Beckhoff OPC UA - Problem with "Publish Request Pump"

I have a problem with 7.6.6 Ignition and the opc server ua Beckhoff.
I have 2 types of Beckhoff PLCs connected to a scada:
BC9191 (very slow)
CX9001 (fast)

Communication with PLCs CX9001 running smoothly.
For communication with the BC9191, the readings of information running smoothly, but writes are not possible (error BAD QUALITY).

I get the following error:
PublishRequestPump Received failure notification for request ‘[PublishRequest requestHandle=70]’.

StatusCode[Severity=Bad, Subcode=Bad_InternalError]: ServiceFault: StatusCode[Severity=Bad, Subcode=Bad_InternalError]
at com.inductiveautomation.xopc.client.stack.TCPClientChannel.validateResponseType(
at com.inductiveautomation.xopc.client.stack.TCPClientChannel.receiveMessage(
at com.inductiveautomation.xopc.common.stack.UAChannel$1DeliverMessage.deliver(
at com.inductiveautomation.xopc.common.stack.UAChannel$
at com.inductiveautomation.xopc.client.stack.SerialExecutionQueue$RunnableExecutor.execute(
at com.inductiveautomation.xopc.client.stack.SerialExecutionQueue$RunnableExecutor.execute(
at com.inductiveautomation.xopc.client.stack.SerialExecutionQueue$
at java.util.concurrent.Executors$ Source)
at Source)
at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$201(Unknown Source)
at java.util.concurrent.ScheduledThreadPoolExecutor$ Source)
at java.util.concurrent.ThreadPoolExecutor.runWorker(Unknown Source)
at java.util.concurrent.ThreadPoolExecutor$ Source)
at Source)

Very Strange:
If I connect to the server Beckhoff with the software “UA Expert” in parallel with Ignition.
I can read and write without problems with UA Expert and Ignition!

Is this is a problem of timer?

Sorry for my english

There’s two things happening that Ignition’s UA client isn’t handling very well.

  1. For some reason the UA session is getting closed and Ignition’s UA client isn’t expecting that, which it handles it very poorly…

  2. For some reason the UA server is returning ServiceFault’s with Bad_InternalError in response to PublishRequests, which Ignition’s UA client isn’t expecting, and which it handles very poorly.

I’ll dig around and see if there’s any loggers or settings that might shine more light on this…

Do you know if the slower server happens to have any limits on the number of concurrent PublishRequests or anything like that?

To my knowledge, the server has no limit on the number of concurrent PublishRequests.

The Beckhoff company is currently looking at the problem, but they have not yet found a solution.

In Attached file, the opc server configuration.
With connection settings