Hi! I am trying to read Modbus RTU messages over the serial line. There is a library called pyModbus which is very usable for Modbus TCP/IP, but the support for RTU is a bit lacking. The main problem seems to be the specification of Modbus RTU. RTU messages are separated by time gaps that are at least 3.5 characters long. Unfortunately, this is the only reliable way for separating messages - there is no length header or something similar. The specification does define a trailing CRC, though. As far as I understand the event loop, I will not be able to guarantee that will not miss such a gap - the reactor might be busy handling some other events while the gap arrives. But when the reactor assigns some time to read from the serial port, I'd like it to read until the gap is detected. My current attempt is to override the doRead method of SerialPort by something like this: def doRead(self): port = self.fileno() result = [] while True: # The following code is stolen from fdesc.readFromFD try: output = os.read(port, 1) if output: result.append(output) else: break except (OSError, IOError), ioe: if ioe.args[0] in (errno.EAGAIN, errno.EINTR): return else: return CONNECTION_LOST if not result: return CONNECTION_DONE else: self.protocol.rawDataReceived("".join(result)) If the serial port is configured with a timeout about the size of the gap, os.read() should return an empty string, thus indicating that the message is complete. Most of the code above is a copy of t.i.fdesc.readFromFD. The main difference is that I don't read 8192 bytes, but only one at a time, to detect a time gap. Does this make sense? Do you think it would be possible to replace the constant 8192 in readFromFD by a parameter (that defaults to 8192)? This would allow me to replace the content of the while loop with a simple function call. TIA and best regards, Albert -- Albert Brandl Weiermayer Solutions GmbH | Abteistraße 12, A-4813 Altmünster phone: +43 (0) 720 70 30 14 | fax: +43 (0) 7612 20 3 56 web: http://www.weiermayer.com