Applies To:
  • CitectSCADA 6.00 Service Pack A
  • CitectHMI 6.00 Service Pack A

When more than one display client starts up at the same time and tries to connect to an IOServer that is communicating with an OPC device, one display client will show #COMS for the OPC device and in the Kernel >> Units window show the value of "Client Using" as <noserver>. The display client showing this is the one that takes that little bit longer to startup than the other display client. This display client will never display this OPC device data until it [display client] or the IOServer is restarted.

The problem does not occur at all if you start up a display client after the other display client has finished starting up and is communicating with the IOServer successfully.

This can be a potential problem when you restart all clients with:

Shutdown("All Clients",ProjectRestartGet(), 2)

I have tried my project in CitectSCADA v6 and it does not exhibit this behaviour at all, it only exists when running v6 SPA.

Further testing shows that this issue does not happen for the MODNET and DISK PLC IODevices.

Citect has confirmed this to be a problem in CitectHMI/SCADA versions 6 SPA for all tag based drivers, e.g. OPC. No register based drivers have this issue. This problem has been fixed in version 6.1.

tag driver, display client, race condition, #COMS