Request an address claim

[johekb]

Is it possible from application side to force a new address claim sequence to start?

[ad3154]

It depends on what you want to achieve here.

Do you:

1. Want to send a request for the address claim PGN to the global address so that everyone else claims?

Or

2. Want to restart an internal control function's actual claim process for some reason?

Both of these would be a bit unusual, though.

[johekb]

Thanks for the answer.
I would like to do number 1 as you describe above. What I would like to achieve is to make sure that the network participants are still available. I.e. how can I detect if a participants is disconnected. I know that there are application level protocols like status and heartbeat messages but I would like to achieve it on network level.

[johekb]

Any update on this question, i.e. how do I force a address claim?

[ad3154]

There has historically not been a way to do it, so I have added one in this PR: #553 as I can see it being valuable.

[ad3154]

This is now merged into main

[johekb]

That sounds very good. Thanks for the work!

[johekb]

@ad3154 I tested your change and it almost worked. I can see the the request messages goes out and that the participants on the network starts sending their addresses. All except the internal control function that asked for the request, it seems that this control function is not getting the request, is there a "filter" that the issuer is not getting the message that was sent out?

[ad3154]

Makes sense that the internal control function never "receives" it... I will see what I can do about that as well.

[ad3154]

Can you try commit 3bb8672 on branch adrian/request-address-claim-force and see if you get better results? Essentially it enables the stack to "receive" transmitted requests for the address claim PGN, which should only really cause an effect if an internal control function is in the completed state of the address claim routine, and should trigger it to re-claim its address.

[johekb]

I have tested your change and it seems to work now. Thanks!

[johekb]

@ad3154 Do you think that your patch can be merged to main in the near future?

[ad3154]

Yeah, shouldn't be long, just waiting on a code review from one of the other contributors.