OPC Problems: OPC servers running in background but not visible for clients
In a current project we are running with the OPC servers Kepware and CET (for IEC61850). These two OPC servers are running on a pair of connectivity servers.
The system has been running without problem for month, where the OPC servers were visible for the other clients.
The only thing that has been changed in the meantime are the user accounts, role etc.
This has been changed on our domain controllers.
We first saw the problem when we were enabling the redundant OPC server for the IEC61850 communication. Suddenly it was not possible to select the Alarm Server under the IEC61850 service group in our Event Collector service. When we were trying to point to the correct AE server instance, it was simply not possible. A screendump of this can be seen in the attached document as Figure 1.
As a comment to this, the configuration of the service providers for the OPC servers under the service group is setup correctly.
We have 4 connectivity servers, and it is only for the two connecticity servers were we have the OPC servers, were this problem occurs.
In 800xa engineering workplace under each service provider in the service group that does not work, it is possible to override the AE server under the Special Configuration tab. This is not a thing that we do, but because of this problem, we tried to override the AE server.
When this is tried, an error message pops up. This error message can be seen in the attached document as Figure 2.
A second problem that probably is caused be the same error, can be seen in a service group containing our OPCDA providers for the OPC servers. Under the Special Configuration tab for the service providers, is it possible to select the correct OPC server identity, ProgID, but when it is applied, it changes to a strange error code. This can be seen in the attached document as Figure 3.
We have of course tried to investigate the problem, and have among other things played a bit with a Matrikon OPC explorer. When we search for OPC servers on one of the connectivity servers were we have a problem, an error message pops up. This message can be seen in Figure 4.
In some of our last investigation we have found out that if we power up the connectivity servers that has the problem with all network cards disconnected, it seems like all OPC servers are running as they should. When we then connect the connectivity servers to the network and the domain controllers, all OPC servers are becoming invisible. Please see Figure 5 for a screendump with network cards disabled and Figure 6 for a screendump showing the Server properties when the networks are connected again.
Based on this we think that the problem is caused by some group policy settings which is changing the DCOm configuration in the domain controllers. The error message in Figure 2 does also indicate that it is some DCOM configuration that causes the problem.
Have anybody experienced this before, or know what can be wrong?
Thanks in advance for the answers, and sorry for the long explanation.
You changed user accounts, and the error message ( 2nd screen shot ) says the configured user identity is incorrect.
Basically, after you change the user account settings, your services will keep running until they have to be restarted, at which point they fall over. You can either go back and reset all the user settings, ( which may or may not work ) or go into the services control panel and change the user/password settings for the affected services to the new username/passwords ( which is likely to be much easier )
You probably want to reboot everything, ( one server at a time ) and force a group policy update. It also sounds like some servers might be running with old stored security credentials.
You also need to check the DCOM settings for those services to ensure remote users are still allowed to access them.