March 2020
Beginner
592 pages
14h 14m
English
Important operations that are performed on a service should always be logged. Logging information needs to be categorized to be of any real value. A common list of categories includes debug, info, warning, error, and fatal. Logging information should be collected by a central log aggregation service and not be stored on an individual node of the cluster. Aggregated logs are easy to parse and filter for relevant information. This information is essential to quickly pinpoint the root cause of a failure or unexpected behavior in a distributed system consisting of many moving parts, running in production.
Read now
Unlock full access