Internal Redirect Loop (Crawler Report)

What’s actually happening

A site crawler — Screaming Frog, Sitebulb, Ahrefs — walks an internal URL through its redirects and lands back on a URL it already visited. It marks the URL as a loop and stops; the destination is never crawled or indexed. The browser version of this is ERR_TOO_MANY_REDIRECTS, but a crawler catches it even when one path happens to resolve, because it tracks every hop. Almost always two rules disagreeing: a server rewrite (.htaccess, Nginx) pushing one direction while the CMS canonical or another rule pushes back.

Common causes

• Trailing-slash rules fight: the server strips the slash, then the CMS or a second rule re-adds it, so /page → /page/ → /page forever.
• An .htaccess redirect and a WordPress/CMS canonical redirect target opposite URLs for the same page.
• www→non-www (or http→https) handled by two rules whose order makes them undo each other.
• A redirect map entry points old→new while another entry (or a wildcard) points new→old.
• Case normalization: one rule lowercases the path, another rule or a stored link uppercases part of it.

How to fix it

1. Print the full hop chaincurl -sIL https://example.com/products/widget shows every Location: header in order. Read down the list until you see a URL repeat — the two lines bracketing the repeat are the rules fighting. That's your loop, named exactly.
2. Find the two rules and keep oneLoops are a pair, not a single bad rule. Check .htaccess / Nginx server block AND the CMS (WordPress Settings → Permalinks/Site URL, redirect plugins). Decide the one canonical form — say, https + non-www + trailing slash — and delete or rewrite whichever rule contradicts it.
3. Consolidate slash and host handling into one ruleDo trailing-slash, www, and protocol in a single combined redirect to the final form, not three sequential ones that can ping-pong. One RewriteRule (or one Nginx return 301) that emits the canonical URL in one shot can't loop against itself.
4. Align the CMS canonical with the serverMake the CMS Site/Home URL and any redirect-plugin target match the exact form the server enforces. If the server forces a trailing slash, the CMS must not redirect it away.
5. Recrawl the single URL to confirmRe-run the crawler on just that URL (Screaming Frog: list mode). It should resolve in one or two hops to a 200 with no loop flag before you recrawl the whole site.