> ## Documentation Index
> Fetch the complete documentation index at: https://help-center-starter-replace-template-content.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Build retries and recovery

> How Mintlify handles failed builds, retries, and recovery paths.

When a Mintlify build fails, the previous successful deployment continues to serve traffic. Your live site stays up even when a build fails.

## Automatic retries

Mintlify does not automatically retry failed builds — a failed build means there's a problem that needs to be fixed (usually a syntax error or misconfiguration). Fix the issue and push a new commit to trigger a new build.

## Manual retries

If a build failed due to a transient issue (network timeout, external resource unavailable), you can retry from the dashboard:

1. Go to **\[Project] → Deployments**.
2. Find the failed deployment.
3. Click **Retry**.

Only retry if you believe the failure was transient — retrying a build with the same code will fail again if the root cause isn't fixed.

## Recovering from a bad deployment

If a deployment succeeded but introduced a problem (broken layout, incorrect content, broken links), you have two recovery paths:

**Option 1: Revert in Git**
Revert the problematic commit in your repository. Mintlify deploys the reverted version automatically.

```bash theme={null}
git revert HEAD
git push
```

**Option 2: Redeploy a previous version**
In **\[Project] → Deployments**, find the last known-good deployment and click **Redeploy**.

## Monitoring deployments

Set up GitHub Actions notifications (see [Slack notifications](/getting-started/integrations/connect-slack)) to alert your team when a deployment fails. Catching failures early prevents readers from hitting broken preview links or outdated content.