What is a React error boundary?

A React error boundary is a component that catches JavaScript errors in its child component tree, logs the error, and renders a fallback UI instead of crashing the entire page. WISEROWS wraps every widget and major section in its own error boundary for maximum isolation.

What happens when a component fails?

The failed component is replaced with a recovery card that shows a brief error description and offers three actions: Retry (re-renders the component), Reload (refreshes the section), and Report (sends error details to the team). The rest of the page continues working normally.

Can one broken widget crash the whole dashboard?

No. Each dashboard widget is wrapped in its own error boundary. If a chart widget throws an error, only that widget shows a recovery card. All other widgets, the navigation, the sidebar, and the header continue working. You can even interact with the rest of the page while the broken widget is in recovery state.

What is graceful degradation?

Graceful degradation means the system continues to function at a reduced capacity rather than failing completely. In WISEROWS, if a non-critical feature (like a dashboard widget or sidebar panel) fails, the core functionality remains fully operational. Missing configuration silently disables features rather than crashing.

How does the retry action work?

When you click Retry on a recovery card, the error boundary resets its error state and attempts to re-render the failed component. This often resolves transient issues like network race conditions or temporary data inconsistencies. If the component fails again, the recovery card reappears.

Are error boundaries nested?

Yes. WISEROWS uses a hierarchy of error boundaries — page-level, section-level, and widget-level. If a widget boundary catches an error, only the widget is affected. If an error somehow escapes the widget boundary (which is rare), the section boundary catches it. The page boundary is the last resort, ensuring the app never shows a blank white screen.

What about errors in event handlers?

React error boundaries do not catch errors in event handlers. WISEROWS handles this with try-catch blocks in all event handlers, combined with the global error handler. Errors in event handlers are caught, logged to Sentry, and handled with appropriate user feedback (like error toasts) without crashing any component.

Does error recovery work with real-time data?

Yes. When a component recovers (via retry), it reconnects to its Convex reactive query and receives the latest data. There is no stale state after recovery — the component always shows current data. Real-time subscriptions resume automatically.

What does the Report action do?

The Report action sends the full error details — stack trace, component tree, user context, and browser information — to Sentry for the engineering team to investigate. It also shows a confirmation message so you know the report was sent. This helps the team identify and fix recurring issues.

How does WISEROWS handle missing configuration?

If a feature requires configuration that is not set (like a missing Sentry DSN or AI API key), it silently disables rather than crashing. The rest of the application works normally. This is part of the graceful degradation philosophy — missing config never causes errors, it just means that specific feature is unavailable.

Error Recovery

One widget fails. The rest keeps working.

React error boundaries isolate every component. When a widget throws an error, it shows a recovery card — retry, reload, or report. The rest of your page stays fully operational.

Active Users

2,847

+12.4%

Revenue

Component Error

This widget encountered an unexpected error. Other widgets are unaffected.

Retry

Report

Recent Activity

Blog post published2m ago

New entity created5m ago

Workflow completed8m ago

Quick Settings

Dark mode

Notifications

Auto-save

Component-level isolation

Every widget, panel, and section is wrapped in its own React error boundary. Errors are caught at the smallest possible scope. A failing chart widget cannot take down your sidebar, navigation, or any other widget on the page.

Each dashboard widget has its own error boundary
Sidebar, header, and navigation have independent boundaries
Failed components render recovery cards, not blank space
Surrounding components continue receiving real-time updates
Error state is local — no global state corruption
Nested boundaries catch errors at the closest level

Error Boundary Hierarchy

Page BoundaryLast resort

Section Boundary

Widget OK

Caught!

Widget OK

Three recovery actions

Every recovery card offers three clear actions. Most issues resolve with a single retry. For persistent errors, the report action sends full diagnostics to the engineering team.

Retry — resets error state and re-renders the component
Reload — refreshes the section with fresh data from the server
Report — sends stack trace, component tree, and context to Sentry
Recovery reconnects to Convex reactive queries automatically
No stale state after recovery — always shows current data
Action buttons are keyboard-accessible

Retry

Re-render the component. Resolves transient issues like race conditions.

Reload

Fetch fresh data from the server and re-render from scratch.

Report

Send full error details — stack trace, context, browser info — to the team.

Graceful degradation

Not everything needs to work for the system to be useful. Missing Sentry config? Error tracking silently disables. No AI API key? Content generation shows a setup prompt. Missing integration credentials? The connector pauses. Nothing crashes.

Missing config silently disables features, never crashes
Optional features degrade independently
Core functionality works even when optional services are down
Fire-and-forget for non-critical operations (analytics, telemetry)
Error tracking itself is fail-safe — Sentry errors never break the app
Reduced functionality is always better than no functionality

Degradation Examples

SENTRY_DSNnot set

Error tracking disabledApp works normally

AI_API_KEYnot set

AI features show setup promptData management works

POSTHOG_KEYnot set

Analytics disabledAll features work

SANITY_TOKENnot set

CMS publishing pausedContent editing works

Boundary levels

Recovery actions

Full-page crashes

<1s

Recovery time

How it feels

Errors are contained, not contagious.

A chart widget hits an edge case? You see a small recovery card in that widget's spot — with a retry button that usually fixes it in one click. Meanwhile, your kanban board is still updating, your sidebar is still navigating, and your AI chat is still responding. One hiccup does not become a full-page crash.

Frequently asked questions

Your data is incomplete. Fix it.

Define your schema. Import your data. Let AI enrich every record. See quality scores improve in real time. Free to start, no credit card.