Best posts made by joaoabs

joaoabs

Hi,

During my intense troubleshooting with my RPI GW I had to recompile MySensors GW code dozens of times.
There were some points where I didn't know what build options I had at that moment, so I had to rebuild again just to be sure.

My suggestion for feature would be to include the build flags defined somewhere so the user could remember what was defined by the compilation time. This could be either pasted as a comment in the mysensors.conf file or a printout when a special flag is indicated.

Something like:

$sudo ./bin/mysgw
Usage: mysgw [options]

Options:
  -c, --config-file          Config file. [/etc/mysensors.conf]
  -h, --help                 Display a short summary of all program options.
  -q, --quiet                Quiet mode, disable log messages written to the terminal.
  --daemon                   Run as a daemon.
  --gen-soft-hmac-key        Generate and print a soft hmac key.
  --gen-soft-serial-key      Generate and print a soft serial key.
  --gen-aes-key              Generate and print an aes encryption key.
  --compiled-flags           Displays a short summary of the compiling flags when this binary was created
$
$sudo ./bin/mysgw --complied-options
My Sensors GW for RPI compiled at 2020-12-02 with the following flags:
--my-gateway=mqtt 
--my-controller-ip-address=192.168.2.180 
--my-mqtt-publish-topic-prefix=mysensors-out 
--my-mqtt-subscribe-topic-prefix=mysensors-in 
--my-mqtt-client-id=MySensorsGW 
--my-transport=rfm69 
--my-signing=software 
--my-signing-request-signatures 
--my-rfm69-frequency=433 
--my-is-rfm69hw 
--my-signing-debug 
--my-rfm69-encryption-enabled 
--my-mqtt-user=mysensorsuser 
--my-mqtt-password=mysensorspassword
(Other options were default. Consult documentation for further information)

What do you think? Would it be useful? This could also be useful for OrangePI or any other *nix system.

joaoabs

For the ones looking for a library that makes this DHT sketch work, mysensors have compiled a set of libraries that include this DHT.h and many others. Just go to https://www.mysensors.org/about/arduino#optional---install-external-mysensors-examples (yes, the name is not intuitive, maybe the keyword library would make it more obvious?), or jump directly to here and follow the standard library instructions. It took me two hours to find this, hope it saves time for future followers of mysensors like myself.

Good luck!

joaoabs

Well, one of the problems when buying from china is that it takes so long to get the stuff that we sometimes forgot what we ordered

It seems my RFM69's are in fact RFM69HW (eventhough the chip has a "H").

After reconfiguring the sensebender code and the raspberry configure command accordingly I got dialogue between them!

Will now try encryption

Update: It seems it's not as simple as just adding the "#define MY_RFM69_ENABLE_ENCRYPTION" in the arduino code and "#define MY_RFM69_ENABLE_ENCRYPTION" in the ./configure command for the RPI GW... Will try to sort it out and if not able to make it work, will open a different thread.

Update2: It seems it really is after all. Just needed a reboot, and the signing personalization will be re-used

joaoabs

Hi,

Got it working!
It was a strange mix of RFM69/RFM69HW devices (all soldered into the nrf2rfm69 board - so couldn't tell which type it was) and also a faulty RFM69. Also, it seems the signing consumes much processing power (and memory) to the little at328 chips. A sketch with more than 65% memory usage may fail some signatures, so not recommended to put many functionalities in the same node, specially when running my_debug.
Simple sketches just with a sht21 sensor are working fine with signing and RFM69 encryption.

Copying the correct HMAC, soft-serial and AES to the mysensors.conf file and with RPI reboots after each config/make/makeinstall cycle solved the problem.

Thanks,
Joaoabs

joaoabs

@frits I think you are right!

I admit initially I didn't thought it could be that because I have a slim node working 100%, but after your comment, I checked it and there is in fact a shunt between INT0 and DIO0 (I must have done it last year or so and didn't remember). The other nodes I've been trying don't have it, so that's likely the case.

Maybe the slim node page could have some reference to this?

I'll solder some shunts and will re-try this evening, but I'm sure that was the problem.

My slim nodes PCB's are red, but were ordered in 2018. Not sure what PCB release they are, though.

Thank you very much!

joaoabs

Hi,

By the PCB silk I was wrongly thinking that the module could be on of "RFM69" or "RFM69H" or "RFM69W" and "RFM69HW".

Now if I understood correctly the diagram, the module can only be either the "RFM69W" (low power 13dBm) or the "RFM69HW" (high power 20dBm).

Since this module has the "H" marked it should be "RFM69HW" and that is confirmed by the other side of it, because it has the two additional black ICs.

Thanks!

joaoabs

Hello,

I'm sorry to insist on this one, but I'm a bit lost....

I could understand that the recommended bootloader is MiniCore. I installed it in my Arduino IDE and after several tries-and-errors I believe I was able to burn the MiniCore bootloader by using an USBasp2 connected to the ICSP on the slim board:

/home/jabss/Desktop/arduino-1.8.2/hardware/tools/avr/bin/avrdude -C/root/.arduino15/packages/MiniCore/hardware/avr/2.0.1/avrdude.conf -v -patmega328p -cusbasp -Pusb -e -Ulock:w:0x3f:m -Uefuse:w:0xff:m -Uhfuse:w:0xd7:m -Ulfuse:w:0xe2:m 

avrdude: Version 6.3, compiled on Jan 17 2017 at 11:00:16
         Copyright (c) 2000-2005 Brian Dean, http://www.bdmicro.com/
         Copyright (c) 2007-2014 Joerg Wunsch

         System wide configuration file is "/root/.arduino15/packages/MiniCore/hardware/avr/2.0.1/avrdude.conf"
         User configuration file is "/home/jabss/.avrduderc"
         User configuration file does not exist or is not a regular file, skipping

         Using Port                    : usb
         Using Programmer              : usbasp
         AVR Part                      : ATmega328P
         Chip Erase delay              : 9000 us
         PAGEL                         : PD7
         BS2                           : PC2
         RESET disposition             : dedicated
         RETRY pulse                   : SCK
         serial program mode           : yes
         parallel program mode         : yes
         Timeout                       : 200
         StabDelay                     : 100
         CmdexeDelay                   : 25
         SyncLoops                     : 32
         ByteDelay                     : 0
         PollIndex                     : 3
         PollValue                     : 0x53
         Memory Detail                 :

                                  Block Poll               Page                       Polled
           Memory Type Mode Delay Size  Indx Paged  Size   Size #Pages MinW  MaxW   ReadBack
           ----------- ---- ----- ----- ---- ------ ------ ---- ------ ----- ----- ---------
           eeprom        65    20     4    0 no       1024    4      0  3600  3600 0xff 0xff
           flash         65     6   128    0 yes     32768  128    256  4500  4500 0xff 0xff
           lfuse          0     0     0    0 no          1    0      0  4500  4500 0x00 0x00
           hfuse          0     0     0    0 no          1    0      0  4500  4500 0x00 0x00
           efuse          0     0     0    0 no          1    0      0  4500  4500 0x00 0x00
           lock           0     0     0    0 no          1    0      0  4500  4500 0x00 0x00
           calibration    0     0     0    0 no          1    0      0     0     0 0x00 0x00
           signature      0     0     0    0 no          3    0      0     0     0 0x00 0x00

         Programmer Type : usbasp
         Description     : USBasp, http://www.fischl.de/usbasp/

avrdude: auto set sck period (because given equals null)
avrdude: AVR device initialized and ready to accept instructions

Reading | ################################################## | 100% 0.00s

avrdude: Device signature = 0x1e950f (probably m328p)
avrdude: erasing chip
avrdude: auto set sck period (because given equals null)
avrdude: reading input file "0x3f"
avrdude: writing lock (1 bytes):

/home/jabss/Desktop/arduino-1.8.2/hardware/tools/avr/bin/avrdude -C/root/.arduino15/packages/MiniCore/hardware/avr/2.0.1/avrdude.conf -v -patmega328p -cusbasp -Pusb -Uflash:w:/root/.arduino15/packages/MiniCore/hardware/avr/2.0.1/bootloaders/empty/empty.hex:i -Ulock:w:0x0f:m 

avrdude: Version 6.3, compiled on Jan 17 2017 at 11:00:16
         Copyright (c) 2000-2005 Brian Dean, http://www.bdmicro.com/
         Copyright (c) 2007-2014 Joerg Wunsch

         System wide configuration file is "/root/.arduino15/packages/MiniCore/hardware/avr/2.0.1/avrdude.conf"
         User configuration file is "/home/jabss/.avrduderc"
         User configuration file does not exist or is not a regular file, skipping

         Using Port                    : usb
         Using Programmer              : usbasp
         AVR Part                      : ATmega328P
         Chip Erase delay              : 9000 us
         PAGEL                         : PD7
         BS2                           : PC2
         RESET disposition             : dedicated
         RETRY pulse                   : SCK
         serial program mode           : yes
         parallel program mode         : yes
         Timeout                       : 200
         StabDelay                     : 100
         CmdexeDelay                   : 25
         SyncLoops                     : 32
         ByteDelay                     : 0
         PollIndex                     : 3
         PollValue                     : 0x53
         Memory Detail                 :

                                  Block Poll               Page                       Polled
           Memory Type Mode Delay Size  Indx Paged  Size   Size #Pages MinW  MaxW   ReadBack
           ----------- ---- ----- ----- ---- ------ ------ ---- ------ ----- ----- ---------
           eeprom        65    20     4    0 no       1024    4      0  3600  3600 0xff 0xff
           flash         65     6   128    0 yes     32768  128    256  4500  4500 0xff 0xff
           lfuse          0     0     0    0 no          1    0      0  4500  4500 0x00 0x00
           hfuse          0     0     0    0 no          1    0      0  4500  4500 0x00 0x00
           efuse          0     0     0    0 no          1    0      0  4500  4500 0x00 0x00
           lock           0     0     0    0 no          1    0      0  4500  4500 0x00 0x00
           calibration    0     0     0    0 no          1    0      0     0     0 0x00 0x00
           signature      0     0     0    0 no          3    0      0     0     0 0x00 0x00

         Programmer Type : usbasp
         Description     : USBasp, http://www.fischl.de/usbasp/

avrdude: auto set sck period (because given equals null)
Writing | ################################################## | 100% 0.00s

avrdude: 1 bytes of lock written
avrdude: verifying lock memory against 0x3f:
avrdude: load data lock data from input file 0x3f:
avrdude: input file 0x3f contains 1 bytes
avrdude: reading on-chip lock data:

Reading | ################################################## | 100% 0.00s

avrdude: verifying ...
avrdude: 1 bytes of lock verified
avrdude: reading input file "0xff"
avrdude: writing efuse (1 bytes):

Writing | ################################################## | 100% 0.00s

avrdude: 1 bytes of efuse written
avrdude: verifying efuse memory against 0xff:
avrdude: load data efuse data from input file 0xff:
avrdude: input file 0xff contains 1 bytes
avrdude: reading on-chip efuse data:

Reading | ################################################## | 100% 0.00s

avrdude: verifying ...
avrdude: 1 bytes of efuse verified
avrdude: reading input file "0xd7"
avrdude: writing hfuse (1 bytes):

Writing | ################################################## | 100% 0.01s

avrdude: 1 bytes of hfuse written
avrdude: verifying hfuse memory against 0xd7:
avrdude: load data hfuse data from input file 0xd7:
avrdude: input file 0xd7 contains 1 bytes
avrdude: reading on-chip hfuse data:

Reading | ################################################## | 100% 0.00s

avrdude: verifying ...
avrdude: 1 bytes of hfuse verified
avrdude: reading input file "0xe2"
avrdude: writing lfuse (1 bytes):

Writing | ################################################## | 100% 0.00s

avrdude: 1 bytes of lfuse written
avrdude: verifying lfuse memory against 0xe2:
avrdude: load data lfuse data from input file 0xe2:
avrdude: input file 0xe2 contains 1 bytes
avrdude: reading on-chip lfuse data:

Reading | ################################################## | 100% 0.00s

avrdude: verifying ...
avrdude: 1 bytes of lfuse verified

avrdude done.  Thank you.

avrdude: AVR device initialized and ready to accept instructions

Reading | ################################################## | 100% 0.00s

avrdude: Device signature = 0x1e950f (probably m328p)
avrdude: NOTE: "flash" memory has been specified, an erase cycle will be performed
         To disable this feature, specify the -D option.
avrdude: erasing chip
avrdude: auto set sck period (because given equals null)
avrdude: reading input file "/root/.arduino15/packages/MiniCore/hardware/avr/2.0.1/bootloaders/empty/empty.hex"
avrdude: writing flash (2 bytes):

Writing | ################################################## | 100% 0.07s

avrdude: 2 bytes of flash written
avrdude: verifying flash memory against /root/.arduino15/packages/MiniCore/hardware/avr/2.0.1/bootloaders/empty/empty.hex:
avrdude: load data flash data from input file /root/.arduino15/packages/MiniCore/hardware/avr/2.0.1/bootloaders/empty/empty.hex:
avrdude: input file /root/.arduino15/packages/MiniCore/hardware/avr/2.0.1/bootloaders/empty/empty.hex contains 2 bytes
avrdude: reading on-chip flash data:

Reading | ################################################## | 100% 0.04s

avrdude: verifying ...
avrdude: 2 bytes of flash verified
avrdude: reading input file "0x0f"
avrdude: writing lock (1 bytes):

Writing | ################################################## | 100% 0.01s

avrdude: 1 bytes of lock written
avrdude: verifying lock memory against 0x0f:
avrdude: load data lock data from input file 0x0f:
avrdude: input file 0x0f contains 1 bytes
avrdude: reading on-chip lock data:

Reading | ################################################## | 100% 0.00s

avrdude: verifying ...
avrdude: 1 bytes of lock verified

avrdude done.  Thank you.

Just for reference, the parameters were:
Board: Atmega328
Bootloader: No (only worked this way, but still not sure what this means: If is already a bootloader on the chip, or if we want to write a bootloader)
Variant: 328P
BOD: Disabled
Clock: 8Mhz internal (here I think its what we want to configure on the target chip - I wasn't sure what was about because it could also have been the clock frequency of the programmer)
Compiler LTO: Disabled
Programer: USBASP

Anyway, after this supposed success message, I believe I have a MiniCore bootloader in the chip and now it should be able to be programed with the AVRISPMKII (with the usb/serial programing adapter I use to program my mini-pros).

However, I'm not able to get it working. Maybe I'm choosing the wrong definitions in the arduino IDE.
As board definition, I'm using the same as when burning the bootloader:

Board: Atmega328 (from the minicore subset)
Bootloader: Yes
Variant: 328P
BOD: Disabled
Clock: 8Mhz internal
Compiler LTO: Disabled
Programer: AVRISPMKII

For instance, to load the ascii table example (for testing purposes), I get this error:

Insert Code Build options changed, rebuilding all
Sketch uses 2330 bytes (7%) of program storage space. Maximum is 32256 bytes.
Global variables use 308 bytes (15%) of dynamic memory, leaving 1740 bytes for local variables. Maximum is 2048 bytes.
/home/jabss/Desktop/arduino-1.8.2/hardware/tools/avr/bin/avrdude -C/root/.arduino15/packages/MiniCore/hardware/avr/2.0.1/avrdude.conf -v -patmega328p -carduino -P/dev/ttyUSB0 -b38400 -D -Uflash:w:/tmp/arduino_build_636369/ASCIITable.ino.hex:i 

avrdude: Version 6.3, compiled on Jan 17 2017 at 11:00:16
         Copyright (c) 2000-2005 Brian Dean, http://www.bdmicro.com/
         Copyright (c) 2007-2014 Joerg Wunsch

         System wide configuration file is "/root/.arduino15/packages/MiniCore/hardware/avr/2.0.1/avrdude.conf"
         User configuration file is "/home/jabss/.avrduderc"
         User configuration file does not exist or is not a regular file, skipping

         Using Port                    : /dev/ttyUSB0
         Using Programmer              : arduino
         Overriding Baud Rate          : 38400
avrdude: stk500_recv(): programmer is not responding
avrdude: stk500_getsync() attempt 1 of 10: not in sync: resp=0x00
avrdude: stk500_recv(): programmer is not responding
avrdude: stk500_getsync() attempt 2 of 10: not in sync: resp=0x00
avrdude: stk500_recv(): programmer is not responding
avrdude: stk500_getsync() attempt 3 of 10: not in sync: resp=0x00
avrdude: stk500_recv(): programmer is not responding
avrdude: stk500_getsync() attempt 4 of 10: not in sync: resp=0x00
avrdude: stk500_recv(): programmer is not responding
avrdude: stk500_getsync() attempt 5 of 10: not in sync: resp=0x00
avrdude: stk500_recv(): programmer is not responding
avrdude: stk500_getsync() attempt 6 of 10: not in sync: resp=0x00
avrdude: stk500_recv(): programmer is not responding
avrdude: stk500_getsync() attempt 7 of 10: not in sync: resp=0x00
avrdude: stk500_recv(): programmer is not responding
avrdude: stk500_getsync() attempt 8 of 10: not in sync: resp=0x00
avrdude: stk500_recv(): programmer is not responding
avrdude: stk500_getsync() attempt 9 of 10: not in sync: resp=0x00
avrdude: stk500_recv(): programmer is not responding
avrdude: stk500_getsync() attempt 10 of 10: not in sync: resp=0x00

avrdude done.  Thank you.

Problem uploading to board.  See http://www.arduino.cc/en/Guide/Troubleshooting#upload for suggestions.

Any hint on what I may be doing wrong?
Thanks,

joaoabs

Hi,

The mysgw seems to be creating a link to the wrong TTY.
When I connect the arduino gateway, I get the /dev/ttyUSB1 device.

Insert Code Herepi@nettemp:~/MySensors $ ls /dev/tty*
/dev/tty    /dev/tty15  /dev/tty22  /dev/tty3   /dev/tty37  /dev/tty44  /dev/tty51  /dev/tty59  /dev/tty9
/dev/tty0   /dev/tty16  /dev/tty23  /dev/tty30  /dev/tty38  /dev/tty45  /dev/tty52  /dev/tty6   /dev/ttyAMA0
/dev/tty1   /dev/tty17  /dev/tty24  /dev/tty31  /dev/tty39  /dev/tty46  /dev/tty53  /dev/tty60  /dev/ttyUSB0
/dev/tty10  /dev/tty18  /dev/tty25  /dev/tty32  /dev/tty4   /dev/tty47  /dev/tty54  /dev/tty61  /dev/ttyUSB1
/dev/tty11  /dev/tty19  /dev/tty26  /dev/tty33  /dev/tty40  /dev/tty48  /dev/tty55  /dev/tty62  /dev/ttyprintk
/dev/tty12  /dev/tty2   /dev/tty27  /dev/tty34  /dev/tty41  /dev/tty49  /dev/tty56  /dev/tty63
/dev/tty13  /dev/tty20  /dev/tty28  /dev/tty35  /dev/tty42  /dev/tty5   /dev/tty57  /dev/tty7
/dev/tty14  /dev/tty21  /dev/tty29  /dev/tty36  /dev/tty43  /dev/tty50  /dev/tty58  /dev/tty8

Then I start the mysgw and a new tty is created: "/dev/ttyMySensorsGateway", which looks nice. However, the mysgw returns lots of errors:

mysgw: Starting gateway...
mysgw: Protocol version - 2.2.0
mysgw: Serial port /dev/ttyMySensorsGateway (115200 baud) created
mysgw: MCO:BGN:INIT GW,CP=RNNGL---,VER=2.2.0
mysgw: TSF:LRT:OK
mysgw: TSM:INIT
mysgw: TSF:WUR:MS=0
mysgw: !TSM:INIT:TSP FAIL
mysgw: TSM:FAIL:CNT=1
mysgw: TSM:FAIL:DIS
mysgw: TSF:TDI:TSL
mysgw: TSM:INIT
mysgw: !TSM:INIT:TSP FAIL
mysgw: TSM:FAIL:CNT=2
mysgw: TSM:FAIL:DIS
mysgw: TSF:TDI:TSL

Then I go check what is wrong an I see the pointer wasn't created towards "/dev/ttyUSB1", but "/dev/pts/1":

pi@nettemp:~/MySensors $ ls -larth /dev/ttyMySensorsGateway 
lrwxrwxrwx 1 root root 10 Apr 26 18:27 /dev/ttyMySensorsGateway -> /dev/pts/1

Taking a look into the /dev/ttyUSB1 directly I can see it is fine, the problem is the GW software on the Raspberry PI that is pointing towards the wrong tty:

pi@nettemp:~ $ stty -F /dev/ttyUSB1 cs8 115200 ignbrk -brkint -icrnl -imaxbel -opost -onlcr -isig -icanon -iexten -echo -echoe -echok -echoctl -echoke noflsh -ixon -crtscts
pi@nettemp:~ $ cat /dev/ttyUSB1 
0;255;3;0;9;0 MCO:BGN:INIT GW,CP=RNNGAS--,VER=2.2.0
0;255;3;0;9;0;255;3;0;9;0 MCO:BGN:INIT GW,CP=RNNGAS--,VER=2.2.0
0;255;3;0;9;26 SGN:PER:OK
0;255;3;0;9;64 SGN:INI:BND OK
0;255;3;0;9;67 TSM:INIT
0;255;3;0;9;69 TSF:WUR:MS=0
0;255;3;0;9;77 TSM:INIT:TSP OK
0;255;3;0;9;79 TSM:INIT:GW MODE
0;255;3;0;9;82 TSM:READY:ID=0,PAR=0,DIS=0
0;255;3;0;9;87 MCO:REG:NOT NEEDED
0;255;3;0;14;Gateway startup complete.
0;255;0;0;18;2.2.0
0;255;3;0;9;91 MCO:BGN:STP
0;255;3;0;9;98 MCO:BGN:INIT OK,TSP=1

Any idea of how to make this work?

Where can I see the alternatives for the "
./configure --my-gateway=serial --my-serial-is-pty --my-serial-pty=/dev/ttyMySensorsGateway --my-rf24-channel=69 --my-transport=nrf24 --my-serial-groupname=tty --my-config-file=/etc/mysensors.dat" command?

Thanks,

joaoabs

Hello,

The GW sketch is the simplest possible. It is the SerialGateway.ino example just with the added flags for the signing.
It seems to have enough resources

Sketch uses 21440 bytes (66%) of program storage space. Maximum is 32256 bytes.
Global variables use 1401 bytes (68%) of dynamic memory, leaving 647 bytes for local variables. Maximum is 2048 bytes.

The node sketch is the Relay example sketch, where I then added, one by one, the code from other examples such as Door/Win/Button (tripwire), Temperature (DS18B20), Temp/Hum (DHT22) and Motion (HC-SR501).

I have a pretty basic version control, so for each new sensor type I was adding I was just creating another file. Since all the signature process was done (and written in the eeprom), I went back to the basics (sketch with Relay + Door) with the signing flags enabled and could confirm it worked with signatures (so it must be something I added). I then went version-by-version, enabling the flags and confirmed it was still working as I was progressing.
I've now reached a point where I just have the last sensor to add (Motion HC-SR501) but the available memory is concerning:

Build options changed, rebuilding all
Sketch uses 28030 bytes (91%) of program storage space. Maximum is 30720 bytes.
Global variables use 1542 bytes (75%) of dynamic memory, leaving 506 bytes for local variables. Maximum is 2048 bytes.
Low memory available, stability problems may occur.

I'll leave as it is, (disabling the signing debug, the memory use drops to 71%). Just out of curiosity, I checked the memory use on the original node sketch (without signing debug) and it was 73%.

Still not sure if it was an issue related with the memory or with the additional code I added for the Motion Sensor.
However, the strange conclusion is that it was really on the node, and not on the gateway as we thought.

I'll leave it on during the night in order to see if it hangs or something.

Thanks for your help!
Joaoabs

joaoabs

Hi,
Any particular reason for the values of the thermal cut-offs? (73ºC, 10A)?
This temperature (73ºC) makes it difficult to solder, I've burned a couple when trying to solder it... And why 10A? Shouldn't we get a lower current? I know the other fuse is limiting the current to 0.2A, but if we have a lower current on this one we may get cheaper components... Wouldn't this one, for instance, do the same function? https://www.ebay.com/itm/10-Pcs-Fuji-Microtemp-Thermal-Fuse-115-TF-Cutoff-1A-250V-New/322657006381

If the PCB is inside a box, without any lossen wires, couldn't we limit the temperature to a sliglhly higher value (115ºC)?

Thanks

joaoabs

So I managed to successfully run MySensorsGW in a RPI4 with 64bit OS, based on the theoretical premise that a 32bit binary can run in a 64bit system.

My project idea is to have a central device running my house automation. For that, I use and RPI4 4GB with IoT stack witch is a set of docker images each one with a specific function. This way, I'll be running MQTT Mosquito, NodeRed, zigbee2MQTT, inFluxDB and Graphana, each one in its own docker container. IoT stack also includes other images like transmission torrent, nextcloud, PiHole, etc. I might also use some of it later on.
Anyway, all of this requires a lot of effort from the little RPI4 so it is important to be able to use all of its resources. Using a 64bit OS is one way to have a more performant system and to be able to address all the RAM (4GB in my case).

So, having mysgw in this RPI was crucial to maintain a small footprint. It would be great to have the MySensors GW as a docker container as well, but for now this workaround would have to sufice.

Compiling mswg in RPI 64bit OS is not working (as per the above), but compiling it in a 32bit RPI OS and then copying a couple of files to the RPI 64bit OS it works, at least in my case. My msgw is configured as a MQTT GW, so if yours is different, your millenage may vary.

./configure --my-gateway=mqtt --my-controller-ip-address=192.168.2.180 --my-mqtt-publish-topic-prefix=mysensors-out --my-mqtt-subscribe-topic-prefix=mysensors-in --my-mqtt-client-id=MySensorsGW --my-transport=rfm69 --my-signing=software --my-signing-request-signatures --my-rfm69-frequency=433 --my-is-rfm69hw --my-signing-debug --my-rfm69-encryption-enabled --my-mqtt-user=mysensorsuser --my-mqtt-password=mysensorspassword

(Now I can change the --my-controller-ip-address to 127.0.0.1. This way, if for some reason I want to change the controller IP address, I'll not need to re-compile)

Anyway, so the process is basically compiling everything in the 32bit RPI OS, (configure, make, sudo make install), configure the /etc/sensors.conf accordingly and then coping these files to the same path in the 64bit RPI OS:

/usr/lib/arm-linux-gnueabihf/libstdc++.so.6
/etc/mysensors.conf
/etc/systemd/system/mysgw.service
/usr/local/bin/mysgw

Run it as normal:

sudo systemctl start mysgw.service

Enjoy

Dec 04 12:34:57 INFO  Starting gateway...
Dec 04 12:34:57 INFO  Protocol version - 2.3.2
Dec 04 12:34:57 DEBUG MCO:BGN:INIT GW,CP=RPNGLS-X,FQ=NA,REL=255,VER=2.3.2
Dec 04 12:34:57 DEBUG SGN:PER:OK
Dec 04 12:34:57 DEBUG SGN:INI:BND OK
Dec 04 12:34:57 DEBUG TSF:LRT:OK
Dec 04 12:34:57 DEBUG TSM:INIT
Dec 04 12:34:57 DEBUG TSF:WUR:MS=0
Dec 04 12:34:57 DEBUG TSM:INIT:TSP OK
Dec 04 12:34:57 DEBUG TSM:INIT:GW MODE
Dec 04 12:34:57 DEBUG TSM:READY:ID=0,PAR=0,DIS=0
Dec 04 12:34:57 DEBUG MCO:REG:NOT NEEDED
Dec 04 12:34:57 DEBUG MCO:BGN:STP
Dec 04 12:34:57 DEBUG MCO:BGN:INIT OK,TSP=1
Dec 04 12:34:57 DEBUG GWT:RMQ:CONNECTING...
Dec 04 12:34:57 DEBUG connected to 192.168.2.180
Dec 04 12:34:57 DEBUG GWT:RMQ:OK
Dec 04 12:34:57 DEBUG GWT:TPS:TOPIC=mysensors-out/0/255/0/0/18,MSG SENT
Dec 04 12:34:57 DEBUG TSM:READY:NWD REQ
Dec 04 12:34:57 DEBUG SGN:SGN:NREQ=255
Dec 04 12:34:57 DEBUG ?TSF:MSG:SEND,0-0-255-255,s=255,c=3,t=20,pt=0,l=0,sg=0,ft=0,st=OK:
Dec 04 12:44:37 DEBUG TSF:MSG:READ,80-80-255,s=255,c=3,t=7,pt=0,l=0,sg=1:
Dec 04 12:44:37 DEBUG TSF:MSG:BC
Dec 04 12:44:37 DEBUG TSF:MSG:FPAR REQ,ID=80
Dec 04 12:44:37 DEBUG TSF:PNG:SEND,TO=0
Dec 04 12:44:37 DEBUG TSF:CKU:OK
Dec 04 12:44:37 DEBUG TSF:MSG:GWL OK
Dec 04 12:44:37 DEBUG SGN:SKP:MSG CMD=3,TYPE=8
Dec 04 12:44:38 DEBUG TSF:MSG:SEND,0-0-80-80,s=255,c=3,t=8,pt=1,l=1,sg=0,ft=0,st=OK:0
Dec 04 12:44:41 DEBUG TSF:MSG:READ,80-80-0,s=255,c=3,t=24,pt=1,l=1,sg=1:1
Dec 04 12:44:41 DEBUG SGN:SKP:MSG CMD=3,TYPE=24
Dec 04 12:44:41 DEBUG TSF:MSG:PINGED,ID=80,HP=1
Dec 04 12:44:41 DEBUG SGN:SKP:MSG CMD=3,TYPE=25
Dec 04 12:44:41 DEBUG TSF:MSG:SEND,0-0-80-80,s=255,c=3,t=25,pt=1,l=1,sg=0,ft=0,st=OK:1
Dec 04 12:45:40 DEBUG TSF:MSG:READ,80-80-0,s=20,c=3,t=16,pt=0,l=0,sg=1:
Dec 04 12:45:40 DEBUG SGN:SKP:MSG CMD=3,TYPE=16
Dec 04 12:45:40 DEBUG SGN:SKP:MSG CMD=3,TYPE=17
Dec 04 12:45:41 DEBUG TSF:MSG:SEND,0-0-80-80,s=255,c=3,t=17,pt=6,l=25,sg=1,ft=0,st=OK:<NONCE>
Dec 04 12:45:41 DEBUG SGN:NCE:XMT,TO=0
Dec 04 12:45:41 DEBUG TSF:MSG:READ,80-80-0,s=20,c=1,t=0,pt=7,l=5,sg=1:14.0
Dec 04 12:45:41 DEBUG SGN:BND:NONCE=607CD378A13D593F9957DAC6077B64FEDC18063A8A3FC7EA50AAAAAAAAAAAAAA
Dec 04 12:45:41 DEBUG SGN:BND:HMAC=F0DA2FBCD8AED09526D29E9C6AF7873076303A8E6F6F0FAB258BE29031DA29EB
Dec 04 12:45:41 DEBUG SGN:VER:OK
Dec 04 12:45:41 DEBUG GWT:TPS:TOPIC=mysensors-out/80/20/1/0/0,MSG SENT
Dec 04 12:45:42 DEBUG TSF:MSG:READ,80-80-0,s=40,c=3,t=16,pt=0,l=0,sg=1:
Dec 04 12:45:42 DEBUG SGN:SKP:MSG CMD=3,TYPE=16
Dec 04 12:45:42 DEBUG SGN:SKP:MSG CMD=3,TYPE=17
Dec 04 12:45:42 DEBUG TSF:MSG:SEND,0-0-80-80,s=255,c=3,t=17,pt=6,l=25,sg=1,ft=0,st=OK:<NONCE>
Dec 04 12:45:42 DEBUG SGN:NCE:XMT,TO=0
Dec 04 12:45:43 DEBUG TSF:MSG:READ,80-80-0,s=40,c=1,t=1,pt=7,l=5,sg=1:64.0
Dec 04 12:45:43 DEBUG SGN:BND:NONCE=AB465A4A24BDDF89CC30C3F30F4F4B519ECDDC2843FF725874AAAAAAAAAAAAAA
Dec 04 12:45:43 DEBUG SGN:BND:HMAC=62C90515061BA556B4CF4F7ECB3FE224E54BE1ED8A2522AB8E8A1FA192B738C4
Dec 04 12:45:43 DEBUG SGN:VER:OK
Dec 04 12:45:43 DEBUG GWT:TPS:TOPIC=mysensors-out/80/40/1/0/1,MSG SENT

pi@raspberrypi:~/MySensors $ uname -a
Linux raspberrypi 5.4.79-v8+ #1373 SMP PREEMPT Mon Nov 23 13:32:41 GMT 2020 aarch64 GNU/Linux

Cheers

joaoabs

@joaoabs Got it. It doesn't compile if we don't replace the RFM69_MAX_PACKET_LEN (0x40ul)

And I can confirm that the procedure above worked. Compiled directly in a RPI4 with 64bit OS.

pi@raspberrypi:~ $ file /usr/local/bin/mysgw
/usr/local/bin/mysgw: ELF 64-bit LSB pie executable, ARM aarch64, version 1 (SYSV), dynamically linked, interpreter /lib/ld-linux-aarch64.so.1, BuildID[sha1]=beb479db20a0f93b0fc004832e06c8213d779bb6, for GNU/Linux 3.7.0, with debug_info, not stripped