V3 Windows Sensor – Modernized Implementation

V3 Windows Sensor – Modernized Implementation

Overview

The Windows sensor has been upgraded to a modern, streamlined design.
This update makes the sensor faster, more reliable, and easier to manage, while keeping the same functionality and behavior you are already familiar with.

Key Benefits

  • Simpler management – fewer scheduled tasks, easier configuration

  • Faster scans – improved script execution without external delays

  • Better troubleshooting – structured logs and unified state data

  • Improved reliability – reduced dependencies, more resilient design

Key Improvements at a Glance

Area

Previous Design

New Design

Area

Previous Design

New Design

Task Scheduling

Multiple interdependent tasks in Windows Task Scheduler

One single task (“GYTPOL Hourly”), all operations handled internally

Script Execution

Depended on external validation, sometimes causing delays or failures

Scripts validated locally for faster and more reliable execution

Logs

Plain text files, unstructured and harder to filter

Structured JSON logs in a local database, easy to query and export

State Data

Scattered across several files

Consolidated into a single database for a clear, consistent view

Dependencies

Required extra supporting components

Self-contained, simpler to deploy and maintain

Task Scheduling

Before:

  • Several scheduled tasks were created under the gytpol folder in Task Scheduler.

  • Each operation (scans, reporting, upgrades) had its own task, creating complexity.

att_4_for_42140025.png

Now:

  • A single scheduled task named GYTPOL Hourly handles all operations.

  • Simpler structure, easier management, fewer points of failure.

image-20250910-081001.png

Script Execution

Before

  • Scripts depended on external validation.

  • This sometimes caused delays or failures during scans in certain environments.

Now

  • Scripts are validated locally by the sensor before running.

  • This ensures faster and more reliable execution, while maintaining code integrity.

Logging

Before

  • Logs stored as plain text in the Logs folder.

  • Easy to open, but unstructured and difficult to filter.

Now

  • Logs are stored in JSON format in a lightweight local logs database.

  • Structured, searchable, exportable.

  • Faster troubleshooting and easier integration with external tools.

Examples

Run the below commands from C:\Program Files\WindowsPowerShell\Modules\gytpol using CMD launched as Administrator:

  • Stream last hour of logs:

    client.exe -tail logs
  • Output logs from the last hour:

    client.exe logs
  • Last 2 hours (7200 seconds):

    client.exe -since 7200 logs > logs.json
  • All logs:

    client.exe -since 0 logs

When downloading logs from the UI, the system generates a zip archive containing:

  • JSON logs

  • Alerts overview (CSV)

  • Version history (CSV)

  • Task assignments (CSV)

  • Scan history (CSV)

  • Configuration details (CSV)

Sensor State

Before

  • State information (scan times, tasks, licenses, settings) was spread across multiple files.

Now

  • All data is stored in a single local sensor database.

  • This provides a clear and consistent view of the Agent’s configuration and activity.

Check with:

client.exe status

Example output:

Version : 3.1.1.0 Type : win-x64 Hostname : DC Endpoint UUID : b26212d7-some-UUID-here-c99506b67b18 Agent UUID : 52174389-some-UUID-here-9dd3ca6f0b27 Installation ID : 332449917 Labels : [none] Custom proxy : [none] First scan start : 2025-08-28T12:40:26 Last scan start : 2025-08-31T11:17:12 Task request : [default] Last tasks request : 2025-08-31T13:47:01

Other Useful Commands

  • Test connectivity (respects proxy configuration):

    client.exe connect
  • Schedule re-scan on next run:

    client.exe rescan

Summary

The updated sensor design delivers the same functionality as before, but with a much more efficient and reliable foundation:

  • One task instead of many in Task Scheduler

  • Faster scans with local script validation

  • Structured logs for better diagnostics

  • Unified state data for clear visibility

  • Self-contained design with fewer dependencies

This results in simpler management, greater reliability, and faster troubleshooting for your team.