All pages
Powered by GitBook
1 of 3

Loading...

Loading...

Loading...

Log Analytics Cross-Tenant Search

Learn how to configure Azure Lighthouse to enable cross-tenant searches in Microsoft Log Analytics.

In order to execute cross-tenant queries against a Microsoft Azure Log Analytics Workspace, the proper permissions first need to be configured. This can be done using Azure Lighthouse, a free service that assists customers in managing multiple Azure tenants. In this case, it is used to assign role-based access control (RBAC) permissions to grant service principals permissions across tenants.

What follows are the instructions to set up Azure Lighthouse to enable the Anvilogic Azure integration to query across Log Analytics Workspaces in different Azure tenants.

Terminology

  • Provider - The tenant that is providing the service (in which the Anvilogic ADX cluster was deployed).

  • Customer - The tenant that the provider needs access to. This contains the Log Analytics Workspaces that will be searched.

There is only one provider, but there can be many customers.

Other Considerations

At the moment, Microsoft does not support resource-level permissions. Their guidance is to place active DENY permissions for the Anvilogic service principal on any resources in the Customer Resource Group that you don't want it to be able to access.

Alternatively, you can move the Log Analytics Workspace to it's own resource group using the . This is a non-destructive change and would not impact the workspace (i.e. it can be done while in production).

Microsoft also recommends that customers have only one Log Analytics Workspace per region. If customers are using multiple, that is an anti-pattern from Microsoft's perspective. For more information, see .

Azure Costs Estimates

Unified Detect for Azure supports both Azure Log Analytics, Azure Data Explorer (ADX), and Microsoft Fabric.

Installing Anvilogic's UD for Azure creates a new Azure Data Explorer cluster in your environment that is used to manage objects to run the Unified Detect framework.

During the set up process, a VM is created that will manage the Data Explorer Cluster. The default size upon our automated installation of that VM is a Standard_E2ads_v5 (Medium 8vCPUs) in a production cluster with SLA. This can be changed at any time if the amount of detections you have running requires more compute resources.

Review your billing configurations for ADX pricing tiers that control cluster management to ensure proper scaling expectations and configuration for the Anvilogic service to not get terminated.

Azure Resource Mover
https://learn.microsoft.com/en-us/azure/azure-monitor/logs/workspace-design
See Estimated cluster sizes and Cluster size costs.

Estimated cluster sizes

The table below assumes each deployed job run averages 1 minute and every rule deployed has the specified job run frequency. In reality, you could have a mix of how long the jobs take to run and how often they run. The table below is a guideline to be used for estimating capacity, and is based on the Azure Data Explorers default concurrency limits, which is the number of cores multiplied by 10.

3 Concurrency job runs are reserved for adhoc jobs executed from the Azure TI Builder view when creating or editing a threat identifier. The remaining jobs are reserved for deployed rules.

Other KQL queries being run outside of Azure UD also contribute towards this search concurrency and can cause throttled jobs if the cluster is operating near full utilization.

Cluster size
Azure ADX concurrency limit
Job run frequency (in minutes)
Deployed rules limit

Standard_E2ads_v5

20

5

80

Standard_E2ads_v5

20

15

240

Standard_E2ads_v5

20

Cluster size costs

The table shows the estimated monthly cost for various cluster sizes.

The estimated monthly and annual costs do not include additional storage costs. To determine the additional storage costs, use Microsoft Azure pricing calculator in the Microsoft documentation.

Cluster size
Number of cores
Estimated monthly cost
Estimated annual cost

Standard_E2ads_v5

2

$512

$6,144

Standard_E4ads_v5

4

$1,024

$12,288

Standard_E8ads_v5

8

Anvilogic on Azure

Anvilogic implementation with Azure (Data Explorer, Log Analytics, and Fabric).

Architecture Diagram

Below is the generic architecture digram for how Anvilogic works on top of Azure.

This supports both Azure Log Analytics, Azure Data Explorer (ADX), and Fabric workspaces.

We support querying a Log Analytics Workspace in a different tenant than the Anvilogic Azure Data Explorer Cluster.

  • In order to execute cross-tenant queries against a Microsoft Azure Log Analytics Workspace, the proper permissions first need to be configured.

  • This can be done using , a free service that assists customers in managing multiple Azure tenants.

Questions around cost? Review .

Diagram:

PDF Download:

Frequently Asked Questions (FAQs)

What gets installed in my Azure environment?

The following infrastructure will be created in the resource group you create for Anvilogic. We use an to deploy the infrastructure.

  • User managed identity (gives permissions to access key vault, and ADX tables)

  • Azure Key Vault

Does Anvilogic's integration incur any Azure costs?

Yes, there are costs associated with running the Anvilogic Resource Group in Azure, specifically on the Azure Data Explorer hosting cluster. These costs depend on the compute required to execute detections within your environment.

The average costs of running Anvilogic's required infrastructure in Azure range from $6,000-$25,000 per year annually depending on how many detections you will be running.

What costs money?

During the set up process, a VM is created that will manage the Data Explorer Cluster. The default size upon our automated installation of that VM is a Standard_E8ads_v5 (Medium 8vCPUs).

Calculate Costs:

What permissions do I need to create/use to install Anvilogic’s Azure integration?

You will be creating the following:


  1. Create new App registration for Anvilogic

  2. Create a new secret in the new App that was created in Step 1

Can you query Data Explorer, Log Analytics, and Fabric?

Yes, Anvilogic supports searching and running detection against any data source inside of an Azure LA, ADX cluster, or Fabric workspace.

You will need to give the Anvilogic app service principal permissions to query any of the ADX, LA clusters, or Fabric workspaces you want Anvilogic searches and detections to use.

For Microsoft Fabric you need to create a workspace, leverage an event stream under real time intelligence, and the destination from the event stream MUST be a KQL Database.

The will then be able to query the KQL database.

Currently, we do not support querying a Log Analytics Workspace in a different tenant than the Anvilogic Azure Data Explorer Cluster.

How does the Anvilogic platform query our LA, ADX, or Fabric Clusters?

We connect into your ADX cluster and then use the Microsoft to initiate a query to any other LA, ADX cluster, or Fabric workspace that our app service principal have access to.

For Microsoft Fabric you need to create a workspace, leverage an event stream under real time intelligence, and the destination from the event stream MUST be a KQL Database.

The will then be able to query the KQL database.

Currently, we do not support querying a Log Analytics Workspace in a different tenant than the Anvilogic Azure Data Explorer Cluster.

What if Azure isn't my primary SIEM and I have a hybrid set up?

Since Anvilogic supports multiple SIEM/Data Lakes, you can configure all of the events if interest (EOIs) generated from detection queries to also write a copy back to your primary Alert Lake or EOI data store. That can be located in any of the other support platforms (ex. Splunk, Snowflake).

For example - if Splunk is your primary SIEM, then you can configure all of your Azure detection results to also send a copy of the event of interest (EOI) back to the Anvilogoic index in Splunk. The Anvilogic platform handles all of this EOI routing for you.

Can you help bring alert data into Azure for us?

Yes, Anvilogic can help retrieve alerts/signals from SaaS security tools (ex. Proofpoint, Wiz, Crowdstrike, etc.) and can ingest those into the Anvilogic table in ADX for correlation.

Can you help bring raw data into Azure for us?

No, Anvilogic does not support raw data ingestion into ADX, LA, or Fabric. Data must already be present in those environments.

Anvilogic only supports raw data ingestion for Azure Snowflake.

Do you provide parsers for un-normalized data?

Yes, we provide hundreds of out-of-the-box parsers that can be used to normalize your security data inside of ADX,LA, or Fabric.

What is the Anvilogic Alert Table in ADX?

Anvilogic Alert table in ADX will store the output from all detections that are running within the App container environment.

This is a fully normalized set of signals that we call “events of interest” that can be used to escalate activity to your SOAR or can be used as a hunting index to create Threat Scenario correlations.

Do you collect the alerts stored in the Anvilogic Alert Table in ADX?

Alerts are stored inside of your Azure table in ADX you specify during the setup.

The Anvilogic AI-Insights (ex. Hunting, Tuning, Health) package requires a copy of these events to be collected and stored by Anvilogic. If enabled, a copy of those events will be collected into Anvilogic.

Do you integrate with SOAR?

Yes, Anvilogic can integrate with most SOARs via REST API through either a push or a pull method.

30

480

Standard_E2ads_v5

20

60

960

Standard_E4ads_v5

40

5

180

Standard_E4ads_v5

40

15

540

Standard_E4ads_v5

40

30

1,080

Standard_E4ads_v5

40

60

2,160

Standard_E8ads_v5

80

5

380

Standard_E8ads_v5

80

15

1,140

Standard_E8ads_v5

80

30

2,280

Standard_E8ads_v5

80

60

4,560

Standard_E16ads_v5

160

5

780

Standard_E16ads_v5

160

15

2,340

Standard_E16ads_v5

160

30

4,680

Standard_E16ads_v5

160

60

9,360

Standard_D32d_v4

320

5

1,580

Standard_D32d_v4

320

15

4,740

Standard_D32d_v4

320

30

9,480

Standard_D32d_v4

320

60

18,960

$2,050

$24,600

Standard_E16ads_v5

16

$4,099

$49,188

Standard_D32d_v4

32

$7,781

$93,372

In this case, it is used to assign role-based access control (RBAC) permissions to grant service principals permissions across tenants.

Click here to learn more.

ADX Cluster, database, and tables Azure Container App/environment/jobs/instance

  • Log analytics workspace for the container app

  • Azure Container app registry & cache

  • Visit Azure Pricing Page -> Type in "azure data explorer" under products

  • In the Instance section, type "E8ads"

  • Refer to Azure Costs Estimates for more details.

    Create an Anvilogic resource group

  • Go through our integration set up on the Anvilogic Platform

  • Azure Lighthouse
    Azure Costs Estimates
    563KB
    Azure Reference Architecture.pdf
    PDF
    Open
    Azure ARM template
    cluster command
    cluster command
    cluster command
    Anvilogic's Integration with Azure ADX, LA, Fabric.