Monitoring, Diagnostics, and Troubleshooting
Troubleshooting questions reward a structured method: define the symptom, gather data, form a hypothesis, test the cheapest/most likely cause first, and verify the fix. Know the monitoring layers: facility (power, cooling, environmental sensors), infrastructure (switches, PDUs), and server (BMC, OS, application). Metrics include temperature, power draw, link errors, disk SMART data, and latency.
Key concepts: differentiate symptom from cause (a server rebooting may be a power or thermal issue, not an OS bug); use logs and timestamps across layers; understand alert severity and escalation; and know that intermittent faults often trace to cabling, connectors, or environmental conditions. Familiarity with ping/traceroute, link status, error counters, and BMC event logs is expected.
Traps: replacing hardware before confirming the fault; ignoring environmental data when diagnosing random reboots; and assuming a green status light means healthy (it may only mean powered). Also watch for questions that confuse a monitoring alert with a root cause.
Worked example / rule of thumb: a server that reboots under load - check inlet temperature and PSU logs before swapping components; a thermal or power event usually leaves a BMC entry. Rule of thumb: change one variable at a time and document the result; if the symptom moves with the component, you found the cause.