thermal: van-thermal daemon (CPU+NVMe monitor, alert, log)
Single sysfs sample loop drives three jobs off one daemon (modeled on van-failover): - live state at /run/van-thermal/state.json; new Cockpit "Temperatures" card reads it in the existing 5s loop (no per-refresh `sensors` shell-out) - journal alerts on ok<->warn<->crit crossings, with clear_margin hysteresis - throttled, self-rotating CSV history at /var/log/van-thermal.csv Sensors resolved by hwmon name+label (coretemp/Package id 0, nvme/Composite), never by hwmonN index (not stable across boots). Defaults: CPU warn 80 / crit 95, NVMe warn 65 / crit 70. deploy.sh installs + enables the service. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This commit is contained in:
Andreas Wrede
co-authored by
Claude Opus 4.8
Claude Opus 4.8
parent
35466f0141
commit
ef6c1ce182
@@ -0,0 +1,207 @@
|
||||
#!/usr/bin/env python3
|
||||
"""van-thermal — temperature monitor for the campervan router (wayback).
|
||||
|
||||
One small daemon that does three jobs off a single sysfs sample loop:
|
||||
1. publishes /run/van-thermal/state.json (the Cockpit "Temps" card reads this,
|
||||
exactly like van-failover's state.json — no shelling out to `sensors` per refresh)
|
||||
2. warns to the journal on threshold crossings, with hysteresis so a sensor hovering
|
||||
on the line doesn't spam (journalctl -u van-thermal)
|
||||
3. appends throttled CSV history to /var/log/van-thermal.csv with self-rotation
|
||||
|
||||
Sensors are resolved by hwmon *name* + *label* at runtime, never by hwmonN index
|
||||
(that number is assigned at boot and is not stable). Stdlib only.
|
||||
"""
|
||||
|
||||
import json
|
||||
import os
|
||||
import sys
|
||||
import time
|
||||
from datetime import datetime, timezone
|
||||
from pathlib import Path
|
||||
|
||||
CONFIG_PATH = os.environ.get("VAN_THERMAL_CONFIG", "/etc/van-thermal/config.json")
|
||||
STATE_DIR = Path("/run/van-thermal")
|
||||
STATE_PATH = STATE_DIR / "state.json"
|
||||
HWMON = Path("/sys/class/hwmon")
|
||||
|
||||
DEFAULTS = {
|
||||
"sample_interval": 10, # seconds between sysfs reads (alerting cadence)
|
||||
"log_interval": 60, # seconds between CSV rows (history cadence)
|
||||
"log_path": "/var/log/van-thermal.csv",
|
||||
"log_max_bytes": 5 * 1024 * 1024, # rotate to .1 past this, keep one old file
|
||||
"sensors": [
|
||||
{"name": "cpu", "hwmon": "coretemp", "label": "Package id 0",
|
||||
"warn": 85, "crit": 95, "clear_margin": 5},
|
||||
{"name": "nvme", "hwmon": "nvme", "label": "Composite",
|
||||
"warn": 65, "crit": 70, "clear_margin": 5},
|
||||
],
|
||||
}
|
||||
|
||||
LEVELS = ("ok", "warn", "crit")
|
||||
|
||||
|
||||
def log(msg, level="info"):
|
||||
# systemd journal severity prefixes (sd-daemon); shows up via journalctl -p.
|
||||
pri = {"info": "<6>", "warn": "<4>", "crit": "<2>"}.get(level, "<6>")
|
||||
print(pri + msg, flush=True)
|
||||
|
||||
|
||||
def load_config():
|
||||
cfg = dict(DEFAULTS)
|
||||
try:
|
||||
with open(CONFIG_PATH) as f:
|
||||
cfg.update(json.load(f))
|
||||
except FileNotFoundError:
|
||||
log(f"config {CONFIG_PATH} not found, using built-in defaults")
|
||||
except Exception as e:
|
||||
log(f"config {CONFIG_PATH} unreadable ({e}), using built-in defaults", "warn")
|
||||
return cfg
|
||||
|
||||
|
||||
def find_hwmon(name):
|
||||
"""Return the hwmon dir whose name matches, or None. Re-resolved on demand
|
||||
because the hwmonN index can shift across boots / module reloads."""
|
||||
for d in sorted(HWMON.glob("hwmon*")):
|
||||
try:
|
||||
if (d / "name").read_text().strip() == name:
|
||||
return d
|
||||
except OSError:
|
||||
continue
|
||||
return None
|
||||
|
||||
|
||||
def read_temp(spec, cache):
|
||||
"""Read one sensor's temperature in °C, or None if unavailable.
|
||||
|
||||
spec: {hwmon, label}. Resolves hwmon dir + the tempN whose *_label matches,
|
||||
falling back to temp1 when the chip exposes no labels."""
|
||||
d = cache.get(spec["hwmon"])
|
||||
if d is None or not d.exists():
|
||||
d = find_hwmon(spec["hwmon"])
|
||||
cache[spec["hwmon"]] = d
|
||||
if d is None:
|
||||
return None
|
||||
|
||||
input_file = None
|
||||
want = spec.get("label")
|
||||
if want:
|
||||
for lbl in sorted(d.glob("temp*_label")):
|
||||
try:
|
||||
if lbl.read_text().strip() == want:
|
||||
input_file = lbl.with_name(lbl.name.replace("_label", "_input"))
|
||||
break
|
||||
except OSError:
|
||||
continue
|
||||
if input_file is None:
|
||||
input_file = d / "temp1_input" # label not found / chip is label-less
|
||||
|
||||
try:
|
||||
return int(input_file.read_text().strip()) / 1000.0
|
||||
except (OSError, ValueError):
|
||||
# hwmon may have re-enumerated; drop the cache so next pass re-resolves.
|
||||
cache[spec["hwmon"]] = None
|
||||
return None
|
||||
|
||||
|
||||
def classify(temp, spec, prev_level):
|
||||
"""Level with hysteresis: step up at the threshold, step down only after
|
||||
dropping clear_margin below it, so a sensor on the line doesn't oscillate."""
|
||||
if temp is None:
|
||||
return prev_level
|
||||
warn, crit = spec["warn"], spec["crit"]
|
||||
margin = spec.get("clear_margin", 5)
|
||||
if temp >= crit:
|
||||
return "crit"
|
||||
if temp >= warn:
|
||||
return "crit" if prev_level == "crit" and temp > crit - margin else "warn"
|
||||
if temp >= warn - margin:
|
||||
return prev_level if prev_level in ("warn", "crit") else "ok"
|
||||
return "ok"
|
||||
|
||||
|
||||
def announce(name, temp, old, new):
|
||||
if old == new:
|
||||
return
|
||||
rising = LEVELS.index(new) > LEVELS.index(old)
|
||||
sev = {"crit": "crit", "warn": "warn", "ok": "info"}[new]
|
||||
arrow = "rose to" if rising else "fell back to"
|
||||
log(f"{name} {arrow} {new.upper()} ({temp:.1f}°C)", sev if rising else "info")
|
||||
|
||||
|
||||
def rotate_log(path, max_bytes):
|
||||
try:
|
||||
if path.exists() and path.stat().st_size > max_bytes:
|
||||
path.replace(path.with_suffix(path.suffix + ".1"))
|
||||
except OSError as e:
|
||||
log(f"log rotate failed: {e}", "warn")
|
||||
|
||||
|
||||
def write_csv(cfg, readings):
|
||||
path = Path(cfg["log_path"])
|
||||
rotate_log(path, cfg["log_max_bytes"])
|
||||
new = not path.exists()
|
||||
try:
|
||||
path.parent.mkdir(parents=True, exist_ok=True)
|
||||
with open(path, "a") as f:
|
||||
if new:
|
||||
f.write(",".join(["time"] + sum(
|
||||
[[f"{s['name']}_c", f"{s['name']}_lvl"] for s in cfg["sensors"]], [])) + "\n")
|
||||
row = [datetime.now(timezone.utc).isoformat(timespec="seconds")]
|
||||
for s in cfg["sensors"]:
|
||||
r = readings[s["name"]]
|
||||
row += ["" if r["temp"] is None else f"{r['temp']:.1f}", r["level"]]
|
||||
f.write(",".join(row) + "\n")
|
||||
except OSError as e:
|
||||
log(f"csv write failed: {e}", "warn")
|
||||
|
||||
|
||||
def write_state(readings):
|
||||
STATE_DIR.mkdir(parents=True, exist_ok=True)
|
||||
payload = {
|
||||
"updated": datetime.now(timezone.utc).isoformat(timespec="seconds"),
|
||||
"sensors": readings,
|
||||
}
|
||||
tmp = STATE_PATH.with_suffix(".tmp")
|
||||
tmp.write_text(json.dumps(payload))
|
||||
tmp.replace(STATE_PATH) # atomic swap so the Cockpit reader never sees a partial file
|
||||
|
||||
|
||||
def main():
|
||||
cfg = load_config()
|
||||
cache = {}
|
||||
levels = {s["name"]: "ok" for s in cfg["sensors"]}
|
||||
last_log = 0.0
|
||||
log(f"van-thermal up: sampling every {cfg['sample_interval']}s, "
|
||||
f"logging every {cfg['log_interval']}s to {cfg['log_path']}")
|
||||
|
||||
while True:
|
||||
readings = {}
|
||||
for s in cfg["sensors"]:
|
||||
temp = read_temp(s, cache)
|
||||
new = classify(temp, s, levels[s["name"]])
|
||||
if temp is not None:
|
||||
announce(s["name"], temp, levels[s["name"]], new)
|
||||
levels[s["name"]] = new
|
||||
readings[s["name"]] = {
|
||||
"temp": None if temp is None else round(temp, 1),
|
||||
"level": new, "warn": s["warn"], "crit": s["crit"],
|
||||
}
|
||||
|
||||
try:
|
||||
write_state(readings)
|
||||
except OSError as e:
|
||||
log(f"state write failed: {e}", "warn")
|
||||
|
||||
now = time.monotonic()
|
||||
if now - last_log >= cfg["log_interval"]:
|
||||
write_csv(cfg, readings)
|
||||
last_log = now
|
||||
|
||||
time.sleep(cfg["sample_interval"])
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
try:
|
||||
main()
|
||||
except KeyboardInterrupt:
|
||||
sys.exit(0)
|
||||
Reference in New Issue
Block a user