Regarding Time Out Error In Python OPCUA when read data from SIEMENS WinCC SCADA

[rajnishnationfirst]

Hi ,

I am using free OPCUA Python for getting values from WinCC Siemens SCADA.

I am facing TIME OUT ERROR issue in 2 cases .

Case 1) During calling get_value() on multiple nodes in a infinite while (True) loop Like .

Code : ( The below code runs continuously and don't have any sleep interval in the while loop )

while True :
     .......node1.get_value()

     .....Some processing 

      .....node2.get_value()

      .....Some processing 

      .....node2.get_value()

The above loop runs continously.

Case 2) I am calling 22 threads using a THREAD pool which runs parallely and one thread has nearly 150 Tags ( Nodes ) , so 22 thread will have nearly 3300 Tags in total.

a) Here EACH Thread in the 22 ThreadPool connects to WinCC Siemens SCADA and using each THREAD using node.read_raw_history(start_time , end_time , 0 ) for each of 150 Tags ...

b) All 22 threads run parallely uses 3300 Nodes and node.read_raw_history(start_time , end_time , 0 ) to be called all all 3300 Tags . every minute .

Also i have set 3600000 Milliseconds as the timeout in OPCUA client.py Python file as shown below .

Also at my WinCC SCADA it has 300000 milliseconds in the WINCC Siemens SCADA configuration file .

What are possible ways to avoid this kind of problem .

1. Do i need to also set the TIMEOUT at back End WinCC side also as 3600000 Milliseconds as same as in the CLIENT.py file ( Now at WInCC SCADA its 300000 milliseconds)

2. Do i need to change the python version ?

3. Also how to restore the connection if a TIMEOUT error / exception occurs and again the program should automatically connect the WinCC SCADA and start fetching the DATA continuously as normal. ?

Pls help me with both the cases that i have raised in the above 2 points .

Thanks Rajnish Vishwakarma.

[AndreasHeine]

polling data in an infinite while loop with node class "get_value" is the worst possible constellation and also with 22 threads 3300 nodes - you flood the network with requests (i would bet 90% of the values does not change from loop to loop) and propably you overwhelm the opc ua server with garbage requests... result timeout!

This is how you make a DDoS Attack on a OPC UA Server ;)

if you want to read lots of data you should use a subscription or the client class "get_values" method where you can read lots of nodes at once (but there is a messagesize/buffersize limit) https://github.com/FreeOpcUa/python-opcua/blob/131ad8ceffeb7510b7bb3d9db1742896dbf74196/opcua/client/client.py#L658

get_value -> one request single node (in your case 3300 requests as fast as your pc can throw them...) get_values -> one request multiple nodes at once subscription -> you will get notified for each value change of a node

it is also possible that you run into a raise condtion by your own threading code...

[vsagarmb]

Hello @AndreasHeine

I am trying to read values from a Kepware UA server and I get the following error.

Traceback (most recent call last): File "opc-client.py", line 106, in File "opc-client.py", line 41, in client_run_v1 File "opcua\common\node.py", line 155, in get_value File "opcua\common\node.py", line 164, in get_data_value File "opcua\common\node.py", line 275, in get_attribute File "opcua\client\ua_client.py", line 347, in read File "opcua\client\ua_client.py", line 83, in send_request File "concurrent\futures_base.py", line 447, in result concurrent.futures._base.TimeoutError

I have tried both get_value() and the get_values() methos and get the same exception. Do you have any pointers for me on how this can be resolved.

Thanks so much in advance.

-Sagar

[AndreasHeine]

@vsagarmb it would be better to open a own issue and fill out the issuetemplate, the error itself can have various reasons...