Introducing PSwatchdog: Keep Your PowerShell Processes Running

Production automation is great — right up until a process silently dies at 3 AM and nobody notices until morning. PSwatchdog solves that by monitoring your processes, services, and script jobs and restarting them automatically whenever they stop.

The Problem: Unattended Processes Die

Long-running PowerShell jobs are everywhere: background workers, scheduled data pipelines, local API proxies, file watchers. They all share one flaw — they eventually crash, and without something watching over them, they stay dead until a human notices.

The old-school workaround looks something like this:

while ($true) {
    Start-Process "C:\workers\ingest.exe"
    Wait-Process -Name "ingest"
    Write-Host "Process died — restarting..."
    Start-Sleep -Seconds 5
}

It works, but it ties up a shell window, offers no visibility, and falls apart the moment you need to manage more than one process. PSwatchdog gives you a proper solution.

Installation

Install-Module -Name PSwatchdog -Scope CurrentUser
Import-Module PSwatchdog

Watching Your First Process

Import-Module PSwatchdog

# Start a watchdog that restarts "worker.exe" whenever it stops
$wd = Start-Watchdog -ProcessName "worker" `
                     -Command "C:\workers\worker.exe" `
                     -Arguments "--config prod.json" `
                     -RestartDelay 10

Write-Host "Watchdog running with ID: $($wd.Id)"

PSwatchdog monitors the process in the background. If worker.exe exits for any reason — crash, OOM kill, unhandled exception — it waits 10 seconds and starts it again automatically.

Watching a Windows Service

For Windows services you don’t even need the command path. Just tell PSwatchdog the service name:

# Keep the "DataIngestSvc" service alive
Watch-Service -Name "DataIngestSvc" -RestartDelay 15

Every 15 seconds after the service enters a stopped state, PSwatchdog will attempt to restart it — far faster than waiting for an on-call engineer to wake up.

Visibility: See What’s Being Watched

Get-Watchdog

This lists all active watchdog monitors and their current status so you always know what’s being protected.

Limiting Restart Attempts

Sometimes a process is crashing for a good reason — a bad config, a missing dependency, a downstream service that’s fully down. Endlessly restarting it just masks the real problem. Use -MaxRestarts to cap the attempts:

Start-Watchdog -ProcessName "flaky-app" `
               -Command "C:\apps\flaky-app.exe" `
               -MaxRestarts 3

After three unsuccessful restarts PSwatchdog gives up, letting your alerting system surface the real issue rather than hiding it under a restart loop.

Stopping a Watchdog

When you no longer need a monitor, shut it down cleanly:

Stop-Watchdog -WatchdogId $wd.Id

Real-World Use Cases

CI / CD agent resilience — keep a self-hosted build agent running without relying on the host OS’s service recovery settings.

Local dev proxies — restart a local reverse proxy or mock API server the moment it crashes, so you don’t lose your development flow.

Data pipeline workers — ensure background jobs that ingest, transform, or export data keep running even if they hit transient errors.

Edge / IoT deployments — on machines where there is no ops team watching dashboards, PSwatchdog gives your scripts the same restart behaviour you’d expect from a proper service manager.

Conclusion

PSwatchdog is a small module with a focused job: make sure your PowerShell processes don’t stay dead. Whether you’re running background workers, Windows services, or long-lived scripts, it keeps things running without any manual intervention.

Install-Module -Name PSwatchdog -Scope CurrentUser

Explore the source and contribute on GitHub and read the full command reference on the PSwatchdog module page.