If there would be any that do that without your consent it would be widely known since the source code is available for GPL apps.
For commercial apps it isn't but there are a lot of people that monitor their net traffic and that would be out in no time either.
I don't even see a reason to run a firewall at all with the mainstream broadband setup since the router or modem normally does NAT and can also be setup for anything you need if it is a halfway decend piece of hardware.

I would like an application firewall so that I can control what applications have access to the internet, and my network.

I am not content with relying on others monitoring their network traffic, or trawling source code.

Perhaps this isn't a common way of thinking amongst linux users, and perhaps that contributes to the situation where there isn't a product on the linux platform that suits my requirements. No demand, no supply.

But for me, I don't want to find out that one of my applications is accessing the internet without my consent at some point in time after the fact when someone notices while running ethereal or hacking code. I want to know about it when it happens and I want to stop it there and then.

koan

Muff or LinuxSeeker, (or any other ex-Windows user),

This is an old thread, but I decided to reply and continue it because I just recently installed Debian on my Dell laptop and wanted to find out how are all the experienced Linux users protect themselves against rough application and I am not talking about viruses, worms, etc. but applications that look, feel and smell innocent, you installed it based on what it promised to do for you but one day a little devilish part of the application decides to collect some info about you, your computer, files, what have you and send all that info to an IP address somewhere on the Internet?

So, this thread started in 2004, 9 years ago. It seems to me,that at that time apparently there were no rough applications written for Linux but what is the case today in this matter? Is it is same? Rough elements out there want to harm only Windows users not Linux users?

Or has someone(s) finally decided to write a firewall app. for Linux that can alert the user when for example a music player application suddenly wants to transmit some data to a mysterious IP address for no apparent or obvious reason?

While using linux doesn't eliminate the risk of an application doing something untoward, the nature of open source means that hiding this type of capability is much harder.

For example, if you decided you like Amarok as a music player, you can download the source code yourself and go through it line by line and confirm it doesn't have any code that could be considered malicious. Compare this to Windows Media Player - the same process would be much harder.

No one is expecting you to do this, but most open source projects have many people involved, each contributing and reviewing code and so it would be difficult for one member to do anything malicious. It does happen however, but infrequently enough that it is big news when it does.

The risk is dramatically reduced compared to an operating system that is closed source, but open in the sense that you can install anything.

As far as I am aware, there isn't an application level firewall that can control access from individual apps, and this is in part because there is little demand for one.

To avoid dodgy applications, don't install applications from unknown sources (use the package manager).

Muff,

Thank you for your reply and your explanation.

If I understand it correctly the security of computers run on Linux is mostly based on not science or technology but trust.

It is a very interesting concept, and new for me, who has been using Windows since the early 90-ies, started with Windows 3.1. Right now I have a dual boot system and has been testing and tasting Linux to see if or when I could switch to it and abandon Windows for good.

I would be just curious for one thing. Here is a popular, not open source but free chat software for example. Recently, I was shocked to notice that one day ZoneAlarm popped up a window that said this software wants to launch Windows Explorer and it wants to connect to the Internet. I don't have a clue whether the same app. does similar things in Linux, I haven't installed it on Linux yet. Anyway, I clicked on "Deny" in the pop up window and I used the app for the original purpose as usual. But how an experienced Linux user would act in a similar scenario? I know, he wouldn't install this app because it is not an open source one, but sometimes one might have a justified need to install a non open source free software or one that you have to pay for.

So, am I on the right track when I assume that the only solution for situation like this in Linux is a Sandbox-like app. Is there a reliable one out there?