Runbook
|
JC pontificates about technology
An occasional series.
|
In a computer network, a runbook is a glorified checklist: a manual of routine procedures that the system administrator carries out when maintaining the system, and handling special requests and contingencies on the network. It allows other operators to effectively manage and troubleshoot a system. Through runbook automation, these processes can be carried out using software tools in a predetermined manner.
It's list of standard protocols and checklists for maintaining complex but dumb machinery, and as such has natural attraction for chief operating officers, who look upon their charges in exactly those terms. so expect to see runbooks applied to the operations department, especially when it is going through a period of stress or significant change.
Judge for yourself the extent to which the machine-age dogma has infected operations management orthodoxy, by this paragraph from Wikipedia:
- “Operational runbooks may be tied to IT Infrastructure Library (ITIL)[1] incidents to allow repeatable processes[2] supporting specific aspects of the service catalog. The runbook is typically divided into routine automated processes and routine manual processes. The runbook catalog begins with an index of processes covered and may be broken down in outline form to align the processes to the major elements they support in the service catalog.”
If you are applying a runbook effectively to a large organisation of people, you have already drastically mis-allocated your resources, since meatware is a really crappy and expensive method of carrying out repetitive tasks.