Most doomsday scenarios reduce to the same practical question: can your team keep working when something big and shared fails? A one page checklist covers it. The items below come from CISA's lessons on the Ukraine attack, GPS dependence and SATCOM, plus a vendor update outage.
Why one checklist
Power cuts, a GPS disruption, a cloud region failing and a bad vendor update look different. What breaks for a small team is similar: you lose access, lose time sync, lose tools, and lose the ability to coordinate. Preparing for the shared effects covers many scenarios.
The checklist
People and contact
- A printed or offline list of who decides, who fixes and how to reach them without your chat tool.
- A named backup for each role.
Access
- Break-glass accounts for your main identity provider and cloud console, stored securely and tested.
- Multi-factor authentication methods that do not all depend on one device or one network.
Data
- Backups kept somewhere your normal credentials cannot delete, with a restore you have actually tested. CISA's Ukraine alert shows attackers targeting recovery by wiping systems and disabling devices.
Operations
- A written manual mode for each critical process. CISA advises developing and exercising contingency plans for safely operating or shutting down processes if a control system is breached.
- Runbooks stored offline.
Dependencies
- A list of providers whose failure would stop you, and what you do if each fails. See Systemic cyber risk: a concentration checklist.
- A second time source, since CISA says GPS is in many cases the sole source of timing.
Communication
- A template for telling customers what is happening, what is affected and when you will update.
Rehearse it
A checklist you never practice fails. Hold a short exercise twice a year: pick one scenario, such as no cloud console for four hours, and walk through it. Write down what surprised you and fix those items.
Keep it small
Do not try to prepare for every scenario. Prepare for loss of access, loss of data, loss of time, and loss of a key vendor. Those four cover most of what the sources describe.
Sample scenarios for the rehearsal
Use short, concrete prompts. Our cloud console is unreachable for four hours. Our identity provider is down and nobody can sign in. Time is wrong on half our servers. A vendor update has broken one in ten laptops. For each, ask who notices, who decides, what customers see and what the first three actions are.
Keep the session to an hour, and write down fixes with owners and dates. The point is not a perfect answer. It is to find the gap while the stakes are low, such as a break-glass account that no one remembers the location of.
Quick check
What is the first thing to prepare?
A way to reach people and critical consoles without your normal tools, since loss of access slows everything else.
Sources
- CISA alert IR-ALERT-H-16-056-01 on the Ukraine power grid attack
- CISA: Positioning, Navigation, and Timing
- CISA and FBI advisory AA22-076A on SATCOM networks
Last reviewed: 19 October 2026.
