What do I do if I notice unauthorized activity in my AWS account?

Last updated: 2020-08-03

I see resources that I don't remember creating in the AWS Management Console.


I received a notification that my AWS resources or account might be compromised. What should I do?

Short description

If you observe unauthorized activity within your AWS account, or you believe that an unauthorized party has accessed your account, then do the following:


Change your AWS account root user password and the passwords of any IAM users

It's a best practice to change your passwords on a regular basis to avoid unauthorized use of your account.

To change your AWS account root user password, see Changing the AWS account root user password.

To change the password of an IAM user, see Managing passwords for IAM users.

For more information, see Best practices for security, identity and compliance. For security documentation specific to an AWS service, see AWS Security Documentation.

For information on best practices to consider when securing your account and its resources, see What are some best practices for securing my AWS account and its resources?

Rotate and delete all AWS access keys

If you find AWS access keys that you no longer need or didn't create, delete the access keys.

If your application currently uses an access key, replace the key with a new one:

  1. First, create a second key. Then, modify your application to use the new key.
  2. Disable (but do not delete) the first key.
  3. If there are any problems with your application, reactivate the key temporarily. When your application is fully functional, and the first key is in the disabled state, then delete the first key.
  4. Delete any IAM users that you didn't create.

Treat AWS access keys the same way you treat an account password:

For instructions on determining which IAM user created a resource and restricting access, see How can I troubleshoot unusual resource activity with my AWS account?

Delete any potentially unauthorized IAM users

  1. Open the IAM console.
  2. Choose Users in the navigation pane.
  3. Choose each IAM user from the list, and then check under Permissions policies for a policy called AWSExposedCredentialPolicy_DO_NOT_REMOVE. If the user has this attached policy, you must delete the user.
  4. Delete any IAM users that you didn't create.
  5. Change the password for IAM users that you created and want to keep.

Delete any unrecognized or unauthorized resources

Sign in to your AWS account, and then check to make sure that all the resources on your account are resources that you launched. Be sure to check all AWS Regions, even Regions where you never launched AWS resources. Pay special attention to the following:

To delete Lambda functions and layers, do the following:

  1. Open the Lambda console.
  2. In the navigation pane, choose Functions.
  3. Select the functions that you want to delete.
  4. For Actions, choose Delete.
  5. In the navigation pane, choose Layers.
  6. Select the layer that you want to delete.
  7. Choose Delete.

For information on how to delete a resource associated with a particular AWS service, see How do I terminate all my resources before closing my AWS account?

Contact AWS Support

If you received a notification from AWS about your account, sign in to the AWS Support Center, and then respond to the notification.

If you can't sign in to your account, use the Contact Us form to request help from AWS Support.

If you have any questions or concerns, create a new AWS Support case in the AWS Support Center.

Note: Don't include sensitive information in your correspondence, such as AWS access keys, passwords, or credit card information.

Use AWS Git projects to scan for evidence of unauthorized use

AWS offers Git projects that you can install to help you protect your account:

  • Git Secrets can scan merges, commits, and commit messages for secret information (that is, access keys). If Git Secrets detects prohibited regular expressions, it can reject those commits from being posted to public repositories.
  • Use AWS Step Functions and AWS Lambda to generate Amazon CloudWatch Events from AWS Health or by Trusted Advisor. If there is evidence that your access keys are exposed, the projects can help you to automatically detect, log, and mitigate the event.