Server Health Monitor

A live terminal dashboard that shows CPU, RAM, GPU, storage, Docker container status, Fail2ban bans, and SDDM session state — all updating every 2 seconds.

---

Install the Script

sudo nano /usr/local/bin/server-health

Paste the full script:

#!/bin/bash
# server-health — Bazzite server status dashboard
# Live mode: updates every 2 seconds · Ctrl+C to exit

R=$'\033[0m'; DIM=$'\033[2;37m'; BLD=$'\033[1m'
CYN=$'\033[1;36m'; BLU=$'\033[1;34m'; WHT=$'\033[1;37m'
GRN=$'\033[1;32m'; YEL=$'\033[1;33m'; RED=$'\033[1;31m'

THICK="${CYN}$(printf '━%.0s' {1..72})${R}"
THIN="${DIM}$(printf '─%.0s' {1..72})${R}"
REFRESH=2
SLOW_EVERY=4

row()     { printf " ${BLU}%-10s${R} %b\n" "$1" "${*:2}"; }
dot_ok()  { printf "${GRN}●${R}"; }
dot_warn(){ printf "${YEL}●${R}"; }
dot_err() { printf "${RED}●${R}"; }
dot_dim() { printf "${DIM}●${R}"; }

pct_dot() {
  local p=$1 warn=${2:-70} err=${3:-90}
  if (( p >= err )); then dot_err
  elif (( p >= warn )); then dot_warn
  else dot_ok; fi
}

trap 'tput cnorm; printf "\n"; exit 0' INT TERM
[ -t 1 ] && tput civis

# Detect AMD/Intel GPU sysfs node
GPU_DEV=""
for _card in /sys/class/drm/card*/; do
  _vendor=$(cat "${_card}device/vendor" 2>/dev/null)
  [[ "$_vendor" == "0x1002" || "$_vendor" == "0x8086" ]] && \
    { GPU_DEV="${_card}device"; break; }
done

_cpustat() { awk '/^cpu /{print $2,$3,$4,$5,$6,$7,$8; exit}' /proc/stat; }
prev_cs=$(_cpustat); cpu_pct=0

# ── Slow refresh data (network, services) ─────────────────────────
WAN_IP="..."; F2B="0"; NPM_DOT="$(dot_dim)"; NPM_ST="?"
LOOPS=0

refresh_slow() {
  WAN_IP=$(curl -s --max-time 4 https://api.ipify.org 2>/dev/null || echo "unreachable")

  F2B=$(sudo fail2ban-client status sshd 2>/dev/null \
    | grep 'Currently banned' | grep -oE '[0-9]+' || echo "0")

  _npm=$(curl -s -o /dev/null -w "%{http_code}" --max-time 2 http://localhost:81 2>/dev/null)
  if [[ "${_npm:-000}" != "000" ]]; then
    NPM_DOT=$(dot_ok); NPM_ST="reachable"
  else
    NPM_DOT=$(dot_err); NPM_ST="down"
  fi

  _nc=$(curl -s -o /dev/null -w "%{http_code}" --max-time 2 https://localhost:9443 -k 2>/dev/null)
  if [[ "${_nc:-000}" != "000" ]]; then
    AIO_DOT=$(dot_ok); AIO_ST="reachable"
  else
    AIO_DOT=$(dot_err); AIO_ST="down"
  fi
}

refresh_slow

# ── Main loop ──────────────────────────────────────────────────────
while true; do
  (( LOOPS++ ))
  (( LOOPS % SLOW_EVERY == 0 )) && refresh_slow

  # CPU %
  curr_cs=$(_cpustat)
  read -r u1 n1 s1 i1 w1 q1 x1 <<< "$prev_cs"
  read -r u2 n2 s2 i2 w2 q2 x2 <<< "$curr_cs"
  total=$(( (u2-u1)+(n2-n1)+(s2-s1)+(i2-i1)+(w2-w1)+(q2-q1)+(x2-x1) ))
  idle=$(( i2-i1 ))
  (( total > 0 )) && cpu_pct=$(( 100 - 100*idle/total )) || cpu_pct=0
  prev_cs="$curr_cs"

  # RAM %
  read -r mem_total mem_avail < <(awk '/MemTotal/{t=$2} /MemAvailable/{a=$2} END{print t,a}' /proc/meminfo)
  mem_used=$(( mem_total - mem_avail ))
  mem_pct=$(( 100*mem_used/mem_total ))
  mem_used_gb=$(awk "BEGIN{printf \"%.1f\", ${mem_used}/1048576}")
  mem_total_gb=$(awk "BEGIN{printf \"%.1f\", ${mem_total}/1048576}")

  # GPU (AMD/Intel via hwmon)
  gpu_line="n/a"
  if [[ -n "$GPU_DEV" ]]; then
    _hwmon=$(ls "${GPU_DEV}/hwmon" 2>/dev/null | head -1)
    if [[ -n "$_hwmon" ]]; then
      _t=$(cat "${GPU_DEV}/hwmon/${_hwmon}/temp1_input" 2>/dev/null)
      [[ -n "$_t" ]] && gpu_line="${_t: : -3}°C"
    fi
  fi

  # Storage
  read -r disk_used disk_total disk_pct < <(df -BG / | awk 'NR==2{gsub(/G/,""); print $3,$2,$5+0}')
  read -r data_used data_total data_pct < <(df -BG /srv/nextcloud-data 2>/dev/null | awk 'NR==2{gsub(/G/,""); print $3,$2,$5+0}' || echo "? ? 0")

  # SDDM session
  sess=$(loginctl list-sessions --no-legend 2>/dev/null \
    | awk '$2==1000 && $4!="-" {print $5; exit}')
  [[ -z "$sess" ]] && sess="idle (headless)"

  # Docker containers
  docker_status=$(docker ps --format "{{.Names}} {{.Status}}" 2>/dev/null)

  printf '\033[H\033[2J'
  printf "%s\n" "$THICK"
  printf " ${CYN}${BLD}BAZZITE SERVER STATUS${R}  ${DIM}$(date '+%Y-%m-%d %H:%M:%S')${R}\n"
  printf "%s\n" "$THICK"

  cpu_dot=$(pct_dot $cpu_pct)
  mem_dot=$(pct_dot $mem_pct)
  disk_dot=$(pct_dot $disk_pct)
  data_dot=$(pct_dot $data_pct)

  row "CPU"     "${cpu_dot} ${cpu_pct}%"
  row "RAM"     "${mem_dot} ${mem_used_gb}G / ${mem_total_gb}G  (${mem_pct}%)"
  row "GPU"     "${gpu_line}"
  printf "%s\n" "$THIN"
  row "OS disk"   "${disk_dot} ${disk_used}G / ${disk_total}G  (${disk_pct}%)"
  row "Data disk" "${data_dot} ${data_used}G / ${data_total}G  (${data_pct}%)"
  printf "%s\n" "$THIN"
  row "Session" "${sess}"
  row "WAN IP"  "${WAN_IP}"
  row "Fail2ban" "${F2B} banned"
  printf "%s\n" "$THIN"
  row "NPM"     "${NPM_DOT} ${NPM_ST}"
  row "AIO"     "${AIO_DOT:-$(dot_dim)} ${AIO_ST:-?}"

  printf "%s\n" "$THIN"
  while IFS= read -r line; do
    name=$(awk '{print $1}' <<< "$line")
    stat=$(cut -d' ' -f2- <<< "$line")
    if [[ "$stat" == Up* ]]; then d=$(dot_ok)
    elif [[ "$stat" == Exited* ]]; then d=$(dot_err)
    else d=$(dot_warn); fi
    printf "  %s %-35s %b\n" "$d" "$name" "${DIM}${stat}${R}"
  done <<< "$docker_status"

  printf "\n${DIM} refresh ${REFRESH}s · Ctrl+C to exit${R}\n"
  sleep "$REFRESH"
done

Make it executable:

sudo chmod 755 /usr/local/bin/server-health

---

Create a Sudoers Drop-in File

Because the script runs privileged commands (like sudo fail2ban-client) and requires root access to fetch full system data, you should configure a sudoers drop-in to allow running the script without being prompted for a password.

echo "$(whoami) ALL=(root) NOPASSWD: /usr/local/bin/server-health" | sudo tee /etc/sudoers.d/server-health
sudo chmod 0440 /etc/sudoers.d/server-health

---

Create a Convenience Alias

To make launching the dashboard even easier, add an alias to your shell profile so you can simply type health from anywhere:

echo 'alias health="sudo server-health"' >> ~/.bashrc
source ~/.bashrc

(If you are using Zsh, replace .bashrc with .zshrc in the above commands.)

---

Run the Dashboard

If you set up the alias, simply run:

health

Otherwise, run the script directly:

sudo server-health

Press Ctrl+C to exit.

---

What It Shows

Section	Details
CPU	Current usage %, colour-coded (green/yellow/red)
RAM	Used / Total in GB + percentage
GPU	Temperature (AMD/Intel via sysfs hwmon)
OS disk	Usage on the OS drive (/)
Data disk	Usage on /srv/nextcloud-data
Session	Current SDDM session type or "idle (headless)"
WAN IP	Your current public IP address
Fail2ban	Number of currently banned IPs
NPM	Nginx Proxy Manager reachability
AIO	Nextcloud AIO master container reachability
Docker containers	All running containers with live status