baseStriker: Office 365 Security Fails To Secure 100 Million Email Users

By Yoav Nathaniel, Customer Success Manager, Avanan

We recently uncovered what may be the largest security flaw in Office 365 since the service was created. Unlike similar attacks that could be learned and blocked, using this vulnerability hackers can completely bypass all of Microsoft’s security, including its advanced services - ATP, Safelinks, etc.

The name baseStriker refers to the method hackers use to take advantage of this vulnerability: splitting and disguising a malicious link using a tag called the <base> URL tag.

So far we have only seen hackers using this vulnerability to send phishing attacks, but but it is also capable of distributing ransomware, malware and other malicious content.

How a baseStriker Attack Works

The attack sends a malicious link, that would ordinarily be blocked by Microsoft, past their security filters by splitting the URL into two snippets of HTML: a base tag and a regular href tag. Here's a short video showing how it works:

Traditional Phish: This html email would be blocked because the URL is known to be malicious.

When scanning this, Office 365 sees the malicious URL, performs a lookup against a list of known bad links, and blocks it. Office 365 Safelink, for customers that purchased ATP, also replaces the URL with a Safelink URL and prevents the end-user from going to the phishing site.

Phish using baseStriker method: This email, however, has the same malicious link presented to the end-user but is let through because the email filters are not handling the <base> HTML code correctly.

In this example, Office 365 only performs the lookup on the base domain, ignoring the relative URL in the rest of the body. Because only part of the URL is tested, it mistakenly appears to not exist in the malicious URL database and the email is let through. Furthermore, Safelinks does not replace the malicious link, and the user get the original malicious link, can click it to get right to the phishing page.

In a nutshell, this attack method is the email equivalent of a virus that blinds the immune system. So even if the attack is already known, Microsoft does not have a way to see it and lets it through.

Are you vulnerable?

We have tested the vulnerability on several configurations and found that anyone using Office 365 in any configuration is vulnerable. If you are using Gmail, you don’t have this issue. If you are protecting Office 365 with Mimecast you are secure. Proofpoint is also vulnerable – if you are using Proofpoint you also have this problem.

Here’s a summary of our findings:

What can you do?

As of the time of writing, there still is no fix so there's no configuration you can make in your Office 365. We have notified Microsoft and Proofpoint and will update if we learn more.

Because this vulnerability is already known to hackers, an immediate first step would be to notify your end-users and reinforce the risk of phishing attacks.

We are recommending customers enable multi-factor authentication to make it harder to take over their account. This will not protect from malware and other types of phishing, but will help with credential harvesting.

Finally, for users of Gmail and Office 365, even if you are not vulnerable to this attack, we always recommend adding a layer of email security for malware, phishing, and account take-over to protect from the sophisticated attacks that the default security does not block. As this is not the first attack that has found a way past default security measures and it will not be the last.

Updates

• 5/1/2018: Avanan identified attackers are leveraging a critical vulnerability in Microsoft Office 365 email service that allows them to completely bypass O365 built in security
• 5/2/2018 11:00am: Avanan reported this issue to Microsoft
• 5/2/2018 11:00am: Avanan tested Gmail and it does not suffer from this vulnerability
• 5/2/2018 11:30am: Avanan tested Mimecast and Proofpoint.
  • Mimecast is fine.
  • Proofpoint has the same vulnerability. Therefore, if you use Proofpoint you are not secured. We informed Proofpoint at 11:44am EDT on May 2nd, 2018.