After the Bluetooth mesh dfu is completed, the distribution network information of the node is cleared

[llly]

I have implemented DFU functionality in my Bluetooth mesh project and used device A as the distributor and device B as the node to be updated. I have already configured both devices and they can communicate with each other. Then I use the distributor to send new firmware to the node. After the node firmware is successfully updated, the original distribution network information of the node is cleared. At this point, the distributor and the node cannot communicate and need to manually reconfigure the distribution network. How can the original distribution network data be retained during DFU updates?

[Hung Bui]

Hi Illy, 
We will need to check if you have declared your model correctly. 
Could you show us how you define it  ?
Have you tried to test with for example the chat sample in our SDK ? To see if you also have the wrong metadata ? 
I would suggest to follow exactly the way we declare the model in the samples. 

[llly]

I added the manufacturer's custom model in the same way as in the mesh_demo example

And I tested it with the mesh chat example, I didn't make any changes, but the hash value in the upgrade package is still inconsistent with the calculated hash value. Can you test it yourself?

[Hung Bui]

Hi Illy, 
Could you send me the project files you used to compile the mesh chat sample ? By default it doesn't support DFU, right ?

[llly]

mesh_chat.rar

I use ncs2.9.1

[Hung Bui]

Hi Illy, 
Thanks a lot for the project. Now I have managed to reproduce the issue. 
I found that it's was the order of Company ID and Model ID was the issue. If I invert them I have the same hash as the auto generation: 

It's quite obvious that the auto generation messed up the Company ID and the Model ID. 

I have reported this internally. 

For now please just use the manual way of generating hash. 

[llly]

Okay thank you, so there are errors in automatic build, but the hash calculated by the device is correct, the hash generated by the shell command is also correct right? So now I need to use shell commands to obtain the correct hash value for the dfu process

When will these two bugs be fixed？

If this problem has existed for a long time, why hasn't anyone noticed it before? I think this problem is easily encountered when performing mesh DFU