Take the 2-minute tour ×
Information Security Stack Exchange is a question and answer site for Information security professionals. It's 100% free, no registration required.

I'm sitting at a cafe, using their public Wi-Fi. I don't go on sensitive sites; I basically go to a music site and listen to a live stream, for which there's no login. I don't check my email or visit any password-protected sites. Some rare times, I visit Stack Overflow, for which I'm automatically logged in. I'm using Windows 7 64-bit with AVG Internet Security.

  1. Can someone hack into my computer and see/download my files?

  2. Is there anything risky about what I'm doing?

share|improve this question
1  
Consider reviewing the answers to: security.stackexchange.com/questions/6946/… –  Luca Filipozzi May 14 '12 at 3:58
    
As for sites where you are already logged in, beware of Session Hijacking which doesn't even require the "logging in" process to work against you. –  Alexander Bird May 18 '12 at 1:01
add comment

5 Answers

up vote 5 down vote accepted
  1. Maybe. The question is not whether there is a possibility that someone might be able to hack into your computer; the answer to that question is always yes, it might be possible. If you are connected to a network, we can't rule out the possibility that someone might be able to attack you. Rather, the right question is, how big is the risk?

    In your case, if you use good security practices, the risk is pretty modest. Good security practices are things like turning on automatic updates to keep all programs updated with the latest security patches, using anti-virus, using a firewall (should be enabled by default on Windows 7), and not visiting any site that requires a login over public Wi-Fi.

  2. Not especially risky. It sounds like your practices are relatively safe.

    The biggest risk has to do with visiting Stack Overflow: because you are using a public Wi-Fi network, and because Stack Overflow does not use SSL (HTTPS), if someone there were malicious, they could steal your Stack Overflow login credentials and take control of your Stack Overflow account. They might not be able to learn your Stack Overflow password, if you never type it into your browser while using public Wi-Fi, but they could still learn the authentication cookie that's stored in your user and that's used to log you in. At that point, once they have the authentication cookie, they have control of your Stack Overflow account.

    However, this risk is relatively modest, because it's just your Stack Overflow account, and let's be honest, it's probably not the end of the world if someone steals your Stack Overflow account. The Stack Overflow folks know about this security risk, and they've been asked to adopt HTTPS to mitigate this risk, but they have declined to adopt HTTPS because they consider the risk to be of low-severity and they consider Stack Overflow accounts to be not important enough to be worth worrying about very much.

share|improve this answer
3  
note that we cycle cookies pretty aggressively, so they'd have to catch the "end" of a browsing session, not just a piece in the middle. And if you explicitly log off at the end, that also clears / resets the cookie, mitigating risk. –  Jeff Atwood May 14 '12 at 16:20
1  
Stack Overflow, Stack Overflow, Stack Overflow...Seems to be more like ALL StackExchange sites (including security.stackexchange.com as well) –  August May 18 '12 at 19:46
add comment

I'd argue that anytime you're in a situation where someone can inject themselves into the data stream between the user (you) and any server/service provider/content provider, you incur a higher level of risk than not.

First, let's look at the attack surface of your laptop (i'm assuming you're referring to a laptop here). Every computer can be a client or a server. In many cases, it can be both. There's increased risk when your laptop runs as some sort of server for an application because running as a server usually implies that the laptop is running a service that is accessible by someone. The risk becomes exaggerated when server software isn't properly patched or when there's a new attack vector that the vendor has not issued patches for. Do a Google search for services like BIND, Sendmail, SSHD, and you'll start to see that there are some vulnerabilities that are remotely exploitable. Fundamentally, this means that when your laptop is accessible by someone and there's a vulnerable version of some server software, then it's possible for an attacker to gain access to your computer. This access might be limited to certain parts or it could provide the attacker with admin (root) rights. It depends on the vulnerability and the service. Most OS's today usually come with some form of host-based firewall. However, some people tend to turn off firewalls. Some other people might open access to certain applications, especially if running a server service is authorized.

Second, another attack vector is network services. When someone can inject themselves into the data stream, an attacker can modify traffic in transit. Considering that most public hotspots don't implement high levels of security and wifi is provided as-is without any warranty. If you look hard enough, you may find the hotspot and trace it back to some sort of switch/router. In such instances, an attacker can easily inject themselves into the data stream and start sniffing sessions. A more advanced attacker could easily make themselves into a proxy and launch man in the middle (MITM) attacks where all sessions pass by their systems. SSL/TLS might provide data encryption, but an attacker that's injected themselves into the data stream could easily see the contents of encrypted traffic as well.

To answer your question, can someone hack into your computer and see what you've downloaded? It's certainly possible. But I'd argue that a malicious attacker will likely attempt to install some sort of malware on your system to extract more valuable data in the long term than a access list of downloaded software.

Is what you're doing risky? Yes - every action comes with risk. Any public network simplifies the launch of an attack. Consider the ease at which anyone can setup a wifi hotspot. You'll likely see wireless networks like "Free Wifi" or "Free Hotspot" in many locations. Many users are smart enough not to connect to those networks. However, what if you went to a coffee shop and saw a wifi network called "ATT" or labeled as some coffeeshop name? The same people that avoid "Free wifi" networks may choose to connect to someone that looks more official.

With that said, if you have host-based firewalls enabled, consistently patch OS and applications (i.e. Flash, Acrobat, Office, etc), use some sort of AV (not really effective against modern malware (i.e. 0day) but still affords some level of protection), don't enable unnecessary services, and perhaps use a VPN connection, the risk of compromise will be lower overall. At the same time, attackers are becoming more organized and are after money. So if you don't access banking sites, access or store data that is sensitive, or otherwise use the laptop in such a way to not access sensitive data, then you'll be at lower risk of losing important data (i.e. identity stolen, loss of corporate data, etc).

share|improve this answer
add comment

I don't go on sensitive sites

You don't, but your computer does

You don't, but the software you are using might go to the most security sensitive servers: the repositories were they automatically download software updates.

Theses repositories are the most possibly security sensitive "sites" for you because if they were hacked (modified) or if some impostor were able to impersonate these servers, you would get, and possiby install, modified software. This corrupt software might then steal your passwords, copy your files, etc.

So the answer is:

Is the automatic update protocol secure? (for every program I use)

(The answer is a question, and I don't know its answer.)

I don't go on sensitive sites

You don't know where you are going with your web browser

You can't know where you are going to "go" when using a Web browser.

You might browse only nicekittenphotos.com, but this site might include resources from other domains (and you can't know that before going to nicekittenphotos.com - even if you have checked before, the site might have changed).

For example, many sites include scripts from google-analytics.com (http://www.google-analytics.com/ga.js). This implies that your www.google-analytics.com cookies are sent.

This works for any website were you have cookies. A rogue Internet gateway could collect your non-secure cookies on any website.

If you are logged in Google, your logins appears in some HTTP pages returned by Google and YouTube, so it would possible to learn your Google account name.

If these HTTP resources were modified by a rogue gateway, this could impact other more important web sites. If these HTTP resources are stored in the disk cache, this could impact other web sites for a long time.

You can mitigate this threat by emptying your browser cache after you disconnect from public Wifi.

I suggest that you use private browsing mode.

You could also use Tor, to trade the risk of a rogue Wifi AP to the risk of a rogue Tor exit node.

share|improve this answer
    
In linux/unix, this isn't a risk as updating system-wide software generally (a) requires super-user permissions and (b) goes through a package-manager/repository that checks for valid, trusted cryptographic signatures on everything. So unless the attacker has already compromised the repository you trust (or tricked you into trusting a bad repo), the threat you mention is non-existent (and if they have they can just push bad packages in the trusted repo). Even simple web-browser/android/iOS add-ons/apps use strong cryptographical signatures that is checked before installing an update. –  dr jimbob May 14 '12 at 19:06
    
(Granted I'm sure some applications will automatically download data in an insecure manner that potentially could be used for an attack somehow; and I'm not sure how windows does auto updating applications; would not be the least bit surprised if no signing protocol is required by the developer or if any installed application could just silently update itself.) –  dr jimbob May 14 '12 at 19:10
add comment

In addition to all of these great suggestions I would like to add if you are running Windows 7, then it would be a great idea to choose the public network profile in the Windows firewall, this will help a little bit more as it closes off certain ports from access on your machine. File and print sharing for example, gets blocked on the public profile, so when you join the network and Windows asks what type of network you are joining, choose the public option when you are on a public wi-fi network.

This should help a little bit more and definitely understand that whatever sites you are visiting if they require a password, even if you are not entering it in, and it has been saved on your machine. The information is being sent through the air and can be captured if the other person has the right tools. So basically if you are really concerned about that type of risk, don't go to sites that require a password, that should help minimize the risk.

share|improve this answer
    
ok, thank you for the recommendation. –  frenchie May 18 '12 at 10:35
add comment

One more scenario just for the sake of completeness and being paranoic:

1.Can someone hack into my computer and see/download my files?

2.Is there anything risky about what I'm doing?

Yes and Yes

For 1. When you connect on a public wifi dont have any idea where you connect your computer either. An attacker could be imitating the AP and force you to connect to his computer instead of the original access point.

2. Many browsers or plugins connect automatically to popular services (so if for example you have a plugin that downloads your twitter feed its as if you are connected to twitter)

You dont actually know which site you visit.

Solutions

1.For the transport layer (are my data safe when they are transported?):

2.When you give your password visit only https sites through a trusted ssh tunel or tor and make sure you cant be redirected to another domain

3.For the attacks on your computer layer: make sure that you dont have any daemons listening to any ports so that an attacker cant communicate with your computer and that you dont have any malware ofc

4.Be sure that the access point is what it advertizes itself to be (I dont know how this can be done)

To be realistic As far as your average starbucks goes unless you carry something importand enough in your laptop or you have something to loose (a hacked facebook account usually is not something disastrous) visiting only https sites and refraining from messing with online money on public wifi should be acceptably enough and the ssh tunel method should be almost perfect.

share|improve this answer
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

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