Problem Statement: Disk performance issues like high I/O wait times or latency can severely affect application performance. Many teams only detect this when applications start failing, rather than proactively monitoring it.
Detailed Scenario: A lightweight Python tool that periodically checks disk I/O stats and latency using iostat or /proc/diskstats. If latency or utilization exceeds a threshold, the tool should log the event and optionally send an alert.
Use Case Approach: Collect disk I/O statistics from iostat, comprae with threshold, log and alert.
Tools and Modules: shutil
══════════════ ⭑ ⭑ ⭑ ⭑ ⭑ ══════════════
Approach:
- Collect disk I/O statistics from iostat or /proc/diskstats
- Parse read/write operations per second, average queue size, and average wait time
- Compare with configurable thresholds
- Log to file and print summary in the console
══════════════ ⭑ ⭑ ⭑ ⭑ ⭑ ══════════════
Usage
./diskio_monitor.py --interval 2 --iterations 0 --util-threshold 80 --await-threshold 50
--interval : seconds between samples (default 2)
--iterations: number of updates, 0 = run forever
--util-threshold: mark devices with util >= this %
--await-threshold: mark devices with await >= this ms
