Hi,
I recently have run into a problem where an end device, it rests itself rather frequently and everytime it resets it randomly chooses an ext address for itself (it is not using TI chip :) ), causes the coordinator to run out of space in its association table so that the coordinator declares not to accept any new end device associations. Given that I am not able to change end device code, I have been looking into having the coordinator maybe time out end device entries that it no longer sees traffic/data requests.
I have found two functions in AssocList.h that seem to related to what I want to do, the AssocGetNextInactiveNode and AssocRemove. So my questions are: the fact that such functions are there probably means that zstack is already doing some purging work with inactive devices. If so, are there some parameters somewhere I can configure to control the purging operations? If not, can I call the AssocGetNextInactiveNode function to find the idle devices then call AssocRemove to get rid of them by myself?
Unfortunately that part of zstack is only provided as library, no source code, only minimum documentatin. Coming here seeking expert advices is my last option. BTW, the zstack I am using is 1.4.3-1.2.1. I have taken a look at the newest 2.3.3 code and found that each association table entry now actually has an "age" field. Unfortunately for this project I am limited to CC2430/31 chips so I am stuck with the 1.4.3 zstack.
Many thanks in advance for answering my questions.
LJ