Page did not respond (timeout or connection error)
http_timeoutShort answer
No HTTP status came back: the connection was refused, reset, or the server did not answer within the crawler's timeout. Google treats this as a server error. Check that the host is reachable and how long the page takes to respond; if it is a single slow page, fix its queries or caching.
Why it matters
A page that never answers cannot be crawled or indexed, and from Google's side it looks like an outage. Persistent timeouts lower the crawl rate for the host and eventually remove the URL from the index. Timeouts on a handful of pages usually point at an expensive page (heavy database query, external API call during render) rather than the server as a whole.
How Glimana detects it
The crawler stores status 0 when no response arrived before the timeout, the connection failed, or TLS could not be negotiated. The rule fires on status 0. The Crawls page shows how many fetches failed in the crawl.
How to fix it
If many pages are affected, the server, DNS or firewall is the cause — check uptime monitoring, the DNS records, and whether a firewall or bot protection is dropping the crawler's requests (Glimana's crawler identifies itself in the user agent). If one or two pages are affected, profile those pages: look for slow queries, uncached external calls, or very large responses.
Install Query Monitor and open the slow page while logged in; it lists queries above 0.05 s and external HTTP calls. Cache the result or move the call to a cron job.
Timeouts come from app proxies and third-party scripts that block rendering; check app embeds.
Add request timing to logs; set an upper bound on external calls during render and cache their results.
How the fix is verified
Affected URLs are re-fetched; the task closes when each returns an HTTP status.