Broken Link Checker

What this tool does

The Broken Link Checker finds every link on a page: anchors, images (including srcset), scripts, stylesheets, icons, frames and media. It checks each one and reports:

• Broken links: 404 and 410, server errors, hosts that don’t exist, and redirect loops.
• Redirects: links that redirect, with the final address, so you can update the link.
• Missing anchors: #fragment links whose target id no longer exists on the page.
• Insecure resources: http:// images, scripts and styles on an https:// page, which browsers block or warn about.
• Blocked or unclear results: sites that refused the checker or timed out, so you can check them by hand.

It can check one page, or crawl a site (up to 100 pages) and check everything it finds. Export the results as CSV, with the page and line number of every link.

When to use it

• After a migration, redesign or URL change: find internal links still pointing at old addresses.
• Before a launch: catch links to staging, localhost or unfinished pages.
• As routine maintenance: external links rot over time, as other sites move or disappear.
• When an AEM page has been copied between sites or languages: references to the source site often survive.

How to use it

1. Choose the input.
   • Live site: enter a page, and pick This page or Crawl the site.
   • Paste HTML: for pages behind a login. Paste the source and the page’s address, and the links are checked live.
2. Set the options.
   • Up to N pages limits a crawl.
   • Stay in this folder keeps it inside, say, /blog/.
   • Untick Images, scripts and CSS to check only links.
   • Untick Links to other sites to check only your own site.
3. Press Check links. Results appear as they come in, and you can Stop at any time.
4. Read the list. It opens on Problems; switch to Broken, Redirects, Blocked or unclear or All, or search by address or link text. Open On … page under any result to see every page and line it appears on.
5. Press Export CSV for a spreadsheet with one row per link per page.

What the results mean

Result	Meaning	What to do
200	Works	Nothing
Redirects → 200	Works after one or more redirects	Point the link at the final address, especially for permanent (301/308) redirects
404 / 410	The page doesn’t exist	Fix the link, or redirect the old address to the right page
5xx	Server error	Often temporary; check again later, then tell the site owner
No connection	The host doesn’t exist or refused to connect	The domain may have expired; remove or replace the link
Redirect loop	Keeps redirecting	Fix the redirect rules
401 / 403 / 429 / 999	The site refused the checker	Usually bot protection. Open it in a browser to confirm
Timed out	No answer in 8 seconds	Slow or down; check again later
#anchor not found	The page works, but the id the link points to doesn’t exist	Update the fragment, or restore the id

The checker asks each link with a quick HEAD request, and confirms any failure with a full GET, because some servers answer HEAD wrongly.

How the crawl works

The crawl starts at your page and follows ordinary links (not nofollow) to pages on the same host and protocol, skipping files such as PDFs and images. It reads up to the page limit, checks every link it finds, and checks #anchors against every page it read. To keep the load on your server low, it fetches a few pages and links at a time.

Limits

• It reads HTML as served, without running JavaScript. Links added by scripts in the browser aren’t found.
• Up to 100 pages and 2,000 distinct links per run. The checker allows 60 batches of 6 links a minute per visitor; if you hit that limit, it pauses and carries on by itself.
• It doesn’t log in or fill in forms: it checks what a visitor following the link would reach.
• A link that works for the checker can still fail for visitors in other countries, or behind a cookie wall.