abort incomplete multipart uploads to remove them, then continue.

Step 4: Delete the empty bucket

Once the bucket has no current objects, no noncurrent versions, no delete markers, and no incomplete multipart uploads, delete it.

AWS CLI
s3cmd
Boto3

Replace [BUCKET-NAME] with the name of your bucket.

Delete the empty bucket with AWS CLI

aws s3api delete-bucket --bucket [BUCKET-NAME]

Remove the empty bucket with s3cmd

s3cmd rb s3://[BUCKET-NAME]

Delete the empty bucket with Boto3

import os
import boto3
from botocore.client import Config

boto_config = Config(
    region_name='US-EAST-04A',
    s3={'addressing_style': 'virtual'}
)

s3 = boto3.client(
    's3',
    endpoint_url='https://cwobject.com',
    aws_access_key_id=os.environ['ACCESS_KEY_ID'],
    aws_secret_access_key=os.environ['SECRET_ACCESS_KEY'],
    config=boto_config
)

s3.delete_bucket(Bucket='[BUCKET-NAME]')

A successful deletion returns no output. If you still get BucketNotEmpty, re-run the listings from each step above to find what remains.

Prevent BucketNotEmpty errors

To keep buckets ready for deletion at any time, apply a bucket lifecycle policy that includes:

An AbortIncompleteMultipartUpload rule. This is the most common cause of bucket-delete failures and the only way to keep abandoned uploads from accumulating without active intervention.
A NoncurrentVersionExpiration rule on versioned buckets. Without it, every overwrite leaves a noncurrent version that must be cleaned up before deletion.

Bucket lifecycle policies: Automate the cleanup of versions and incomplete multipart uploads.
Abort incomplete multipart uploads: Detailed procedures for listing and aborting MPUs.
Versioned buckets: Manage versioned bucket workflows with rclone and the AWS CLI.
Manage buckets: Bucket creation and management with S3-compatible clients.

​Prerequisites

​Step 1: Delete current objects

​Bulk deletion performance

​Step 2: Delete noncurrent versions and delete markers

​Step 3: Abort incomplete multipart uploads

​Step 4: Delete the empty bucket

​Prevent BucketNotEmpty errors

​Related

Prerequisites

Step 1: Delete current objects

Bulk deletion performance

Step 2: Delete noncurrent versions and delete markers

Step 3: Abort incomplete multipart uploads

Step 4: Delete the empty bucket

Prevent BucketNotEmpty errors

Related