Release 3.0.0 is now available for general use.

[janiversen]

We are pleased to announce the version 3.0.0 have finally after a long wait been released.

Please remark that there a number of incompatible changes in the API, but they are hopefully all documented.

We look forward to receive comments/suggestions and even better pull requests.

[MAKOMO]

Thanks for all your hard work! Is there a summary or list of the main API changes to quickly lift projects from 2.x to 3.x?

[janiversen]

Thanks. Sorry but the only record is CHANGELOG which is not very helpful. I started to make a list, but realised that a lot of changes had happened earlier so my list would be incomplete.

I just upgraded HomeAssistant Modbus integration (just locally for test), and the changes was quite easy to identify (with pylint and testing), I think I spent around 2 hours in total on the upgrade.

[alexandrecuer]

Starting to test the 3.0.0 Release on a ubuntu 22

I had to remove the .sync in the imports and to replace unit by slave in read_holding_registers, read_input_registers, read_discrete_inputs, write_registers, write_coil, read_coils

anyway, I usually import in all my services the logging library for my own needs, and all the entries are duplicated

2022-10-13 09:25:28,901 DEBUG modbus unit 2 > 1.00(h) écrit sur registre 100 09:25:28 DEBUG modbus:58 modbus unit 2 > 1.00(h) écrit sur registre 100 2022-10-13 09:25:28,901 DEBUG 1 promux slave(s) configured 09:25:28 DEBUG modbus:93 1 promux slave(s) configured 2022-10-13 09:25:59,103 DEBUG ................................ 09:25:59 DEBUG modbus:214 ................................ 2022-10-13 09:25:59,103 DEBUG modbus RTU connexion opened port /dev/ttyUSB0 - 9600 bauds 09:25:59 DEBUG modbus:58 modbus RTU connexion opened port /dev/ttyUSB0 - 9600 bauds 2022-10-13 09:25:59,137 INFO {'unamed_30002': 21.0, 'unamed_30003': 21.5, 'unamed_30004': 21.4, 'unamed_30005': 21.0, 'unamed_30006': -3276.8, 'unamed_30007': -3276.8} 09:25:59 INFO modbus:223 {'unamed_30002': 21.0, 'unamed_30003': 21.5, 'unamed_30004': 21.4, 'unamed_30005': 21.0, 'unamed_30006': -3276.8, 'unamed_30007': -3276.8} 2022-10-13 09:25:59,143 ERROR Could not connect to MQTT 09:25:59 ERROR modbus:60 Could not connect to MQTT 2022-10-13 09:25:59,144 DEBUG All queries finished - modbus connection is closed 09:25:59 DEBUG modbus:228 All queries finished - modbus connection is closed 2022-10-13 09:25:59,144 DEBUG -------------------------------- 

[alexandrecuer]

The library is not duplicating the log messages, so probably have something to do with your app.

I just found that you are configuring the logging within the library. So indeed it's coming from my app but I am not used to the log configuration being in the library.

Actually all the services I create in my app inherit from a skeleton class, which autoconfigures the log. In order not to have duplicates, I have to deactive this autoconfiguration in all the services using pymodbus.

[janiversen]

The logging was also configured in earlier versions, pr #1106 is a discussion how to make logging independent. I am not sure if you have a problem (for that please open an issue) or it is something else ?

[alexandrecuer]

OK the configuration is not the problem
The duplication occurs when using logging.StreamHandler

loghandler = logging.StreamHandler() self._log.addHandler(loghandler) 

I was using StreamHandler to send the log to the screen, but it's not necessary

[janiversen]

so there are no problem?

[alexandrecuer]

Yes there are no problem at all I think :-)
I can't recall why I was using StreamHandler in my app, so removing it is a simplification

[alexandrecuer]

The new forwarder is nice

I had to change a few things to make it work with a serial rtu bus :

#!/usr/bin/env python3 """Pymodbus RTU Serial synchronous forwarder. """ import argparse import logging from pymodbus.client import ModbusSerialClient from pymodbus.datastore import ModbusServerContext from pymodbus.datastore.remote import RemoteSlaveContext from pymodbus.server import StartTcpServer def run_forwarder(): """Run forwarder setup.""" port_server, port_client, slaves = get_commandline() client = ModbusSerialClient(method='rtu', port=port_client, baudrate=9600) if slaves: store = {} for i in slaves: store[i] = RemoteSlaveContext(client, unit=i) else: store = RemoteSlaveContext(client) context = ModbusServerContext(slaves=store, single=False) # start forwarding client and server client.connect() StartTcpServer(context=context, address=("192.168.1.29", port_server), allow_reuse_address=True) # loop forever FORMAT = "%(asctime)-15s %(levelname)-8s %(module)-15s:%(lineno)-8s %(message)s" logging.basicConfig(format=FORMAT) _logger = logging.getLogger() def get_commandline(): """Read and validate command line arguments""" parser = argparse.ArgumentParser(description="Command line options for examples") parser.add_argument( "--log", choices=["critical", "error", "warning", "info", "debug"], help='"critical", "error", "warning", "info" or "debug"', type=str, ) parser.add_argument( "--port", help="the port to use", type=int, ) parser.add_argument( "--port_client", help="the port to use", type=str, ) parser.add_argument( "--slaves", help="list of slaves to forward", type=int, nargs="+", ) args = parser.parse_args() # set defaults _logger.setLevel(args.log.upper() if args.log else logging.INFO) if not args.port: args.port = 5020 if not args.port_client: args.port_client = '/dev/ttyUSB0' return args.port, args.port_client, args.slaves if __name__ == "__main__": run_forwarder()

There is an error reported in mixin.py line 113. I just check the mixin.py file but nothing seem wrong...

python3 modbusTCPserver_b.py --port_client /dev/ttyUSB0 --slave 1 2 3 --log debug 17:10:06 DEBUG selector_events:54 Using selector: EpollSelector 17:10:18 DEBUG async_io:127 Socket [('192.168.1.29', 5020)] opened 17:10:18 DEBUG async_io:349 TCP client connection established [('192.168.1.117', 57555)] 17:10:21 DEBUG async_io:222 Handling data: 0x0 0x0 0x0 0x0 0x0 0x6 0x2 0x3 0x0 0x0 0x0 0x3 17:10:21 DEBUG socket_framer:171 Processing: 0x0 0x0 0x0 0x0 0x0 0x6 0x2 0x3 0x0 0x0 0x0 0x3 17:10:21 DEBUG factory:220 Factory Request[ReadHoldingRegistersRequest': 3] 17:10:21 DEBUG remote:51 validate[3] 0:3 17:10:21 ERROR mixin:113 Please do not use unit=, convert to slave=. 17:10:21 DEBUG transaction:130 Current transaction state - IDLE 17:10:21 DEBUG transaction:134 Running transaction 1 17:10:21 DEBUG transaction:291 SEND: 0x2 0x3 0x0 0x0 0x0 0x3 0x5 0xf8 17:10:21 DEBUG base:199 New Transaction state "SENDING" 17:10:21 DEBUG transaction:312 Changing transaction state from "SENDING" to "WAITING FOR REPLY" 17:10:21 DEBUG transaction:428 Changing transaction state from "WAITING FOR REPLY" to "PROCESSING REPLY" 17:10:21 DEBUG transaction:327 RECV: 0x2 0x3 0x6 0x7 0x6d 0x0 0xd9 0x0 0xda 0xc9 0x9b 17:10:21 DEBUG rtu_framer:193 Getting Frame - 0x3 0x6 0x7 0x6d 0x0 0xd9 0x0 0xda 17:10:21 DEBUG factory:352 Factory Response[ReadHoldingRegistersResponse': 3] 17:10:21 DEBUG rtu_framer:116 Frame advanced, resetting header!! 17:10:21 DEBUG transaction:516 Adding transaction 2 17:10:21 DEBUG rtu_framer:252 Frame - [b'\x02\x03\x06\x07m\x00\xd9\x00\xda\xc9\x9b'] not ready 17:10:21 DEBUG transaction:528 Getting transaction 2 17:10:21 DEBUG transaction:234 Changing transaction state from "PROCESSING REPLY" to "TRANSACTION_COMPLETE" 17:10:21 DEBUG async_io:298 send: [ReadHoldingRegistersResponse (3)]- b'000000000009020306076d00d900da' 

[alexandrecuer]

and when disconnecting the client, using modbus doctor:

17:44:13 DEBUG async_io:356 TCP client disconnected [('192.168.1.117', 58058)] 17:44:13 ERROR async_io:100 Handler for stream [('192.168.1.117', 58058)] has been canceled 

should'nt this be reported as an INFO, or DEBUG not ERROR ?

[dhoomakethu]

Thanks @janiversen for taking this on your shoulder. Apologies for not being much of help this time around. Well done.