TL;DR: When a deployment rolls back, check the validator results first; the config never reached your targets, so the bug is in the data, not in the app. Always attach validators to AppConfig configuration profiles: a JSON Schema validator for structure, and a Lambda validator for cross-field business rules. Keep validators fast and deterministic, since a slow or flaky Lambda validator blocks every deployment.

```text
AWS AppConfig auto-rolls back deployments that fail JSON Schema or Lambda validators
```

## Fix
1. When a deployment rolls back, check the validator results first; the config never reached your targets, so the bug is in the data, not in the app.
   Expected: this specific failure stops.
2. You can attach optional validators to a configuration profile; during deployment AppConfig evaluates them and automatically rolls back the change if validation fails.
   Expected: this specific failure stops.

## Details
Always attach validators to AppConfig configuration profiles: a JSON Schema validator for structure, and a Lambda validator for cross-field business rules. Keep validators fast and deterministic, since a slow or flaky Lambda validator blocks every deployment. The JSON Schema validator checks structure before any target sees the data, while a Lambda validator can enforce semantic rules (for example rejecting a flag combination that would break checkout).

## When to use
You hit exactly this: AWS AppConfig auto-rolls back deployments that fail JSON Schema or Lambda validators in AWS.

## When not to use
A different error, or the same symptom in a different tool. This page only covers the failure above.

## Compatibility
AWS.

## Edge cases
- Context from the original thread: Official docs (AWS AppConfig User Guide): documents the validator gotcha that trips agents whose deployments roll back with no obvious cause.