# INFRASTRUCTURE HEALTH REPORT
## Laboratory Exercise 01

### 1. Objective
To monitor the health of an Ubuntu application server using Linux CLI monitoring tools and identify possible CPU, memory and disk I/O bottlenecks.

### 2. Tools Used
- top – CPU and process monitoring
- free – Memory monitoring
- df – Disk space monitoring
- uptime – System load monitoring
- vmstat – Virtual memory and system statistics
- iostat – Disk I/O monitoring
- sar – System activity reporting
- stress-ng – Controlled workload generation

### 3. Observations

The system was monitored under normal and simulated load conditions.

Under normal conditions, CPU utilization and memory consumption were relatively low, and disk operations were performing normally.

CPU stress was generated using stress-ng. This increased CPU utilization and reduced CPU idle time.

Memory stress was generated to observe changes in available RAM and swap usage.

Disk I/O stress was generated to observe changes in disk throughput, request latency and device utilization.

### 4. Performance Analysis

CPU:
High CPU utilization combined with low idle time indicates CPU contention.

Memory:
Reduced available memory and increased swap activity indicate memory pressure.

Disk:
Increased disk latency, utilization and I/O wait indicate possible disk I/O contention.

The resource showing sustained abnormal measurements is considered the probable performance bottleneck.

### 5. Monitoring Thresholds

Threshold 1 – CPU Utilization:
Generate a warning when CPU utilization exceeds 80% continuously for 5 minutes.

Justification: Sustained high CPU utilization can indicate insufficient processing capacity and may increase application response time.

Threshold 2 – Available Memory:
Generate a warning when available physical memory falls below 10% of total RAM continuously for 5 minutes.

Justification: Low available memory increases the likelihood of swapping, which can degrade application performance.

### 6. Conclusion

Linux monitoring utilities were used to examine CPU, memory, disk capacity, system load and disk I/O performance.

Controlled workloads were generated to observe resource utilization under stressed conditions.

The collected measurements were analyzed to identify probable performance bottlenecks. Appropriate monitoring thresholds were defined to detect resource exhaustion before it affects application availability.