Preventing Duplicate Async Operations in .NET MAUI | Community Feed

Async makes concurrency easy to introduce even when it was never intended: a double-tapped Save sends two requests, a refresh overlaps a reconnect, two navigation commands stack the same page. Jorge Perales Diaz explains why an IsBusy flag isn’t always enough and builds reusable patterns for deciding what should happen when an operation is requested while it’s already running.

What you’ll learn

Why IsBusy can fail — separating UI state from concurrency control, and the check-then-set window across multiple entry points
A reusable async guard — a SemaphoreSlim-based AsyncExecutionGuard with WaitAsync(0) to reject duplicates, result wrappers, and full cancellation flow
Choosing an execution policy — reject, wait, cancel-previous, queue, share, or allow — mapped to Save, search, navigation, uploads, and initialization
Single-flight and keyed locks — sharing one in-flight task for token refresh and app init, plus per-resource ConcurrentDictionary gates
Search done right — latest-wins cancellation plus debouncing, and why client-side guards still need server-side idempotency keys

Read the full article for the guard implementation, strategy comparison tables, concurrency tests, and a production Save ViewModel.