# statuspage example configuration
#
# Everything in this file is commented out on purpose. Uncomment (and adjust)
# the bits you need. See the README for the full reference.
#
# Save a copy as config.local.yaml and start the server with:
#
#     ./bin/statuspage -c config.local.yaml
#
# or point Docker at it (see docker-compose.yml).

# title: "My machines"            # shown in the header and browser tab
#                                  # default: "Status Page"

# interactive: true                # allow on-demand probes via the refresh
#                                  # button on each machine card
#                                  # default: false

# shared_metric_window: false      # align all metric plot x-axes to the same
#                                  # window (the last ~24h) so shapes can be
#                                  # compared directly
#                                  # default: false

# ---------- global ping defaults ----------
# ping:
#   interval: 10s                  # how often to probe        default: 10s
#   timeout: 2s                    # per-probe timeout         default: 2s
#   tcp_port: 22                   # port to TCP-connect to    default: 22
#                                  # (only IPv4 hosts are probed)

# ---------- global ssh defaults ----------
# ssh:
#   interval: 10m                  # how often to run the script   default: 10m
#   user: root                     # ssh user                      default: root
#   port: 22                       # ssh port                      default: 22
#   key: |                         # PEM private key (the only supported auth)
#     -----BEGIN OPENSSH PRIVATE KEY-----
#     ...
#     -----END OPENSSH PRIVATE KEY-----
#   script: |                      # shell script to run over ssh
#     # status chips (name:status:info)
#     echo "hostname:on:$(hostname)"
#     echo "kernel:on:$(uname -sr)"
#     echo "vulkan:on:$(vulkaninfo --summary 2>/dev/null | sed -n '/deviceName/p' | head -1 | cut -d: -f2-)"
#     echo "docker:$(systemctl is-active docker 2>/dev/null | sed 's/active/on/; s/inactive/off/; s/failed/down/'):docker daemon"
#     echo "nfs:$(systemctl is-active nfs-server 2>/dev/null | sed 's/active/on/; s/inactive/off/; s/failed/down/'):nfs-server"
#
#     # metrics (name:metric:value)
#     echo "cpu_pct:metric:$(top -bn1 2>/dev/null | awk '/%Cpu/ {printf "%.1f", 100-$8}')"
#     echo "load1:metric:$(cut -d' ' -f1 /proc/loadavg)"
#     echo "ram_pct:metric:$(free -m 2>/dev/null | awk '/^Mem:/ {printf "%.1f", $3/$2*100}')"
#     echo "ram_used:metric:$(free -m 2>/dev/null | awk '/^Mem:/ {print $3}')MB"
#     echo "disk_root:metric:$(df -h / 2>/dev/null | awk 'NR==2 {print $5}' | tr -d '%')"
#     echo "boot_days:metric:$(awk '{printf "%.2f", $1/86400}' /proc/uptime)"
#     echo "cpu_temp:metric:$(sensors 2>/dev/null | awk '/Package id 0:/ {print $4; exit}' | tr -d '+°C')"
#     echo "gpu_temp:metric:$(nvidia-smi --query-gpu=temperature.gpu --format=csv,noheader 2>/dev/null | tr -d ' ')"
#     echo "gpu_util:metric:$(nvidia-smi --query-gpu=utilization.gpu --format=csv,noheader 2>/dev/null | tr -d ' %')"
#
#     # free-form section, shown raw in the machine modal
#     echo "---"
#     echo "uptime: $(uptime -p)"
#     echo "load: $(cat /proc/loadavg)"
#     echo "mem: $(free -h | awk '/^Mem:/ {print $3 "/" $2}')"
#     echo "disk: $(df -h / | awk 'NR==2 {print $4 " free"}')"
#     echo "top: $(ps --no-headers -eo %cpu,comm --sort=-%cpu | head -4 | tr '\n' ';')"
#
#   Script format, one line per item:
#     name:status:info    -> a check chip (status: on / off / down / anything)
#     name:metric:value   -> a numeric metric, accumulated into the history db
#     ---                 -> everything after the first such line is shown raw
#
#   If key or script is empty, the ssh check is skipped for that machine.

# ---------- suggested y-axis bounds for metric plots ----------
# metrics:
#   cpu_pct:
#     min: 0                       # soft bounds: the plot only extends the
#     max: 100                     # axis if the data goes outside
#   ram_pct:
#     min: 0
#     max: 100
#   disk_root:
#     min: 0
#     max: 100
#   gpu_util:
#     min: 0
#     max: 100
#   gpu_temp:
#     min: 0
#     max: 105
#   boot_days:
#     min: 0

# ---------- machines grouped by section ----------
# groups:
#   - name: Aula 3
#     machines:
#       - host: a3-dott1.example.net
#       - host: a3-dott2.example.net
#         name: dottorandi 2      # optional display name (defaults to host)
#         ping:                    # per-machine overrides, inherit the rest
#           interval: 30s
#           tcp_port: 22
#         ssh:
#           interval: 1m
#           user: root
#
#   # host patterns are expanded at load time. Inside a {...} group:
#   #   {1,2,3}      enumeration            -> 1, 2, 3
#   #   {2..5,8..10} inclusive numeric range -> 2, 3, 4, 5, 8, 9, 10
#   #   {a3,a4}      labels (any non-range item is kept literally)
#   # several groups multiply in cartesian fashion:
#   #   server-{a3,a4}-{1..2}.example.org
#   #     -> server-a3-1.example.org, server-a3-2.example.org,
#   #        server-a4-1.example.org, server-a4-2.example.org
#   # leading zeros are preserved: {01..03} -> 01, 02, 03
#   # a name: applies to every expanded machine.
#   #
#   # Bad patterns fail at startup: unmatched/nested braces, ranges with
#   # non-integer endpoints or extra dots (e.g. {1..5..10}), and descending
#   # ranges. A single pattern may expand to at most 10000 machines.
#   - name: Server room
#     machines:
#       - host: server-{a3,a4}-{1..10}.example.org
#         ssh:
#           interval: 1m

# ---------- machines without a group (shown under "Machines") ----------
# machines:
#   - host: router.example.net
#     name: router
#     ping:
#       interval: 5s
#       tcp_port: 80