- Identify the actual resource constraint.
- Create a usable recovery point.
- Choose a suitable maintenance window.
Choose a proportionate step
The six VPS plans increase CPU, memory and NVMe allocation in specific steps. Compare the numbers instead of assuming every resource doubles. Review dedicated hardware separately if physical isolation or specialised ports matter.
Region and term remain explicit. The review shows the selected recurring price and any adjustment.
Protect the current state
Create a named snapshot and verify the latest backup. Use an application-consistent export for databases. Keep console access and the important service configuration available.
Do not make resizing the first response to an unexplained data or disk issue. Preserve data and diagnose the underlying fault.
Review the change
Choose Resize in the server overview. Inspect CPU, memory, storage and cost before confirming. If credit is used, check the deduction and remaining balance.
Avoid changing the OS, application and machine size simultaneously. A controlled change is easier to assess.
Verify the guest and application
Confirm reachability and recognised resources. Check the application, data writes and background tasks. Expanding a virtual disk may still require a volume or filesystem change in the guest; use the procedure for that filesystem.
Observe a representative workload and compare it with the previous baseline.
Treat a smaller disk as a migration
Disk reduction is different from adding resources. If the intended destination cannot safely use the current disk layout, migrate to a new machine and verify the copy. Keep the original available until the rollback window closes.