Creating the CloudWatch Logs IAM role Creating the user import CSV file Creating and running the user import job Viewing import job results Requiring imported users to reset their passwords

Importing users into user pools from a CSV file

When you have an external identity store and the time to prepare your user pool for new local users, a bulk user import from a comma-separated values (CSV) file can be a low-effort, low-cost option for a migration to an Amazon Cognito user pool. A CSV file import is a process of downloading and populating a template file, then handing off the file to your user pool in an import job. You can use a CSV import to quickly create test users. You can also programmatically populate the file with read API requests to your external identity store, followed by parsing their details and attributes into write operations to the file.

The import process sets values for all user attributes except password. Password import is not supported, because security best practices require that passwords are not available as plain text, and we don't support importing hashes. This means that your users must change their passwords the first time they sign in. Your users are in a RESET_REQUIRED state when imported using this method.

You can set your users' passwords with an AdminSetUserPassword API request that sets the Permanent parameter to true. CSV import doesn't contribute to the billed monthly active users (MAUs) in your user pool. However, password-reset operations do generate MAUs. To manage costs when you import large numbers of users that might not be immediately active, set up your application to prompt users for a new password when they sign in and receive the RESET_REQUIRED challenge.

Note

The creation date for each user is the time when that user was imported into the user pool. Creation date is not one of the imported attributes.

Steps to create a user import job

Create an Amazon CloudWatch Logs role in the AWS Identity and Access Management (IAM) console.
Create the user import .csv file.
Create and run the user import job.
Upload the user import .csv file.
Start and run the user import job.
Use CloudWatch to check the event log.
Require the imported users to reset their passwords.

More resources

Cognito User Profiles Export Reference Architecture for exporting user accounts between user pools

Topics

Creating the CloudWatch Logs IAM role
Creating the user import CSV file
Creating and running the Amazon Cognito user pool import job
Viewing the user pool import results in the CloudWatch console
Requiring imported users to reset their passwords

Creating the CloudWatch Logs IAM role

If you're using the Amazon Cognito CLI or API, then you need to create a CloudWatch IAM role. The following procedure describes how to create an IAM role that Amazon Cognito can use to write the results of your import job to CloudWatch Logs.

Note

When you create an import job in the Amazon Cognito console, you can create the IAM role at the same time. When you choose to Create a new IAM role, Amazon Cognito automatically applies the appropriate trust policy and IAM policy to the role.

To create the CloudWatch Logs IAM role for user pool import (AWS CLI, API)

Sign in to the AWS Management Console and open the IAM console at https://console.aws.amazon.com/iam/.
Create a new IAM role for an AWS service. For detailed instructions, see Creating a role for an AWS service in the AWS Identity and Access Management User Guide.
1. When you select a Use case for your Trusted entity type, choose any service. Amazon Cognito isn't currently listed in service use cases.
2. In the Add permissions screen, choose Create policy and insert the following policy statement. Replace REGION with the AWS Region of your user pool, for example us-east-1. Replace ACCOUNT with your AWS account ID, for example 111122223333.
```
{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Effect": "Allow",
            "Action": [
                "logs:CreateLogGroup",
                "logs:CreateLogStream",
                "logs:DescribeLogStreams",
                "logs:PutLogEvents"
            ],
            "Resource": [
                "arn:aws:logs:REGION:ACCOUNT:log-group:/aws/cognito/*"
            ]
        }
    ]    
}
```
Because you didn't choose Amazon Cognito as the trusted entity when you created the role, you now must manually edit the trust relationship of the role. Choose Roles from navigation pane of the IAM console, then choose the new role that you created.
Choose the Trust relationships tab.
Choose Edit trust policy.

Paste the following policy statement into Edit trust policy, replacing any existing text:


{
        "Version": "2012-10-17",
        "Statement": [
            {
                "Effect": "Allow",
                "Principal": {
                    "Service": "cognito-idp.amazonaws.com"
                },
                "Action": "sts:AssumeRole"
            }
        ]
    }

Choose Update policy.
Note the role ARN. You'll provide the ARN when you create your import job.

Creating the user import CSV file

Before you can import your existing users into your user pool, you must create a comma-separated values (CSV) file that contains the users that you want to import, and their attributes. From your user pool, you can retrieve a user import file with headers that reflect the attribute schema of your user pool. You can then insert user information that matches the formatting requirements in Formatting the CSV file.

Downloading the CSV file header (console)

Use the following procedure to download the CSV header file.

To download the CSV file header

Go to the Amazon Cognito console. You might be prompted for your AWS credentials.
Choose User Pools.
Choose an existing user pool from the list.
Choose the Users tab.
In the Import users section, choose Create an import job.
Under Upload CSV, select the template.csv link and download the CSV file.

Downloading the CSV file header (AWS CLI)

To get a list of the correct headers, run the following CLI command, where USER_POOL_ID is the user pool identifier for the user pool you'll import users into:


aws cognito-idp get-csv-header --user-pool-id "USER_POOL_ID"

Sample response:


{
    "CSVHeader": [
        "name",
        "given_name",
        "family_name",
        "middle_name",
        "nickname",
        "preferred_username",
        "profile",
        "picture",
        "website",
        "email",
        "email_verified",
        "gender",
        "birthdate",
        "zoneinfo",
        "locale",
        "phone_number",
        "phone_number_verified",
        "address",
        "updated_at",
        "cognito:mfa_enabled",
        "cognito:username"
    ],
    "UserPoolId": "USER_POOL_ID"
}

Formatting the CSV file

The downloaded user import CSV header file looks like the following string. It also includes any custom attributes you have added to your user pool.


cognito:username,name,given_name,family_name,middle_name,nickname,preferred_username,profile,picture,website,email,email_verified,gender,birthdate,zoneinfo,locale,phone_number,phone_number_verified,address,updated_at,cognito:mfa_enabled

Edit your CSV file so that it includes this header and the attribute values for your users, and is formatted according to the following rules:

Note

For more information about attribute values, such as proper format for phone numbers, see Working with user attributes.

The first row in the file is the downloaded header row, which contains the user attribute names.
The order of columns in the CSV file doesn't matter.
Each row after the first row contains the attribute values for a user.
All columns in the header must be present, but you don't need to provide values in every column.
The following attributes are required:
- cognito:username
- cognito:mfa_enabled
- email_verified or phone_number_verified
  - At least one of the auto-verified attributes must be true for each user. An auto-verified attribute is an email address or phone number that Amazon Cognito automatically sends a code to when a new user joins your user pool.
  - The user pool must have at least one auto-verified attribute, either email_verified or phone_number_verified. If the user pool has no auto-verified attributes, the import job will not start.
  - If the user pool only has one auto-verified attribute, that attribute must be verified for each user. For example, if the user pool has only phone_number as an auto-verified attribute, the phone_number_verified value must be true for each user.
  Note
  For users to reset their passwords, they must have a verified email or phone number. Amazon Cognito sends a message containing a reset password code to the email or phone number specified in the CSV file. If the message is sent to the phone number, it is sent by SMS message. For more information, see Verifying contact information at sign-up.
- email (if email_verified is true)
- phone_number (if phone_number_verified is true)
- Any attributes that you marked as required when you created the user pool
Attribute values that are strings should not be in quotation marks.
If an attribute value contains a comma, you must put a backslash (\) before the comma. This is because the fields in a CSV file are separated by commas.
The CSV file contents should be in UTF-8 format without byte order mark.
The cognito:username field is required and must be unique within your user pool. It can be any Unicode string. However, it cannot contain spaces or tabs.
The birthdate values, if present, must be in the format mm/dd/yyyy. This means, for example, that a birthdate of February 1, 1985 must be encoded as 02/01/1985.
The cognito:mfa_enabled field is required. If you've set multi-factor authentication (MFA) to be required in your user pool, this field must be true for all users. If you've set MFA to be off, this field must be false for all users. If you've set MFA to be optional, this field can be either true or false, but it can't be empty.
The maximum row length is 16,000 characters.
The maximum CSV file size is 100 MB.
The maximum number of rows (users) in the file is 500,000. This maximum doesn't include the header row.
The updated_at field value is expected to be epoch time in seconds, for example: 1471453471.
Any leading or trailing white space in an attribute value will be trimmed.

The following list is a example CSV import file for a user pool with no custom attributes. Your user pool schema might differ from this example. In that case, you must provide test values in the CSV template that you download from your user pool.


cognito:username,name,given_name,family_name,middle_name,nickname,preferred_username,profile,picture,website,email,email_verified,gender,birthdate,zoneinfo,locale,phone_number,phone_number_verified,address,updated_at,cognito:mfa_enabled
John,,John,Doe,,,,,,,johndoe@example.com,TRUE,,02/01/1985,,,+12345550100,TRUE,123 Any Street,,FALSE
Jane,,Jane,Roe,,,,,,,janeroe@example.com,TRUE,,01/01/1985,,,+12345550199,TRUE,100 Main Street,,FALSE

Creating and running the Amazon Cognito user pool import job

This section describes how to create and run the user pool import job by using the Amazon Cognito console and the AWS Command Line Interface (AWS CLI).

Topics

Importing users from a CSV file (console)
Importing users (AWS CLI)

Importing users from a CSV file (console)

The following procedure describes how to import the users from the CSV file.

To import users from the CSV file (console)

Go to the Amazon Cognito console. You might be prompted for your AWS credentials.
Choose User Pools.
Choose an existing user pool from the list.
Choose the Users tab.
In the Import users section, choose Create an import job.
On the Create import job page, enter a Job name.
Choose to Create a new IAM role or to Use an existing IAM role.
1. If you chose Create a new IAM role, enter a name for your new role. Amazon Cognito will automatically create a role with the correct permissions and trust relationship. The IAM principal that creates the import job must have permissions to create IAM roles.
2. If you chose Use an existing IAM role, choose a role from the list under IAM role selection. This role must have the permissions and trust policy described in Creating the CloudWatch Logs IAM role.
Choose Create job to submit your job, but start it later. Choose Create and start job to submit your job and start it immediately.
If you created your job but didn't start it, you can start it later. In the Users tab under Import users, choose your import job, then select Start. You can also submit a StartUserImportJob API request from an AWS SDK.
Monitor the progress of your user import job in the Users tab under Import users. If your job doesn't succeed, you can select the Status value. For additional details, select View the CloudWatch logs for more details and review any issues in the CloudWatch Logs console.

Importing users (AWS CLI)

The following CLI commands are available for importing users into a user pool:

create-user-import-job
get-csv-header
describe-user-import-job
list-user-import-jobs
start-user-import-job
stop-user-import-job

To get the list of command line options for these commands, use the help command line option. For example:


aws cognito-idp get-csv-header help

Creating a user import job

After you create your CSV file, create a user import job by running the following CLI command, where JOB_NAME is the name you're choosing for the job, USER_POOL_ID is the user pool ID for the user pool into which the new users will be added, and ROLE_ARN is the role ARN you received in Creating the CloudWatch Logs IAM role:


aws cognito-idp create-user-import-job --job-name "JOB_NAME" --user-pool-id "USER_POOL_ID" --cloud-watch-logs-role-arn "ROLE_ARN"

The PRE_SIGNED_URL returned in the response is valid for 15 minutes. After that time, it will expire and you must create a new user import job to get a new URL.

Example Sample response:


{
    "UserImportJob": {
        "Status": "Created",
        "SkippedUsers": 0,
        "UserPoolId": "USER_POOL_ID",
        "ImportedUsers": 0,
        "JobName": "JOB_NAME",
        "JobId": "JOB_ID",
        "PreSignedUrl": "PRE_SIGNED_URL",
        "CloudWatchLogsRoleArn": "ROLE_ARN",
        "FailedUsers": 0,
        "CreationDate": 1470957431.965
    }
}

Status values for a user import job

In the responses to your user import commands, you'll see one of the following Status values:

Created - The job was created but not started.
Pending - A transition state. You have started the job, but it has not begun importing users yet.
InProgress - The job has started, and users are being imported.
Stopping - You have stopped the job, but the job has not stopped importing users yet.
Stopped - You have stopped the job, and the job has stopped importing users.
Succeeded - The job has completed successfully.
Failed - The job has stopped due to an error.
Expired - You created a job, but did not start the job within 24-48 hours. All data associated with the job was deleted, and the job can't be started.

Uploading the CSV file

Use the following curl command to upload the CSV file containing your user data to the presigned URL that you obtained from the response of the create-user-import-job command.


curl -v -T "PATH_TO_CSV_FILE" -H "x-amz-server-side-encryption:aws:kms" "PRE_SIGNED_URL"

In the output of this command, look for the phrase "We are completely uploaded and fine". This phrase indicates that the file was uploaded successfully.

Describing a user import job

To get a description of your user import job, use the following command, where USER_POOL_ID is your user pool ID, and JOB_ID is the job ID that was returned when you created the user import job.


aws cognito-idp describe-user-import-job --user-pool-id "USER_POOL_ID" --job-id "JOB_ID"

Example Sample response:


{
    "UserImportJob": {
        "Status": "Created",
        "SkippedUsers": 0,
        "UserPoolId": "USER_POOL_ID",
        "ImportedUsers": 0,
        "JobName": "JOB_NAME",
        "JobId": "JOB_ID",
        "PreSignedUrl": "PRE_SIGNED_URL",
        "CloudWatchLogsRoleArn":"ROLE_ARN",
        "FailedUsers": 0,
        "CreationDate": 1470957431.965
    }
}

In the preceding sample output, the PRE_SIGNED_URL is the URL that you uploaded the CSV file to. The ROLE_ARN is the CloudWatch Logs role ARN that you received when you created the role.

Listing your user import jobs

To list your user import jobs, use the following command:


aws cognito-idp list-user-import-jobs --user-pool-id "USER_POOL_ID" --max-results 2

Example Sample response:


{
    "UserImportJobs": [
        {
            "Status": "Created",
            "SkippedUsers": 0,
            "UserPoolId": "USER_POOL_ID",
            "ImportedUsers": 0,
            "JobName": "JOB_NAME",
            "JobId": "JOB_ID",
            "PreSignedUrl":"PRE_SIGNED_URL",
            "CloudWatchLogsRoleArn":"ROLE_ARN",
            "FailedUsers": 0,
            "CreationDate": 1470957431.965
        },
        {
            "CompletionDate": 1470954227.701,
            "StartDate": 1470954226.086,
            "Status": "Failed",
            "UserPoolId": "USER_POOL_ID",
            "ImportedUsers": 0,
            "SkippedUsers": 0,
            "JobName": "JOB_NAME",
            "CompletionMessage": "Too many users have failed or been skipped during the import.",
            "JobId": "JOB_ID",
            "PreSignedUrl":"PRE_SIGNED_URL",
            "CloudWatchLogsRoleArn":"ROLE_ARN",
            "FailedUsers": 5,
            "CreationDate": 1470953929.313
        }
    ],
    "PaginationToken": "PAGINATION_TOKEN"
}

Jobs are listed in chronological order from last created to first created. The PAGINATION_TOKEN string after the second job indicates that there are additional results for this list command. To list the additional results, use the --pagination-token option as follows:


aws cognito-idp list-user-import-jobs --user-pool-id "USER_POOL_ID" --max-results 10 --pagination-token "PAGINATION_TOKEN"

Starting a user import job

To start a user import job, use the following command:


aws cognito-idp start-user-import-job --user-pool-id "USER_POOL_ID" --job-id "JOB_ID"

Only one import job can be active at a time per account.

Example Sample response:


{
    "UserImportJob": {
        "Status": "Pending",
        "StartDate": 1470957851.483,
        "UserPoolId": "USER_POOL_ID",
        "ImportedUsers": 0,
        "SkippedUsers": 0,
        "JobName": "JOB_NAME",
        "JobId": "JOB_ID",
        "PreSignedUrl":"PRE_SIGNED_URL",
        "CloudWatchLogsRoleArn": "ROLE_ARN",
        "FailedUsers": 0,
        "CreationDate": 1470957431.965
    }
}

Stopping a user import job

To stop a user import job while it is in progress, use the following command. After you stop the job, it cannot be restarted.


aws cognito-idp stop-user-import-job --user-pool-id "USER_POOL_ID" --job-id "JOB_ID"

Example Sample response:


{
    "UserImportJob": {
        "CompletionDate": 1470958050.571,
        "StartDate": 1470958047.797,
        "Status": "Stopped",
        "UserPoolId": "USER_POOL_ID",
        "ImportedUsers": 0,
        "SkippedUsers": 0,
        "JobName": "JOB_NAME",
        "CompletionMessage": "The Import Job was stopped by the developer.",
        "JobId": "JOB_ID",
        "PreSignedUrl":"PRE_SIGNED_URL",
        "CloudWatchLogsRoleArn": "ROLE_ARN",
        "FailedUsers": 0,
        "CreationDate": 1470957972.387
    }
}

Viewing the user pool import results in the CloudWatch console

You can view the results of your import job in the Amazon CloudWatch console.

Viewing the results

The following steps describe how to view the user pool import results.

To view the results of the user pool import

Sign in to the AWS Management Console and open the CloudWatch console at https://console.aws.amazon.com/cloudwatch/.
Choose Logs.
Choose the log group for your user pool import jobs. The log group name is in the form /aws/cognito/userpools/USER_POOL_ID/USER_POOL_NAME.
Choose the log for the user import job you just ran. The log name is in the form JOB_ID/JOB_NAME. The results in the log refer to your users by line number. No user data is written to the log. For each user, a line similar to the following appears:
- [SUCCEEDED] Line Number 5956 - The import succeeded.
- [SKIPPED] Line Number 5956 - The user already exists.
- [FAILED] Line Number 5956 - The User Record does not set any of the auto verified attributes to true. (Example: email_verified to true).

Interpreting the results

Successfully imported users have their status set to "PasswordReset".

In the following cases, the user will not be imported, but the import job will continue:

No auto-verified attributes are set to true.
The user data doesn't match the schema.
The user couldn't be imported due to an internal error.

In the following cases, the import job will fail:

The Amazon CloudWatch Logs role cannot be assumed, doesn't have the correct access policy, or has been deleted.
The user pool has been deleted.
Amazon Cognito is unable to parse the .csv file.

Requiring imported users to reset their passwords

The first time each imported user signs in and enters any password, they are required to enter a new password. The following procedure describes the user experience in a custom app with local users after you import a CSV file. If your users sign in with the hosted UI, Amazon Cognito prompts them to set a new password when they first sign in.

Requiring imported users to reset their passwords

In your app, silently attempt sign-in for the current user with InitiateAuth using a random password.
Amazon Cognito returns a NotAuthorizedException when PreventUserExistenceErrors is enabled. Otherwise, it returns PasswordResetRequiredException.
Your app makes a ForgotPassword API request and resets the user's password.
1. The app submits the username in a ForgotPassword API request.
2. Amazon Cognito sends a code to the verified email or phone number. The destination depends on the values you provided for email_verified and phone_number_verified in your CSV file. The response to the ForgotPassword request indicates the destination of the code.
  
  Note
  Your user pool must be configured to verify emails or phone numbers. For more information, see Signing up and confirming user accounts.
3. Your app displays a message to your user to check the location where the code was sent, and prompts your user to enter the code and a new password.
4. The user enters the code and new password in the app.
5. The app submits the code and new password in a ConfirmForgotPassword API request.
6. Your app redirects your user to sign-in.

Warning Javascript is disabled or is unavailable in your browser.

To use the Amazon Web Services Documentation, Javascript must be enabled. Please refer to your browser's Help pages for instructions.

Document Conventions

User migration trigger import

Attributes