# Redirect a whole domain to another from your _redirects file

> A _redirects rule whose source is a full URL, such as https://old.example.com/* to https://new.example.com/:splat, now redirects that whole domain to the new one, as long as the old domain is claimed and bound to your site.

Date: November 3, 2026 · Type: Added · Products: CDN · Version: v0.0.132

Source: https://changelog.evolvingedge.ai/2026/11/domain-level-redirects/

When a site moves to a new domain, links to the old one should land on the new one. Until now, a [`_redirects`](https://changelog.evolvingedge.ai/2026-09-26-redirects-from-your-redirects-file/) rule whose source was a whole URL was skipped. Now it works, in the same format Netlify uses:

```
https://old.example.com/*   https://new.example.com/:splat   301!
https://www.example.com/*   https://example.com/:splat       301!
```

## Before a rule works

- **The old domain must be [claimed](https://changelog.evolvingedge.ai/2026-10-01-prove-you-own-your-domain/) and bound to your site.** Your site's rules only see requests for domains bound to it, so a rule can never redirect a domain you haven't proved is yours.
- **The build lists the domains your rules need.** It can't see which domains are bound, so it names them instead: `_redirects: domain-level rules for old.example.com; each fires only once that domain is claimed and bound to this site`.

## How rules are matched

- **Domain rules come before path rules**, wherever they are in the file. It differs for those of you migrating from Netlify which uses file order: a catch-all such as `/* /index.html 200` above a domain rule redirects here, but rewrites on Netlify.
- **The scheme counts.** An `http://` rule matches plain HTTP only. Visitors are moved to HTTPS first, so an `http://` rule on its own never fires, and the build warns about it. Write `https://` rules.
- **Case, port and a trailing dot in the visitor's address are ignored.**
- **Only 301 and 302.** A domain rule can't rewrite (200) or proxy. The build refuses a domain rule with 200 or 404.
- **Rules that would loop are skipped**, such as a rule that sends a domain back to itself. Domain rules that send visitors round in a ring, such as www to the bare domain and back, stop the build.

## Limits

- **Level 0 sites only**, like everything in `_redirects`.
- **Later deploys don't move the old domain.** It stays bound to the deploy it was bound to. Its redirects keep working while that deploy's rule is right, but changing the rule means binding the domain again.
- **A rule for a domain that isn't bound isn't refused.** The domain may be bound after the deploy, so the build names it rather than stopping.

## What you need to do

Nothing, unless you're moving a domain. To redirect one, add a rule like the ones above to `_redirects`, claim the old domain and bind it to your site, then deploy.