We need to think how to address this. The difficulty is that (obviously) swarm cannot know what each derived node requires. One option might be to make warn_limit a protected method of ConnectionHandler, allowing each node to specify its own limit, if necessary.
In the dhtnode, we recently increased the size of the record batch buffers to 1Mb.
We're seeing this warning in swarm firing now: https://github.com/sociomantic-tsunami/swarm/blob/v5.x.x/src/swarm/node/connection/ConnectionHandler.d#L301
We need to think how to address this. The difficulty is that (obviously) swarm cannot know what each derived node requires. One option might be to make
warn_limita protected method ofConnectionHandler, allowing each node to specify its own limit, if necessary.