| 开发者 | devdome |
|---|---|
| 更新时间 | 2026年9月18日 00:45 |
| PHP版本: | 7.4 及以上 |
| WordPress版本: | 7.1 |
| 版权: | GPLv2 or later |
| 版权网址: | 版权信息 |
post_content) of published posts in public post types (attachments excluded), meaning <a href> and <img src>, including markup inside block comments. Internal and external URLs are both checked.
Not scanned: widgets, menus, theme and customizer options, post meta and custom fields, srcset candidates, oEmbed and iframe targets, the bodies of reusable or synced patterns that are not inlined in the post, and drafts, private or scheduled content. This is not a whole-site crawler.
Why Use DevDome Link Monitor?
/wp-content/plugins/ and activate it, or install it from the Plugins screen.Yes. Every <a href> and <img src> in the saved content of published posts is checked, whether it points to your own site or to another one. Widgets, menus, custom fields and theme options are not scanned.
No. Scanning only reads your content and records results. Content changes only when you choose Edit URL or Unlink for a specific link, and only if you have WordPress permission to edit every post that contains it.
Some websites block automated requests, rate-limit them, require a login, or time out now and then. DevDome Link Monitor keeps those results separate as "Unverified" or "Could not verify" instead of calling the link broken. Use Re-check now, or open the link yourself.
Because it refuses to guess. A link is only marked broken after failing twice in checks separated in time; timeouts, DNS failures and TLS errors are shown as "Unverified" rather than broken; and 401/403/429/999 responses are shown as "Could not verify". What is left is much more likely to be genuinely broken. It is a conservative policy, not a guarantee - a target that answers 200 with a "page not found" page still counts as OK.
Yes, always. Every request for a missing URL is recorded with its path, hit count, last referrer, last user agent and first and last seen time, with human hits and bot hits kept apart.
404 logging adds one indexed lookup and one write, and only on requests that are already 404s. The scan runs in small chunks with low default budgets and at most two links per host per tick, but it does run on your server and does use PHP workers, so a scan is not free. Lower the batch sizes in Settings on small hosts.
It matches the user-agent string against known crawler patterns. That is all it can do: user agents are trivially forged, so treat "human" as "not obviously a crawler". Requests with no user agent are counted as bots. The Settings tab names the classifier that produced the split.
They use WP-Cron, which only fires when someone visits your site (or when a real cron job calls wp-cron.php). On a low-traffic site a scheduled scan can run late.
No. Link checks go from your server straight to the sites you link to. Nothing is sent to DevDome unless you connect a DevDome account, and even then only aggregate counts (links checked, healthy, broken, redirects) for the optional email summary. Never URLs, anchor text, post titles or 404 paths.
Yes. Network activation provisions the sites it can reach in one request; any remaining site (and any site created later while the plugin is network-active) creates its own tables on its first load. Settings, the 404 log and scan results are per site.