Category: Azure

Automate and Deploy Microsoft Defender Advanced Threat Protection (MDATP) via PowerShell

A few days ago, I needed to on-board Azure Windows Server VMs with Microsoft Windows Defender Advanced Threat Protection, or in short, MDATP. Sometimes Azure Security Center (ASC) has issues with on-boarding VMs and deploying the MDATP agent. As a result, I wrote the following PowerShell script that will download the MDATP.cmd file from my Azure Blob container and install it locally to the VM. This script allows you to automate it for many VMs to the scope of a Resource Group.

Now there are a few assumptions here…

  1. Download the MDATP.cmd file from the Defender Security Center portal
  2. Remove the requirement for user consent for the MDATP execution
  3. Upload the modified MDATP file to an Azure Blob container
  4. Generated a SAS URI for the MDATP file

There are many examples on the Internet on how to go step #4. Maybe in time I will do another post.

To remove the requirement of the MDATP agent to execute based on user interaction/consent can be done by removing, or commenting out the following lines of code. Launch the MDATP.cmd file within Notepad, and add a “:” before each line of code from lines 9 through 19, except line 14. Should look something like this.

Now, update and run the following PowerShell code. You can validate the VM is calling back to the Defender Security Center portal or by running the MDATPClientAnalyzer on the VM.

#update resource as needed
$resourcegroup = "YOUR_RESOURCE_GROUP"
#get only Windows Server VMs
$vms = Get-AzVM -ResourceGroupName $resourcegroup | Where-Object {$_.StorageProfile.OSDisk.OSType -eq "Windows"} | Select-Object Name
foreach ($vm in $vms)
{
    #friendly start message to indicate which server has started
    Write-Host "Server $vm has started..."
    #create folder, do not display error if folder already exists
    New-Item -Path "C:\" -Name "MDATP" -ItemType "directory" -ErrorAction SilentlyContinue
    #download MDATP.cmd file from Storage Account with SAS URI. Execute the cmd file. Passing "Y" to continue with installation.
    Invoke-WebRequest -Uri "YOUR_URI_SAS" -OutFile WindowsDefenderATPLocalOnboardingScript.cmd; Start-Process -FilePath "C:\MDATP\WindowsDefenderATPLocalOnboardingScript.cmd" -Verb RunAs
    #sleep for 5 seconds
    Start-Sleep -Seconds 5
    #restart-server
    Restart-Computer -ComputerName $vm
    #friendly finished message to indicate which server has completed and will now reboot
    Write-Host "Server $vm has completed, reboot initiated..."
}

Azure Conditional Access – Report-Only Mode

What is Conditional Access Report-Only mode?

Not too long ago, Azure Conditional Access introduced a new feature that allows Azure Active Directory administrators to test conditional access policies and its impact without actually enforcing the policy — ie. Report-Only mode.

For starters, Azure Conditional Access allows Azure Active Directory (Azure AD) administrators to enforce a specific set of conditions to be satisfied before a user (or group) can access specific resources within Azure. For example, a policy could be something simple as, ‘Enforce all users to go through MFA in order to gain access to the Azure portal“. Or, “Only users connected to the On-Premises network can gain access to the Azure portal.

Conditional Access is one of the many layers of implementing a Zero-trust network/environment. More on that in another post…

Implementing Conditional Access policies introduces a lot of challenges to end-users as sometimes it is difficult to determine the level of impact to the end-user(s). Report-Only mode allows for a Conditional Access administrators and the policy to determine the level of impact to users before actually enforcing the policy.

In the example below, I have created a Conditional Access policy with the following conditions and controls before a user can gain access to an Azure application. In this example, that application is Azure Portal.

Overview:

All Guest Users, must go through MFA in order to be granted access to Azure Management Portal. — pretty simple. Let’s see how this is setup, and the effects of Report-Only mode.

To begin, navigate to the Azure Active Directory service within the Azure Portal. Some base requirements, you need to have an Azure AD P1 or P2 license, and you the administrator must have Conditional Access Administrator (Azure AD role) as a minimum.

Create a policy, and give it some name followed by providing various requirements/conditions/controls.

Next we need to specify the user/users/group that this policy will be applicable to (or not, see the Exclude function).

 

Next, we need to specify the application this policy will be applicable to. Here I have selected the Azure Management Portal (Microsoft Azure Management) as the Cloud app.

 

Next, we need to either block or grant access to the users and the application once they pass the controls. In this case, the user must go through MFA in order to gain access to the Cloud app.

 

Finally we will save our configuration and leave the policy as Report-Only.

 

Now we can navigate to the Sign-In logs, and audit and validate our policy. Again, since this is a Report-Only policy, we can see the level of impact it would have caused to our end-users.

For more on Azure AD Conditional Access, please feel free to visit the following Microsoft URL, https://docs.microsoft.com/en-us/azure/active-directory/conditional-access/.

Get Azure Global Administrators

Recently a customer asked me to retrieve all users that have Global Administrator access to their Azure environment. The PowerShell code below will allow you to query the Azure environment against Azure Active Directory (AAD). Nothing new here or unique here, but this quick two-liner should do the trick. 😉

$role = Get-AzureADDirectoryRole | Where-Object {$_.displayName -eq 'Company Administrator'}
Get-AzureADDirectoryRoleMember -ObjectId $role.ObjectId | Sort-Object DisplayName | Select-Object DisplayName, UserPrincipalName, ObjectId 

Hope this was helpful!

Azure Bastion – What is it?

A little late to the party but nevertheless, I wanted to quickly show what is and how to use Azure Bastion. Azure Bastion is still in “public preview” but the solution is mature enough to start implementing now. Azure Bastion reduces the risk significantly in comparison to your traditional jumpbox approach, as it forces users to authenticate over SSL/443.

So what is Azure Bastion? Azure Bastion is a fully managed service by Azure/Microsoft that allows you to RDP and/or SSH into any Azure VM. Azure Bastion allows you to connect to your Azure VMs over HTML5-based browsers and using SSL.

Key Benefits:

  • Protection against 0-Day exploits:
    • Because Bastion sits at the perimeter of your VNet, you do not need to worry about hardening each of your VMs (although you should harden everything!!) The Azure platform will protect you keeping Azure Bastion hardened and is always up-to-date.
  • No Public IP(s) required for your Azure VMs:
    • By using Bastion, you can remove PIPs from your Azure VMs and can force your users to go through the Bastion host to connect to your VMs in your Azure environment.
  • Remote Sessions over SSL:
    • Since Azure Bastion uses HTML5 modern browsers, users can RDP/SSH over SSL (443) enabling you traverse corporate firewalls securely.
  • Simplified Management of NSGs:
    • Since Bastion is fully managed PaaS service by Azure, you no longer need to apply Network Security Groups (NSGs) on your Bastion subnet.  Since Bastion connects to your VMs over a Private IP, you can configure your NSGs to allow RDP/SSH from Azure Bastion only.

Architecture:

Source: Microsoft, https://docs.microsoft.com/en-us/azure/bastion/bastion-overview

The architecture diagram above shows use the workflow how Azure Bastion works.

  • The Bastion host is deployed within a VNet and with its own dedicated subnet
  • Users can connect with any modern HTML5 browser
  • No Public IPs on Azure VMs

Base Requirements:

Before getting setup with Azure Bastion there are some key things to know for example.

  • You must have a Virtual Network
  • The VNet must have a subnet dedicated for Bastion and its name must be “AzureBastionSubnet”
  • It is always recommended to have the subnet with a /27 CIDR. It is easy to grow your subnet as needed, much more difficult to shrink. Always start small and grow as needed.
  • No User Defined Routing (UDR) or Network Security Groups (NSG) can be applied to the subnet.

 

Next step, how to deploy and configure Azure Bastion. If you want to get started with Azure Bastion, you can enroll with the Public Preview here, https://aka.ms/BastionHost.

Azure Policy – Audit for Network UDR Changes

Azure Policy has been available for some time now, but for folks getting start with Cloud Governance, Azure Policy is a service in Azure that allows you to manage, assign, and create custom policies. These policies can be used to enforce a global set of rules or specific set of controls for a specific environments, ie. less control and governance in a “development” environment. This allows resources to stay compliant with you enterprise standards. Azure policies can enforce different rules, from Denying specific services, for example, ensuring only resources are built within a specific region, ie. resources can only be built within the Canadian regions. Conversely, rather than enforcing, policies can also be configured to Audit, where resources will be marked if they are not compliant, for example, a Storage Account is not configured with secure transfer.

Before diving into the policy itself, I want to quick go over the types of conditions that are available, and that can be used to enforce different compliance rules. The following table shows how different policy effects work with the condition evaluation for the resulting compliance state. Although you don’t see the evaluation logic in the Azure portal, the compliance state results are shown. The compliance state result is either compliant or non-compliant.

Resource StateEffectPolicy EvaluationCompliance State
ExistsDeny, Audit, Append, DeployIfNotExist, AuditIfNotExist*TrueNon-Compliant
ExistsDeny, Audit, Append, DeployIfNotExist, AuditIfNotExist*FalseCompliant
NewAudit, AuditIfNotExist*TrueNon-Compliant
NewAudit, AuditIfNotExist*FalseCompliant
  • *The Append, DeployIfNotExist, and AuditIfNotExist effects require the IF statement to be TRUE. The effects also require the existence condition to be FALSE to be non-compliant. When TRUE, the IF condition triggers evaluation of the existence condition for the related resources.

Source: https://docs.microsoft.com/en-us/azure/governance/policy/assign-policy-portal

In this example today, I want to show a real world example, where a customer recently asked to monitor any changes being made to their UDRs (User Defined Routes/Routing).

The following will continuously monitor all UDRs in the environment. If any changes are made to a single UDR table, it will be audited and its changes will be tracked. Once the policy is enabled, you can see it in action by creating/modifying a UDR.

“policyRule”: {
   “if”: {
      “anyOf”: [
       {
            “source”: “action”,
            “like”: “Microsoft.Network/routeTables/*”
       }
    ]
    },
   “then”: {
   “effect”: “audit”
    }
}

See below for the compliance once a change has been made to a UDR. Once you drill down to the event, the user, the activity log, you can then see the exact changes that were made to the UDR.

I hope this was helpful! 

Forcefully Revoke Azure AD User Session Access – Immediately

Sometimes it is critical to revoke a user’s Azure AD session for whatever reason it may be. You can always delete the user from Azure AD, however if the user is connected via PowerShell, the user’s token may not expire for a few more minutes, or maybe hours, depending on the token TTLs settings… So what can you do? You can forcefully revoke a user’s token session by using the following PowerShell cmdlet, “Revoke-AzureADUserAllRefreshToken“. Due to Microsoft’s ever changing Azure modules, I have tested this solution within the Azure Cloud Shell, and not on a local machine with PowerShell ISE with the AZ or RM modules.

First we need to identify which user will have its access revoked. Based off of the Revoke cmdlet, you will need to specify the “ObjectID” parameter, and the user’s ObjectID can be found within the Azure AD blade as seen below:

For additional information you can view the user’s access by executing the following cmdlet, “Get-AzRoleAssignment -ObjectId <>

Once we have identified the user and its ObjectID, we first need to connect to Azure AD, this is done by running the following cmdlet, “Connect-AzureAD -TenantId <>“. With my experience you need to specify the TenantID. Once you have connected, and verified your device, you can now run the Revoke cmdlet, as seen below, the following cmdlet needs to be executed, “Revoke-AzureADUserAllRefreshToken -ObjectId <>“. The Revoke cmdlet will not provide any details if the operation was successful, however it will throw an error if something did not go right — yes, very helpful, right? 🙂

By running this Revoke cmdlet, the user has now lost all access to its Azure AD account and any active sessions, either via the Azure Portal UI, or PowerShell will be immediately revoked. 🙂

What’s an Azure Service Principal and Managed Service Identity?

One of the general recommendations I always suggest to customers and their environments it leverage Azure Managed Service Identities (or MSI) over the traditional Service Principal (SP). Of course, the question then becomes, well what is the difference? When should I use a Service Principal and when should I use a Managed Service Identity?

In short, the difference is pretty clear. However, let’s make sure we understand what a Service Principal is, and what are they intended for…

What is a Service Principal (SP)?

In Azure, and many cloud environments, Service Principals carry the most weight with regards to access to the environment. Service Principals are an identity created for the use of applications, hosted services and automated tools to access Azure resources. This access is and can be restricted by assigning roles to the service principal(s).

What is a Managed Service Identity (MSI)?

With Managed Identities, there are two types of identities, system-assigned managed identity and user-assigned managed identity.

  1. System-assigned Managed Identity: is created and enabled directly on an Azure service. When enabled, Azure will automatically create the identity for the instance within Azure AD and will ensure it is trusted by the subscription where the instance resides. After the Azure resource/instance is created, the system-assigned managed identity credentials are provisioned onto the instance. The lifecycle is also directly attached to the Azure instance/resource and cannot be shared with other resources. In other words, the managed identity is created within Azure AD when the resource is created, and similarly the managed identity is deleted from Azure AD when the associated resource is deleted.
  1. User-assigned Managed Identity: is created as a standalone Azure resource, where the managed identity is created by the Azure AD administrator. The managed identity is trusted within the subscription and can also be assigned and shared with multiple Azure resources. The lifecycle is not dependent on the Azure resource, in other words, when the Azure resource is deleted, the managed identity is not deleted until the Azure AD administrator manually deletes the identity.

With MSI’s Azure automatically rotates/rolls the credentials every 46 days, Microsoft provides a workflow diagram on how MSIs work with Azure VM’s and other various Azure resources. See the diagram below to understand the credential rotation workflow.

Source: https://docs.microsoft.com/en-us/azure/active-directory/managed-identities-azure-resources/overview

What is the difference?

In short, when considering to use an MSI (Managed Service Identity) or a SP (Service Principal), also consider using a MSI for the reasons below.

MSI’s, managed the creation and automatically roll over the service principal for you. This is done by Azure in the background and requires no human/customer intervention. These credentials are rotated/rolled over every 46 days, this is a default behaviour/policy.

Use an MSI when and where available. Azure continues to grow their list of MSI’s and which resources can work with MSI’s, you can find the list HERE.

 

 

For a complete overview on MSI’s please visit Microsoft’s documentation HERE.

Deploy an Azure Cloud Witness for your Failover Cluster Quorum for Windows Server 2016 & 2019 with PowerShell

For the longest time, when deploying a cluster with Windows Server, you only had the two options,

  1. Using a dedicated disk for the quorum, or
  2. Configuring an SMB file-share as the quorum witness

With Server 2016 and 2019, there is now a third option, Cloud Witness. The Cloud Witness leverages Azure Blob storage to provide that additional cluster/quorum vote.

Before showing you how this is done, one should understand the purpose of a witness/quorum is with respect to a failover cluster.

When one or more members of a cluster stops reporting to the other cluster members, there is a vote. The vote ensures that there is no split-vote, and ensures the cluster has a true owner. For example, in a two node cluster, if each node believe it is the owner, then this will cause a “split-brain”. In short, neither node will ever agree it is the owner (or not). This is where a quorum is required to determine who is the owner by providing the third vote, ie. majority. This ensures the cluster has a true owner by having the majority of votes. Each member gets a vote, plus the quorum.

Why this matters, in the even there is no quorum, a node from the cluster can be evicted and as a result will suspend all application services to prevent data corruption by more than one system writing data without the cluster services coordinating data writes and access. Depending on policies, VMs running on the ejected cluster member will either suspend operations or be migrated to other nodes before being ejected.

Below is a step-by-step guide on how to configure the Azure Blob storage as the Cloud Witness.

Assumptions:

  • The Azure Blob storage account has already been created,
  • The cluster with at least 2 nodes already exists.

Launch the PowerShell console as Administrator, and execute the following cmdlet:

Set-ClusterQuorum -CloudWitness -AccountName "storage_account_name" -AccessKey "primary_access_key"

Now if we go back to the Failover Manager console we can see we have successfully configured cluster with a Cloud Witness.

In conclusion, deploying a Cloud Witness for a Failover Cluster is very simple, and in case of power outage in one datacenter, maintenance on a node, etc. then the entire cluster and its members (nodes) are all given an equal opportunity. Not only is it recommended and a requirement for 2-node clusters, but for any number of nodes, having a quorum is key ensuring high-availability.  As mentioned, there are the traditional options such as using a dedicated disk or a file-share (SMB) as the cluster witness. However with Azure Blob storage with its 16×9 uptime, we can always ensure the quorum witness is online and available.

Deploy an Azure Cloud Witness for your Failover Cluster Quorum for Windows Server 2016 & 2019

For the longest time, when deploying a cluster with Windows Server, you only had the two options,

  1. Using a dedicated disk for the quorum, or
  2. Configuring an SMB file-share as the quorum witness

With Server 2016 and 2019, there is now a third option, Cloud Witness. The Cloud Witness leverages Azure Blob storage to provide that additional cluster/quorum vote.

Before showing you how this is done, one should understand the purpose of a witness/quorum is with respect to a failover cluster.

When one or more members of a cluster stops reporting to the other cluster members, there is a vote. The vote ensures that there is no split-vote, and ensures the cluster has a true owner. For example, in a two node cluster, if each node believe it is the owner, then this will cause a “split-brain”. In short, neither node will ever agree it is the owner (or not). This is where a quorum is required to determine who is the owner by providing the third vote, ie. majority. This ensures the cluster has a true owner by having the majority of votes. Each member gets a vote, plus the quorum.

Why this matters, in the even there is no quorum, a node from the cluster can be evicted and as a result will suspend all application services to prevent data corruption by more than one system writing data without the cluster services coordinating data writes and access. Depending on policies, VMs running on the ejected cluster member will either suspend operations or be migrated to other nodes before being ejected.

Below is a step-by-step guide on how to configure the Azure Blob storage as the Cloud Witness.

Assumptions:

  • The Azure Blob storage account has already been created,
  • The cluster with at least 2 nodes already exists.

Launching the Failover Manager within Windows Server manager, connect to the cluster, and do the following. Right click the cluster object and select More Actions > Configure Cluster Quorum Settings…

Next select the Advanced Quorum configuration..

Ensure we have all the nodes selected, as seen below.

Next, select the Configure a Cloud Witness:

Now we need to get our Azure Blob storage account name, and its primary account key. This can be retrieved from the Azure portal.

Now validate the settings and complete the configuration.

Now if we go back to the Failover Manager console we can see we have successfully configured cluster with a Cloud Witness.

In conclusion, deploying a Cloud Witness for a Failover Cluster is very simple, and in case of power outage in one datacenter, maintenance on a node, etc. then the entire cluster and its members (nodes) are all given an equal opportunity. Not only is it recommended and a requirement for 2-node clusters, but for any number of nodes, having a quorum is key ensuring high-availability.  As mentioned, there are the traditional options such as using a dedicated disk or a file-share (SMB) as the cluster witness. However with Azure Blob storage with its 16×9 uptime, we can always ensure the quorum witness is online and available.