Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Moose]

I just wanted to check the following

1) you're not connecting your CM19 via a USB hub to your Pi

2) what amperage is your Pi power supply putting out. Remember unlike the CM15 your CM19 is powered solely from the Pi board. Is the power supply a genuine Raspberry Pi supplied one.

Yes, the CM19 is plugged into the USB on the Pi, directly, no hub
Yes, it is the CanaKit supplied PS.

[Moose]

Well, not really much change. The CM19 is the most likely problem, I think. After a restart of service, I get about 10-30 sec. of functionality, then it dies. The Cm19 led still does nothing. If nothing else, at least my configuration is staying put. Modules and module labels kept disappearing. Thatnks so much for your help and time.

Yes it's pointing to the CM19 ok. Just one last check. While HG is running again unplug the CM19 and plug it back in again. Copy/paste the following

sudo systemctl status homegenie.service and paste the results.

pi@raspberrypi:~ $ sudo systemctl status homegenie.service
● homegenie.service - Homegenie server
   Loaded: loaded (/etc/systemd/system/homegenie.service; enabled; vendor preset: enabled)
   Active: active (running) since Sat 2020-02-15 16:36:50 EST; 1min 1s ago
 Main PID: 806 (mono)
    Tasks: 68 (limit: 2200)
   Memory: 79.8M
   CGroup: /system.slice/homegenie.service
           └─806 /usr/bin/mono /home/pi/homegenie/HomeGenie.exe --log

Feb 15 16:37:51 raspberrypi mono[806]: libusb: warning [add_to_flying_list] failed to arm first timerfd (errno 9)
Feb 15 16:37:51 raspberrypi mono[806]: 2020-02-15 16:37:51.7149 Error System.Exception: Submit Async Read Failed.
Feb 15 16:37:51 raspberrypi mono[806]:   at CM19Lib.Driver.CM19.ReadData () [0x00029] in <a99ee7f64cb6456ca801bb9afe6c67c6>:0
Feb 15 16:37:51 raspberrypi mono[806]:   at CM19Lib.Cm19Manager.ReaderTask () [0x00008] in <a99ee7f64cb6456ca801bb9afe6c67c6>:0
Feb 15 16:37:51 raspberrypi mono[806]: libusb: warning [add_to_flying_list] failed to arm first timerfd (errno 9)
Feb 15 16:37:51 raspberrypi mono[806]: 2020-02-15 16:37:51.7149 Error System.Exception: Submit Async Read Failed.
Feb 15 16:37:51 raspberrypi mono[806]:   at CM19Lib.Driver.CM19.ReadData () [0x00029] in <a99ee7f64cb6456ca801bb9afe6c67c6>:0
Feb 15 16:37:51 raspberrypi mono[806]:   at CM19Lib.Cm19Manager.ReaderTask () [0x00008] in <a99ee7f64cb6456ca801bb9afe6c67c6>:0
Feb 15 16:37:51 raspberrypi mono[806]: libusb: warning [add_to_flying_list] failed to arm first timerfd (errno 9)
Feb 15 16:37:51 raspberrypi mono[806]: 2020-02-15 16:37:51.7149 Error System.Exception: Submit Async Read Failed.
Feb 15 16:37:51 raspberrypi mono[806]:   at CM19Lib.Driver.CM19.ReadData () [0x00029] in <a99ee7f64cb6456ca801bb9afe6c67c6>:0
Feb 15 16:37:51 raspberrypi mono[806]:   at CM19Lib.Cm19Manager.ReaderTask () [0x00008] in <a99ee7f64cb6456ca801bb9afe6c67c6>:0
Feb 15 16:37:51 raspberrypi mono[806]: libusb: warning [add_to_flying_list] failed to arm first timerfd (errno 9)
Feb 15 16:37:51 raspberrypi mono[806]: 2020-02-15 16:37:51.7149 Error System.Exception: Submit Async Read Failed.
Feb 15 16:37:51 raspberrypi mono[806]:   at CM19Lib.Driver.CM19.ReadData () [0x00029] in <a99ee7f64cb6456ca801bb9afe6c67c6>:0
Feb 15 16:37:51 raspberrypi mono[806]:   at CM19Lib.Cm19Manager.ReaderTask () [0x00008] in <a99ee7f64cb6456ca801bb9afe6c67c6>:0
Feb 15 16:37:51 raspberrypi mono[806]: libusb: warning [add_to_flying_list] failed to arm first timerfd (errno 9)
Feb 15 16:37:51 raspberrypi mono[806]: 2020-02-15 16:37:51.7149 Error System.Exception: Submit Async Read Failed.

[petera]

I just wanted to check the following

1) you're not connecting your CM19 via a USB hub to your Pi

2) what amperage is your Pi power supply putting out. Remember unlike the CM15 your CM19 is powered solely from the Pi board. Is the power supply a genuine Raspberry Pi supplied one.

Yes, the CM19 is plugged into the USB on the Pi, directly, no hub
Yes, it is the CanaKit supplied PS.

If you have access to an alternative power supply do try it before purchasing a replacement X10 controller. I do suggest using a CM15 with the Pi 3 b+ bit it's entirely up to you.

I've ruled out any problems with your HG install so all you should need to do is purchase a replacement X10 controller and you're good to go. Use this post as a troubleshooting reference for future issues.

[Moose]

I just wanted to check the following

1) you're not connecting your CM19 via a USB hub to your Pi

2) what amperage is your Pi power supply putting out. Remember unlike the CM15 your CM19 is powered solely from the Pi board. Is the power supply a genuine Raspberry Pi supplied one.

Yes, the CM19 is plugged into the USB on the Pi, directly, no hub
Yes, it is the CanaKit supplied PS.

If you have access to an alternative power supply do try it before purchasing a replacement X10 controller. I do suggest using a CM15 with the Pi 3 b+ bit it's entirely up to you.

I've ruled out any problems with your HG install so all you should need to do is purchase a replacement X10 controller and you're good to go. Use this post as a troubleshooting reference for future issues.

Will do. Thanks again.

[Moose]

So, I installed AHP with the CM19A drivers and my CM19A works fine, red led and all. I guess I might have to try and find another power supply, to be certain. I do not have any with that connector.

Another break-through,  I logged into HG without the CM19A connected, and guess what? I do not get the constant string of disconnect and reconnect messages.

[bkenobi]

I have only skimmed this thread but I can confirm that power issues will cause unpredictable behavior.  After swapping an underpowered power supply for a well above spec unit I experienced many fewer issues.   

[petera]

I have only skimmed this thread but I can confirm that power issues will cause unpredictable behavior.  After swapping an underpowered power supply for a well above spec unit I experienced many fewer issues.   

That's why I asked early on in the troubleshooting if the CM19 and the transceiver were confirmed as working. As soon as HG tries to load the CM19 driver there's a draw of power by the CM19 which is clearly not sufficient to continue the loading process and the connection is being dropped. HG will continue to try establish a connection with the CM19 hence the connect/reconnect showing on the UI.

I tend not to use those PSUs bundled with those SBCs. While the rating displayed may be 2.5ah most of them would be lucky to deliver 2ah or less. This is not noticeable when using a CM15 as it is powered independently but will definitely fall over with a CM19.

My concluding advice was to try an alternative PSU before buying a new X10 controller but my preference would be to have the best of both worlds and use a CM15.

Of course knowing that the CM19 was operational before starting the process would have really helped. God bless AHP 

[brobin]

So, I installed AHP with the CM19A drivers and my CM19A works fine, red led and all. I guess I might have to try and find another power supply, to be certain. I do not have any with that connector.

Another break-through,  I logged into HG without the CM19A connected, and guess what? I do not get the constant string of disconnect and reconnect messages.

This one will give you 3.5 amps and likely solve your problem.  amzn.to/321h8Le

[Moose]

So, I installed AHP with the CM19A drivers and my CM19A works fine, red led and all. I guess I might have to try and find another power supply, to be certain. I do not have any with that connector.

Another break-through,  I logged into HG without the CM19A connected, and guess what? I do not get the constant string of disconnect and reconnect messages.

This one will give you 3.5 amps and likely solve your problem.  amzn.to/321h8Le

I ordered that one this AM.

[Moose]

That's why I asked early on in the troubleshooting if the CM19 and the transceiver were confirmed as working.
My concluding advice was to try an alternative PSU before buying a new X10 controller but my preference would be to have the best of both worlds and use a CM15.
Of course knowing that the CM19 was operational before starting the process would have really helped. God bless AHP  [/quote]

I did answer as accurately as I could.  The CM19 was new to me, but I did know that the TM751 and the rest of my X10 worked well.
I did not know how to test the CM19 until this morning. A buddy asked me to plug it into my PC. It did not do much so I went looking for a driver. The AHP was the only one available. That is why I asked what is "normal" for the led function. I was suspicious!

Thanks again. We got somewhere, eventually.

[brobin]

I'm surprised that the 2.5A supply you have isn't enough so I suspect it isn't living up to its label.  The CM19 uses less than .5 amps and the Pi3B+ also typically draws less than .5 amps so 2.5 amps should provide more than enough headroom unless you have other USB devices connected to it.  Let us know if the new PS solves the problem.

[petera]

That's why I asked early on in the troubleshooting if the CM19 and the transceiver were confirmed as working.
My concluding advice was to try an alternative PSU before buying a new X10 controller but my preference would be to have the best of both worlds and use a CM15.
Of course knowing that the CM19 was operational before starting the process would have really helped. God bless AHP 

I did answer as accurately as I could.  The CM19 was new to me, but I did know that the TM751 and the rest of my X10 worked well.
I did not know how to test the CM19 until this morning. A buddy asked me to plug it into my PC. It did not do much so I went looking for a driver. The AHP was the only one available. That is why I asked what is "normal" for the led function. I was suspicious!

Thanks again. We got somewhere, eventually.
[/quote]

I have to assume that the majority of users here have at some stage used AHP or are still using it and are trying to migrate to something different. Personally I’ve never used it but it would appear to be the “gold standard” as far as X10 equipment testing is concerned. Obviously you are one of a rare breed here. 

When it comes to single board computers available amperage becomes an issue. Not so on a PC as you have now experienced. CM19 is now proven to work so your only enemy will be amperage or the lack of it.

When troubleshooting I always advise eliminating the obvious first. That’s what had you in a loop constantly reinstalling Raspbian and hoping for a different outcome. Now you are a little wiser on what needs to be eliminated first.

The exercise you carried out yesterday was not in vain. You now know how simple it is to reinstall HG without having to go through the whole image creation routine. Two simple commands and you’re there.

[Tuicemen]

My 3b+ reports under voltage conditions all the time and is visible if a monitor & keyboard is connected.
I don't get cm19, mono, libusb or system exceptions or warnings messages as moose does.
You can free up some power utilized by the HDMI connection a couple of ways.

Code: [Select]

vcgencmd display_power 0

Code: [Select]

tvservice --offOne of these commands may provide the bit of extra power required for the CM19.
Interestingly the under voltage condition of the Pi 3B+ is well documented and I've not yet found a solution to mine.
Wiggling the USB power connection will cause the under voltage condition so I suspect a bad solder joint at the mini USB connection.