CompTIA Server+ (SK0-005)TroubleshootingMedium
A server administrator receives an alert that a server's disk I/O utilization is consistently at 100%, causing applications to be extremely slow and unresponsive. Further investigation reveals that a single logical drive, part of a RAID 5 array, is showing high read/write latency. All other drives in the array appear healthy according to the RAID controller's utility. What is the MOST likely cause of this issue?
- AA failing individual disk drive within the RAID 5 array.
- BNetwork saturation on the server's NIC.
- CInsufficient server RAM for the workload.
- DA faulty RAID controller cache module.
Show answer & explanationAnswer & explanation
Correct answer: A. A failing individual disk drive within the RAID 5 array.
High disk I/O on a single logical drive with high latency on one specific physical drive within a RAID 5 array, while others are healthy, strongly suggests that the individual drive is failing or experiencing severe performance degradation. RAID 5 can tolerate a single drive failure, but a struggling drive can cause significant performance bottlenecks as the controller tries to work around its issues.
Why the other options are wrong
- B. Network saturation would cause network-related slowdowns and high NIC utilization, not specific high disk I/O and latency on a storage array.
- C. Insufficient RAM would typically manifest as high CPU utilization (due to excessive swapping) or application crashes, not specifically high disk I/O on a single logical drive with one failing physical disk.
- D. A faulty RAID controller cache would typically impact all I/O across the array, not just one logical drive or one physical disk's latency.
Degraded RAID 5 Performance
Significant slowdown in a RAID 5 array's read/write operations, often caused by a single failing or degraded drive, even if the array is still functional.
- RAID 5 can operate with one failed drive, but performance is severely impacted.
- A 'failing' drive might not be completely dead but struggling, causing high latency and retries.
- Monitoring individual drive health and I/O statistics is crucial for early detection.
Memory trick: RAID 5: One slow drive makes the whole array a snail!