For example, we continuously monitor our systems to check that they are working as intended and to detect and fix errors. Also, when a specific feature becomes unavailable or crashes, reviewing the log information that was created prior to that event allows us to analyze the problem and quickly make the feature available again.