Search indexing
Search engines decide whether and when to index a page. Correct technical signals make pages crawlable; they do not guarantee rankings, impressions or recovery.
Installation checks
- Set the exact final HTTPS
TECHPEDIA_PUBLIC_URL. - Confirm article GET and HEAD return the expected status and canonical URL.
- Inspect returned HTML without JavaScript: the title and article body must exist.
- Check
/robots.txtand/sitemap.xmlthrough the real public proxy. - Confirm active articles are listed, not drafts, admin pages or missing articles.
- Check canonical/hreflang links use real Techdata relationships.
- Submit the sitemap in Search Console; inspect representative URLs and request indexing for corrected public articles where appropriate.
History, editor, account, search and administrator pages are intentionally excluded. A noindex report for those paths is not an article indexing defect. Missing articles return 404 even when a creation UI is displayed.
Old URLs
Legacy MediaWiki paths, underscore/title variants and deleted pages need separate analysis. Redirect only when there is a real equivalent target. Do not redirect every 404 to the homepage or remove all noindex rules to improve a coverage count. Maintain real moved-name and alias redirects.
The sitemap uses stored update times; pages with dynamic boxes/system components may omit lastmod rather than invent a date. HTML validators use full-response ETags because edition UI, boxes and links can change independently of revisions.
Compare Search Console by page/query and date, not only a single aggregate chart. Check actual HTTP responses, deploy history, server availability and editorial changes before attributing a decline to one code change.
References: Google sitemap guidance, canonical URLs, JavaScript SEO.