ghostbuster 1.0.2

Eliminate dangling elastic IPs by performing analysis on your resources within all your AWS accounts.

Table of Contents

• Table of Contents

• Ghostbuster

  • The problem

  • Project Features

  • Installing Ghostbuster

  • Using Ghostbuster

  • Example Output

  • Setting up your AWS accounts

  • Setting up Cloudflare (Optional)

  • Authors

  • License

Ghostbuster

Eliminate dangling elastic IPs by performing analysis on your resources within all your AWS accounts.

Ghostbuster obtains all the DNS records present in all of your AWS accounts (Route53), and can optionally take in records via CSV input, or via Cloudflare.

After these records are collected, Ghostbuster iterates through all of your AWS Elastic IPs and Network Interface Public IPs and collects this data.

By having a complete picture of the DNS records (from route53, file input or cloudflare) and having a complete picture of the AWS IPs owned by your organization, this tool can detect subdomains that are pointing to dangling elastic IPs (IPs you no longer own).

The problem

When you are deploying infrastructure to AWS, you may spin up EC2 instances which have an IP associated with them. When you create DNS records pointing to these IPs, but forget to remove the DNS records after the EC2 instance has been given a new IP or destroyed, you are susceptible to subdomain takeover attacks.

There has been a great amount of research done on elastic IP takeovers, where it is possible for attackers to continually claim elastic IPs until they obtain an IP associated with a subdomain of the company they are targeting.

While AWS frequently bans accounts that are attempting to perform this attack pattern, no long term fix has been released by AWS.

The impact of dangling elastic IP subdomain takeover attacks are more serious than a typical subdomain takeover where you can only control the content being served. With dangling elastic IP takeovers, it is possible for an attacker to do the following:

• Claim SSL certificates for the subdomain

• Listen for traffic on all ports (potentially discovering sensitive information still being sent to the subdomain)

• Run server-side scripts with the ability to steal HTTPOnly cookies, typically leading to a one-click account takeover attack when cookies are scoped to *.domain.com

Project Features

• Dynamically iterates through each AWS profile configured in .aws/config

• Pulls A records from AWS Route53

• Pulls A records from Cloudflare (optional)

• Pulls A records from CSV input (optional)

• Iterate through all regions, a single region, or a comma delimitted list of regions

• Obtains all Elastic IPs associated with all of your AWS accounts

• Obtains all Public IPs associated with all of your AWS accounts

• Cross checks the DNS records, with IPs owned by your organization to detect potential takeovers

• Slack Webhook support to send notifications of takeovers

Important:

In order for this tool to be effective, it must have a complete picture of your AWS environment. If it does not have a complete picture, it will lead to false positive findings.

Installing Ghostbuster

Installing Ghostbuster is as simple as running: pip install ghostbuster. The CLI tool will then be accessible via the ghostbuster command.

This requires a Python 3.x environment.

Using Ghostbuster

❯ ghostbuster scan aws --help
Usage: ghostbuster scan aws [OPTIONS]

  Scan for dangling elastic IPs inside your AWS accounts.

Options:
  --profile TEXT          Specify a specific AWS profile to run ghostbuster
                          on.

  --skipascii             Skip printing the ASCII art when starting up
                          Ghostbuster.

  --json                  Only return a JSON object.
  --slackwebhook TEXT     Specify a Slack webhook URL to send notifications
                          about potential takeovers.

  --records PATH          Manually specify DNS records to check against.
                          Ghostbuster will check these IPs after checking
                          retrieved DNS records. See records.csv for an
                          example.

  --cloudflaretoken TEXT  Pull DNS records from Cloudflare, provide a CF API
                          token.

  --allregions            Run on all regions.
  --exclude TEXT          Comma delimited list of profile names to exclude.
  --regions TEXT          Comma delimited list of regions to run on.
  --help                  Show this message and exit.

Example Commands

Run Ghostbuster with access to Cloudflare DNS records, send notifications to a Slack webhook, iterate through every AWS profile configured in .aws/config or .aws/credentials for all AWS regions

❯ ghostbuster scan aws --cloudflaretoken APIKEY --slackwebhook https://hooks.slack.com/services/KEY --allregions

Run Ghostbuster with a manually input list of subdomain A records (see records.csv in this repo for example file):

❯ ghostbuster scan aws --records records.csv

You can specify specific regions using --regions set to a comma delimited list of regions i.e. us-east-1,us-west-1.

Example Output

❯ ghostbuster scan aws --cloudflaretoken whougonnacall
Obtaining all zone names from Cloudflare.
Obtaining DNS A records for all zones from Cloudflare.
Obtained 33 DNS A records so far.
Obtaining Route53 hosted zones for AWS profile: default.
Obtaining Route53 hosted zones for AWS profile: account-five.
Obtaining Route53 hosted zones for AWS profile: account-four.
Obtaining Route53 hosted zones for AWS profile: account-four-deploy.
Obtaining Route53 hosted zones for AWS profile: account-two-deploy.
Obtaining Route53 hosted zones for AWS profile: account-one-deploy.
Obtaining Route53 hosted zones for AWS profile: account-three-deploy.
Obtaining Route53 hosted zones for AWS profile: account-six.
Obtaining Route53 hosted zones for AWS profile: account-seven.
Obtaining Route53 hosted zones for AWS profile: account-one.
Obtained 124 DNS A records so far.
Obtaining EIPs for region: us-east-1, profile: default
Obtaining IPs for network interfaces for region: us-east-1, profile: default
Obtaining EIPs for region: us-east-1, profile: account-five
Obtaining IPs for network interfaces for region: us-east-1, profile: account-five
Obtaining EIPs for region: us-east-1, profile: account-four
Obtaining IPs for network interfaces for region: us-east-1, profile: account-four
Obtaining EIPs for region: us-east-1, profile: account-four-deploy
Obtaining IPs for network interfaces for region: us-east-1, profile: account-four-deploy
Obtaining EIPs for region: us-east-1, profile: account-two-deploy
Obtaining IPs for network interfaces for region: us-east-1, profile: account-two-deploy
Obtaining EIPs for region: us-east-1, profile: account-one-deploy
Obtaining IPs for network interfaces for region: us-east-1, profile: account-one-deploy
Obtaining EIPs for region: us-east-1, profile: account-three-deploy
Obtaining IPs for network interfaces for region: us-east-1, profile: account-three-deploy
Obtaining EIPs for region: us-east-1, profile: account-six
Obtaining IPs for network interfaces for region: us-east-1, profile: account-six
Obtaining EIPs for region: us-east-1, profile: account-seven
Obtaining IPs for network interfaces for region: us-east-1, profile: account-seven
Obtaining EIPs for region: us-east-1, profile: account-one
Obtaining IPs for network interfaces for region: us-east-1, profile: account-one
Obtained 415 unique elastic IPs from AWS.


Takeover possible: {'name': 'takeover.assetnotecloud.com', 'records': ['52.54.24.193']}

Setting up your AWS accounts

The first step is creating keys or roles in your AWS accounts that grant the privileges necessary to read Route53 records and describe elastic addresses and EC2 network interfaces.

1. To create a new IAM user in AWS, visit the following URL: https://console.aws.amazon.com/iam/home#/users$new?step=details

2. Choose Access key - Programmatic access only, and click Next: Permissions.

3. Click Attach existing policies directly and then click Create policy.

4. Click JSON and then paste in the following policy:

{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Sid": "GhostbusterPolicy",
            "Effect": "Allow",
            "Action": [
                "ec2:DescribeAddresses",
                "ec2:DescribeNetworkInterfaces",
                "route53:ListResourceRecordSets",
                "route53:ListHostedZonesByName",
                "route53:GetTrafficPolicyInstance",
                "route53:GetTrafficPolicy"
            ],
            "Resource": "*"
        }
    ]
}

5. Click Next: Tags and then Next: Review.

6. Set the name of the policy to be GhostbusterPolicy.

7. Click Create Policy.

8. Go to https://console.aws.amazon.com/iam/home#/users$new?step=permissions&accessKey&userNames=ghostbuster&permissionType=policies

9. Select GhostbusterPolicy.

10. Click Next: Tags and then Next: Review.

11. Click on Create user and setup the AWS credentials in your .aws/credentials file.

Repeat the above steps for each AWS account you own.

This tool will work with however you’ve setup your AWS configuration (multiple keys, or cross-account assume role profiles). This is managed by boto3, the library used to interface with AWS.

An example configuration looks like this:

.aws/credentials:

[default]
aws_access_key_id = AKIAIII...
aws_secret_access_key = faAaAaA...

.aws/config:

[default]
output = table
region = us-east-1

[profile account-one]
role_arn = arn:aws:iam::911111111113:role/Ec2Route53Access
source_profile = default
region = us-east-1

[profile account-two]
role_arn = arn:aws:iam::911111111112:role/Ec2Route53Access
source_profile = default
region = us-east-1

[profile account-three]
region = us-east-1
role_arn = arn:aws:iam::911111111111:role/Ec2Route53Access
source_profile = default

Alternatively, instead of having roles which are assumed, you can also configure the .aws/credentials file to have a list of profiles and assocaited keys with scoped access.

Once your AWS configuration has been set with all the accounts in your AWS environment, you can then run the tool using the following command:

Setting up your AWS permissions for –roles

Ghostbuster can use roles instead of profiles which removes the need of having so many credentials at one place. However, roles approach will require creating additional IAM policies within you organisation.

Roles can be used via --roles roles.csv or --autoroles flags. To setup --roles to work, one needs to create GhostbusterTargetAccountRole role named exactly like that in every account that is being scanned:

{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Sid": "GhostbusterTargetAccountRole",
            "Effect": "Allow",
            "Action": [
                "ec2:DescribeAddresses",
                "ec2:DescribeNetworkInterfaces",
                "ec2:DescribeRegions",
                "route53:ListResourceRecordSets",
                "route53:ListHostedZonesByName",
                "route53:GetTrafficPolicyInstance",
                "route53:GetTrafficPolicy"
            ],
            "Resource": "*"
        }
    ]
}

The ghostbuster will be running in lambda/ec2/whatever with ghostbuster role - let’s call it ghostbuster master role. The master role must be able to assume TargetAccountRoles.

resource "aws_iam_policy" "ghostbuster_target_account_roles" {
  name        = "ghostbuster_target_account_roles"
  path        = "/"
  description = "Allow inspecting DNS and elastic IP data."

  policy = jsonencode({
    "Version" : "2012-10-17",
    "Statement" : [
      {
        "Effect" : "Allow",
        "Action" : "sts:AssumeRole",
        "Resource" : ["arn:aws:iam::*:role/GhostbusterTargetAccountRole"]
    }]
  })

  tags = {
    Project = "ghostbuster"
  }
}

Setting up your AWS permissions for –autoroles

Automatic account discovery requires additional permission compared to –roles. After setting up –roles to work, consider adding following IAM policy in an account that has organisation overview:

{
    sid = "BaseAccess"

    actions = [
      "organizations:DescribeAccount",
      "organizations:ListAccounts"
    ]

    resources = ["*"]
    effect    = "Allow"
  }
}

Then, in the account that is running ghostbuster, attach following policy to ghostbuster master role so it can assume the organisation lookup role:

{
    "Version" : "2012-10-17",
    "Statement" : [
      {
        "Effect" : "Allow",
        "Action" : "sts:AssumeRole",
        "Resource" : ["arn:aws:iam::ORG_LOOKUP_ACCOUNT_ID:role/ghostbuster-org-role"]
    }]
}

Don’t forget to replace ORG_LOOKUP_ACCOUNT_ID with actual account ID.

Setting up Cloudflare (Optional)

If you want Ghostbuster to pull in all the A records that you have set in Cloudflare, you will have to setup an API token that can read zones.

https://dash.cloudflare.com/profile/api-tokens

Setup a Cloudflare API token like shown in the screenshot below:

Once you have obtained this API token, make a note of it somewhere (password manager). In order to use it with Ghostbuster, pass it in via the cloudflaretoken argument.

Authors

• Shubham Shah - Initial work - github

See also the list of contributors who participated in this project.

License

GNU Affero General Public License

Testing Ghostbusters

1. Clone this repo.

2. Install virtualenv using pip3 install virtualenv

3. Create a virtual environment using virtualenv venv

4. Activate virtual environment using . venv/bin/activate

5. Install ghostbuster by going to root of repo. pip3 install --editable .

6. Make your changes and run the ghostbuster command.