This post is an overview of AWS Security Hub and the value it brings to a security operations (SECOPS) organization for monitoring and alerting on AWS cloud activity.
What is Security Hub?
Security Hub serves as a single pane of glass through which you can view security findings reported by AWS GuardDuty, Inspector, Macie, Access Analyzer, and Firewall Manager.
For more general info, see the Security Hub FAQs page.
Security Findings
More than this, Security Hub automatically aggregates, organizes, and assigns risk to these findings, allowing your security operations (SECOPS) team to monitor events occurring in your cloud. Using the AWS Security Finding Format (ASFF), these events can be piped to CloudWatch Events where they can be consumed by other tools to enable automation.
This finding format is significant, because it represents the first (to my knowledge) time that AWS has tried to tie the findings from all of their security tools together. Security Hub itself does a decent job of allowing one to analyze a finding by drilling down into the relevant service console, but there is a limit to the number of services that can plug into Security Hub at the moment. With the ASFF, it is possible for any tool to produce a finding that Security Hub can consume.
Governance
Security Hub also offers an overview of an organization’s compliance with particular security frameworks or ad-hoc configuration baselines leveraging AWS Config. Out-of-the-box, Security Hub will display adherence to the CIS AWS Foundations and PCI DSS benchmarks, as of this writing (2020-02-21). As with security findings, Security Hub simplifies the view into these governance standards to allow one to quickly assess organizational adherence.
Why Should I Use SecurityHub?
I believe the first and most obvious benefit of Security Hub is quickly disseminating information to security operations and compliance teams. However, the real value of Security Hub moving forward is the orchestration of security findings via all of the other security tools using the ASFF.
How Should I Use SecurityHub?
Every organization is going to answer this question differently. I want to offer some pointers from my own lessons learned.
Designate A Master Security Monitoring Account
A master security monitoring account eases the pain of consuming findings across many accounts and regions. An architecture I recommend is to designate one account as the master security monitoring account and invite all of your other organizational accounts into it.
This is easily done in the console by navigating to Settings > Accounts > Add accounts:
Member accounts can be added by creating a CSV with the following format:
Account ID,Email
000111222333,[email protected]
111222333444,[email protected]
222333444555,[email protected]
...
These can be then bulk added to Security Hub by navigating to Settings > Accounts > Add accounts > Upload List (.csv):
After clicking Add Accounts, Security Hub will display the accounts to be added. Click Next. Now invite these accounts by selecting the “Invitation not yet sent” drop-down, select all of the accounts you want to invite, and then click Actions > Invite:
You’ll be prompted for a message to send to the account owner and then you can invite the account. The annoying part now is that you must login to each account and accept the invitation. For a large organization, this can take quite some time :(
If there’s a smarter way to do this, please tweet me @unl0ckd!
Rollout AWS Organization Policy to prevent tampering
Unfortunately, principals with sufficient access in each account you invite to Security Hub could disable it. AWS Organizations service control policy can be used to prevent this. An example SCP might look something like this, but do not use this example verbatim without testing in your environment!
{
"Version": "2012-10-17",
"Statement": [
{
"Sid": "PreventTamper",
"Effect": "Deny",
"Action": [
"account:*",
"organizations:*"
],
"Resource": [
"*"
]
}
]
}
Enable Security Standards
Depending on your organizational needs, you may want to enable the AWS CIS Foundations or PCI-DSS security standards in Security Hub. This is easily done by navigating to Security Standards from the Security Hub homepage.
For these standards to fully work, AWS Config must be enabled in each monitored account
Conclusion
If you’ve followed along with this article, your SECOPS team should now be able to view aggregated security and compliance information across all of your AWS accounts from one pane of glass.
I want to give AWS credit where it’s due: I just described enabling security monitoring for an entire organization’s AWS presence in a five-minute blog post due to the engineering effort that went into Security Hub. That is remarkable to me. Kudos, Security Hub team!