Hello,
I configured NUT adding to read my UPS data via USB. It looks to work fine because I can see some new entities in HA but I cannot understand what entity should I use to verify if my UPS is actively supporting the power.
I can see these sensors:
sensor.zinto1000_battery_charge 100 state: Online
unit_of_measurement: %
friendly_name: Zinto1000 Battery Charge
device_class: battery
sensor.zinto1000_input_voltage 229.1 state: Online
unit_of_measurement: V
friendly_name: Zinto1000 Input Voltage
icon: mdi:flash
sensor.zinto1000_load 15 state: Online
unit_of_measurement: %
friendly_name: Zinto1000 Load
icon: mdi:gauge
sensor.zinto1000_status Online state: Online
unit_of_measurement:
friendly_name: Zinto1000 Status
icon: mdi:information-outline
sensor.zinto1000_status_data OL state: Online
unit_of_measurement:
friendly_name: Zinto1000 Status Data
icon: mdi:information-outline
sensor.zinto1000_ups_type offline / line interactive state: Online
unit_of_measurement:
friendly_name: Zinto1000 UPS Type
icon: mdi:information-outline
Pull the plug and watch the changes. This was how I found the various state changes in NUT config. Note, I send NUT via MQTT and not directly into HA, so my sensor names and attributes are non-standard. But I think the state info is same as you will see in direct NUT configuration:
- alias: 'Home power outage'
trigger:
- platform: state
entity_id: sensor.ups_status
to: 'OB DISCHRG'
action:
service: notify.pushover
data:
message: "Power outage detected by UPS: {{ (state_attr('sensor.ups_status', 'battery_runtime') | float / 60) | round(0) }} minutes of battery power remaining."
data:
priority: 1
sound: gamelan
- alias: 'Home power recharging'
trigger:
- platform: state
entity_id: sensor.ups_status
to: 'OL CHRG'
from: 'OB DISCHRG'
action:
service: notify.pushover
data:
message: 'Power restored detected by UPS, recharging'
data:
priority: 1
sound: gamelan
- alias: 'Home power restored no recharging'
trigger:
- platform: state
entity_id: sensor.ups_status
to: 'OL'
from: 'OB DISCHRG'
action:
service: notify.pushover
data:
message: 'Power restored detected by UPS, no recharging done'
data:
priority: 1
sound: gamelan
- alias: 'Home power restored recharging started'
trigger:
- platform: state
entity_id: sensor.ups_status
to: 'OL CHRG'
from: 'OL DISCHRG'
action:
service: notify.pushover
data:
message: 'Power restored detected by UPS, recharging started'
data:
priority: 1
sound: gamelan
- alias: 'Home power restored after recharge'
trigger:
- platform: state
entity_id: sensor.ups_status
to: 'OL'
from: 'OL CHRG'
action:
service: notify.pushover
data:
message: 'UPS recharged'
data:
priority: 1
sound: gamelan
Yes, I have a python script that loops every 30 seconds, executes upsmon, formats the output as JSON and publishes to MQTT.
I was having communications timeout issues with NUT and CyberPower UPS, so I went down this route so I could debug that issue. Never changed it. I sort of like using MQTT for my sensors general anyway.
Code is pretty ugly, apologies. This python code is running on a raspberry pi zero plus that is USB attached to cyberpower UPS. On boot launches a tmux session that waits for network to be ready and then runs the python script that stays in loop querying NUT and publishing status to MQTT.
----- code on raspberry pi to launch tmux session running python code on boot
Notes: had to change first line, do the wait for network to startup and run tmux as pi not root
sudo chmod +x /etc/rc.local
sudo -s
vi /etc/rc.local
#!/bin/bash
#
#######!/bin/sh -e
#
# rc.local
#
# This script is executed at the end of each multiuser runlevel.
# Make sure that the script will "exit 0" on success or any other
# value on error.
#
# In order to enable or disable this script just change the execution
# bits.
#
# By default this script does nothing.
# Print the IP address
#_IP=$(hostname -I) || true
#if [ "$_IP" ]; then
# printf "My IP address is %s\n" "$_IP"
#fi
# wait for networks to start
STATE="error";
while [ $STATE == "error" ]; do
#do a ping and check that its not a default message or change to grep for something else
STATE=$(ping -q -w 1 -c 1 `ip r | grep default | cut -d ' ' -f 3` > /dev/null && echo ok || echo error)
#sleep for 2 seconds and try again
sleep 2
done
su pi -c 'tmux new-session -d -s watts'
su pi -c 'tmux send-keys \'/home/pi/wattsup/wattsup_script_v01.py' C-m'
su pi -c 'tmux new-session -d -s ups'
su pi -c 'tmux send-keys cd\ /home/pi C-m'
su pi -c 'tmux send-keys /home/pi/upsmon/upsmon_script_v01.py C-m'
exit 0
----- in HA, this is the MQTT sensor
on home assistant
sensors.yaml
# nut ups monitor
- platform: mqtt
name: "UPS Status"
state_topic: "upsmon/status"
value_template: "{{ value_json.ups_status }}"
json_attributes:
- "timestamp"
- "input_voltage"
- "ups_status"
- "battery_runtime"
- "battery_runtime_low"
- "battery_voltage"
- "output_voltage"
- "battery_charge"
- "ups_load"
- "ups_test_result"
- "ups_model"
- "ups_vendorid"
- "ups_serial"
----- # python code that loops reading NUT upsc, converts to JSON and publishes to MQTT
#!/usr/bin/env python
# upsmon_script_v01.py
# 2019-09-16
#
# using mqtt example from here: https://techtutorialsx.com/2017/04/14/python-publishing-messages-to-mqtt-topic/
import sys
import os
import subprocess
import datetime
import time
import re
import paho.mqtt.client as mqttClient
import logging
import logging.handlers
import csv
import json
import pprint
####################
# mqtt server IP address
mqttAddress = "192.168.2.13"
# mqtt topics
upsmon_topic = "upsmon/status"
# global flag to indicate connection to mqtt server
Connected = False
####################
# Logging setup
LOG_FILENAME = 'upsmon.log'
# Set up a specific logger with our desired output level
my_logger = logging.getLogger('upsmon')
my_logger.setLevel(logging.INFO)
# setup log format
formatter = logging.Formatter(fmt='%(asctime)s %(levelname)-8s %(message)s',
datefmt='%Y-%m-%d %H:%M:%S')
# Add the log message handler to the logger
handler = logging.handlers.RotatingFileHandler(
LOG_FILENAME, backupCount=5)
handler.setFormatter(formatter)
my_logger.addHandler(handler)
# Roll over on application start
my_logger.handlers[0].doRollover()
def onConnect(client, userdata, flags, rc):
global Connected
# connect to mqtt server
if rc == 0:
Connected = True
else:
Connected = False
def main():
global Connected
my_logger.info("startup")
# set up connection to mqtt server
client = mqttClient.Client("media-pi-01-upsmon")
client.onConnect = onConnect
client.connect(mqttAddress)
client.loop_start()
# pause to connection to complete
time.sleep(0.5)
my_logger.info("connected to mqtt server")
try:
while True:
# print("=========================== start")
retrieve_time = datetime.datetime.now().strftime("%Y%m%d%H%M%S")
# get output from nut upsc command
p = subprocess.Popen('upsc cyberpower1@localhost', stdout=subprocess.PIPE, stderr=subprocess.PIPE, shell=True)
rawline = p.communicate()
# print(type(rawline))
# pprint.pprint(rawline)
my_logger.debug("At : " + retrieve_time + ", Received string from shell: [[" + rawline[0] + "]]")
if len(rawline[0]) > 0:
try:
# check if NUT is not communicating with UPS
if rawline[0].find("battery") == -1 :
dct_string = "{\"timestamp\": " + retrieve_time + ", \"ups_status\": \"OFFLINE\"}"
client.publish(upsmon_topic, dct_string, 1)
my_logger.error("NUT not communicating with UPS: " + dct_string + ", rawline = " + rawline[0])
time.sleep(15.0)
continue
vals = rawline[0].split("\n")
dct = {}
for x in vals:
if len(x) == 0 :
continue
if x.find(":") == False :
continue
if x.find("driver") <> -1 :
continue
if x.find("device.type") <> -1 :
continue
if x.find("battery.type") <> -1 :
continue
if x.find("ups.timer") <> -1 :
continue
if x.find("ups.delay") <> -1 :
continue
if x.find("battery.charge.low") <> -1 :
continue
# if x.find(".low") <> -1 :
# continue
if x.find(".beeper") <> -1 :
continue
if x.find(".mfr.date") <> -1 :
continue
if x.find("battery.charge.warning") <> -1 :
continue
if x.find("device.mfr") <> -1 :
continue
if x.find("device.model") <> -1 :
continue
if x.find("device.serial") <> -1 :
continue
if x.find("ups.productid") <> -1 :
continue
if x.find("battery.mfr.date") <> -1 :
continue
if x.find("battery.voltage.nominal") <> -1 :
continue
if x.find("ups.realpower.nominal") <> -1 :
continue
if x.find("input.voltage.nominal") <> -1 :
continue
# if x.find("ups.serial") <> -1 :
# continue
if x.find("ups.mfr") <> -1 :
continue
# if x.find("ups.test.result") <> -1 :
# continue
# if x.find("ups.model") <> -1 :
# continue
tup = x.split(":")
# bush league code to strip spaces from the two elements
tup = [tup[0].strip().replace(".", "_"), tup[1].strip()]
#since list has only two items, use this trick to convert the list to a dictionary
dct.update(dict([tup]))
# print x
dct.update({'timestamp': retrieve_time})
dct_string = str(dct).replace("'","\"")
# print(dct_string)
# print("----- publishing to mqtt server")
client.publish(upsmon_topic, dct_string, 1)
except Exception as e:
my_logger.error("Exception triggered: " + str(e))
# print("=========================== end")
time.sleep(15.0)
except KeyboardInterrupt:
client.disconnect()
client.loop_stop()
my_logger.info("keyboard interrupt")
if __name__ == '__main__':
main()
-----
Thanks! NUT is one of the really amazing open source projects. You just sit there and say âthank you!â and âwow!â to the amazing folks that created it and maintain it.
Back to my original answer to your first query. Since I rely on HAâs events and the value that NUT publishes as the UPS status, âOL CHRGâ, âOL DISCHRGâ, âOLâ ⌠etc⌠If you do the same, I do recommend you test your UPS and the status changes that NUT passes along with various changes in state by the UPS. I am not sure, but I suspect that if you go the same route, the state change sequences and status string returned can be UPS specific. In the case of my CyberPower units, until I looked in detail during tests, I was missing some of the routes the UPS went from âon batteryâ back to âfully charged and back on mainsâ.
Just a question: If I am not mistaken the status changes not immediately in HA. Is there any parameter to set the read interval fo the values in NUT or HA?
The pollinterval in UPS.CONF controls the rate at which NUT queries the UPS, this is where I had my problem with my CyberPower units, the default value of 2 seconds caused my UPS to quit talking to NUT over the USB connection. I ended up with a value of 15 seconds and it has been solid for over a year of continuous operation. I did not try to tune it down to a smaller value, so I do not know what the minimum is for the CyberPower:
[cyberpower1]
driver = usbhid-ups
port = auto
desc = "CyberPower BRG850AVRLCD"
pollinterval = 15
Since I am doing a rather inefficient polling loop of 15 seconds in my python code that polls NUTâs upsc utility and publishes to MQTT, Home Assistant will see the changes somewhere between 15 and 30 seconds after an event occurs at the UPS. This was what I saw during my testing.
If you use the direct HA to NUT integration, I looked at the HA documentation and I do not see a mention of the default sensor scan interval for NUT, nor do I see that it supports the âscan_interval:â configuration key. Again I recommend doing testing with your configuration. And you can also look in GITHUB for the HA code for NUT and see if you can figure out itâs scanning interval.
and watch the changes. This was how I found the various state changes in NUT config. Note, I send NUT via MQTT and not directly into HA, so my sensor names and attributes are non-standard. But I think the state info is same as you will see in direct NUT configuration: