Is there any way to permanently set USB0 to one node and USB1 to different node?

ButsonJ · 26 October 2021 21:10

Hi all,

I have node red setup on a raspberry pi4b. Everything works, but the usb to serial connection. In the raspberry pi I specified each usb in by adding their attributes to the 10-usb-serial.rules file.
I can identify each cable by its unique Serial number. Here is an example (where UniqueSN is the
actual Serial Number):
KERNEL=="ttyUSB1", ATTRS{serial}=="UniqueSN01"
KERNEL=="ttyUSB0", ATTRS{serial}=="UniqueSN02"

I have it set like this as each usb has a different job and I can not have the data conflicting. When I go to the dashboard, if only one usb-serial connection is taking in data , both the serial nodes default to the port bringing in the data. I can not have this.
Is there any way to permanently set say USB0 to one node and USB1 to different node?
Here is a image where you can see both the connections are being set to the port receiving the data:

Jean-Luc · 26 October 2021 21:35

Here is what I did with 2 USB interfaces, 1x for RS232 and 1x for Modbus. This to have always the right information from the right interface

sudo cat /etc/udev/rules.d/99-usb-serial.rules
SUBSYSTEM=="tty", ATTRS{idVendor}=="10c4", ATTRS{idProduct}=="ea60",SYMLINK+="tty-onduleur"
SUBSYSTEM=="tty", ATTRS{idVendor}=="1a86", ATTRS{idProduct}=="7523",SYMLINK+="modbus"

For example it creates a symbolic link like this

ls /dev/tty-onduleur -l
lrwxrwxrwx 1 root root 7 oct.  23 09:34 /dev/tty-onduleur -> ttyUSB0

And inside NR flows editor, i use this one for Modbus

and this one for the other RS232

like this, never mix data.
I hope this will help you.

Colin · 27 October 2021 08:32

Here is a link to a starter for writing udev rules, but it is quite heavy stuff.

ButsonJ · 27 October 2021 14:12

Hi Jean-Luc,

Thanks for the quick reply. I tried this already where I added SYMLINK+="ttyUBS_COM1" and SYMLINK+="ttyUSB_COM2" to each cables attributes respectively. For some reason the dashboard does not pickup the set names, and when I manually change the name, the dashboard sets both serial inputs (although displaying different names) to take the input from the single working connection.
Here is an example image of it:

I also updated the .rules file:
KERNEL=="ttyUSB1", ATTRS{idProduct}=="2303", ATTRS{idVendor}=="067b", ATTRS{serial}=="CTAQf103Y23", SYMLINK+="ttyUSB_COM2"
KERNEL=="ttyUSB0", ATTRS{idProduct}=="2303", ATTRS{idVendor}=="067b", ATTRS{serial}=="BZBPv103Y23", SYMLINK+="ttyUSB_COM1"

For some reason when I set the name in node red to the SYMLINK name I added, It does not connect. But if I leave it as the default USB0 or USB1 like depicted in the original post image it is able to connect. I think that the rules I set are correct, have you had issues connecting them in node red before?

Colin · 27 October 2021 15:01

Unplug the devices then run
tail -f /var/log/syslog
and plug them in one at a time. Check that it recognises them and creates the symlinks.

Jean-Luc · 27 October 2021 15:09

We will try to understand, what does it look like if you do this command ?

ls /dev/ttyUSB_COM1 -l and the same for ttyUSB_COM2

and this one when the 2 interfaces connected
lsusb

did you launch this order as well ? I can imagine
sudo udevadm trigger

ButsonJ · 27 October 2021 15:29

Hmm thanks for the assistance,
It looks like my rules file is not going through properly.

When I run the ls command:

~$ ls /dev/ttyUSB_COM2 -l
ls: cannot access '/dev/ttyUSB_COM2': No such file or directory

~$ lsusb
Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 001 Device 004: ID 067b:23c3 Prolific Technology, Inc. 
Bus 001 Device 003: ID 067b:23c3 Prolific Technology, Inc. 
Bus 001 Device 002: ID 2109:3431 VIA Labs, Inc. Hub
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub

Also I did launch the order.

when checking the usb's connected it is clear the name has not changed.

~$ ls -l /dev/ttyUSB*
crw-rw---- 1 root dialout 188, 0 Oct 27 16:27 /dev/ttyUSB0
crw-rw---- 1 root dialout 188, 1 Oct 27 16:27 /dev/ttyUSB1

ButsonJ · 27 October 2021 15:40

Hi Colin,

Thanks for the suggestion!
When I followed your steps, I see the device is recognized. The serial number and make are all the same from what I have in the rules file. :

Oct 27 16:30:34 raspberrypiscale2and3 kernel: [  233.294632] usb 1-1.2: new full-speed USB device number 5 using xhci_hcd
Oct 27 16:30:34 raspberrypiscale2and3 kernel: [  233.441730] usb 1-1.2: New USB device found, idVendor=067b, idProduct=23c3, bcdDevice= 3.05
Oct 27 16:30:34 raspberrypiscale2and3 kernel: [  233.441751] usb 1-1.2: New USB device strings: Mfr=1, Product=2, SerialNumber=3
Oct 27 16:30:34 raspberrypiscale2and3 kernel: [  233.441770] usb 1-1.2: Product: USB-Serial Controller 
Oct 27 16:30:34 raspberrypiscale2and3 kernel: [  233.441788] usb 1-1.2: Manufacturer: Prolific Technology Inc. 
Oct 27 16:30:34 raspberrypiscale2and3 kernel: [  233.441805] usb 1-1.2: SerialNumber: CTAQf103Y23
Oct 27 16:30:34 raspberrypiscale2and3 kernel: [  233.446657] pl2303 1-1.2:1.0: pl2303 converter detected
Oct 27 16:30:34 raspberrypiscale2and3 kernel: [  233.449241] usb 1-1.2: pl2303 converter now attached to ttyUSB0
Oct 27 16:30:34 raspberrypiscale2and3 mtp-probe: checking bus 1, device 5: "/sys/devices/platform/scb/fd500000.pcie/pci0000:00/0000:00:00.0/0000:01:00.0/usb1/1-1/1-1.2"
Oct 27 16:30:34 raspberrypiscale2and3 mtp-probe: bus: 1, device: 5 was not an MTP device
Oct 27 16:30:34 raspberrypiscale2and3 watchdog[608]: device eth0 did not receive anything since last check
Oct 27 16:30:34 raspberrypiscale2and3 mtp-probe: checking bus 1, device 5: "/sys/devices/platform/scb/fd500000.pcie/pci0000:00/0000:00:00.0/0000:01:00.0/usb1/1-1/1-1.2"
...
Oct 27 16:30:45 raspberrypiscale2and3 watchdog[608]: device eth0 did not receive anything since last check
Oct 27 16:30:46 raspberrypiscale2and3 kernel: [  245.524569] usb 1-1.1: new full-speed USB device number 6 using xhci_hcd
Oct 27 16:30:46 raspberrypiscale2and3 watchdog[608]: device eth0 did not receive anything since last check
Oct 27 16:30:46 raspberrypiscale2and3 kernel: [  245.661731] usb 1-1.1: New USB device found, idVendor=067b, idProduct=23c3, bcdDevice= 3.05
Oct 27 16:30:46 raspberrypiscale2and3 kernel: [  245.661754] usb 1-1.1: New USB device strings: Mfr=1, Product=2, SerialNumber=3
Oct 27 16:30:46 raspberrypiscale2and3 kernel: [  245.661773] usb 1-1.1: Product: USB-Serial Controller 
Oct 27 16:30:46 raspberrypiscale2and3 kernel: [  245.661790] usb 1-1.1: Manufacturer: Prolific Technology Inc. 
Oct 27 16:30:46 raspberrypiscale2and3 kernel: [  245.661807] usb 1-1.1: SerialNumber: BZBPv103Y23
Oct 27 16:30:46 raspberrypiscale2and3 kernel: [  245.666513] pl2303 1-1.1:1.0: pl2303 converter detected
Oct 27 16:30:46 raspberrypiscale2and3 kernel: [  245.669196] usb 1-1.1: pl2303 converter now attached to ttyUSB1
Oct 27 16:30:46 raspberrypiscale2and3 mtp-probe: checking bus 1, device 6: "/sys/devices/platform/scb/fd500000.pcie/pci0000:00/0000:00:00.0/0000:01:00.0/usb1/1-1/1-1.1"
Oct 27 16:30:46 raspberrypiscale2and3 mtp-probe: bus: 1, device: 6 was not an MTP device
Oct 27 16:30:46 raspberrypiscale2and3 mtp-probe: checking bus 1, device 6: "/sys/devices/platform/scb/fd500000.pcie/pci0000:00/0000:00:00.0/0000:01:00.0/usb1/1-1/1-1.1"
Oct 27 16:30:46 raspberrypiscale2and3 mtp-probe: bus: 1, device: 6 was not an MTP device
Oct 27 16:30:47 raspberrypiscale2and3 watchdog[608]: device eth0 did not receive anything since last check

It is connecting to the USB but it does not seem like the rules are being applied

Colin · 27 October 2021 16:03

Wrong product.

Colin · 30 October 2021 09:14

Did that help?

ButsonJ · 1 November 2021 14:53

Hi Colin, I was out of town for the past 5 days so I was not able to update this ticket. It did help. I am able to see both cables are now connected in node red! Now I am just trying to test the data. I will update the forum later today if I am able to get data from both.

ButsonJ · 1 November 2021 20:03

Thanks a bunch for the help everyone! Crazy that such a silly mistake caused the issue.

Colin · 1 November 2021 20:29

I find I often get the easy bit wrong. Typos are very difficult to detect oneself. The brain sees what you think is there, not what is actually there.

hotswapster · 1 November 2021 23:50

You could probably use the exec node to see which ports are available and use this to enable that part of your flow?

Oh m a side note, it would be nice if we could choose serial device from /dev/serial/by-id To save using the udev rules

dceejay · 2 November 2021 10:50

the search we use wraps the serialport library find function which eventually calls

github.com

RIAEvangelist/serialport-js/blob/1795892ea025a1b3fdf4eb04d5adee6d011669d3/lib/find.js#L10


      
          'use strict';
          
          /**
           * Module dependencies.
           */
          const fs = require('fs'),
            path = require('path');
          
          /**
           * Find all registered serial ports by-id
           *
           * @return {Promise<Object[]|Error>}
           */
          function find () {
            const serialPath = this.paths.linux.serial;
            const readLink = (file, filePath) => (
              new Promise((resolve, reject) => {
                fs.readlink(filePath, (error, link) => {
                  if (error) {
                    reject(error);

hotswapster · 2 November 2021 12:16

Thanks @dceejay for the response. The code you've quoted doesn't mean a great deal to me due to my limited programming abilities. It did however, prompt me to try using by-id instead of just the tty port. It did work like this. Which makes me think it's easier to assign port by-id from now on.

Does this seem like a reasonable approach? (I have multiple ports on my node-red instances also).

Colin · 2 November 2021 13:41

I haven't been able to find out how the system allocates serial by_id. Do you know how it is done?

ButsonJ · 2 November 2021 14:04

That would work, but I want to make it cable specific as it will be used in an area where other people are working. If a cable gets unplugged I wanted it to work regardless of which port it is in hence using the cable's serial as an identifier.

ButsonJ · 2 November 2021 14:07

This is what we originally tried, but if one connection was not receiving data, Node-red would automatically switch it to the other connection for some reason. We found this on two different raspberry pi setups running node red with 2 serial cables. This was not an option for us as something like this could easily slip under the radar and cost us greatly. Again I am not sure why it auto switches when data is not being received.

hotswapster · 2 November 2021 14:22

lsusb shows the id:

~ $ lsusb
Bus 006 Device 002: ID 1a86:7523 QinHeng Electronics HL-340 USB-Serial adapter

Note the 1a86:7523.

Then look for the device by id:

~ $ ls /dev/serial/by-id
usb-0658_0200-if00 usb-1a86_USB2.0-Ser_-if00-port0

Note in my case there are two devices. One has the vendor id of 0658, the other has the vendor id of 1a86.

So the port I want to use is /dev/serial/by-id/usb-1a86_USB2.0-Ser_-if00-port0

Topic		Replies	Views
Custom serial node General	0	448	25 October 2019
Serial port / dev / ttyUSB0 General	5	1138	30 January 2020
How can I have diffrent nodes modbus and serial node use the same usb port? General modbus	4	364	12 August 2022
RPI USB allocation General	39	1898	26 February 2020
Best way to connect multiple serial UART sensors to Node-Red? General	3	1437	27 December 2020

Is there any way to permanently set USB0 to one node and USB1 to different node?

Related topics