Versions Compared


  • This line was added.
  • This line was removed.
  • Formatting was changed.

In this article

Table of Contents
excludeOlder Versions|Additional Resources|Related Links|Related Information


S3 File Writer is a Write-type Snap that reads a binary data stream from its input view and writes it to an S3 file destination. If you provide values for File permissions, the Snap sets the permissions to the file. 

  • This Snap has the ability to use an MD5 checksum that automatically checks for data integrity and corruption while uploading the file. 
  • The current Snap functionality supports AWS S3 Cloud Service and is applicable for AWSGovCloud setup.

Image RemovedImage Added



Supported Features

Works in Ultra Task Pipelines.

Limitations and Known Issues


Snap Views

View TypeView FormatNumber of ViewsExamples of Upstream and Downstream SnapsDescription
Input Document
  • Min: 0
  • Max: 1
  • CSV Formatter
  • JSON Formatter
  • XML Formatter
  • File Reader
Any binary data stream.
  • Min: 0
  • Max: 1
  • File Reader
  • Mapper

If an output view is open and the file write action is successful, the output view provides a document with information on the filename, result, and original data. An example is:

Code Block
        "filename": "s3:///mybucket/qatest/user_manual.json",
        "result": "overwritten",
        "original": {
            "content-type" : "application/json"



  • Min: 1

  • Max: 1


The error view contains error, reason, resolution and stack trace. For more information, see Handling Errors with an Error Pipeline

Optional Configuration

IAM Roles for Amazon EC2

To access S3 files from Groundplex nodes hosted in the EC2 environment without specifying Access-key ID and Secret key in AWS S3 account configured for the Snap, enable the ‘IAM_CREDENTIAL_FOR_S3’ feature. When you enable this feature, the IAM credential stored in EC2 metadata is used to access S3 buckets.

To enable the IAM_CREDENTIAL_FOR_S3 feature:

  1. Open Manager.
  2. Open the Snaplexes tab of the project that contains the EC2-based Groundplex.
  3. Click the Groundplex to open its Properties.
  4. Open the Node Properties tab.
  5. Add a new row in the Global properties section.
  6. Specify jvm_options as the Key and -DIAM_CREDENTIAL_FOR_S3=TRUE as the Value.

    Restart the JCC (node) to apply the changes. For more information about IAM Roles, refer to IAM Roles for Amazon EC2.

Behavior Change

titleBehavior Change

In the 4.23 release, the Snap used to write an S3 file and then set ACL if the ACL property is not empty. If the file write is successful and the ACL setting fails, both output document in the output view and error document in the error view were displayed. This has been fixed in 4.24 release by enabling the Snap to configure ACL when writing the fileTherefore, the Snap produces an error document in the error view if the ACL configuration fails, irrespective of the file write action passing or failing.

Snap Settings


Field names marked with an asterisk ( * ) are mandatory.

Field NameField Type Description



Specify the name for the Snap. You can modify this to be more specific, especially if you have more than one of the same Snap in your pipelinePipeline.

Default ValueS3 File Writer
ExampleS3 File Writer

File name*


Specify the URL for the S3 file, from where the binary data is to be read.


The file name must start with s3:///. You can use the suggest feature to view the list of buckets, subdirectories and files. Bucket names are suggested if the property is empty or s3:///. Once you select a bucket, it can list subdirectories and files immediately below the bucket. Names of subdirectories end with a forward slash ("/"). The suggest feature is not supported if the properties in the S3 Dynamic account are parameters.


The provided account must have 'read' access to the specified S3 bucket in order to read the file successfully.

Using Expressions:

This property can be an expression with the "=" button pressed.

For example, if the File property is "s3:///mybucket/out_" + + ".csv" then the evaluated filename is s3:///mybucket/out_2013-11-13T00:22:31.880Z.csv.


Paste code macro

For region names and their details, see AWS Regions and Endpoints.

titleRegion Name

Region name is optional only if the region is us-east-1. In all other cases the region name must be specified based on the syntax above. For example, mybucket@eu-west-1. 

For more information about regions, see AWS Regions and Endpoints.

Default Value: s3:///


  • s3:///
  • _filename (A key/value pair with "filename" key should be defined as a pipeline parameter)
  • $filename (A key/value pair with "filename" key should be defined the input document)
Suggest fully-qualified file namesCheckbox

Select this checkbox to include the region and authority of the S3 bucket in the associated paths that appear in the Suggestion box.


We recommend you to use fully-qualified suggestions from the Suggestions list if you are using an instance in the gov cloud.

Default Value: Not Selected

File action*

Dropdown list

Specify the action to perform if the file already exists. The available options are:

  • OVERWRITE - The Snap attempts to write the file without checking for the file's existence for a better performance, and the "fileAction" field will be "overwritten" in the output view data.
  • IGNORE The Snap will not overwrite the file and will do nothing but write the status and file name to its output view.
  • ERROR-  The error displays in the Pipeline Run Log. If an error view is defined, the error will be written there as well.

Even though it is listed, Append is not supported for the S3 file protocol.

Default ValueOVERWRITE 

Write empty fileCheckbox

Select this checkbox to write an empty file when the incoming binary document has empty data. If there is no incoming document at the input view of the Snap, no file is written regardless of the value of the property.

Default Value: Not selected

Write header file


Select this checkbox to The binary data stream in the input view may contain header information about the binary data in the form of a document with key-value-pair map data. If this property is checked, the Snap writes a header file whose name is generated by appending ".header" to the value of the File name property. The same header information is also included in the output view data, as shown in the "Expected output" section above, under the key "original". Note that if the header has no keys other than Content-Type or Content-Encoding, the .header file will not be written

Default Value: Not selected

Validate after write


Select this checkbox to allow the Snap to check if the file exists after writing the file.

Default Value: Not selected 

Number of retriesInteger/Expression

Specify the maximum number of retry attempts the Snap must make when it fails to write. If the value is larger than 0, the Snap first stores the input data in a temporary local file before writing to the target file.


Ensure that the local drive has sufficient free disk space as large as the expected target file size.

If the value is larger than 0, the Snap first downloads the target file into a temporary local file. If any error occurs during the download, the Snap waits for the time specified in the Retry interval and attempts to download the file again from the beginning. When the download is successful, the Snap streams the data from the temporary file to the downstream Pipeline. All temporary local files are deleted when they are no longer needed.

Minimum value: 0

Default Value: 0
Example: 3

Retry interval (seconds)Integer/Expression

Specify the minimum number of seconds for which the Snap must wait before attempting recovery from a network failure.

Minimum value: 1

Default Value: 1

Buffer size(MB)


Specify the data (in MB) to load into the S3 bucket, at a time.

Default Value: 10 MB

  • The minimum data size you can upload is 6 MB.

  • The maximum data size you can upload is limited to 10000 times the buffer size.

  • To upload S3 files that are more than 100 GB in size, we recommend you to set the Buffer size to 100 MB or more. Also, set the Maximum upload threads to 10; otherwise, many parallel HTTP connections may be opened, leading to the following error: "AmazonClientException: Unable to execute HTTP request: Timeout waiting for connection from pool".

Refer to Upload Part for more information on uploading to S3. 

Maximum upload threadsInteger

Specify the maximum number of threads to be used for the concurrent multipart upload. The minimum value allowed is 1.

Default Value: 10


To upload S3 files that are more than 100 GB in size, we recommend you to set the Maximum upload threads to 10. Also, set the Buffer size to 100 MB or more; otherwise, many parallel HTTP connections may be opened, leading to the following error: "AmazonClientException: Unable to execute HTTP request: Timeout waiting for connection from pool".

Access Control List

Use this field set to define the Access Control List (ACL) to the specified S3 file. This filed set contains the following fields:

  • Grantee
  • Read
  • View permissions
  • Full control


The account that you use for the Snap must have the required permissions to set the ACLs for the S3 object it is writing to.


Select a grantee from the suggested list or enter a valid email address or a canonical ID associated to an AWS account. Canonical AWS ID can be obtained in the Security Credentials page of the AWS console.


A grantee can be an AWS account or one of the predefined Amazon S3 groups. The following note is an excerpt from AWS document:

"An email grantee is a grantee identified by their email address and authenticated by an Amazon system. email grants are internally converted to the canonical user representation when creating the ACL. If the grantee changes their email address, it will not affect existing Amazon S3 permissions. Adding a grantee by email address only works if exactly one Amazon account corresponds to the specified email address. If multiple Amazon accounts are associated with the email address, an AmbiguousGrantByEmail error message is returned. This happens rarely, but usually occurs if a user created an Amazon account in the past, forgotten the password, and created another Amazon account using the same email address. If this occurs, the user should contact Amazon customer service to have the accounts merged. Alternatively, grant user access specifying the canonical user representation."


  • "Everyone"
  • "Authenticated user"
  •  an email address used to create AWS account
  • a canonical AWS ID, e.g. "1700891f3927e316dc4c9e18c789b32131880f48d3e03ac110aaf695b212573e"

"Everyone” option allows anyone in the world to access the file (authenticated or anonymous). Even when the bucket is protected with permission and if the file operation in the Snap under ACL->Grantee is set to 'Everyone', any user (authenticated or anonymous) can access the file. So, we highly recommend not to use this option as its unsafe. 

titleEmail Address Grantee Access

You can use Email addresses to specify a grantee only in the following regions:

  • US East (N. Virginia)
  • US West (N. California)
  • US West (Oregon)
  • Asia Pacific (Singapore)
  • Asia Pacific (Sydney)
  • Asia Pacific (Tokyo)
  • EU (Ireland)
  • South America (São Paulo)

For more information, see Specifying Grantee.

ReadCheckboxGrants permission to read the file.
View permissionsCheckboxGrants permission to read the ACL.
Full controlCheckboxGrants full control to the file.
User-defined object metadataUse this field set to define key-value pairs for user-defined object metadata of an S3 object. For more information about user-defined object metadata, see Using Metadata

This field set contains the following fields:

  • Key
  • Value

Specify the key name of the object metadata.


The key names of the object metadata are case-insensitive.  AWS S3 converts them to lower-case and prefixes them with “x-amz-meta-” when displayed in the AWS S3 web console.

When the S3 File Reader Snap reads an S3 file, this metadata is shown in the header of the output binary data, and the key names are displayed in lower-case without the prefix “x-amz-meta-”.

ValueString/ExpressionSpecify the value for the key entered above.
Object tags

Use this field set to define key-value pairs for object tags of an S3 object. Object tags enable you to categorize existing and new objects using key-value combinations. For details about the object tags, see Object Tagging. This field set contains the following fields:

  • Key
  • Value

Specify the key name of the object tag.


The key names of object tags are case-sensitive. When the S3 File Reader Snap reads an S3 file, these object tags are displayed in the header of the output binary data. If a key name of an object tag is the same as another in the header, it is prefixed with “tag_”.

See example Providing User-defined Object Metadata and Object Tags using the S3 File Writer Snap below for more information. 

ValueString/ExpressionSpecify the value for the key entered above.

Snap Execution

Dropdown list

Multiexcerpt include macro
pageSOAP Execute


Providing User-defined Object Metadata and Object Tags using the S3 File Writer Snap

This example is a basic use case for the S3 File Writer Snap. It also demonstrates how you can configure the Snap with custom object metadata and object tags to classify the data. 

In the sample Pipeline, the S3 File Writer Snap is configured as follows with the User-defined object metadata and Object tags

The following is a preview of the output data from the S3 File Writer Snap: 

When the S3 File Reader Snap reads this data, it picks up the user-defined object metadata and object tags we defined, as show below: 

Typical Configuration

Key configuration of the Snap lies in how the values are passed. Values can be passed to the Snap:

Without Expressions

For example, the File name is passed directly to the Snap in the image below.

With Expressions

The File name is passed using Pipeline Parameters:

Download the Pipeline.


patterns*.slp, *.zip

See Also

Excerpt Include
Binary Snap Pack
Binary Snap Pack