Before you touch production: a short operational checklist for an infrastructure change

[rychy]

Прочети на български

Not every infrastructure change needs a long document, but almost every one benefits from a short operational checklist. It is a way to reduce impulsive mistakes right before the change.

A minimal checklist that is worth it

• is it clear what exactly will change
• is there a quick rollback option
• is it clear what needs to be watched right after the change
• is there a dependent component that is easy to miss
• if something goes wrong, is it clear what the first safe move is

A checklist like this does not automatically make the change good, but it sharply reduces the number of silly omissions. That is reason enough for it to exist as a separate topic.

Related topics

• Maintenance windows
• The troubleshooting process