Monitoring the KVM virtualization platform
When we talk about running a system that is performing any kind of processing, we quickly come to the problem of monitoring and making sure that our system runs inside a given set of parameters.
When we create a system that is running a workload, it will inevitably produce some kind of data on everything that is happening. This data can be almost infinite in its scope – a server that is just online, without a single useful task running will create some kind of log or service data, such as the amount of used memory, services that are starting or stopping, the amount of disk space left, devices that are connecting and disconnecting, and more.
When we start running any useful task, the logs will only get larger.
Having a good and verbose log means that we can find what is going on at this instant with the system; is it running correctly and do we need to do something to make it run better? If something unexpected happens, logs...