If your architecture permits it, it's not a bad idea to dump and save the raw page on every crawl. It'll be insanely useful for debugging and being able to assign blame )