Please tell me more about this PR. I've added the comment originally because the RPC method seemed to be called multiple times. This PR removes the comment and says nothing about any findings.

I didn't find any RPC being re-sent in all my tests and PRs and debuggings. I didn't test specifically for this, but I am confident my loggers would have caught a duplicate RPC. Feel free to close this if you suspect duplicate RPCs still exist 👍

Changed NetworkTime's _submit_sync_success() to the following:

@rpc("any_peer", "reliable", "call_remote")
func _submit_sync_success():
	var peer_id = multiplayer.get_remote_sender_id()
	
	if not _synced_clients.has(peer_id):
		_synced_clients[peer_id] = true
		after_client_sync.emit(multiplayer.get_remote_sender_id())
		_logger.info("Peer %s is synced" % [peer_id])
	else:
		_logger.warning("Peer %s is synced again" % [peer_id])

Resulting logs:

[DBG][@42][#1][netfox::NetworkTime] Client #1524895070 is now on time!
[INF][@42][#1][netfox::NetworkTime] Peer 1524895070 is synced
[WRN][@42][#1][netfox::NetworkTime] Peer 1524895070 is synced again

Unfortunately the _submit_sync_success() method is triggered twice, and based on the code, it is only ever called through an RPC. So this issue seems to persist, closing PR.

Fixes still welcome though!

The RPC is not triggered twice, the codeblock sending the RPC is.

The above image is with only a server and a client.

I found the source of the problem: NetworkTime.start() starts twice. See

By removing the highlighted code (at join() as well), everything works smoothly.

There are many solutions for this issue. The easiest solution is to put a boolean has_started inside NetworkTime.start() so when it detects true, it returns. lol this already exists with _active, but it doesn't become true in time, because of the await race condition.

Edit: Fixed in below PR