I am using Datasocket as a means of communicating test status's between 4 systems on a network. The basics are working fine - each system publishes its status onto the LAN while subscribing to the status's of the other 3. So far so good. However I am now trying to build in some robustness into the app to ensure that each system works in a predetermined manner when things go wrong ie if one of the systems is switched off or if the LAN goes down, etc. I've tried looking at the Datasocket error clusters but am not sure what's going on - for example m/c 1 is doing DS reads and m/c 2 is doing writes and the error cluster on the reading app is being looked at:
If the datasocket Server is not running I get error 63
If the
DSS is started but no DS write app is running I get error 42
If the DS write app is now started - no error (this would be the normal scenario and all would be well)
If the DS write app is now stopped no error is flagged (!!??)
If the DSS is now stopped still get no error (!!!??)
But if I close the DSS and leave the DS write app running I get an error 63.
If I unplug the LAN cable from the writing m/c I get errors 0, 54 and 56 on the reading app (can't see any reason why different codes appear).
I don't seem to be getting consistent results. Any suggestions as to how best to handle datasocket errors would be appreciated.
Thanks,
Niel.
