Here are configuration files to reproduce this issue with this commit 37e07d2 in master in occurence.
The problem occurs in occurrence in run number 7 during the 1st second of the simulation when the number of UDP packets generated is very far from the number of packets pushed in the MAC queue as illustrated in figure below for wpanHost[0].


While with the same parameters of the scenario and only the seed that is supposed to change (run number 6) we get a more coherent simulation state as shown in the figure below.


I wonder where they passed the generated packets in run number 7 between the application layer and the MAC layer.
I was getting statistics (as on the number of failed transmission attempts shown in the figure below) from Ieee802154Mac not very credible which would finally be linked to this problem; the mac layer spends its time doing nothing (IDLE state) when it is supposed to have data to process.

Here are configuration files to reproduce this issue with this commit 37e07d2 in master in occurence.
The problem occurs in occurrence in run number 7 during the 1st second of the simulation when the number of UDP packets generated is very far from the number of packets pushed in the MAC queue as illustrated in figure below for




wpanHost[0].While with the same parameters of the scenario and only the seed that is supposed to change (run number 6) we get a more coherent simulation state as shown in the figure below.
I wonder where they passed the generated packets in run number 7 between the application layer and the MAC layer.
I was getting statistics (as on the number of failed transmission attempts shown in the figure below) from
Ieee802154Macnot very credible which would finally be linked to this problem; the mac layer spends its time doing nothing (IDLEstate) when it is supposed to have data to process.