> ## Documentation Index
> Fetch the complete documentation index at: https://docs.vertracloud.app/llms.txt
> Use this file to discover all available pages before exploring further.

# Web Analytics Errors: 5xx, 4xx and Failing Routes

> The Errors sub-tab shows server errors (5xx) over time, which routes fail most and why 4xx errors are optional and noisy.

The Errors sub-tab shows which requests failed, how often, and on which routes, so you can tell a real problem from background noise. It is available on every plan.

Same data through the API: [Application errors](https://docs.vertracloud.app/api-reference/endpoint/apps/analytics/errors.md).

## 5xx: your app has a problem

A status starting with **5** (`500`, `502`, `503`, `504`) means the server failed to answer. Visitors see these as broken pages, so they are always shown.

Common causes: an exception in your code, a database that does not answer, an app that is restarting or ran out of memory, or a request that took too long. Pair this tab with the [application logs](https://docs.vertracloud.app/api-reference/endpoint/apps/logs.md) to find the cause.

## 4xx: optional

A status starting with **4** (`400`, `401`, `403`, `404`, `429`) means the request itself was refused or pointed at something that does not exist. These are hidden by default; turn on **Include 4xx** to see them.

> [!WARNING]
>
> **4xx is noisy.** Bots and scanners constantly probe addresses like `/wp-login.php` or `/.env`, which produces a stream of `404` that says nothing about your site's health. A high 4xx count is normal on any public site. Look at it only when you are chasing something specific, such as a broken link on your own pages or a `401` your login should not return.

## What the tab shows

* **Totals by class:** how many 5xx (and 4xx, if included) in the period.
* **Errors over time:** a chart, so you can see whether the errors started with a deploy or at a given hour.
* **Routes with most errors:** for each route, the error **count** and the error **rate**. Use both: a route with 3 errors out of 3 requests (100%) is a different problem from one with 300 out of 100,000 (0.3%).
* **Status distribution:** how the errors split between `500`, `502`, `504`, `404` and so on. Each class points to a different cause.

## Investigating

1. Look for the moment the line rises and compare it with your [deploys](https://docs.vertracloud.app/guides/rollback-and-zero-downtime-deploys.md).
2. Click the worst route to filter everything by it.
3. On Pro and above, open [Requests](https://docs.vertracloud.app/web-analytics/requests.md) to see the exact failing requests and note a request ID.
4. Read the application logs for that time.

Next: [Performance](https://docs.vertracloud.app/web-analytics/performance.md).


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.