Check how to set up an early warning system for crawler and extractor issues.
Xavier has suggested checking the logs on extractor job backlog
CC noted the downloads_success_feed table gives an idea of if feed fetching was healthy (i.e.group by date to see how many feeds are fetched per day)
Evaluate if it is necessary to create API point for early warning monitoring