"Help! My Server is on Fire (Probably)" - A DevOps Engineer's Guide to Panic-Driven Statistics.

Christopher Akinsanmi — Wed, 16 Apr 2025 09:19:02 +0000

The Problem: Servers Behaving Badly

You wake up at 3 AM to an alert: "CPU at 99%!" Your heart rate spikes. Is it a crypto miner? A runaway cron job? Did someone deploy to production on a Friday? Again?
As a DevOps engineer, you need answers—fast. But logging into a server and running a dozen commands is exhausting. You need one magical script to tell you everything:

How much CPU is melting?
Why is memory disappearing like my motivation on Monday?
Who filled up the disk? (Spoiler: It’s always logs.)

So, I present to you: server-stats.sh—the Swiss Army knife of server diagnostics, wrapped in anxiety-reducing colors.

Introducing server-stats.sh – Because Guessing is Not a Monitoring Strategy

This script does everything you wish top and htop did, but without the weird keybindings you always forget.

What It Tells You (Before You Have a Mental Breakdown)
"Is the CPU on Fire?"

Total CPU usage (so you know whether to panic now or in 5 minutes).

"Why is Memory Gone?"

Free vs. Used (with percentages, because math is hard at 3 AM).

"Who Ate My Disk Space?"

Lists disk usage (and turns red if it’s critical, just like your face when you see 95% usage).

"Top 5 Processes Trying to Kill Your Server"

CPU hogs (looking at you, Java).
Memory vampires (probably Node.js with 500 open dependencies).

Bonus: "Is Anyone Else Even Awake?"

Shows logged-in users (so you know who to blame).
Failed login attempts (because hackers love weekends too).

The Script (AKA "The Hero We Need")

#!/bin/bash

# server-stats.sh - A comprehensive server performance statistics script

# Color codes for output
RED='\033[0;31m'
GREEN='\033[0;32m'
YELLOW='\033[0;33m'
BLUE='\033[0;34m'
MAGENTA='\033[0;35m'
CYAN='\033[0;36m'
NC='\033[0m' # No Color

# Function to display section headers
section_header() {
    echo -e "\n${MAGENTA}=== $1 ===${NC}"
}

# Function to convert bytes to human readable format
human_readable() {
    local bytes=$1
    local units=("B" "KB" "MB" "GB" "TB" "PB")
    local unit=0

    while (( bytes > 1024 )) && (( unit < ${#units[@]} - 1 )); do
        bytes=$((bytes / 1024))
        ((unit++))
    done

    echo "$bytes ${units[$unit]}"
}

# Get OS information
section_header "System Information"
echo -e "${CYAN}Hostname:${NC} $(hostname)"
echo -e "${CYAN}OS:${NC} $(grep PRETTY_NAME /etc/os-release | cut -d '"' -f 2)"
echo -e "${CYAN}Kernel:${NC} $(uname -r)"
echo -e "${CYAN}Uptime:${NC} $(uptime -p | sed 's/up //')"
echo -e "${CYAN}Current Time:${NC} $(date)"

# CPU Usage
section_header "CPU Usage"
total_cpu_usage=$(top -bn1 | grep "Cpu(s)" | sed "s/.*, *\([0-9.]*\)%* id.*/\1/" | awk '{print 100 - $1"%"}')
echo -e "${CYAN}Total CPU Usage:${NC} $total_cpu_usage"
load_avg=$(uptime | awk -F 'load average: ' '{print $2}')
echo -e "${CYAN}Load Average (1, 5, 15 min):${NC} $load_avg"

# Memory Usage
section_header "Memory Usage"
free_output=$(free -m)
total_mem=$(echo "$free_output" | grep Mem: | awk '{print $2}')
used_mem=$(echo "$free_output" | grep Mem: | awk '{print $3}')
free_mem=$(echo "$free_output" | grep Mem: | awk '{print $4}')
mem_percent=$((used_mem * 100 / total_mem))

echo -e "${CYAN}Total Memory:${NC} ${total_mem}MB"
echo -e "${CYAN}Used Memory:${NC} ${used_mem}MB (${mem_percent}%)"
echo -e "${CYAN}Free Memory:${NC} ${free_mem}MB"

# Swap Usage
swap_total=$(echo "$free_output" | grep Swap: | awk '{print $2}')
swap_used=$(echo "$free_output" | grep Swap: | awk '{print $3}')
if [ "$swap_total" -gt 0 ]; then
    swap_percent=$((swap_used * 100 / swap_total))
    echo -e "${CYAN}Total Swap:${NC} ${swap_total}MB"
    echo -e "${CYAN}Used Swap:${NC} ${swap_used}MB (${swap_percent}%)"
else
    echo -e "${CYAN}Swap:${NC} Not configured"
fi

# Disk Usage
section_header "Disk Usage"
df_output=$(df -h --total | grep -vE '^tmpfs|udev')
echo "$df_output" | awk -v cyan="${CYAN}" -v nc="${NC}" -v green="${GREEN}" -v red="${RED}" '
    NR==1 {printf "%s%-20s %s%-10s %s%-10s %s%-10s %s%-10s %s%s\n", cyan, "Filesystem", cyan, "Size", cyan, "Used", cyan, "Avail", cyan, "Use%", cyan, "Mounted on", nc}
    NR>1 && $1 != "total" {
        color=($5+0 > 80) ? red : green;
        printf "%-20s %-10s %-10s %-10s %s%-10s%s %-10s\n", $1, $2, $3, $4, color, $5, nc, $6
    }
    $1 == "total" {
        printf "%s%-20s %s%-10s %s%-10s %s%-10s %s%-10s %s%s\n", cyan, $1, cyan, $2, cyan, $3, cyan, $4, cyan, $5, cyan, $6, nc
    }'

# Top Processes by CPU
section_header "Top 5 Processes by CPU Usage"
echo -e "${CYAN}%-10s %-10s %-10s %-10s %s${NC}" "PID" "USER" "CPU%" "MEM%" "COMMAND"
ps -eo pid,user,%cpu,%mem,comm --sort=-%cpu | head -n 6 | tail -n 5 | awk '{printf "%-10s %-10s %-10s %-10s %s\n", $1, $2, $3, $4, $5}'

# Top Processes by Memory
section_header "Top 5 Processes by Memory Usage"
echo -e "${CYAN}%-10s %-10s %-10s %-10s %s${NC}" "PID" "USER" "MEM%" "CPU%" "COMMAND"
ps -eo pid,user,%mem,%cpu,comm --sort=-%mem | head -n 6 | tail -n 5 | awk '{printf "%-10s %-10s %-10s %-10s %s\n", $1, $2, $3, $4, $5}'

# User Information
section_header "User Information"
echo -e "${CYAN}Logged in Users:${NC}"
who
echo -e "\n${CYAN}Failed Login Attempts (last 24h):${NC}"
lastb -a --time-format iso | head -n 10

# Network Information
section_header "Network Information"
echo -e "${CYAN}Active Connections:${NC}"
ss -tulnp | head -n 10

# Security Information
section_header "Security Information"
echo -e "${CYAN}Last 10 Security Events:${NC}"
grep -i 'error\|fail\|denied' /var/log/auth.log | tail -n 10 2>/dev/null || echo "No security logs found or permission denied"

echo -e "\n${GREEN}Server stats collected successfully!${NC}"

How to Use This Script (Before You Cry)

Save it as server-stats.sh

nano server-stats.sh # Paste the script, then [Ctrl]+[X], [Y], [Enter]

Make it executable (because permissions are annoying)
chmod +x server-stats.sh
Run it when you suspect disaster
./server-stats.sh
Interpret the results

Green = "Maybe I can go back to bed?"
Yellow = "I should check this… later."
Red = "Wake up the team (and maybe a therapist)."

Final Thoughts

Servers are like pets—they misbehave when you least expect it. But with server-stats.sh, you can:

✅ Diagnose problems in seconds (instead of Googling htop commands).
✅ Blame the right processes (looking at you, apache2).
✅ Pretend you’re a hacker (even though you just ran a Bash script).

So next time your server acts up, stay calm, run the script, and panic efficiently.

🚀 Happy debugging! (And may your uptime be high.)

Forem: Christopher Akinsanmi

"Help! My Server is on Fire (Probably)" - A DevOps Engineer's Guide to Panic-Driven Statistics.

The Problem: Servers Behaving Badly

Introducing server-stats.sh – Because Guessing is Not a Monitoring Strategy

How to Use This Script (Before You Cry)

Final Thoughts