tailscale / tailscale/tailscale
Terrible performance (packet loss) with Mullvad exit node enabled
- Dominant language
- Go
- Stars
- 36.5k
- Forks
- 3.2k
- Avg merge
- 2d 3h
- Merged PRs (30d)
- 123
Description
### What is the issue?
With local (in-country) Mullvad exit node enabled:
```
% ping -c5 quad9.net
PING quad9.net (195.176.255.138): 56 data bytes
Request timeout for icmp_seq 0
Request timeout for icmp_seq 1
64 bytes from 195.176.255.138: icmp_seq=2 ttl=57 time=120.069 ms
64 bytes from 195.176.255.138: icmp_seq=3 ttl=57 time=119.576 ms
64 bytes from 195.176.255.138: icmp_seq=4 ttl=57 time=119.955 ms
--- quad9.net ping statistics ---
5 packets transmitted, 3 packets received, 40.0% packet loss
round-trip min/avg/max/stddev = 119.576/119.867/120.069/0.211 ms
```
With Mullvad exit node disabled:
```
% ping -c5 quad9.net
PING quad9.net (195.176.255.138): 56 data bytes
64 bytes from 195.176.255.138: icmp_seq=0 ttl=51 time=26.557 ms
64 bytes from 195.176.255.138: icmp_seq=1 ttl=51 time=26.606 ms
64 bytes from 195.176.255.138: icmp_seq=2 ttl=51 time=26.867 ms
64 bytes from 195.176.255.138: icmp_seq=3 ttl=51 time=26.588 ms
64 bytes from 195.176.255.138: icmp_seq=4 ttl=51 time=26.473 ms
--- quad9.net ping statistics ---
5 packets transmitted, 5 packets received, 0.0% packet loss
round-trip min/avg/max/stddev = 26.473/26.618/26.867/0.133 ms
```
It is intermittent in nature.
### Steps to reproduce
_No response_
### Are there any recent changes that introduced the issue?
Not sure, maybe to do with the 1.94.1 update possibly. But not 100% certain.
### OS
macOS
### OS version
Tahoe 26.2
### Tailscale version
1.94.1 (app store)
### Other software
_No response_
Contributor guide
Assessment
This issue has not been assessed yet.