I'm in the early thoughts on this, but was wondering if anyone knows the exact lookup/routing behaviour of multiple nodes sharing services with the same hostname. I'm working on a mesh deployment where I expect to have several of the nodes with independent internet connections on different providers. While I don't want to share the WAN to the mesh for obvious reasons, it would be nice to be able to mirror specific sites out to the mesh, I'm thinking like weather.gov and radar.weather.gov from multiple of these if remote (nodes without internet) would automatically select the closest mesh node providing the service with the same name. The reason I'd be looking for this behaviour would be to minimize how much of the mesh such requests need to travel across, yet provide redundancy for failures.
Creating the relay would be trivial with Apache, and I have done so with other web services elsewhere. And I'm relatively certain scripting the the machine to make sure it can reach those servers periodically and adding/removing the mesh share would also be trivial
Creating the relay would be trivial with Apache, and I have done so with other web services elsewhere. And I'm relatively certain scripting the the machine to make sure it can reach those servers periodically and adding/removing the mesh share would also be trivial