When a network or other resource needs to become unavailable in order to maintain it, how do you position this to your boss or customer to see it in a positive light and understand that the inconvenience is necessary for a greater good?
Obviously we’re doing what we can to prevent downtime to begin with (redundancy, fail-overs, etc) but sometimes downtime can not be avoided.
The idea is abstract but can apply to many situations; Service architecture changes, failing hardware replacements, even the need to reboot a machine for a feature/security update.
How do you get someone who argues ‘This is unacceptable, this needs to be available every minute’ to understand that this maintenance is to help that ‘every minute’ availability, and that scheduled is better than unplanned outages?
Responses
This server is going offline, period. Either we schedule it, take precautions to minimize the impact (let users work from home or whatever), or the server will schedule it for us when it breaks. At that point, the impact will be significantly bigger, and the downtime exponentially longer. I know which I’d pick.
We are contractually obligated to update this server. This includes downtime. If you don’t allow us to do this maintenance, all incidents following an issue with this device will be billable.
Create a maintenance schedule. We have the 3rd weekend of every month scheduled for maintenance tasks. We don’t always need downtime, but it is scheduled. We communicate out a week in advance as well as 24hrs out whenever downtime is required. Preventive Maintenance should be a part of everything we do to maintain stability and vendor support. We also do our best to tackle things outside of typical work hours. You will never make everyone happy, but its the majority we gear towards.