# s3_delete>: Delete files in Amazon S3

**s3_delete>** operator deletes files from Amazon S3.


```yaml
+delete_file:
  s3_delete>: my-bucket/my-file.gz
```


```yaml
+delete_with_params:
  s3_delete>:
  bucket: my-bucket
  key: my-file.gz
```

## Secrets

When you don't know how to set secrets, please refer to [Managing Workflow Secret](https://docs.treasuredata.com/articles/pd/about-workflow-secret-management)

* **aws.s3.region, aws.region**
An optional explicit AWS Region in which to access S3. Default is us-east-1.
* **aws.s3.access_key_id, aws.access_key_id**
The AWS Access Key ID to use when accessing S3. When using `credential_provider: assume_role`, this is not required.
* **aws.s3.secret_access_key, aws.secret_access_key**
The AWS Secret Access Key to use when accessing S3. When using `credential_provider: assume_role`, this is not required.


## Options

* **s3_delete>**: BUCKET/KEY
Path to the file in Amazon S3 to delete. Use either the operator command or the combination of `bucket` and `key`.
Examples:



```yaml
  s3_delete>: my-bucket/my-data.gz
```


```yaml
  s3_delete>: my-bucket/file/in/a/directory
```

* **bucket**: BUCKET
The S3 bucket where the file is located. Can be used together with the `key` parameter instead of putting the path on the operator line.
* **key**: KEY
The S3 key of the file to delete. Can be used together with the `bucket` parameter instead of putting the path on the operator line.
* **recursive**: BOOLEAN
Delete all objects with the specified prefix recursively. Default is false.
Examples:



```yaml
  +delete_directory:
    s3_delete>: my-bucket/old-data/
    recursive: true
```

* **objects_per_iteration**: NUMBER
Maximum number of objects to delete per iteration when using recursive mode. Must be between 1 and 1000. Default is 1000.
* **region**: REGION
An optional explicit AWS Region in which to access S3. This may also be specified using the `aws.s3.region` secret. Default is us-east-1.
* **path_style_access**: BOOLEAN
An optional flag to control whether to use path-style or virtual hosted-style access when accessing S3. Default is false.
* **log_each_object**: BOOLEAN
Whether to log each object being deleted. Default is true.
* **credential_provider**: NAME
The credential provider to use for AWS authentication. Supported values are `access_key` (default) and `assume_role`.
Examples:



```yaml
  +delete_file_with_assume_role:
    s3_delete>: my-bucket/my-file.gz
    credential_provider: assume_role
    assume_role_authentication_id: ${auth_id}
```

* **assume_role_authentication_id**: NUMBER
The authentication ID for assume role when using `credential_provider: assume_role`. This corresponds to the `Amazon S3 Import Integration v2` configuration.
How to get authentication_id is written in [Reusing the existing Authentication](https://docs.treasuredata.com/articles/#!int/amazon-s3-import-integration-v2/a/AmazonS3ImportIntegrationv2-reuseAuthenticationReusingtheexistingAuthentication).


## Notes

* If you try to delete a file that doesn't exist, the operation will succeed without errors.