mcsMQTT and RPi 1-Wire hub

[Pete]

Finished the basement Rpi 1Wire hub. Note here it is just an RPi - sensors on the work bench.

This will be covering main floor, basement and one temperature sensor outside.

[ATTACH]67932[/ATTACH]

[ATTACH]67933[/ATTACH]

[mloebl]

Nice! Looking good.

I have sensor envy; I only have one basement temperature sensor.

[Pete]

Got your python script working today.

I had edited it a bit and mtqq wasn't running on the attic RPi.

It appears to be running but not seeing any numbers in the log file.

2018-04-05 21:12:23,971 Starting mqtt-owfs-temp
2018-04-05 21:12:23,972 INFO MODE
2018-04-05 21:12:23,973 DEBUG MODE
2018-04-05 21:12:23,978 Connecting to 127.0.0.1:1883
2018-04-05 21:12:23,983 DeviceList.data is : [['# Server', ' port', ' sensor path'], ['192.168.244.169', ' 4304', ' /28.CFD8F4020000F5'], ['192.168.244.169', ' 4304', ' /26.8CC814000000DC'], ['192.168.244.169', ' 4304', ' /26.993AF50000008F'], ['192.168.244.169', ' 4304', ' /26.993AF50000008F']]
2018-04-05 21:12:23,984 Querying sensor path on # Server: port
2018-04-05 21:12:39,003 Received CONNACK (0, 0)
2018-04-05 21:17:33,055 Starting mqtt-owfs-temp
2018-04-05 21:17:33,056 INFO MODE
2018-04-05 21:17:33,057 DEBUG MODE
2018-04-05 21:17:33,061 Connecting to 127.0.0.1:1883
2018-04-05 21:17:33,067 DeviceList.data is : [['# Server', ' port', ' sensor path'], ['192.168.244.169', ' 4304', ' /28.CFD8F4020000'], ['192.168.244.169', ' 4304', ' /26.8CC814000000'], ['192.168.244.169', ' 4304', ' /26.993AF5000000'], ['192.168.244.169', ' 4304', ' /26.993AF5000000']]
2018-04-05 21:17:33,068 Querying sensor path on # Server: port

Yeah set up a test RPi in the basement with a few sensors. Connected one to the main floor hallway and moved one basement sensor to the middle of the basement.

[mloebl]

Got your python script working today.

I had edited it a bit and mtqq wasn't running on the attic RPi.

It appears to be running but not seeing any numbers in the log file.

2018-04-05 21:12:23,971 Starting mqtt-owfs-temp
2018-04-05 21:12:23,972 INFO MODE
2018-04-05 21:12:23,973 DEBUG MODE
2018-04-05 21:12:23,978 Connecting to 127.0.0.1:1883
2018-04-05 21:12:23,983 DeviceList.data is : [['# Server', ' port', ' sensor path'], ['192.168.244.169', ' 4304', ' /28.CFD8F4020000F5'], ['192.168.244.169', ' 4304', ' /26.8CC814000000DC'], ['192.168.244.169', ' 4304', ' /26.993AF50000008F'], ['192.168.244.169', ' 4304', ' /26.993AF50000008F']]
2018-04-05 21:12:23,984 Querying sensor path on # Server: port
2018-04-05 21:12:39,003 Received CONNACK (0, 0)
2018-04-05 21:17:33,055 Starting mqtt-owfs-temp
2018-04-05 21:17:33,056 INFO MODE
2018-04-05 21:17:33,057 DEBUG MODE
2018-04-05 21:17:33,061 Connecting to 127.0.0.1:1883
2018-04-05 21:17:33,067 DeviceList.data is : [['# Server', ' port', ' sensor path'], ['192.168.244.169', ' 4304', ' /28.CFD8F4020000'], ['192.168.244.169', ' 4304', ' /26.8CC814000000'], ['192.168.244.169', ' 4304', ' /26.993AF5000000'], ['192.168.244.169', ' 4304', ' /26.993AF5000000']]
2018-04-05 21:17:33,068 Querying sensor path on # Server: port

Yeah set up a test RPi in the basement with a few sensors. Connected one to the main floor hallway and moved one basement sensor to the middle of the basement.

Are all the devices behind a hub by any chance? If so, I believe the CSV needs to look like:
HUBPORTID/SENSORID instead of just the sensor ID. For example I believe it looks like:

localhost,4304,22.344245501444/10.9D524D010800,temperature

My hub died so I had to go straight thru, but I believe because how OWFS addresses the sensors, you need the path. If that doesn't work, let me know for sure.

[Pete]

Are all the devices behind a hub by any chance?

Using Maxim 9097 serial devices here.

Just looked here and did not put in temperature.

# Server, port, sensor path
192.168.244.169, 4304, /28.CFD8F4020000
192.168.244.169, 4304, /26.8CC814000000
192.168.244.169, 4304, /26.993AF5000000
192.168.244.169, 4304, /26.993AF5000000

will change it to this:

# Server, port, sensor path
192.168.244.169, 4304, /28.CFD8F4020000, temperature
192.168.244.169, 4304, /26.8CC814000000, temperature
192.168.244.169, 4304, /26.993AF5000000, temperature
192.168.244.169, 4304, /26.993AF5000000, temperature

[ATTACH]67949[/ATTACH]

Digitemp view: (just noticed that all but one are combo sensors)

ROM #0 : 28CFD8F4020000F5
ROM #1 : 268CC814000000DC
ROM #2 : 26993AF50000008F
ROM #3 : 26B3CA1400000096
ROM #4 : 263B3AF5000000BD
ROM #5 : 1D37C907000000FC
Apr 06 01:08:59 Sensor 0 C: 11.44 F: 52.59
Apr 06 01:09:09 Sensor 1 C: 9.25 F: 48.65 H: 65%
Apr 06 01:09:13 Sensor 2 C: 18.94 F: 66.09 H: 49%
Apr 06 01:09:22 Sensor 3 C: 18.78 F: 65.81 H: 50%
Apr 06 01:09:27 Sensor 4 C: 19.62 F: 67.32 H: 46%

[mloebl]

You can also take out the leading slash (I never tested with that), so it can just be:

# Server, port, sensor path
192.168.244.169,4304,28.CFD8F4020000,temperature
192.168.244.169,4304,26.8CC814000000,temperature
192.168.244.169,4304,26.993AF5000000,temperature
192.168.244.169,4304,26.993AF5000000,temperature

The reason I added the season type to the end was so it knew how to process the device type. However I may be able to add auto-detection in the future.

[Pete]

Wondering if this is related to updating from Jessie to Stretch. I see the OWFS 2121 directory but get an odd error running the script.

Testing using the old "in production" one wire network in the attic.

ZNet:/usr/local/mqtt-owfs-temp# python mqtt*.py
Traceback (most recent call last):
File "mqtt-owfs-temp.py", line 289, in <module>
main_loop()
File "mqtt-owfs-temp.py", line 255, in main_loop
ow.init(owserver + ":" + owport)
File "/usr/lib/python2.7/dist-packages/ow/__init__.py", line 224, in init
raise exNoController

Noticed that fuse didn't start in the systemd list. Started it and ran the python script again. Same error as above.

ZNet:/usr/local/mqtt-owfs-temp# ls /mnt/1wire
1D.37C907000000 26.993AF5000000 alarm simultaneous system
26.3B3AF5000000 26.B3CA14000000 bus.0 statistics uncached
26.8CC814000000 28.CFD8F4020000 settings structure

The above only has one temperature only device and rest are combo temperature and humidity sensors. I am going to change the csv file to the temperature sensor only and see what happens.

No change:

ICS-ZNet:/usr/local/mqtt-owfs-temp# python mqtt*.py
Traceback (most recent call last):
File "mqtt-owfs-temp.py", line 289, in <module>
main_loop()
File "mqtt-owfs-temp.py", line 255, in main_loop
ow.init(owserver + ":" + owport)
File "/usr/lib/python2.7/dist-packages/ow/__init__.py", line 224, in init
raise exNoController
ow.exNoController

Looked at your RPi OWFS stuff...my configuration is similiar for owfs.

ICS-ZNet:/usr/local/mqtt-owfs-temp# owread -s 192.168.244.169:4304 /28.CFD8F4020000/temperature

Will try using two forward slashes.

Changed the read to a loopback address and it worked (except noticed that temperature is C).

ZNet:/usr/local/mqtt-owfs-temp# owread -s 127.0.0.1:4304 /28.CFD8F4020000/temperature
8.9375

Went looking at my owfs.

####################### OWSERVER ########################
server: port = 127.0.0.1:4304

For node red I had to change the port to a loopback address as it did not connect with the IP address of the OWFS server.

A side rant here...recently noticed that my Samba shares authentication stuff has changed recently with upgrading Ubuntu and Rpi Jessie to Stretch. This most noticeable on my laptops and using oldest NAS box. It should not have anything to do with OWFS though. Switched my Squeezebox server shares to using NFS instead of Samba last week. Windows whatever OS to samba shares on NAS work fine. It is just Linux.

[mloebl]

I'll stage up a Raspbian to try out. Both mine are running Ubuntu Server, so *possible* there's a subtle difference.

OH, also noticed you are running from a test dir in usr/local/mqtt-owfs-temp . It was designed to be run as a service in a specific location, so bet it's not seeing the config files.

Open up the mqtt-owfs-temp.py and change line 24 to match the path:

config.read("/etc/mqtt-owfs-temp/mqtt-owfs-temp.cfg")

Also make sure the mqtt-owfs-temp.cfg has the right path to the csv as well.

[Pete]

Thank you Mike.

Yeah here running now with RPi Stretch on all of the RPis. That said though for Homeseer running on an Intel iSeries and ARM Pine64 Ubuntu 16.04 64 bit plus utilize Oracle Virtual box for some Windows only stuff (well like Microsoft SAPI).

edited the file /usr/local/mqtt-owfs-temp/mqtt-owfs-temp.py as you have mentioned above.

Line 23 already read as above. IE: the path for the configuration file is at:
/etc/mqtt-owfs-temp/mqtt-owfs-temp.cfg and that is where the configuration file is at.

# Read the config file
config = ConfigParser.RawConfigParser()
config.read("/etc/mqtt-owfs-temp/mqtt-owfs-temp.cfg")

cfg file shows ...so it is OK...

[global]
DEBUG = True
LOGFILE = /var/log/mqtt-owfs-temp.log
MQTT_HOST = 127.0.0.1
MQTT_PORT = 1883
MQTT_TOPIC = /#
MQTT_SUBTOPIC = /1-wire/
POLLINTERVAL = 60
DEVICESFILE = /etc/mqtt-owfs-temp/devices.csv

I am guessing that the node red and your python program should coexist OK; eh?

Rather than running it twice will just do the systemd restart thing...

I left it running with only one 1-wire sensor on while tinkering with the other Pi in the basement.

For the original OWFS installation here followed the Ubuntu DIY for the RPi rather than the RPi one out there. Thinking they are the same anyhow...

For your application followed your step by step on the git hub link you mentioned above.

Killed my basement 1-wire test network yesterday extending the two Temperature / Humidity sensors. I left them in parasitic mode and they worked fine with 5" leads in parasitic mode. Then extended them to 20 - 30 feet or so and one of the two took the network down. (one was a HB and the other was a Midon one). The AAG combos work the best and very tiny combos...maybe 1/2" in diameter.

restarted /started service...same as before...cannot find the 1wire bus.

mqtt-owfs-temp.service - LSB: mqtt-owfs-temp daemon start/stop script
Loaded: loaded (/etc/init.d/mqtt-owfs-temp; generated; vendor preset: enabled)
Active: active (exited) since Fri 2018-04-06 11:51:25 UTC; 10h ago
Docs: man:systemd-sysv-generator(8)
Process: 586 ExecStart=/etc/init.d/mqtt-owfs-temp start (code=exited, status=0/SUCCESS)
CGroup: /system.slice/mqtt-owfs-temp.service

Apr 06 11:51:25 ICS-ZNet systemd[1]: Starting LSB: mqtt-owfs-temp daemon start/stop script...
Apr 06 11:51:25 ICS-ZNet mqtt-owfs-temp[586]: Starting mqtt-owfs-temp
Apr 06 11:51:25 ICS-ZNet systemd[1]: Started LSB: mqtt-owfs-temp daemon start/stop script.
Apr 06 11:51:53 ICS-ZNet OWFS[639]: DEFAULT: owlib.c52) No valid 1-wire buses found

ICS-ZNet:/mnt# ls 1wire
1D.37C907000000
26.993AF5000000
alarm
simultaneous
system
26.3B3AF5000000
26.B3CA14000000
bus.0
statistics
uncached
26.8CC814000000
28.CFD8F4020000
settings
structure

I know it is there cuz Node Red is using it for one sensor. It is 1 temperature sensor and 4 combo sensors (AAG ones).

[Pete]

This morning restarted the daemon...

mqtt-owfs-temp.service - LSB: mqtt-owfs-temp daemon start/stop script
Loaded: loaded (/etc/init.d/mqtt-owfs-temp; generated; vendor preset: enabled)
Active: active (exited) since Sat 2018-04-07 11:12:58 UTC; 13s ago
Docs: man:systemd-sysv-generator(8)
Process: 6466 ExecStop=/etc/init.d/mqtt-owfs-temp stop (code=exited, status=0/SUCCESS)
Process: 6482 ExecStart=/etc/init.d/mqtt-owfs-temp start (code=exited, status=0/SUCCESS)

Apr 07 11:12:58 ICS-ZNet systemd[1]: Starting LSB: mqtt-owfs-temp daemon start/stop script...
Apr 07 11:12:58 ICS-ZNet mqtt-owfs-temp[6482]: Starting mqtt-owfs-temp
Apr 07 11:12:58 ICS-ZNet systemd[1]: Started LSB: mqtt-owfs-temp daemon start/stop script.

debug mqtt-owfs-temp.log

2018-04-07 11:12:59,389 Starting mqtt-owfs-temp
2018-04-07 11:12:59,391 INFO MODE
2018-04-07 11:12:59,391 DEBUG MODE
2018-04-07 11:12:59,396 Connecting to 192.168.244.169:1883

restarted mosquitto

● mosquitto.service - LSB: mosquitto MQTT v3.1 message broker
Loaded: loaded (/etc/init.d/mosquitto; generated; vendor preset: enabled)
Active: active (running) since Sat 2018-04-07 11:19:35 UTC; 12s ago
Docs: man:systemd-sysv-generator(8)
Process: 7682 ExecStart=/etc/init.d/mosquitto start (code=exited, status=0/SUCCESS)
CGroup: /system.slice/mosquitto.service
└─7688 /usr/sbin/mosquitto -c /etc/mosquitto/mosquitto.conf

Apr 07 11:19:35 ICS-ZNet systemd[1]: Starting LSB: mosquitto MQTT v3.1 message broker...
Apr 07 11:19:35 ICS-ZNet mosquitto[7682]: Starting network daemon:: mosquitto.
Apr 07 11:19:35 ICS-ZNet systemd[1]: Started LSB: mosquitto MQTT v3.1 message broker.

mosquitto.log shows:

1523099976: mosquitto version 1.4.14 (build date Mon, 10 Jul 2017 23:48:43 +0100) starting
1523099976: Config loaded from /etc/mosquitto/mosquitto.conf.
1523099976: Opening ipv4 listen socket on port 1883.
1523099976: Opening ipv6 listen socket on port 1883.
1523099979: New connection from 192.168.244.171 on port 1883.
1523099979: New client connected from 192.168.244.171 as d43887d6-d14d-4de8-bb70-6cf0064f7b3f (c1, k60).
1523099987: New connection from 127.0.0.1 on port 1883.

Note that 171 is Homeseer Pro

Restarted python script.

● mqtt-owfs-temp.service - LSB: mqtt-owfs-temp daemon start/stop script
Loaded: loaded (/etc/init.d/mqtt-owfs-temp; generated; vendor preset: enabled)
Active: active (running) since Sat 2018-04-07 11:23:53 UTC; 9s ago
Docs: man:systemd-sysv-generator(8)
Process: 8805 ExecStop=/etc/init.d/mqtt-owfs-temp stop (code=exited, status=0/SUCCESS)
Process: 8818 ExecStart=/etc/init.d/mqtt-owfs-temp start (code=exited, status=0/SUCCESS)
CGroup: /system.slice/mqtt-owfs-temp.service
└─8827 mqtt-owfs-temp

Apr 07 11:23:53 ICS-ZNet systemd[1]: Starting LSB: mqtt-owfs-temp daemon start/stop script...
Apr 07 11:23:53 ICS-ZNet mqtt-owfs-temp[8818]: Starting mqtt-owfs-temp
Apr 07 11:23:53 ICS-ZNet systemd[1]: Started LSB: mqtt-owfs-temp daemon start/stop script.

1523099976: mosquitto version 1.4.14 (build date Mon, 10 Jul 2017 23:48:43 +0100) starting
1523099976: Config loaded from /etc/mosquitto/mosquitto.conf.
1523099976: Opening ipv4 listen socket on port 1883.
1523099976: Opening ipv6 listen socket on port 1883.
1523099979: New connection from 192.168.244.171 on port 1883.
1523099979: New client connected from 192.168.244.171 as d43887d6-d14d-4de8-bb70-6cf0064f7b3f (c1, k60).
1523099987: New connection from 127.0.0.1 on port 1883.
1523099987: New client connected from 127.0.0.1 as mqtt_f336f58.5e98a08 (c1, k60).
1523100233: New connection from 192.168.244.169 on port 1883.
1523100233: New client connected from 192.168.244.169 as mqtt-owfs-temp_8827 (c1, k60).
1523100248: Socket error on client mqtt-owfs-temp_8827, disconnecting.

[Pete]

Seeing the plugin now in Homeseer 3 but only state with a value of 0.

Sub: clients/ICS-ZNet/mqtt-owfs-temp/state 0

[Pete]

Going to move python script over to a microrouter running OpenWRT that is inside of my OmniPro 2 can.

[Pete]

Just a quickie update to the RPi #2 / Red Node / mcsMQTT / 1-Wire endeavor.

Switched to 22/4 alarm cable adding two voltages 5VDC and 12VDC.

The combo Midon Sensor uses 12VDC and the combo Hobby boards sensor uses 5VDC.

Repurposed an old Midon Temp08 for the two voltages using one 12VDC power supply.

Moving test sensors to main floor of home from basement.

[Pete]

Redid my Node Red OWFS stuff for 12 sensors (9 temperature and 3 humidity).

[ATTACH]68269[/ATTACH]

[Pete]

Just a quickie post here relating to using the RPi/9097/1-Wire combo temperature / humidity and temperature sensors.

Currently utilizing V.3.2.20.2 of the plugin and all appears fine for me.

Historically here was using Digitemp ==> temperature.txt file ==> read by a custom script which would create and update temperature / humidity variables every 10 minutes or so.

I have shut off the script and the sharing of the file now and switched over to using just OWFS / mcsMQTT.

Old old stuff used Midon Temp05 and a few Temp08's and Maxim 9097's and cat5e for multiple 1-wire networks here.

Note base here for 1-Wire mini hub is an RPi2 running Stretch connected to a USB (serial) 9097 and utilizing combo Hobby Boards, Midon and AAG sensors and AAG/Midon temperature sensor boards at this time. Temperature sensors are running in parasitic mode while combo sensors are utilizing 5VDC or 12VDC.

Next looking at graphing temperature / humidity values in Homeseer or external to Homeseer using Node Red.

Will post a drawing here of wiring.