Closed erlingrj closed 11 months ago
The error exhibited here:
Federate 1: FATAL ERROR: Expected 15 and got 14.
means that the Destination reaction was invoked a second time with the same input value from a previous invocation. The test case (incorrectly, I believe) has no STP offsets (neither STA nor STAA), despite using decentralized coordination. Somehow, it advances time (possibly to the stop time), believes the input to be present, but either a new input has not arrived or an input is in the process of arriving but the value has not been updated.
I believe this test can be made to pass reliably by introducing STP offsets, as done in the draft PR #2049, but I'm concerned that that may mask a real bug, which could be a race condition involving when inputs are deemed to be present and when their value is updated.
I believe this is addressed by #2118.
DecentralizedP2PUnbalancedTimeoutPhysical.lf
Is failing sometimes:
A successful (?) output looks like this: