DCNSA Level III

Server Maintenance flashcards

Read the question, say the answer to yourself, then turn the card to check. The first 10 of 56 cards are free; the rest come with the subject.

Card 1 of 56

Jump to a card

Complete Flashcard Reference

Review every card in this deck. Open any card to reveal its answer. The first 10 are free; the other 46 come with the subject.

What are the four types of server maintenance, and what triggers each?

Preventive: a fixed plan, acting before any sign of trouble. Predictive: an early warning sign in monitoring data. Corrective: a fault that has already occurred. Periodic: a fixed calendar interval, regardless of the server's condition.

What is the difference between restore and recover?

Restore: returning lost or corrupted data to a working state from a backup, the narrow, data-focused half. Recover: the broader process of returning a server or service to normal operation after a disruption, which may involve restoring data but also rebuilding configuration, reinstalling roles or failing over to another server.

What are the four parts of a network maintenance procedure?

1. Create a plan. 2. Troubleshoot network issues that arise. 3. Install and configure products. 4. Monitor and improve network performance. Skipping any one of them is what turns routine maintenance into an incident of its own.

What are the three interface categories of monitoring tool, and the strength of each?

GUI tools: visual dashboards, easy to read at a glance and to show a non-technical stakeholder. CLI tools: text-based, fast, scriptable, and often the only option over a limited remote connection. PowerShell: automates the same check or fix across dozens of servers with one script.

How do you project how many months remain before a resource fills, from a monitoring trend?

Months remaining = (capacity - current) / rate of change per month. Work out the rate from the trend first (total change / number of intervals between readings), then divide the remaining capacity by it, not the other way round.

What three documents does a well-run server maintenance programme keep?

The server monitoring and maintenance document. The server hardware document. The server back-up document. Together they let a different administrator, or the same one months later, see what state a server is in, what has been done to it and what to do next, without relying on memory.

A 500 GB file server takes a full backup every Sunday and an incremental on the other six days, retained four weeks, each incremental averaging 15 GB. How much backup storage is needed?

Fulls: 4 x 500 GB = 2,000 GB. Incrementals: 6 x 4 = 24 x 15 GB = 360 GB. Total: 2,000 + 360 = 2,360 GB, plus a safety margin. A target sized to the 500 GB of live data would start failing within the first month.

How is availability worked out from MTBF and MTTR?

Availability = MTBF / (MTBF + MTTR). MTBF 1,000 hours and MTTR 5 hours: 1,000 / 1,005 = 99.5%. Halving MTTR to 2.5 hours: 1,000 / 1,002.5 = 99.75%. Repair time is the lever the administrator controls.

With four 2 TB disks, what usable capacity does each RAID level give, and how many failures does it survive?

RAID 0: 8 TB, survives none. RAID 10: 4 TB (half), one failure per mirrored pair. RAID 5: (4 - 1) x 2 = 6 TB, one failure. RAID 6: (4 - 2) x 2 = 4 TB, two failures.

Why is system clock accuracy part of server maintenance?

Authentication: a domain's Kerberos authentication fails if clocks drift too far apart, even when the credentials are correct. Logging: events must be timestamped consistently to be useful. Scheduled tasks also depend on it to run correctly.

What is a disaster, in server maintenance terms?