-
-
Notifications
You must be signed in to change notification settings - Fork 54
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Lowered maximum packet size #594
base: master
Are you sure you want to change the base?
Conversation
Can't this break some existing voxels? |
Sadly it probably can, so it needs to be used with caution and is more of an experiment. A proper solution would involve extending entity mixer so that it can send a property spanning multiple messages. This would also allow for far bigger and higher resolution voxels. Another requirement would be transmitting MTU to server so that it can also respect the MTU value for the given client. |
Also, on a technical level, where does a MTU of 1444 come from? The Ethernet MTU is 1500 bytes. The previous 1492 MTU came from PPPoE, which takes 8 bytes per packet. This is used by many internet providers. This new limit suggests something here is taking either 56 bytes on its own, or 48 bytes on top of PPPoE. What is it? |
There's no technical reason for this value, just that particular setup for which this experiment was for has this limitation. |
Proper solution to this would involve MTU detection when connecting to the server. For fast detection a series of packets could be sent with increasing sizes and MTU would be chosen depending on which ones arrive. |
I think there has to be a technical reason. ISPs don't set limits just because admins are bored. There has to be some sort of underlying reason why it's 1444 and not something else. It could be ISP-side, or it could be client-side. Here's a few possibilities:
So I think it's probably 1452 - 8 = 1444. This would mean probably PPPoE + DS-Lite. A person using a connection that's natively IPv6 and tunneling IPv4 inside IPv6 packets. In this case I think we'd be better off implementing IPv6. Which we should anyway! |
Sadly my ISP is not allowing IPv6 so I'm unable to work on it |
A workaround for this:
#204