An administrator checks a server running a RAID 1 with ten drives and notices three drives have blinking amber lights. Which of the following is the FIRST action the administrator should take?

Reseating drives one at a time resolves potential false alerts caused by loose backplane or SAS/SATA connector seating without risking further degradation of the RAID 1 array. This non-destructive step follows standard troubleshooting methodology by addressing the simplest physical cause first, and RAID 1 can continue operating during the process because mirroring keeps the remaining drives active.

A RAID 1 array is designed to tolerate drive failures and continue serving data, so declaring the server down before confirming actual failure is premature and likely inaccurate.

Shutting down the server and replacing drives before verifying that the drives have truly failed introduces unnecessary downtime and risks disrupting a still-functional array.

Contacting support is appropriate only after local troubleshooting steps are exhausted; escalating before attempting a simple reseat adds unnecessary delay to resolution.