How can we get nearly a million pages reindexed after blocking Googlebot?

A 403 tells crawlers that access is forbidden, so using it for a prolonged period can cause URLs to be dropped or delayed. Make sure important pages now return successful responses consistently, that internal links still point to them, and that robots.txt or firewall rules are not blocking Googlebot. Some people prefer 404 responses for genuinely removed pages, but removed content should not be changed to 404 merely to speed up recrawling.

Create XML sitemaps as soon as possible. Individual URL inspection and indexing requests are intended for a small number of pages, while sitemaps give Google a way to discover and recrawl large URL sets. Split the URLs across multiple sitemap files if necessary and submit a sitemap index. Don’t rely on the pending validation process as the trigger for recovery; validation checks whether the error is gone, while ordinary crawling can proceed separately.

For a site with roughly a million URLs, a sitemap is probably the most practical way to accelerate recovery. Include canonical, currently available URLs and update the files as content changes. If the site uses a common CMS, a sitemap extension may handle generation automatically.