This post is older than 2 years and might not be relevant anymore
More Info: Consider searching for newer posts

[ZIGBEE] Mesh network recovery from one down router

Hi guys,

I am trying to create a zigbee mesh with 10 devices with 1 coordinator and 9 routers (all are nrf52833 based custom board). I can create a mesh network easily with the help of the light control example in the Zigbee SDK 4.0. But there is just one small problem that when i turned down 1 node in the middle, the router right after did not re-route to a possible nearby node and send the message in another path, making the whole network down from that point forward. I waited for 2 minutes straight but still nothing happenned. So my question is that is there something that i need to do for the router to send out route request when it realizes that the parent node of it is down and if so, can you give me some hint for that

Thank you and please reply soon,

Best regards,

Tu

Parents
  • Hi Tu,

    According to the Zigbee spec, coordinators and routers shall participate in route repair, but initiating route repair is something they may provide, and is not mandatory. I'm not certain how this is handled in the SDK, as I assumed the network repaired itself in case of link failure, but from your issue that appears to not be true. I have forwarded this to our Zigbee team and will come back to you when I get a response.

    Best regards,

    Marte

  • Hi Tu,

    How far away from each other are the router devices? Are there remaining routers in radio range?

    Best regards,

    Marte

  • Hi Marte

    Each of them is within 10m radius from the others.

    I want to update some more information that i also try to make the coordinator to send many-to-one routing request and i can see that the node repair is happening. When i capture the message from the node to the coordinator, i can see that it choose a different path. But what i do not understand is that the process of choosing a new route.

    So the question is: Is the network suppose to be in many-to-one routing to have the ability to heal itself?  And is this happen to every other kinds of routing discovery like table routing, etc ?

    Another problem is that when i try to connect, for example, 3 routers around the coordinator directly, then i re-position each of them on a straight line 10m away from each others. The routers in the back of the row did not send route request to find a new way even it fail to send a report attribute message. So How can I fix this ?

    Thank you and please reply soon,

    Sincerely,

    Tu

Reply
  • Hi Marte

    Each of them is within 10m radius from the others.

    I want to update some more information that i also try to make the coordinator to send many-to-one routing request and i can see that the node repair is happening. When i capture the message from the node to the coordinator, i can see that it choose a different path. But what i do not understand is that the process of choosing a new route.

    So the question is: Is the network suppose to be in many-to-one routing to have the ability to heal itself?  And is this happen to every other kinds of routing discovery like table routing, etc ?

    Another problem is that when i try to connect, for example, 3 routers around the coordinator directly, then i re-position each of them on a straight line 10m away from each others. The routers in the back of the row did not send route request to find a new way even it fail to send a report attribute message. So How can I fix this ?

    Thank you and please reply soon,

    Sincerely,

    Tu

Children
Related