Hello:
This is the situation:
- PC_A hosts an OPC server (Applicom or Kepware)
- PC_B hosts an OPC client (Kepware) and the full Labview and DSC 8.20 that I use as OPC client as well.
PC_A and PC_B connected via a cross-over Ethernet cable;
On PC_B I am using Datasocket read, Legacy read multiple tags VI and data binding in the Front Panel to compare results to fetch data on PC_A.
Upon loss of communication (cable disconnect) I have noticed that if the loss is less than 5 min, when I replug the cable, all methods of communication just recover the data correctly.
If the loss of communication is more than 5 min then all 3 methods (DS, Legacy VI and Data Binding) do not recover it upon reconnection which makes me think that the issue has something to do with the Variable Manager service. The other OPC client (Kepware OPC client) does recover it flawlessly. The only way for Labview to recover it is to stop/restart the Variable Engine service but this does not work all the time (error 1053) and I do not know if this is programatically do-able in Labview?
As a conclusion this aspect of the OPC communication did not evolve from Labview 6 and 7 where the only way to recover the communication was to restart the tag engine. It is sad that no effort have been made here to correct this failure.
I have updated my DSC to the last update available on the NI website (end of October 2006).
Changing the setting that tells every x sec the I/O project to try to reconnect/recover the communcation in case of loss is of no influence.
Thank you for your precious insights,
Christophe
