Pretty much what the title says. I have a subscription to Proton as well as Mullvad. I’ve ordered a router and I plan on running Proton on the router and Mullvad on the device I’m using, or vice versa. I would just like to know with absolute certainty that doing this won’t somehow put my Proton account at risk, before I actually do it; I rely on my Proton account for a lot of things. I know there are automated systems in place that detect abuse, especially with respect to DDoS and whatnot. I do not do anything related to DDoS or anything similar, so my account will never be flagged for anything of that nature correctly. I really can’t see how/why daisy chaining with another VPN could reasonably be construed as an abuse of the service, but I actually do worry about that in all honesty, whether or not I should. If there’s any way something like that might happen, I’ll abandon the idea as it would concern my Proton account and figure some other means of accomplishing this. Thanks.
What do you gain by doing this? I trust both proton and mullvad to not fuck up their encryption so attackers can’t read your traffic even through one VPN. The second one doesn’t offer additional security here.
In your setup, proton will only know you use mullvad but not know which sites you visit in the end. Mullvad knows everything just the same as without proton. So the outer VPN doesnt add privacy either.
If you are suspected of a crime, forcing mullvad to disclose your identity/IP is enough and proton doesn’t help.
If you are worried about traffic correlation analysis, then yes 2 VPNs will help. But honestly for normal usage I don’t see the point of 2 VPNs.
And about the DoS fear. Just do it the other way round? Mullvad on the router, proton on the device? From protons perspective you produce the same amount of traffic, it just comes from a mullvad server. The outer VPN is the one where you have increased traffic due to 2 VPNs. But I am pretty sure neither will be a problem and tunneling a VPN through a VPN is not a TOS violation
I am trying to obfuscate my traffic fingerprint as much as possible, yes.
So, does it (roughly) just double the amount of traffic by adding the second one, or…?
Edit:
I’m not sure why I was downvoted. Advanced traffic analysis techniques already exist. I can only imagine that as soon as methods sufficient to fingerprint innocuous use of the clearnet at significant scale become feasible, that is exactly what will happen. I see nothing inherently irrational about having a threat model that makes some reasonable attempt to account for that.