Reads one page from our server and checks up to 50 of its links, listing the ones that return errors, time out or redirect in chains. For site owners who want to fix dead links on a page before visitors find them. It checks one page, it is not a site crawler.
How to use it
Enter the address of the page whose links you want to check. You can leave out https://.
- Type or paste the address of the page and press Run check.
- Read the line under the title: it says how many links the page has and how many were checked.
- Open the list of links that need attention, starting with the broken ones.
- Fix, replace or remove those links on the page, then run the check again.
How Reavlo tests this
This tool runs on our server, because a browser cannot request another site's links. Nothing about the request is stored or logged: not the address, not the links found.
One page, 50 links. The tool reads the HTML of the page you enter and checks its links. It does not follow them to other pages and does not crawl your site. The limit is 50 distinct links per run, in the order they appear on the page; a link that appears twice, or twice with different #fragments, is checked once. If the page has more, the tool says how many were not checked.
How a link is checked. Each link gets a HEAD request, which asks for the status without the page. Only when a site does not support HEAD (status 405 or 501) is one GET request made, and then only the status is read. At most 5 links are checked at the same time, and requests to the same site are spaced apart, so the check does not load a site. Up to 3 redirects are followed and shown. Addresses that point to private networks or unusual ports are not requested and are reported as not checked.
Statuses. 404, 410 and other client errors, and a domain that does not resolve, are broken (critical). A server error (500 and above) and a link that times out, cannot connect or loops are warnings. A link that redirects is information, and two or more redirects in a row are a warning. Status 401, 403, 429 and 999 are reported as "refuses automated checks": many sites, social networks among them, turn robots away while people get through, so the link may work.
Limits. The links are read from the HTML as the server sends it. Links added by JavaScript are not seen, and mailto:, tel: and in-page # links are not checked. A link that works for a robot can still fail for a visitor (a login wall or a region block), and the other way round.
Questions
Is the address I enter stored?
No. The address goes to our server only to read the page and check its links, and neither it, the links nor the answers are stored or written to a log. Analytics records only that the tool ran, never the address.
Does it check my whole site?
No. It checks one page and up to 50 of its links. To check another page, run it again with that address. It is not a crawler and does not follow the links it finds.
Why is a link marked as refusing automated checks?
The site answered with 401, 403, 429 or 999, which many sites send to robots. Open the link in a browser to see whether it works for a person.
Why does the tool use HEAD requests?
A HEAD request asks for the status and headers without the page body, so it costs the other site almost nothing. If a site does not support it, a single GET is made and only the status is read.
Why are some of my links not listed?
Links added by JavaScript, mailto: and tel: links, in-page # links and anything after the first 50 distinct links are not checked.