Shell Scripting — Real-World Projects: Server Health Check Dashboard

#!/usr/bin/env bash
# health_dashboard.sh — Real-time server health overview

set -euo pipefail

# ── Formatting helpers ────────────────────────────────────
RED=$'\033[0;31m' YLW=$'\033[0;33m' GRN=$'\033[0;32m'
BLU=$'\033[0;34m' DIM=$'\033[0;90m' RST=$'\033[0m'
BOLD=$'\033[1m'

hr()     { printf '%s\n' "${DIM}$(printf '─%.0s' {1..60})${RST}"; }
ok()     { printf "  ${GRN}✔${RST} %-35s ${GRN}%s${RST}\n" "$1" "$2"; }
warn()   { printf "  ${YLW}⚠${RST} %-35s ${YLW}%s${RST}\n" "$1" "$2"; }
fail()   { printf "  ${RED}✘${RST} %-35s ${RED}%s${RST}\n" "$1" "$2"; }
section(){ echo; printf "${BOLD}${BLU}  %s${RST}\n" "$1"; hr; }

echo
printf "${BOLD}  SERVER HEALTH DASHBOARD — $(date '+%Y-%m-%d %H:%M:%S')${RST}\n"
printf "  Host: ${HOSTNAME}  |  Uptime: $(uptime -p 2>/dev/null || uptime | awk '{print $3,$4}')\n"

# ── CPU ───────────────────────────────────────────────────
section "SYSTEM RESOURCES"
CPU_IDLE=$(top -bn1 | awk '/^%Cpu/{print $8}' | tr -d '%')
CPU_USED=$(awk "BEGIN{printf \"%.0f\", 100 - ${CPU_IDLE:-0}}")
MEM_TOTAL=$(free -m | awk '/Mem:/{print $2}')
MEM_USED=$(free -m  | awk '/Mem:/{print $3}')
MEM_PCT=$(awk "BEGIN{printf \"%.0f\", ${MEM_USED}*100/${MEM_TOTAL}}")

(( CPU_USED < 70 )) && ok "CPU usage" "${CPU_USED}%" || \
  (( CPU_USED < 90 )) && warn "CPU usage" "${CPU_USED}%" || fail "CPU usage" "${CPU_USED}%"
(( MEM_PCT < 75 )) && ok "Memory usage" "${MEM_USED}/${MEM_TOTAL}MB (${MEM_PCT}%)" || \
  warn "Memory usage" "${MEM_USED}/${MEM_TOTAL}MB (${MEM_PCT}%)"

# Load average
LOAD=$(awk '{print $1}' /proc/loadavg)
CORES=$(nproc)
LOAD_PCT=$(awk "BEGIN{printf \"%.0f\", ${LOAD}*100/${CORES}}")
(( LOAD_PCT < 100 )) && ok "Load average" "${LOAD} (${CORES} cores, ${LOAD_PCT}%)" || \
  warn "Load average" "${LOAD} (${CORES} cores, ${LOAD_PCT}%)"

# ── Disk ──────────────────────────────────────────────────
section "DISK USAGE"
df -h | awk 'NR>1 && /^\// {
  gsub(/%/,"",$5)
  if ($5+0 >= 90) printf "  \033[0;31m✘\033[0m %-25s %s used (%s%%)\n",$6,$3,$5
  else if ($5+0 >= 75) printf "  \033[0;33m⚠\033[0m %-25s %s used (%s%%)\n",$6,$3,$5
  else printf "  \033[0;32m✔\033[0m %-25s %s used (%s%%)\n",$6,$3,$5
}'

# ── Services ──────────────────────────────────────────────
section "SERVICES"
for svc in nginx mysql redis-server; do
  if systemctl is-active --quiet "${svc}" 2>/dev/null; then
    ok "${svc}" "active"
  elif pgrep -x "${svc}" &>/dev/null; then
    ok "${svc}" "running (no systemd)"
  else
    fail "${svc}" "NOT RUNNING"
  fi
done

# ── Database ──────────────────────────────────────────────
section "DATABASE"
if mysql --defaults-file=/etc/myapp/mysql.conf -e "SELECT 1" &>/dev/null; then
  CONNS=$(mysql --defaults-file=/etc/myapp/mysql.conf -BNs \
    -e "SELECT COUNT(*) FROM information_schema.PROCESSLIST")
  ok "MySQL" "${CONNS} connections"
else
  fail "MySQL" "UNREACHABLE"
fi

if redis-cli PING 2>/dev/null | grep -q PONG; then
  KEYS=$(redis-cli DBSIZE 2>/dev/null)
  ok "Redis" "${KEYS} keys"
else
  fail "Redis" "UNREACHABLE"
fi

# ── Recent errors ─────────────────────────────────────────
section "RECENT ERRORS (last 15min)"
ERROR_COUNT=$(find /var/log/myapp -name "*.log" -newer \
  /tmp/.health_check_last 2>/dev/null -exec grep -c ERROR {} + 2>/dev/null | \
  awk '{s+=$1}END{print s+0}')
touch /tmp/.health_check_last
(( ERROR_COUNT == 0 )) && ok "Application errors" "none" || \
  warn "Application errors" "${ERROR_COUNT} in last 15min"

echo

#!/usr/bin/env bash
# monitor.sh — Cron-scheduled version with alerts

set -euo pipefail

ALERT_LOG="/var/log/myapp/alerts.log"
SLACK_URL="${SLACK_WEBHOOK_URL:-}"
THRESHOLD_CPU=85
THRESHOLD_MEM=85
THRESHOLD_DISK=85

alert() {
  local level="$1" msg="$2"
  echo "[$(date --iso-8601=seconds)] [${level}] ${msg}" >> "${ALERT_LOG}"
  [[ -n "${SLACK_URL}" ]] && curl -s -X POST "${SLACK_URL}" \
    -H "Content-Type: application/json" \
    -d "{\"text\":\"[${level}] $(hostname): ${msg}\"}" &>/dev/null
  logger -t health_monitor "${level}: ${msg}"
}

# CPU check
CPU=$(top -bn1 | awk '/^%Cpu/{gsub(/%us,/,"",$2); print 100-$8}' | cut -d. -f1)
(( CPU > THRESHOLD_CPU )) && alert "WARNING" "CPU ${CPU}% (threshold ${THRESHOLD_CPU}%)"

# Memory check
MEM_PCT=$(free | awk '/Mem:/{printf "%.0f", $3/$2*100}')
(( MEM_PCT > THRESHOLD_MEM )) && alert "WARNING" "Memory ${MEM_PCT}% (threshold ${THRESHOLD_MEM}%)"

# Disk checks
df -h | awk 'NR>1 && /^\// {gsub(/%/,"",$5); if ($5+0>THRESH) print $6,$5}' \
  THRESH="${THRESHOLD_DISK}" | while read -r mount pct; do
  alert "WARNING" "Disk ${mount} at ${pct}% (threshold ${THRESHOLD_DISK}%)"
done

# Service checks
for svc in nginx mysql; do
  systemctl is-active --quiet "${svc}" || alert "CRITICAL" "${svc} is NOT running"
done

# Dead-man: if no log entry in 10 min, cron has failed
echo "$(date)" > /tmp/.monitor_heartbeat