RestoreObject - Amazon Simple Storage Service


Restores an archived copy of an object back into Amazon S3

This operation performs the following types of requests:

  • select - Perform a select query on an archived object

  • restore an archive - Restore an archived object

To use this operation, you must have permissions to perform the s3:RestoreObject action. The bucket owner has this permission by default and can grant this permission to others. For more information about permissions, see Permissions Related to Bucket Subresource Operations and Managing Access Permissions to Your Amazon S3 Resources in the Amazon Simple Storage Service Developer Guide.

Querying Archives with Select Requests

You use a select type of request to perform SQL queries on archived objects. The archived objects that are being queried by the select request must be formatted as uncompressed comma-separated values (CSV) files. You can run queries and custom analytics on your archived data without having to restore your data to a hotter Amazon S3 tier. For an overview about select requests, see Querying Archived Objects in the Amazon Simple Storage Service Developer Guide.

When making a select request, do the following:

  • Define an output location for the select query's output. This must be an Amazon S3 bucket in the same AWS Region as the bucket that contains the archive object that is being queried. The AWS account that initiates the job must have permissions to write to the S3 bucket. You can specify the storage class and encryption for the output objects stored in the bucket. For more information about output, see Querying Archived Objects in the Amazon Simple Storage Service Developer Guide.

    For more information about the S3 structure in the request body, see the following:

  • Define the SQL expression for the SELECT type of restoration for your query in the request body's SelectParameters structure. You can use expressions like the following examples.

    • The following expression returns all records from the specified object.

      SELECT * FROM Object

    • Assuming that you are not using any headers for data stored in the object, you can specify columns with positional headers.

      SELECT s._1, s._2 FROM Object s WHERE s._3 > 100

    • If you have headers and you set the fileHeaderInfo in the CSV structure in the request body to USE, you can specify headers in the query. (If you set the fileHeaderInfo field to IGNORE, the first row is skipped for the query.) You cannot mix ordinal positions with header column names.

      SELECT s.Id, s.FirstName, s.SSN FROM S3Object s

For more information about using SQL with S3 Glacier Select restore, see SQL Reference for Amazon S3 Select and S3 Glacier Select in the Amazon Simple Storage Service Developer Guide.

When making a select request, you can also do the following:

  • To expedite your queries, specify the Expedited tier. For more information about tiers, see "Restoring Archives," later in this topic.

  • Specify details about the data serialization format of both the input object that is being queried and the serialization of the CSV-encoded query results.

The following are additional important facts about the select feature:

  • The output results are new Amazon S3 objects. Unlike archive retrievals, they are stored until explicitly deleted-manually or through a lifecycle policy.

  • You can issue more than one select request on the same Amazon S3 object. Amazon S3 doesn't deduplicate requests, so avoid issuing duplicate requests.

  • Amazon S3 accepts a select request even if the object has already been restored. A select request doesn’t return error response 409.

Restoring Archives

Objects in the GLACIER and DEEP_ARCHIVE storage classes are archived. To access an archived object, you must first initiate a restore request. This restores a temporary copy of the archived object. In a restore request, you specify the number of days that you want the restored copy to exist. After the specified period, Amazon S3 deletes the temporary copy but the object remains archived in the GLACIER or DEEP_ARCHIVE storage class that object was restored from.

To restore a specific object version, you can provide a version ID. If you don't provide a version ID, Amazon S3 restores the current version.

The time it takes restore jobs to finish depends on which storage class the object is being restored from and which data access tier you specify.

When restoring an archived object (or using a select request), you can specify one of the following data access tier options in the Tier element of the request body:

  • Expedited - Expedited retrievals allow you to quickly access your data stored in the GLACIER storage class when occasional urgent requests for a subset of archives are required. For all but the largest archived objects (250 MB+), data accessed using Expedited retrievals are typically made available within 1–5 minutes. Provisioned capacity ensures that retrieval capacity for Expedited retrievals is available when you need it. Expedited retrievals and provisioned capacity are not available for the DEEP_ARCHIVE storage class.

  • Standard - S3 Standard retrievals allow you to access any of your archived objects within several hours. This is the default option for the GLACIER and DEEP_ARCHIVE retrieval requests that do not specify the retrieval option. S3 Standard retrievals typically complete within 3-5 hours from the GLACIER storage class and typically complete within 12 hours from the DEEP_ARCHIVE storage class.

  • Bulk - Bulk retrievals are Amazon S3 Glacier’s lowest-cost retrieval option, enabling you to retrieve large amounts, even petabytes, of data inexpensively in a day. Bulk retrievals typically complete within 5-12 hours from the GLACIER storage class and typically complete within 48 hours from the DEEP_ARCHIVE storage class.

For more information about archive retrieval options and provisioned capacity for Expedited data access, see Restoring Archived Objects in the Amazon Simple Storage Service Developer Guide.

You can use Amazon S3 restore speed upgrade to change the restore speed to a faster speed while it is in progress. You upgrade the speed of an in-progress restoration by issuing another restore request to the same object, setting a new Tier request element. When issuing a request to upgrade the restore tier, you must choose a tier that is faster than the tier that the in-progress restore is using. You must not change any other parameters, such as the Days request element. For more information, see Upgrading the Speed of an In-Progress Restore in the Amazon Simple Storage Service Developer Guide.

To get the status of object restoration, you can send a HEAD request. Operations return the x-amz-restore header, which provides information about the restoration status, in the response. You can use Amazon S3 event notifications to notify you when a restore is initiated or completed. For more information, see Configuring Amazon S3 Event Notifications in the Amazon Simple Storage Service Developer Guide.

After restoring an archived object, you can update the restoration period by reissuing the request with a new period. Amazon S3 updates the restoration period relative to the current time and charges only for the request-there are no data transfer charges. You cannot update the restoration period when Amazon S3 is actively processing your current restore request for the object.

If your bucket has a lifecycle configuration with a rule that includes an expiration action, the object expiration overrides the life span that you specify in a restore request. For example, if you restore an object copy for 10 days, but the object is scheduled to expire in 3 days, Amazon S3 deletes the object in 3 days. For more information about lifecycle configuration, see PutBucketLifecycleConfiguration and Object Lifecycle Management in Amazon Simple Storage Service Developer Guide.


A successful operation returns either the 200 OK or 202 Accepted status code.

  • If the object copy is not previously restored, then Amazon S3 returns 202 Accepted in the response.

  • If the object copy is previously restored, Amazon S3 returns 200 OK in the response.

Special Errors

    • Code: RestoreAlreadyInProgress

    • Cause: Object restore is already in progress. (This error does not apply to SELECT type requests.)

    • HTTP Status Code: 409 Conflict

    • SOAP Fault Code Prefix: Client

    • Code: GlacierExpeditedRetrievalNotAvailable

    • Cause: S3 Glacier expedited retrievals are currently not available. Try again later. (Returned if there is insufficient capacity to process the Expedited request. This error applies only to Expedited retrievals and not to S3 Standard or Bulk retrievals.)

    • HTTP Status Code: 503

    • SOAP Fault Code Prefix: N/A

Request Syntax

POST /{Key+}?restore&versionId=VersionId HTTP/1.1 Host: x-amz-request-payer: RequestPayer x-amz-expected-bucket-owner: ExpectedBucketOwner <?xml version="1.0" encoding="UTF-8"?> <RestoreRequest xmlns=""> <Days>integer</Days> <GlacierJobParameters> <Tier>string</Tier> </GlacierJobParameters> <Type>string</Type> <Tier>string</Tier> <Description>string</Description> <SelectParameters> <Expression>string</Expression> <ExpressionType>string</ExpressionType> <InputSerialization> <CompressionType>string</CompressionType> <CSV> <AllowQuotedRecordDelimiter>boolean</AllowQuotedRecordDelimiter> <Comments>string</Comments> <FieldDelimiter>string</FieldDelimiter> <FileHeaderInfo>string</FileHeaderInfo> <QuoteCharacter>string</QuoteCharacter> <QuoteEscapeCharacter>string</QuoteEscapeCharacter> <RecordDelimiter>string</RecordDelimiter> </CSV> <JSON> <Type>string</Type> </JSON> <Parquet> </Parquet> </InputSerialization> <OutputSerialization> <CSV> <FieldDelimiter>string</FieldDelimiter> <QuoteCharacter>string</QuoteCharacter> <QuoteEscapeCharacter>string</QuoteEscapeCharacter> <QuoteFields>string</QuoteFields> <RecordDelimiter>string</RecordDelimiter> </CSV> <JSON> <RecordDelimiter>string</RecordDelimiter> </JSON> </OutputSerialization> </SelectParameters> <OutputLocation> <S3> <AccessControlList> <Grant> <Grantee> <DisplayName>string</DisplayName> <EmailAddress>string</EmailAddress> <ID>string</ID> <xsi:type>string</xsi:type> <URI>string</URI> </Grantee> <Permission>string</Permission> </Grant> </AccessControlList> <BucketName>string</BucketName> <CannedACL>string</CannedACL> <Encryption> <EncryptionType>string</EncryptionType> <KMSContext>string</KMSContext> <KMSKeyId>string</KMSKeyId> </Encryption> <Prefix>string</Prefix> <StorageClass>string</StorageClass> <Tagging> <TagSet> <Tag> <Key>string</Key> <Value>string</Value> </Tag> </TagSet> </Tagging> <UserMetadata> <MetadataEntry> <Name>string</Name> <Value>string</Value> </MetadataEntry> </UserMetadata> </S3> </OutputLocation> </RestoreRequest>

URI Request Parameters

The request uses the following URI parameters.


The bucket name or containing the object to restore.

When using this API with an access point, you must direct requests to the access point hostname. The access point hostname takes the form When using this operation using an access point through the AWS SDKs, you provide the access point ARN in place of the bucket name. For more information about access point ARNs, see Using Access Points in the Amazon Simple Storage Service Developer Guide.

Required: Yes


Object key for which the operation was initiated.

Length Constraints: Minimum length of 1.

Required: Yes


VersionId used to reference a specific version of the object.


The account id of the expected bucket owner. If the bucket is owned by a different account, the request will fail with an HTTP 403 (Access Denied) error.


Confirms that the requester knows that they will be charged for the request. Bucket owners need not specify this parameter in their requests. For information about downloading objects from requester pays buckets, see Downloading Objects in Requestor Pays Buckets in the Amazon S3 Developer Guide.

Valid Values: requester

Request Body

The request accepts the following data in XML format.


Root level tag for the RestoreRequest parameters.

Required: Yes


Lifetime of the active copy in days. Do not use with restores that specify OutputLocation.

Type: Integer

Required: No


The optional description for the job.

Type: String

Required: No


S3 Glacier related parameters pertaining to this job. Do not use with restores that specify OutputLocation.

Type: GlacierJobParameters data type

Required: No


Describes the location where the restore job's output is stored.

Type: OutputLocation data type

Required: No


Describes the parameters for Select job types.

Type: SelectParameters data type

Required: No


S3 Glacier retrieval tier at which the restore will be processed.

Type: String

Valid Values: Standard | Bulk | Expedited

Required: No


Type of restore request.

Type: String

Valid Values: SELECT

Required: No

Response Syntax

HTTP/1.1 200 x-amz-request-charged: RequestCharged x-amz-restore-output-path: RestoreOutputPath

Response Elements

If the action is successful, the service sends back an HTTP 200 response.

The response returns the following HTTP headers.


If present, indicates that the requester was successfully charged for the request.

Valid Values: requester


Indicates the path in the provided S3 output location where Select results will be restored to.


Restore an object for 2 days using the expedited retrieval option

The following restore request restores a copy of the photo1.jpg object from S3 Glacier for a period of two days using the expedited retrieval option.

Host: examplebucket.dummy value Date: Mon, 22 Oct 2012 01:49:52 GMT Authorization: authorization string Content-Length: content length <RestoreRequest> <Days>2</Days> <GlacierJobParameters> <Tier>Expedited</Tier> </GlacierJobParameters> </RestoreRequest>


If the examplebucket does not have a restored copy of the object, Amazon S3 returns the following 202 Accepted response.


If a copy of the object is already restored, Amazon S3 returns a 200 OK response, and updates only the restored copy's expiry time.

HTTP/1.1 202 Accepted x-amz-id-2: GFihv3y6+kE7KG11GEkQhU7/2/cHR3Yb2fCb2S04nxI423Dqwg2XiQ0B/UZlzYQvPiBlZNRcovw= x-amz-request-id: 9F341CD3C4BA79E0 Date: Sat, 20 Oct 2012 23:54:05 GMT Content-Length: 0 Server: AmazonS3

Query an archive with a SELECT request

The following is an example SELECT restore request.

POST /object-one.csv?restore HTTP/1.1 Host: examplebucket.dummy value Date: Date: Sat, 20 Oct 2012 23:54:05 GMT Authorization: <>authorization string</> Content-Length: <>content length</> <RestoreRequest xmlns=""> <Type>SELECT</Type> <Tier>Expedited</Tier> <Description>this is a description</Description> <SelectParameters> <InputSerialization> <CSV> <FileHeaderInfo>IGNORE</FileHeaderInfo> <Comments>#</Comments> <QuoteEscapeCharacter>"</QuoteEscapeCharacter> <RecordDelimiter>\n</RecordDelimiter> <FieldDelimiter>,</FieldDelimiter> <QuoteCharacter>"</QuoteCharacter> </CSV> </InputSerialization> <ExpressionType>SQL</ExpressionType> <Expression>select * from object</Expression> <OutputSerialization> <CSV> <QuoteFields>ALWAYS</QuoteFields> <QuoteEscapeCharacter>"</QuoteEscapeCharacter> <RecordDelimiter>\n</RecordDelimiter> <FieldDelimiter>\t</FieldDelimiter> <QuoteCharacter>\'</QuoteCharacter> </CSV> </OutputSerialization> </SelectParameters> <OutputLocation> <S3> <BucketName>example-output-bucket</BucketName> <Prefix>test-s3</Prefix> <AccessControlList> <Grant> <Grantee xmlns:xsi="" xsi:type="AmazonCustomerByEmail"> <EmailAddress></EmailAddress> </Grantee> <Permission>FULL_CONTROL</Permission> </Grant> </AccessControlList> <UserMetadata> <MetadataEntry> <Name>test</Name> <Value>test-value</Value> </MetadataEntry> <MetadataEntry> <Name>other</Name> <Value>something else</Value> </MetadataEntry> </UserMetadata> <StorageClass>STANDARD</StorageClass> </S3> </OutputLocation> </RestoreRequest>


Amazon S3 returns the following 202 Accepted response.

HTTP/1.1 202 Accepted x-amz-id-2: GFihv3y6+kE7KG11GEkQhU7/2/cHR3Yb2fCb2S04nxI423Dqwg2XiQ0B/UZlzYQvPiBlZNRcovw= x-amz-request-id: 9F341CD3C4BA79E0 x-amz-restore-output-path: js-test-s3/qE8nk5M0XIj-LuZE2HXNw6empQm3znLkHlMWInRYPS-Orl2W0uj6LyYm-neTvm1-btz3wbBxfMhPykd3jkl-lvZE7w42/ Date: Sat, 20 Oct 2012 23:54:05 GMT Content-Length: 0 Server: AmazonS3

See Also

For more information about using this API in one of the language-specific AWS SDKs, see the following: