Python 3.6 x64 OpenOPC

I am developing a tensorflow model to make a inference of a product property. I want to put this model to run on-line reading data from a OPC Server a writing the result back. To do that I am using the OpenOPC project.

Tensorflow just works with Python 3.6 in a 64 bit OS. So I need to make a OPC connection in the same environment.

OpenOPC was originally desinged to work with Python 2.7 in a x86 OS. There are some projects that upgraded it to work with Python 3.4+. I followed the procedure in the following link to install it: https://github.com/joseamaita/openopc120

I am developing my model in a VM with Windows 2012 Server. For OPC Server I am using Matrikon OPC for Simulation.

The list function from OpenOPC seems to be working fine. When I run the following script

import OpenOPC
open_host='10.0.2.15'
open_port='7766'

opc = OpenOPC.open_client(open_host, open_port)
opc.connect('Matrikon.OPC.Simulation.1')

aux=opc.list('Simulation Items.Random')
for item in aux:
    print(item)

opc.close()

I got the following result

Random.ArrayOfReal8
Random.ArrayOfString
Random.Boolean
Random.Int1
Random.Int2
Random.Int4
Random.Money
Random.Qualities
Random.Real4
Random.Real8
Random.String
Random.Time
Random.UInt1
Random.UInt2
Random.UInt4

Which is promising. But when I tried to run the following code to read a value:

import OpenOPC
open_host='10.0.2.15'
open_port='7766'

opc = OpenOPC.open_client(open_host, open_port)
opc.connect('Matrikon.OPC.Simulation.1')

aux = opc.read(['Random.Int4'])
for item in aux:
    print(item)

opc.close()

the OpenOPC gateway service stops and in the event viewer I have the following

Faulting application name: PythonService.exe, version: 0.0.0.0, time stamp: 0x5af24199
Faulting module name: gbda_aut.dll, version: 1.0.8.304, time stamp: 0x4d92bcca
Exception code: 0xc0000005
Fault offset: 0x000000000001b966
Faulting process id: 0xa90
Faulting application start time: 0x01d464c2264cd7ec
Faulting application path: C:\ProgramData\Anaconda3\lib\site-packages\win32\PythonService.exe
Faulting module path: c:\Windows\System32\gbda_aut.dll
Report Id: 69629dee-d0b5-11e8-8120-080027bc5cb4
Faulting package full name: 
Faulting package-relative application ID: ` .

It seems that the problem is related to this gbda_aut.dll. I have download the x64 version of it from

http://gray-box.net/download_daawrapper.php?lang=en

and registered it in the appropiated way.

The problem also happen when I try to use the DCOM mode. When doing that I receive the following.

Problem signature:
Problem Event Name: APPCRASH
Application Name:   pythonw.exe
Application Version:    3.6.5150.1013
Application Timestamp:  5abd3212
Fault Module Name:  gbda_aut.dll
Fault Module Version:   1.0.8.304
Fault Module Timestamp: 4d92bcca
Exception Code: c0000005
Exception Offset:   000000000001b966
OS Version: 6.3.9600.2.0.0.272.7
Locale ID:  1033
Additional Information 1:   e0f5
Additional Information 2:   e0f51433c52d71c63f2bde60bf8cf401
Additional Information 3:   1c6f
Additional Information 4:   1c6f1bc49be4e21c20f4309779c02df6

Does anyone have a idea how to make it work or a better way to do a OPC read using Python?

I have been trying to read data from KEPWARE SERVER using the OpenOPC package for python3.6 and i'm able to do so easily now by following the below given links:

1. https://github.com/Alexhll/OpenOPC-python3.6
2. https://github.com/joseamaita/openopc120

But the issue arises when i'm trying to read data continuously for every 5 secs for all 4 devices. Currently the server is a windows machine on which i have installed all 32bit version thing like python, pywin32, pyro4.

Below given are the logs i'm getting which is mostly timeout while trying to read the data for the device(i see this error mostly after 3-4 hours of continuous awesome run of the python script).

2019-12-14 05:09:08,750 - loggingModule - getOPCData10thDec.py - <module> - 1368 - 170 - Exception Captured in OPC Part. Trying to reconnect.
2019-12-14 05:09:08,765 - loggingModule - getOPCData10thDec.py - opcReconnect - 1368 - 41 - Trying to reconnect to the OPC server. Max Retries: 3
2019-12-14 05:09:08,781 - loggingModule - getOPCData10thDec.py - opcReconnect - 1368 - 44 - Closing the opc connection if any and retrying to connect 0 time
2019-12-14 05:09:08,796 - loggingModule - getOPCData10thDec.py - opcReconnect - 1368 - 47 - Creating a new OPC client
2019-12-14 05:09:08,859 - loggingModule - getOPCData10thDec.py - opcReconnect - 1368 - 49 - Trying to Connect to the OPC server: Kepware.KEPServerEX.V6
2019-12-14 05:09:08,906 - loggingModule - getOPCData10thDec.py - opcReconnect - 1368 - 51 - Connected to the OPC server
2019-12-14 05:09:08,921 - loggingModule - getOPCData10thDec.py - <module> - 1368 - 174 - Available Devices on the OPC:
2019-12-14 05:09:08,937 - loggingModule - getOPCData10thDec.py - <module> - 1368 - 175 - ['_AdvancedTags', '_ConnectionSharing', '_CustomAlarms', '_DataLogger', '_EFMExporter', '_IDF_for_Splunk', '_IoT_Gateway', '_LocalHistorian', '_Redundancy', '_Scheduler', '_SecurityPolicies', '_SNMP Agent', '_System', '_ThingWorx', 'Compressor01', 'Metering-Coriolis-S3S-Active', 'Metering-Coriolis-S3S-Diesel', 'Metering-Coriolis-S3S-Frac', 'Metering-Coriolis-S3S-Water', 'MudCube01', 'MudCube02', 'MudCube03', 'Pason WITS', 'S3S Coriolis Skid']
2019-12-14 05:09:18,869 - loggingModule - getOPCData10thDec.py - readOPCData - 1368 - 101 - Error while reading the OPC data for device: Compressor01
2019-12-14 05:09:18,884 - loggingModule - getOPCData10thDec.py - readOPCData - 1368 - 102 - Callback: Timeout waiting for data 

On further inspecting the issue, i looked at the event viewer logs wherein i found the application error given below pertaining the dll i registered.

Faulting application name: python.exe, version: 3.6.8150.1013, time stamp: 0x5c201b63
Faulting module name: gbda_aut.dll, version: 1.2.8.508, time stamp: 0x4822c190
Exception code: 0xc0000005
Fault offset: 0x0001c088
Faulting process id: 0x1484
Faulting application start time: 0x01d5b30ce3a7a612
Faulting application path: C:\Users\OPCAdmin\AppData\Local\Programs\Python\Python36-32\python.exe
Faulting module path: C:\OpenOPC36\lib\gbda_aut.dll
Report Id: 3fd3dbef-8701-4516-87ce-bc514ee627c2
Faulting package full name: 
Faulting package-relative application ID: 

I'm trying to find some solution to this problem but not able to do anything about it. Below given is the code i'm trying to use when i'm trying to reconnect to the OPC server once i see any timeout issue.

def opcReconnect():
    """
    reconnect to the opc server
    :return: opc client
    """
    global opc
    logger.info("Trying to reconnect to the OPC server. Max Retries: " + str(retry_n_times))
    for i in range(retry_n_times):
        try:
            logger.info("Closing the opc connection if any and retrying to connect " + str(i) + " time")
            opc.remove(opc.groups())
            opc.close()  # closing all old conneciton to the OPC
            logger.info("Creating a new OPC client")
            opc = OpenOPC.client()
            logger.info("Trying to Connect to the OPC server: " + s3sEdgeOPCServer)
            opc.connect(s3sEdgeOPCServer)
            logger.info("Connected to the OPC server")
            return opc
        except ConnectionError as e:
            logger.error('OPC Connection Failure')
            logger.error(e)
        time.sleep(retry_threshold)
    raise Exception

I'm trying to properly close the old opc connection by removing all the opc groups and then closing the connection, only after which i try to reconnect to the server again.