Lately we’ve seen a variety of new tools that help you make sense of your logs. Open-source projects such as Scribe and LogStash, on-premise tools like Splunk, and hosted services such as SumoLogic and PaperTrail. These all help you reduce mass amounts of log data into something more meaningful.
But there’s one thing none of them can you help with. They’re all dependant on the data you actually put into the log. The task of getting that quality and quantity of the data right falls on to you. So when the chips are down and you need to debug your code with partial or missing log data, you may want to cancel that dinner reservation.
To reduce the chances of that happening I wanted to share 5 things you should be constantly mindful of when logging in production.