# Know when your site goes down, a deploy fails or sales stop

> Automate website uptime and health checks across hosting, deploys, traffic, and payments using thelastapi's built-in Mission Control.

**You monitor website uptime and business health by checking the entire visitor journey—from DNS and deploys to error rates, traffic drops, and payment webhooks—rather than relying on a simple ping.** In thelastapi, background health checks run automatically every hour for every active project, flagging downtime, failed builds, broken checkout deliveries, and silent traffic collapses in one central board called Mission Control. You can view these checks in your browser console, query them directly in your terminal using the CLI, or have your coding agent inspect them over MCP.

This guide explains what health checks small businesses actually need, how the automated checks work, and how to spot and fix problems before customers notice.

## What health checks need to cover for a small business

A basic HTTP ping tells you if a web server answers requests, but it cannot tell you if your site is actually working for a customer. A site can return a `200 OK` status on its homepage while its database connection pool is broken, its newest production build failed, or its payment webhooks are failing quietly in the background.

When you run an online business, four distinct things can break:

1. **Availability and SSL.** The site does not resolve, times out, serves an invalid HTTPS certificate, or serves a server error (`5xx`).
2. **Deployments and infrastructure.** A recent git push or CLI deploy failed during the build phase, or a required database or sign-in resource failed to provision.
3. **Visitor experience.** JavaScript errors spike in the browser, visitors experience broken buttons, or page traffic drops abruptly after an unannounced change.
4. **Checkout and payments.** Webhook events cannot reach your application server, leaving customer orders in limbo while visitors try to pay.

Instead of signing up for separate monitoring dashboards, status monitors, and analytics tools, you can track these failure points in one place. You can read more about tracking operations with an agent in [Ask your agent "how is my site doing?" and get a real answer](https://thelastapi.dev/blog/ask-agent-how-is-my-site-doing).

## How the nine automated checks work

Every active project in your thelastapi account is checked automatically about once an hour in the background, a few at a time. The checks only read data—nothing is deployed, charged, or sent. Requests to your live site go out as a regular visitor's would, using the user agent `VibizStack-Health/1.0`.

Each check returns one of four statuses: `ok`, `warn`, `fail`, or `unknown` (used when a project does not use that resource yet, like an app without payments). A business's overall status is always determined by its worst check.

| Check | Fails when (`fail`) | Warns when (`warn`) |
| --- | --- | --- |
| **Site up** | The visitor address answers `5xx`, times out after 10 s, does not resolve, or has an invalid HTTPS certificate | Homepage `/` answers `4xx`, or the site ends on plain `http://` |
| **Last deploy** | The latest production deploy failed (with the build's error reason) | A production deploy has been queued or building for over 30 minutes |
| **Resources** | A resource (database, sign-in, email) failed to set up, with its error | None |
| **Errors** | Spike: 10+ sessions with JavaScript errors in 24 h, at 3x the usual day | 3+ sessions with errors, 5+ rage clicks, or 10+ dead clicks in 24 h |
| **Traffic** | None | Pageviews in the last 24 h are over 60% below the 7-day daily average (once average is 20/day or more) |
| **Sitemap & pages** | A page in `/sitemap.xml` answers `5xx` (up to 20 pages checked) | No sitemap, pages in sitemap answer `404`, or homepage lacks a `<title>` or meta description |
| **Payments** | The project sells in production with no webhook URL, or no event reached the app in 24 h while some failed | Some webhook deliveries failed, or there is no development (test) webhook configured |
| **Limits** | A project cap (daily builds, AI budget, app email) is reached | A cap reaches 80% of its limit |
| **Domain** | None | A custom domain is not live yet (DNS, ownership, or HTTPS pending) |

Every failing or warning check tracks a `since` timestamp, so you can immediately see if your site has been failing for five minutes or three hours. Complete details on every step can be found in the [Mission Control and health checks documentation](https://thelastapi.dev/docs/mission-control).

## Checking site status from your terminal or AI agent

You do not need to keep a web browser open to monitor your projects. You can inspect project health directly from the command line using the `vibiz` CLI.

To list all businesses on your account, sorted worst first with their top problems:

```bash
vibiz health
```

If you only want to see businesses that require immediate attention:

```bash
vibiz health --failing
```

If you made a change and want to run the checks immediately across all businesses rather than waiting for the hourly background cycle, trigger an on-demand check:

```bash
vibiz health run
```

When building with AI agents such as Claude Code, Codex, or Cursor, your agent can call the MCP tool `vibiz_health` with actions `"list"`, `"project"`, or `"run"`. For custom tooling, query the HTTP API directly:

```bash
curl https://thelastapi.dev/v1/health \
  -H "Authorization: Bearer $VIBIZ_API_KEY"
```

The endpoint returns project metrics, error counts, and an array of top problems. For single-project inspection, call `GET /v1/projects/:id/health`.

## Monitoring projects in the Mission Control canvas

In your browser, navigate to console ▸ **Mission Control**. Mission Control presents an account-wide board where each business is represented as a row, laid out across eight consecutive steps that mirror the visitor journey:

1. **Domain & HTTPS:** Verifies your address, DNS resolution, and SSL certificate validity.
2. **Deploy:** Tracks your latest build status and infrastructure provisioning.
3. **Pages & SEO:** Validates `/sitemap.xml`, title tags, and page status codes.
4. **Traffic:** Shows 7-day visitor counts, 30-day trends, and checks for traffic drops.
5. **Experience:** Flags browser errors, dead clicks, and rage clicks captured by session replay.
6. **Checkout & payments:** Confirms payment processing and webhook delivery health.
7. **Sales & revenue:** Displays sales and revenue over the last 30 days and today.
8. **Limits & spend:** Monitors AI model spend against your project daily budgets.

Rows are automatically sorted into sections: **Pinned**, **Failing**, **Warnings**, **Not checked**, and **Healthy**. Any problem shows as a red or amber dot with the explanation right below it.

You can navigate large accounts using trackpad gestures or mouse zoom. Press **F** to search for a business by name, click **Check now** in the toolbar to run checks across all projects, or click on any business row to open the inspector sidebar. The inspector gives you the exact fix instructions and links directly to the page where you resolve the issue.

## What to do when a deploy fails or your site crashes

When the **Site up** or **Last deploy** check flags a failure, you have two fast options to restore service:

### 1. Roll back the deployment instantly

If a new production release breaks your site, roll back to the version before the live one:

```bash
vibiz rollback
```

A production rollback switches your live traffic instantly. Note that only the site's code rolls back; your database tables and rows stay untouched. You can browse all previous deployments first to see what changed across releases:

```bash
vibiz versions                 # view previous deployments, timestamps, and commit messages
```

For more on single-command deployments and managing your releases, see [Deploy a static site with one command](https://thelastapi.dev/blog/deploy-static-site-one-command) and [Edit a live site by deploying only what changed](https://thelastapi.dev/blog/edit-live-site-deploy-changed-files).

### 2. Inspect logs and code diffs

To see what broke between your previous release and the failing one, compare them directly:

```bash
vibiz diff
```

This displays the file changes and text diffs between the live version and the one before it. You can pull the live code down to inspect it locally with `vibiz pull` or examine individual files using `vibiz code cat index.html`.

For apps deployed on the platform, full hosting details, limits, and build commands are covered under [Hosting](https://thelastapi.dev/docs/hosting).

## Next step

If you already have projects running, run `vibiz health` in your terminal or open **Mission Control** in your console to verify your checks. To get started on the platform, create your account at [thelastapi.dev/start](https://thelastapi.dev/start).

## Frequently asked questions

### How often do website uptime and health checks run?

Background health checks run automatically about once an hour for every active project on your account. You can also trigger an immediate check at any time by running `vibiz health run` or clicking "Check now" in Mission Control.

### What triggers a critical "Site up" health check failure?

A site fails the "Site up" check if the visitor address answers with a 5xx server error, times out after 10 seconds, fails DNS resolution, or presents an invalid HTTPS certificate.

### Does a rollback undo database schema migrations or delete rows?

No. Running `vibiz rollback` only reverts your site code to the prior deployment manifest; database rows and schema changes remain exactly as they are.

### How do payment health checks detect broken checkout setups?

The payment check fails if your project is configured to sell in production without a webhook URL, or if no payment event has reached your application in 24 hours while failed deliveries have occurred.

---

Published 2026-11-15 · https://thelastapi.dev/blog/know-when-your-site-goes-down
