SEO

Redirect Loop

Also called Infinite redirect, ERR_TOO_MANY_REDIRECTS

Two or more redirect rules that point back at each other, so the requested page never finishes loading.

Quick facts: Redirect Loop

Category
SEO
Also called
Infinite redirect, ERR_TOO_MANY_REDIRECTS
Level
Intermediate
Affects
Site availability, crawling and indexing, paid traffic
Where to see it
Browser developer tools (Network tab), server and CDN logs, Search Console URL Inspection
In this article4
  1. How a redirect loop works
  2. Why redirect loops matter
  3. Common mistakes with redirect loops
  4. How to break a redirect loop

How a redirect loop works

One rule says address A should be served from address B. Another rule, written somewhere else, says B should be served from A. The browser follows the instruction, gets sent back, follows it again, and eventually gives up with a message about too many redirects. Nothing is served at all — not the page, not a useful error, not even a header a crawler could learn something from.

Loops are almost always a fight between two layers that each believe they own the site’s canonical address. A hosting panel forces HTTPS while a plugin still forces HTTP. A CDN adds a trailing slash while the application strips it. A WordPress site URL points at the bare domain while the server insists on www. Each rule is reasonable alone; together they never agree, so the request never settles.

Why redirect loops matter

A loop is worse than a missing page. A not-found response at least tells a crawler the URL is dead and lets it move on; a loop returns nothing it can use, so the page cannot be crawled, indexed, tested or measured. If the loop sits on a rule that applies site-wide, the whole site can disappear from search and from paid traffic at once.

What makes them dangerous is how well they hide. The owner is logged in, their browser has a cached redirect and a session cookie, and everything looks normal on their screen. Loops that depend on location are worse still: a rule that sends visitors to a country version can loop only for people whose location is read differently, so the site works in Kathmandu and fails for a customer in Sydney.

Common mistakes with redirect loops

The common causes are stacking. A security plugin, the server configuration, the CMS settings and a CDN all handling scheme and host at once. A wildcard rule whose destination also matches its own pattern. A redirect written for a URL that already resolves to the target. A staging environment copied to production with its old domain rules attached.

The common mistake is testing in the wrong conditions — checking while logged in, with a warm cache, or only on the homepage. Loops frequently affect a single template or a single language folder while the front page is perfectly fine.

How to break a redirect loop

Start by reading the raw responses rather than the browser message: your developer tools Network tab, or a command line request that prints headers, will show exactly which two addresses are bouncing. That pair names the two rules in conflict.

Then remove ownership from all but one layer. Decide the single correct combination of scheme, host and trailing slash, enforce it in one place, and switch the duplicate rules off one layer at a time, clearing the CDN cache between tests. Test in a private window on a device that has never visited the site, and repeat the check from another country if you use geographic rules. Once it resolves, keep a record of which layer owns redirects so the next release does not reintroduce the same argument, and re-run a crawl as part of routine technical SEO checks. A loop found by monitoring costs an hour; one found by a client costs a week of traffic.

Do and do not

Do

  • Let one layer alone decide scheme, host and slash
  • Test in a private window on an unvisited device
  • Read raw response headers, not the browser message

Do not

  • Trust a check made while logged in
  • Stack redirect rules across plugin, server and CDN
  • Assume a working homepage means the site is fine

Questions people ask about this

Why does the site work for me but show a redirect loop for others?

Your browser has already cached a redirect and probably holds a login cookie, so it never takes the path a new visitor takes. Some loops also depend on location, device or language headers, which differ from person to person. Always retest in a private window on a device that has never visited the site, ideally from another network.

Does a redirect loop remove pages from Google?

It can. A crawler that cannot reach a page gets no content and no usable status, and if that continues across several crawl attempts the page can be dropped from the index. Short-lived loops usually recover once the rules are fixed and the page is crawled again, but a loop left in place for weeks does real damage.

Which layer should own redirect rules?

Pick one and stick to it. On most small sites the web server or the hosting panel is the right place, with the CMS and any plugins left out of scheme and host decisions entirely. What matters is not which layer you choose but that only one layer decides, so two rules can never disagree about the correct address.

Related terms

Found this useful?

Share it, or ask an AI to summarise it

Back to the glossary

Knowing the term is the easy part

Applying it to your own site and budget is the work. Book a call and I will tell you what actually applies to you.