Learning DevOps: Continuously Deliver Better Software

Centralizing logs with Syslog

Generally speaking, most applications will have a logging facility that will produce text logs that can be written into an arbitrary location on a storage device (normally, defaulting to a local disk partition). Although this is an essential first step, it also produces problems; you need to provide adequate storage for logs and you need to put in place rotation to stop them from growing too large. If they contain vital information, you need to ensure that they are rotated to a safe location, but you may also need to ensure that they remain secure if they contain sensitive information, such as credit card numbers. On top of all of this, you will lose the current logs if you have a disaster on that node.

A good solution ...