Discovering Your Privileged Cloud Provider Entities with CEM

cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Announcements

Welcome Idira to LIVEcommunity!

Our community just got stronger! Explore new identity security resources, connect with experts, and discover more ways to learn and collaborate.

Meet Idira

Idira Success Blogs
13 min read
L4 Transporter
It’s hard to map the road to follow to get to your destination if you don’t know where you are. This adage about life applies to many things well, including cyber security. Understanding the current state of your privileged cloud entities is an important early step in implementing the right identity security and cloud security controls within your organization. In this article, we'll cover how to discover privileged cloud provider entities with CyberArk’s Cloud Entitlements Manager service, and what to do with that information once you have it.

Introduction
There are a few methods of discovering privileged cloud entities using Idira and cloud provider-native services, including Idira DNA, SkyArk and your native IAM reporting capabilities in your provider. All these options have a variety of functions and capabilities, but none more complete than Idira CEM. Organizations who are looking to scan and analyze all types privileged cloud entities including users, roles, groups, and services should strongly consider leveraging CEM.

What is CEM?
Idira Cloud Entitlements Manager (CEM) is comprehensive CIEM service designed to automate the manual and complex process of continuously scanning an organization’s cloud provider organizations and accounts for users, roles and services, among other functions like permissions analysis, least privilege policy creation and automatic credential onboarding to PAM. CEM is API-driven and SaaS delivered, and does not require agents or connectors to be installed on cloud providers, making it non-intrusive to the IT environment – zero-footprint. Cloud providers are scanned in read-only mode as to not impact systems.

Idira CEM is often used at the beginning of an organization’s cloud security or identity security initiative to conduct more advanced analysis and planning as it provides a manipulatable UI to review the cloud entities, as well as provides a CSV report output. It’s also common for organizations to want to understand their privileged landscape prior to implementing a PAM or cloud security solution as it may influence the architecture and security controls of the solution. CEM’s agent-less nature makes it easy to quickly start scanning and analyzing your cloud provider entities.

Idira CEM supports the discovery of the following cloud provider entities:
  • Amazon Web Services (AWS): IAM Users & Access Keys
  • Amazon Web Services (AWS): Federated Users
  • Amazon Web Services (AWS): IAM Roles
  • Amazon Web Services (AWS): IAM Groups
  • Amazon Web Services (AWS): Lambda Functions
  • AWS Elastic Kubernetes Service: Service Accounts
  • Microsoft Azure: Users
  • Microsoft Azure: API Keys
  • Microsoft Azure: Ent. Applications
  • Microsoft Azure: Service Principals
  • Microsoft Azure: Functions
  • Google Cloud Platform: Users
  • Google Cloud Platform: Groups
  • Google Cloud Platform: Service Accounts
  • Google Cloud Platform: Cloud Functions
Discovering with Idira CEM
Getting the Software and License
The first step in performing discovery with CEM is obtaining a license for the service and a tenant created for your organization. Already a CEM customer? Great! Skip this section. Don’t have CEM yet? Consider signing up for our free 30-day trial and follow along with the steps below!

Scanning Permissions
As any scanning software requires, CEM needs some sort of access to use in its process to scan the cloud provider environments (also referred to as workspaces). What CEM requires is essentially read permissions on each cloud provider environment to fetch details about accounts, users, roles, and their assigned permissions. The specific permissions and setup instructions for AWS, Microsoft Azure, Google Cloud and Amazon EKS can be found in their respective embedded links.

Coordinate Internally
One very important non-technical step is to ensure proper coordination internally as you prepare to configure CEM to scan your workspaces. CEM uses native cloud provider APIs to scan the cloud environments and is accessing those workspaces’ directory systems. Before running CEM, it is strongly recommended that deploying organizations assess the related legal requirements both under data protection and telecommunication laws as well as any other applicable laws and regulations on their own behalf.

We recommend that IT security personnel verify that corporate employment policies or local laws do not prohibit the scanning or analysis of data streams for network security purposes, because the tool involves scanning, locating and analyzing corporate access.


Idira CEM is a zero-footprint service, making it non-intrusive and non-agent based, but these API activities may trigger alerts in security applications and tools designed to detect anomalous cloud provider traffic. It’s recommended to alert your Security Operations Teams in advance so that the security work is coordinated and does not disrupt or distract from ordinary security monitoring or management of alerts.

Perform a Test Scan
It’s recommended to perform a test scan on each cloud provider type prior to launching your production scan. Doing so will allow you to become familiar with the CEM UI, the process itself, and the technical prerequisites required to run it. This gives you an opportunity to iron out any kinks in your internal processes or cloud provider permissions, giving your main scan the best chance of success.

Running the Scan
  1. Defining Your Cloud Provider Scope
    Before you configure CEM to perform its ongoing scans, you’ll need to define the scope of cloud provider accounts, subscriptions, projects, or organizations you want to scan (conceptually, not technically in the tool yet). Which instances are we interested in performing discovery against? The answer to that question is going to be dependent on your cloud security goal or initiative. Does your roadmap already break up cloud provider workspaces into segments or phases?  
  2. Configuring the Cloud Provider Workspace and Starting the Scan
    Your scan will automatically begin after you complete the configuration of your cloud provider workspace. So before you start your scan, be sure you’re familiar with the requirements for configuration. CEM will continuously scan your workspace(s) unless you suspend it manually or it encounters a scanning error.
Now We Wait…
Once CEM has kicked off your scan, the service will start scouring the cloud provider workspace(s) and analyzing the cloud entities it discovers. In the Platform management page you’ll see an icon for “scanning in progress” on each of the workspaces being scanned. When complete the icon will change to “scan complete” indicating that the scan is finished, and the results are available. The duration of each scan depends on the scale of the data and the platform, but you can expect the average time to scan to be around 15 minutes.


Analyzing the Data
Now that your scan is complete, what do you do with that data? It’s time to start looking through it. Of which there are two ways you can look at this data. The first being highly targeted – what specific scenario are you looking for to remediate (e.g., I’m only interested in securing my IAM administrator accounts now). The other is by looking at everything and seeing what you uncover, then reacting to that.


To help accelerate your analysis for those specific scenarios, we’ve included some instructions to analyze your CEM results for those common account types from the Cloud entities page (although this same logic can be applied to offline manipulation with the CSV file).
  • Admin
    In the top left-hand corner of the screen, click the “Filter” icon to expand the filter side menu. Under “Exceptional factors” select “Admin”. This will provide you with a complete list of the unique entities which are explicitly granted administrative permissions across their respective workspaces. These entities have permissions to control the entire cloud environment in which they’ve been granted permissions to. View the online documentation for more specifics on how each platform defines admins.
  • Shadow Admin
    In the top left-hand corner of the screen, click the “Filter” icon to expand the filter side menu. Under “Exceptional factors” select “Shadow Admin”. This will provide you with a complete list of the unique entities which have been granted permissions which enable the entity to escalate to become a full-blown administrator across their respective workspaces. These permissions may appear to be benign, but through escalation techniques they too have the ability to control the entire cloud environment in which they’ve been granted permissions to. View the online documentation for more specifics on what permissions grant an entity “Shadow admin” privileges for each cloud provider.
  • High Exposure
    On the column header, sort Exposure “High to Low”. This will provide you with a complete list of the unique entities sorted by the highest risk to the lowest risk. Focus your attention to the “High” risk entities with an exposure rating of 90% or higher. These entities all have admin permissions (per the previous bullet description) and often do not all require them, exposing your organization to unnecessary risk.
  • Medium Exposure
    On the column header, sort Exposure “High to Low”. This will provide you with a complete list of the unique entities sorted by the highest risk to the lowest risk. Focus your attention to the “Medium” risk entities with an exposure rating between 89% to 30%. This is generally the bulk of cloud entities. Most entities above 70% will likely have shadow admin permissions (but will clearly be marked accordingly). These entities typically contain a wide variety of permissions and often do not all require them, exposing your organization to unnecessary risk.
  • Low Exposure
    On the column header, sort Exposure “Low to High” (for this scenario it’s easier to flip it). This will provide you with a complete list of the unique entities sorted by the lowest risk to the highest risk. Focus your attention to the “Low” risk entities with an exposure rating between 0% to 29%. These entities typically contain a wide variety of permissions which are generally not sensitive or are explicitly used be the entity. The goal should be to get as many entities into this category as possible to mitigate your cloud risk.
When you’re looking to understand the broader picture there are some key things we suggest looking for or being aware of. Major influencers of an entities risk include its level of privilege (and whether they’re even necessary), scope of influence and ease of compromise. We’ve already covered the level of privilege and scope of influence in the scenarios described in the previous section, which you should be looking for anyway. In this next section, we’ll look at some of the indicators that are related to ease of compromise, which are typically attributes about the entity.
  • Entities with Shadow Admin Permissions
    As mentioned earlier, entities which contain shadow admin permissions have to escalate to become a full-blown administrator across their respective workspaces. These permissions may appear to be benign, but through escalation techniques they too have the ability to control the entire cloud environment in which they’ve been granted permissions to. These permissions often go unnoticed and under the radar, leading them to have a greater chance of abuse than their explicit admin counterparts.
  • Sensitive Workspace
    Look for entities that have access to or exist on workspace names that you deem sensitive or important. Are there certain cloud provider instances that have regulatory requirements, host critical applications, contain sensitive data? Filter by those workspace names and look for the any of the above-mentioned findings.
  • Personal User Accounts
    Look for user accounts that are a user’s personal name. This can be easier to do when the organization has a clear naming convention. Personal accounts typically lack the same security controls as explicitly defined privileged accounts or just-in-time based access, leaving them vulnerable to credential theft and privilege escalation and abuse.
  • Admin Users without MFA
    Look for user entities that do not have MFA applied to them. This can be done by filtering the cloud entities by the recommendation “secure admin or shadow admin with MFA”. Accounts without MFA are vulnerable to credential theft and privilege escalation and abuse.
  • Admin Users without PAM Controls
    Look for user entities that do not have PAM applied to them. This can be done by filtering the cloud entities by the recommendation “secure admin or shadow admin with PAM” or (if integrated with Privilege Cloud) “onboard an admin or shadow admin user from pending accounts”. Admin account users without PAM are vulnerable to credential theft and privilege escalation and abuse.
  • Admin Entities with Keys
    Look for entities that also have key-based access enabled (access/api keys). Key-based access is often used for programatic interfaces like APIs or CLIs but can also serve as a backdoor for malicious actors and can bypass MFA. Admin users and applications without PAM controls applied to their keys are vulnerable to credential theft and privilege escalation and abuse.
Building Your Strategy
Now that you’ve performed your analysis and have some useful insights, what do you do next? If you haven’t already built one going into this exercise, it’s time to create a strategy for remediating these security issues.

What can you do about these accounts? Generally speaking, there’s only three actions we can take.
  1. Delete the entity: If the entity is old, inactive, and no longer has a valid business reason to exist, delete it. It’s just unnecessarily contributing to your risk footprint.
  2. Remove permissions: If the entity should exist but doesn’t need the administrative privileges or other excessive permissions whatsoever, just remediate them. Entities are often over-privileged out of expediency or laziness.
  3. Secure the access: Depending on the type of entity (freestanding account vs federated user vs. application key) we’ll want to implement the appropriate cloud security controls, like credential management, session isolation, MFA, just-in-time access, session protection and/or audit logging. We cover how to apply the appropriate types of cloud security controls for each access scenario in other cloud security articles on the Success Blog.
Next Steps in Your Journey
Remember, getting to your destination requires knowing where you are. Discovering and analyzing your privileged cloud entities is a key step in that journey. You can’t protect what you don’t know exists. Hopefully this article has empowered you with the technical and process insight and confidence to go forth and discover your privileged cloud entities. Be sure to explore our Introduction to Idira Cloud Entitlements Manager training course too as a visual companion to this article.

Discovery is only one part of the process, if you’re in need of developing a strong plan of action, we recommend checking out our article on Creating Your Identity Security Roadmap with the Idira Blueprint. Of course, if you’re in need of hands-on assistance with an expert to discover your privileged entities, Idira can assist via a predefined outcome in our success plans outcome catalog. If you’re looking for more general information and resources on assessment tools, we recommend checkout out these assessment tool resources.
  • 3274 Views
  • 0 comments
  • 0 Likes
Top Liked Authors