Is it more secure to use a VPN that has open ports?

22/tcp open ssh 
53/tcp open domain 
80/tcp open http 
110/tcp open pop3 
443/tcp open https 
500/tcp open isakmp 
1723/tcp open pptp 
4567/tcp open tram 
8888/tcp open sun-answerbook

Or use the IP assigned by my ISP which has mostly filtered ports?

21/tcp    filtered ftp
53/tcp    open     domain
80/tcp    filtered http
548/tcp   open     afp
1036/tcp  filtered nsstp
3333/tcp  open     dec-notes
20005/tcp open     btx

In other words, if you are a hacker or pentester, would it be easier for you to gain root access to my machine with my vanilla IP or with my VPN IP?

  • Your question is quite unclear. Use for what? What does closed/filtered ports mean and what Open ports does your VPN have and why? – Ben Aug 29 '18 at 18:23
  • Basically I am saying that an "nmap ip" will result in more open ports for my VPN then for my standard IP address. – Cody Rutscher Aug 29 '18 at 18:27
  • 1
    Maybe what he is saying is that, when connected through VPN, he goes to a site like "what's my ip" he sees IP address = "X." When not connected to VPN he goes to "whatsmyip" and sees IP address = "Y." Then when he runs nmap on X he sees ssh open and a bunch of crap, but when he runs nmap on Y he sees nothing open. – hft Aug 29 '18 at 19:03
  • 3
    @CodyRutscher once again, there's not enough information in the question to responsibly guide you. A low-quality VPN service might be a much more serious threat for your privacy than your good ol' ISP line, but it's hard to tell the quality of this one looking at the open port list only. – ximaera Aug 29 '18 at 19:11
  • 1
    Personally, I'd recommend to read "Computer Networks" by Andrew S. Tanenbaum first. – ximaera Aug 29 '18 at 19:20

I think you're confusing a bunch of points together. Most people would consider "privacy" to be equivalent to "anonymity", ie whether or not your traffic can be traced back to your IP address. By and large, what services are being offered by a VPN company (and therefore what ports they have open) have no impact on whether I can trace traffic back to you through the VPN. How many customer service phone numbers my bank has is largely irrelevant to someone trying to trace my money transfers.

If your goal is privacy (ie hiding your IP address), then, umm, don't use your IP address.

Let's back up and talk about your threat model

In comments you say "I basically want to create a machine that is unhackable without social engineering" which is an almost completely separate goal from privacy / anonymity. It's also a much larger goal.

I think the first step in designing your security solution would be to decide what type and level of threat model you're trying to protect against. Some handy starter questions:

  1. What do you want to protect? (Data on your hard drive? Data you send over the network? Your computer against viruses? etc)
  2. Who do you want to protect it from? (someone who finds your laptop unlocked? Amature hackers? Criminal hackers? Nation-state hackers?)
  3. How bad are the consequences if you fail?
  4. How much time, effort, money, and changes to your internet usage behaviours are you willing to invest protecting it? (Are you willing to stop carrying a cell phone? Only use live-boot Linux OSes? Give up facebook and youtube and any other website that breaks if you block javascript?)

Bear in mind that the broader your threat model, the harder your job. It's very easy to say "... against all criminal hacking groups", and then you realize that even hiring a team of 30 security experts won't get you that.

To coin Mike's law of threat modelling:

The universal threat model "Confidential even against my hardware manufacturer and the most sophisticated nation-state hackers" is basically meaningless.

  • 1
    The issue with hiding your IP address behind a VPN is that if someone breaks into that VPN server you're using, then you expose not only your IP address to them, but also all your communications. So you get much less privacy than if simply not hiding. So the right constraint here is: "your goal is privacy AND you're not a person of a particular interest, nobody's going to track you down personally". How those two conditions live together in a same sane threat model is interesting though. – ximaera Aug 29 '18 at 23:05
  • 1
    @ximaera +1 I think there's an assumption baked in there that either the VPN provider is keeping sensitive logs, or that an attacker can get persistent monitoring tools into the VPN backend, right? So the quality of VPN service comes into play, right? – Mike Ounsworth Aug 29 '18 at 23:13
  • 1
    Exactly. More precisely, the second clause. My concern now is whether the author of the question does realise that issue or he's just checking some (sorry, senseless) formal conditions without any threat model in place. – ximaera Aug 29 '18 at 23:22
  • Basically, I am looking for a model that protects me beyond my browser being hooked. I understand how easy it is to get socially engineered. I am more concerned with being the victim of a personal attack that involves zero social engineering. – Cody Rutscher Aug 29 '18 at 23:25
  • 1
    @CodyRutscher AHA! A threat model! Editing that info into the question would actually turn it into a good question. If you can add details about their technical skills, whether they have physical access to your stuff, etc, that would be helpful. – Mike Ounsworth Aug 29 '18 at 23:27

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.