Imagine the following scenario, after a successful phishing campaign, an attacker finds themselves with an active shell on one of your non-sensitive employee’s devices. Using Living-off-the-Land (LotL) tactics (employing built in tools and commands), the attacker starts identifying their next targets and configuration weaknesses to leverage to comprise them. One such vulnerability might be the infamous print spooler bug.
As published on the security conference DerbyCon at (2018), the print spooler bug is a way to gain elevated credentials (a sensitive computer account Kerberos ticket in this case) by exploiting two distinct configurations.
Using these two configurations in tandem, an attacker can “request” the sensitive device Kerberos ticket using the print spooler service through the victim’s device; and because unconstrained Kerberos delegation is enabled, that request will contain the highly sought-after Kerberos authentication ticket (TGT) that allows the attacker to impersonate the domain controller computer account - pretty bad, right?
*Image from https://adsecurity.org/?p=4056
As part of Azure ATP, we introduced the Identity Security Posture assessment that automatically looks for these kinds of risky configurations in your on-premises environment. This assessment and the recommendation are now integrated in Microsoft Secure Score, makes creating a mitigation plan for these recommended actions easier than ever. To view the recommendations you can use the following steps:
Once mitigated, Microsoft Secure Score will reflect the actions taken by increasing your score and mark these improvement actions as “Completed”.
Now, let’s consider another scenario.
After mitigating the print spooler bug , you discover another improvement action called “Stop clear text credential exposure”. After clicking the “Manage” button to access the assessment to view additional details, you discover that a sensitive entity and an accompanying device are authenticating over LDAP simple bind. To make matters even worse, the exposed user is also sensitive (belonging to the Domain admin group)
You then realize that if an attacker can get onto your network, they can “sniff” the transmitted credentials that are in clear text and gain elevated credentials.
To further investigate this case, you will need additional details regarding this type of unsecure communication such as:
To answer these questions, let’s utilize one of MTP’s cool features called Advanced hunting which is now also enriched with Azure ATP activities alongside other security products such as Microsoft Defender ATP, Cloud App Security and Azure AD, and data from Office 365.
We’ll start by going to the Advanced hunting page in the new M365 security portal and from there, we now have access to all the different types of hunting data available for us, including two main tables relevant for this task:
To get started, we can write a query, joining detected logon sessions and network events from both these tables, resulting in a consolidated result that includes the all the unsecure logon activities and their initiating process names.
Using the query result, we can construct a cohesive story:
“On Aug 05, Testuser2 used the LDP.exe executable to generate an unsecure authentication using LDAP Simple Bind from Client5.”
We now have the relevant details to create a full remediation plan for this activity, such as simply blocking LDP.exe or configuring the application to require a more secure authentication while blocking insecure authentications at the domain level.
And there you have it! Using Azure ATP’s Identity Security Posture assessment with Microsoft Secure Score and Advanced hunting, we can build a simple mitigation plan for our organization’s security weak spots, making it that much harder for the bad guy to get in
For more information about Identity Security Posture assessments and MTP, see
* Learn more about Identity security posture assessments.
* Learn more about Microsoft Secure Score
* Azure ATP activities in advanced hunting
Or Tsemah, Senior Product Manager, Azure ATP.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.