Every RMM tells you what broke.
ET Ducky tells you why.

Lower time to resolution on the tickets that other monitoring tools can't explain. ET Ducky captures the kernel event stream on every Windows and Linux endpoint, returning a named root cause with the evidence behind it. Includes the full RMM toolkit so your team can act on the answer in the same dashboard.

Book a Demo

Free tier available • No credit card required

Root-Cause Analysis: SERVER-12Interactive. Open any section
Technician · 21:08Disk alert fired on SERVER-12. What happened?
Root cause: Process BackgroundDownload (PID 37040) entered a tight I/O retry loop at 21:06:13, issuing hundreds of sequential reads in rapid succession, each failing with FileIOFailed. Disk utilization rose from 1.24% to 281%. The process hit a read error and hammered the disk with retries rather than backing off.

The anomaly is self-limiting. Memory stabilised by 21:09:18, so the process either exhausted its retries or crashed. Brief, but severe while it lasted.
Confidence high. 493 DiskIO and 7 FileIO error events converge on this single process inside a 2-minute window.
  • [Correlation] Process ranking by event count: BackgroundDownload(PID:37040) = 500 events (DiskIO/FileIO dominant)
  • [21:06:13.412] Kernel:DiskIO / Read / PID 37040 / 493 events over 2m 14s
  • [21:06:13.418] Kernel:FileIO / FileIOFailed / PID 37040 / 7 events
  • [21:06:14.002] Snapshot / disk / DiskTimePct 1.24% → 281%
  • [21:09:18.771] Snapshot / memory stabilised; no further DiskIO from PID 37040
Raw events retained and replayable. Forward to compliance or attach to a postmortem.
✓ Captured before any fix runs

BITS service state, recorded as Running, startup type Automatic (Delayed).

Pending job list, 1 job captured by GUID before removal, so a purge is reversible rather than final.

To restore:

Start-Service BITS bitsadmin /create <captured-job-name>
The undo is written first. If the remediation makes things worse, you are one command from where you started.

1 · Stop the retry loop

Stop-Process -Id 37040 -Force

2 · Check the target file’s integrity

chkdsk C: /f # requires reboot

3 · Clear the stuck BITS job

bitsadmin /list /verbose bitsadmin /remove <jobid>
Each step is proposed, not executed. Review, edit, or skip. Nothing runs on the endpoint without your approval.

The same alert, two different answers.

Traditional monitoring agents poll WMI counters and forward Event Log entries. Neither captures the kernel event stream that shows what actually caused the problem.

What your RMM tells you

“CPU exceeded 90% on SERVER-12.”

Open a remote session and start guessing.

What ET Ducky tells you

“CPU exceeded 90% on SERVER-12 because the .NET garbage collector ran a full Gen 2 collection lasting 4.2 seconds, triggered by WorkerService.exe exceeding its 2 GB heap limit.”

Fix the root cause, not the symptom.

ETW on Windows and eBPF on Linux give the agent kernel-level visibility on the host. ET Ducky processes that stream on-agent and ships only the curated evidence the dashboard and the AI need. The agent is bounded at the cgroup or service level so it never crowds out the workloads on the host.

Alerts with RCA
Threshold alerts that arrive with the root cause already explained, not just the symptom.
Ticketing & PSA
Built-in ticket queue with time tracking, plus two-way sync to your existing PSA or helpdesk.
Fleet Management
Tag, group, and configure agents at scale. One dashboard for Windows and Linux.
Custom Automations
Rules that fire on alerts, new agents, tickets, or a schedule to run scripts, send notifications, or open tickets.
Distribution Servers
Host installers and scripts on a LAN file hub so agents fetch large payloads locally instead of from the cloud.
Self-Hosted (Local-First)
Run the complete ET Ducky stack on your own infrastructure, with unlimited agents and technicians for one flat price.
AI Included, or Bring Your Own Key
AI analysis is built into every subscription and works out of the box. Teams already paying for Anthropic, OpenAI or Azure can point ET Ducky at that existing capacity instead of buying more.
Linux Agent
Install via .deb, .rpm, .run or a one-line installer, managed by systemd with eBPF-backed capture.
Network Discovery
Find and deploy to LAN machines not yet running an agent.
Out-of-Band Management
Discover and operate Intel AMT / vPro endpoints alongside in-band managed agents.
Data Explorer
Visual query builder for fleet telemetry with saved views and chart export.
Mobile Notifications
Push alerts to Android and iPhone with on-call awareness and escalation.
Application Management
Maintain an approved-software allowlist and review unapproved apps detected across the fleet.
Agent Tags
Group agents by client, site, OS or rollout wave, and scope alert rules to those tags.
Data Retention
Per-tier retention windows, on-agent retention, and the on-demand event upload gate.
Workspaces & Custom Domains
Subdomain-per-workspace routing, custom domains, and ownership transfers.

For IT leaders and MSPs

Cut time-to-resolution on the tickets without an obvious cause.

Half the time on an incident is spent finding root cause. ET Ducky's diagnostic engine removes that work for the issues other monitoring tools can only describe by their symptoms.

Audit-ready evidence on every incident.

Every alert, every detection, every AI-determined root cause is backed by the exact kernel events that triggered it. Forward to compliance, attach to a postmortem, or replay through Smart Reports.

Optionally host your own server, with unlimited agents and technicians.

Cloud hosting is included at no extra charge, or run the complete stack on your own infrastructure for one flat monthly price. No per-agent fees, no agent-count tiers, no seat management. AI analysis is included either way, and teams with existing Anthropic, OpenAI or Azure capacity can supply their own key so inference stays inside infrastructure they already control.

Every feature is in the subscription.

No modules to unlock, no per-feature upsell, and nothing held behind a sales call. Remote access, patching, alerting, automations, behavioral security and AI analysis are all included. The only optional cost is extended data retention, and the default retention window comes as standard. See how the pricing compares to NinjaOne, Atera, ConnectWise, Datto, Kaseya and Syncro.

See How the Pricing Stacks Up

Side-by-side breakdown of ET Ducky, NinjaOne, ConnectWise RMM, Datto RMM, Atera, Kaseya VSA, and Syncro: what each platform includes in its base price, what costs extra, and an interactive calculator for your own fleet size.

Read the comparison

Frequently Asked Questions

What is ET Ducky?

ET Ducky is a cross-platform diagnostics and behavioral-security platform with first-class agents for Windows and Linux. The Windows agent uses Event Tracing for Windows (ETW); the Linux agent uses eBPF programs attached to scheduler and syscall tracepoints. Both correlate kernel events directly on the endpoint, filter out PII before data leaves the machine, and determine the root cause in simple terms so your team can act on it.

Remote desktop, alerting, scripts, automations, and fleet management are included so your team can act on the diagnosis without leaving the dashboard.

How much does ET Ducky cost?

ET Ducky offers a Free tier (bring your own API key for unlimited queries), Professional at $39/month (1,000 AI queries), Business at $99/month (5,000 queries), and Enterprise at $249/month (50,000 queries). All paid plans include 20 free managed agent seats per user. Additional agent seats cost $5/month each with volume discounts down to $2/agent at scale. Local-First self-hosted is $650/month flat with unlimited agents and technicians, and annual billing saves 15%. No credit card is required for the free tier.

How is ET Ducky different from Datadog or other monitoring tools?

Unlike general observability platforms that focus on APM, metrics, and log aggregation, ET Ducky works at the kernel boundary on every endpoint it runs on: ETW on Windows (the same telemetry source Microsoft engineers use internally) and eBPF on Linux. It provides deeper root cause analysis than RMM tools by correlating low-level kernel events and using AI to explain exactly what happened and why, with specific remediation steps. Behavioral-security rules run on the same event stream, so detection of suspicious activity does not require a separate XDR product.

Does ET Ducky send my data to the cloud?

Raw ETW events never leave the machine — even during AI-driven dynamic diagnostic sessions where the AI is actively picking which providers to enable on the host. The on-agent correlation engine filters out PII and proprietary information; the dynamic capture engine counts events into a per-(provider, event id) tally and purges the underlying buffers on tear-down. Only the counts-only summary and the AI's conclusion travel to the cloud — achieving 99.6% bandwidth reduction while keeping sensitive data local. Self-hosted deployment options are also available for organizations that need full on-premises control.

What operating systems does ET Ducky support?

The Windows agent supports Windows 10, Windows 11, and Windows Server 2016 and later, running as a Windows Service under the Local System account (required for ETW kernel access). The Linux agent supports modern distributions on the .deb (Debian, Ubuntu) and .rpm (RHEL, Fedora, Rocky) families, and ships a universal .run installer for everything else. Linux agents run as the unprivileged etducky user under a hardened systemd unit; eBPF capture is supported on kernel 5.4 and later, with a graceful no-op fallback when kernel BTF is missing. The desktop app runs in user mode and is currently Windows-only.

Can I try ET Ducky for free?

Yes. The free tier includes the full desktop app with unlimited local ETW monitoring and interactive troubleshooting. Bring your own API key from Anthropic, OpenAI, or Microsoft Copilot for unlimited AI queries at zero subscription cost. No credit card required.

What deployment options are available?

ET Ducky offers two deployment models: Cloud Hosted, a shared multi-tenant platform included with every subscription at no additional infrastructure charge, and Local-First self-hosted, which runs the full stack on your own infrastructure with unlimited agents and technicians for one flat monthly price. Customers who need single-tenant isolation choose Local-First rather than a dedicated cloud tier. All options include the full feature set including live sessions, remote desktop, multi-agent analysis, and alerting.

Team

Team

Manage your organization's subscriptions, AI provider and usage, agent seats and data retention, and your workspace — all in one place. Pick a section on the left.

Systems

Operational systems

Tickets, Alerts, and Automations are the day-to-day operational surfaces of ET Ducky. They all share the same plumbing — orgs, agents, integrations, and connectors — so you'll see overlapping concepts (a ticket's automation runs an integration-connected script against an agent, for example).

Pick a system on the left to work with it. Patch Management links to Distribution Servers for installer hosting and Script Repositories for vendor scripts, the same way Tickets links to Jira/ServiceNow and Automations links to your Script Repositories (including Agent Generated Scripts).

TicketsCustomer issues + linked Jira/ServiceNow records. AlertsRules that fire on agent telemetry + channel dispatch. AutomationsTrigger-condition-action workflows that act on alerts and tickets. SecurityBehavioral-detection rules, response actions, and EDR/XDR provider coverage per agent. Patch ManagementApplication and OS update management available for configuration across Windows and Linux. Scan, schedule, deploy, and verify updates from one place. Fleet BaselineWhat's normal for each app across your own fleet — the learned per-app behavioral baseline that powers deviation detection. Network DiscoveryMachines on a gateway's LAN and which are not yet running the agent, with deployment coverage and per-VLAN segments.

Alert Monitoring

Real-time system monitoring and notifications

Loading... Auto-refresh: 30s

Settings

Appearance

Theme

Switch between light and dark.

Account

Loading...

Workspace

Your team's tenant on ET Ducky. All organizations, agents, alerts, and data live within a workspace. The workspace URL is the subdomain on etducky.com.

Loading…

Verified Domains

Prove that your workspace controls a domain to enable HTTPS at dashboard.yourdomain.com, claim a matching workspace slug (e.g. acme for acme.com), and surface your verified status on signups from that domain.

Loading…

Integrations

How integrations work

Integrations connect ET Ducky to the third-party systems your team already uses — ticketing, script repositories, file servers, behavioral exceptions. Most integrations are admin-only and live at the organization level.

Pick a category on the left to configure it. New integration types appear here automatically as your fleet evolves.

Automations

Define rules that fire when events happen across your fleet. Trigger scripts, webhooks, and notifications automatically.

0
Active Rules
0
Runs (24h)
Success Rate
0
Org Scripts
Loading automation rules…

Tickets

View and triage ET Ducky tickets. Search by title or device, switch between the board and the table, and right-click (or long-press) a ticket for actions.

All tickets (ET Ducky)

Tickets submitted from agents or the dashboard for this organization.

Loading…

Agent Management

Monitor and manage your ET Ducky monitoring agents.

Asset Management

Managed Agents
$5/month each
Desktop Agents
Included in Pro
Active Now
Heartbeat < 10min
Monthly Agent Fees

All Agents

Name
Type
Status
CPU
Memory
Last Seen
Actions
Loading agents…

OOB Audit Log

Recent out-of-band command history across discovered devices.

No OOB audit records.

Loading organization details...

Team Management

Manage your organization members and invitations.

Organization

Total Members
Active users
Administrators
Manage members and settings
Pending Invites
Awaiting acceptance

Team Members

Email
Role
Joined
Actions
Loading members...
Queries your organization's data in real time
Try asking:
Analyzing your data...

Query Results

Generated SQL


        

Recent Queries

Reports

Active Agents
-
Events Today
-
Live Sessions
-
Correlations (24h)
-
Patterns (24h)
-

CPU & Memory

Disk I/O

Network Throughput

Disk Space

Event Volume

Top Providers

Event Severity

Fleet Overview

Agent Version CPU Memory Disk Queue Net Conns Last Seen
Loading...

Data Explorer

Loading templates...
Loading saved views...

Historical AI Analysis

Suggested questions:
Loading suggestions...

Ask anything about your fleet data

I can analyze your historical metrics, events, correlations, and sessions to find trends, anomalies, and actionable insights.

Highest CPU last week? Common errors today? Fleet health comparison
Uses 1 AI query per analysis • Ready
Loading history...

Knowledge Base