Querying Data Source Logs with Generative AI

Querying Data Source Logs with Generative AI

Querying the logs of your data sources using generative AI can give you deep insights into your users, data, and trends.

Write your data source log search in plain language or run one of the preset queries.

Up to 90 days of log data is searchable.

Barracuda XDR adds data sources on a regular basis. To see if your data source is available, with your account selected, navigate to Intelligence > Log Search. Click the Select Data Source drop down to see the available data sources.

Available data sources

We add data sources on a regular basis. The following are currently available:

Name

Code

Name

Code

Microsoft 365    

o365.audit

Microsoft GCC High          

gcc.o365.audit

AWS Cloudtrail

aws.cloudtrail

Google Workspace

google_workspace

Barracuda Email Protection Essentials

essentials

Barracuda Email Protection

barracudasentinel

Barracuda Email Protection Forensics

forensics

microsoft_windows

windows

microsoft_windows

wmic

Barracuda SecureEdge

barracuda.secure_edge

Barracuda CloudGen

cloudgenfw

Palo Alto 

panw.panos

WatchGuard      

watchguard

Cisco Meraki     

cisco.meraki

SonicWall

sonicwall.firewall

Sophos XG

sophos.xg

Barracuda web application firewall  

barracuda.waas

Cisco Adaptive Security Appliance (ASA)

cisco.asa

Fortinet FortiGate

fortinet.firewall

pfSense

pfsense

Stormshield

stormshield_sns

Ubiquiti UniFi

ubiquiti_unifi

Zscaler

zscaler

Plain language searching

You can write your data source log searches using natural language. Barracuda XDR uses generative AI to query the database.

For example, you could write either of the following:

  • Show me windows login events for user j.smith in the last 5 days

  • Show me windows failed login event counts by user, country and source ip between 9-1-25 and 9-10-25

You can write queries in other languages, however, only English is tested and supported.

You can write and edit queries in SQL. For more information, see https://documentation.campus.barracuda.com/wiki/spaces/SKOUT/pages/519045121.

Preset queries

We also provide some common queries that might be useful for you.

Best practices for searching

  • Narrow Your Date Range Limit your queries to a specific and recent date range for faster results. The system currently limits results to 5,000 records per search.

  • Feedback Use the feedback form at the bottom right to report issues or share suggestions.

Searching the database

Once you have written a natural language search or selected a query, click the Submit button.

The results are displayed at the bottom of the page.

To let us know what you think or make a suggestion, in the top right corner, click the Feedback Feedback.png button.

Exporting the results

The results are displayed at the bottom of the screen. You can export the results to a .CSV file to view later or share with others.

To analyze data source logs
  1. In the Barracuda XDR Dashboard, click Intelligence > Log Search.

  2. If required, in the top right corner, select one of your accounts.

  3. Select your Data Source.
    NOTE Only data sources that can be searched are shown.

  4. Do one of the following:

    • Write your query in natural language.

    • Select a preset query.

  5. Click the Submit button AISubmit.png.

The result of the query is displayed in the Results section.

To see the SQL of the AI query, click SQL SQL button.png.

For more information, see https://documentation.campus.barracuda.com/wiki/spaces/SKOUT/pages/519045121 .

To export the results of the query
  1. Follow the To analyze data source logs procedure, above.

  2. When the results are generated, at the top right of the Results section, click Export CSV ExportCSV.png.

The CSV file downloads to your computer’s download folder.

 

 


We value your feedback.
If you have questions, suggestions, or feedback on our documentation, contact the Campus Product Documentation team.
For general product inquiries or technical support, please contact the global Barracuda Support team.