Data center maintenance tasks [/ˈdætə ˈsɛnər ˈmeɪntənəns ˈtæsks/] n - A running reckoning of the labors that keep a data center reliable. Most failures do not arrive with a bang; they begin with an inspection skipped, a filter never changed.
Of Each Day
- Walk the floor; attend to audible alarms and abnormal LED states.
- Review the BMS and monitoring tool dashboards for temperature and humidity trends.
- Verify the generator and UPS status indicators stand green.
- Read the previous shift’s handoff notes for follow-up items.
Of Each Week
- Inspect CRAC/CRAH filter status; change them when differential pressure runs high.
- Test emergency lighting and EPO signage.
- Review capacity reports for power, cooling, and rack space.
- Spot-check cable management and airflow beneath the raised floors.
Of Each Month
- Exercise the The Standby Engine under load and verify fuel levels.
- Run a battery discharge test on the Uninterruptible power supply.
- Calibrate the sensors: thermal cameras, leak detection, humidity probes.
- Enter new lessons into the runbooks.
Of Each Quarter
- Inspect busway tap boxes and bolted connections for torque and heat damage.
- Clean condenser coils; verify cooling-loop chemistry.
- Review and rotate stock for critical spares.
- Walk a failover drill through with the operations team.
Of the Record
Every task must record who did it, when, and what was found. A plain log lets drift be seen over time, and defends the facility when an audit or an incident review comes calling.