The specification in thaliMobile contains parts like:
* If an attempt to connect to a peer via MPCF whose presence we have previously
* advertised via peerAvailabilityChanged should fail we MUST use our own
* heuristics to decide if we should mark the peer as no longer available.
* Typically we should allow several failed connection attempts before we decide
* to mark the peer as no longer present.
Current implementation will mark peers unavailable after certain threshold, but not based on failed connection attempts.
It should be determined how often connection failure -based unavailability would be needed in real life scenarios.
The text was updated successfully, but these errors were encountered:
I think it's fine to skip this entirely for now. E.g. we can just not implement for NewInfra. As it is thaliMobileNativeWrapper will send up a new peerAvailabilityChanged event when this occurs and that will automatically reset our timers. If we try to connect and thaliMobileNative can't make the connect work then we should get a peerAvailabilityChanged from thaliMobileNativeWrapper telling us the peer is gone.
The specification in thaliMobile contains parts like:
Current implementation will mark peers unavailable after certain threshold, but not based on failed connection attempts.
It should be determined how often connection failure -based unavailability would be needed in real life scenarios.
The text was updated successfully, but these errors were encountered: