This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Bug in z-stack2.5? Desc Res sent to end device in unicast mode even when device is not in neighbour table

Folks,

I am running 2.5 stack and my system looks like this: I am running exact same configuration as "SensorDemo" as I have requirement to collect data from multiple nodes and aggregate in central device.

- Router "A" which also acts as end point, connects to Gateway through router "B".

- On the startup, system runs fine. It ran for almost 12 hours. After that A stops responding.

- On the network analyzer I find that, A is sending match descriptor request to Gateway (Through its router B). Gateway who is also co-ordinator,  sends rsp directly to "A".

- As A is not directly in range of Gateway/co-ordinator, Probably A never gets rsp sent and comes back online.

Does this sound like an error? I also observed the link status for co-ordinator and clearly it has entry for "A", but all 0 which I assume means, not direct link.

Attached the log. A = 0xCDAF Network address 

  • Adding some more data:

    Basically what I see is "A" has "incoming link cost to "B" = 7. Outgoing cost= 0. We do not see any "route reply" from A to B(probably outgoing from A is not working at all!). Right at the time of failure. Is this the reason? Given that "match_desc" requests are coming to co-ordinator, can I assume only the receive functionality of A has issues and most likely at PHY later? Can I rule out any NWK related issues at this point?

    BTW, after resetting A- We see everything working fine- as in "route reply to B which sets new incoming and outgoing values" and everything sets up fine.