I understand that antivirus software is necessary on Windows, but I’ll never understand the existence of Windows Defender. If Microsoft knows enough to prevent a virus that exploits something in windows, why are they putting their effort into an antivirus program, and not fixing ththe problem in Windows? If someone has a good explanation for this, I’d love to hear it.
ok so let’s start with the exploits. Exploit is a bug (problem) in a piece of software that when… umm… “abused” (well the word is just exploited) it allows you to do stuff that you shouldn’t. An exploit could be live from your browser to the program you use to zip files. The top 2 reasons to use an exploit is to either get initial foothold on a machine (e.g. an exploit in a browser that would allow an attacker to execute arbitrary code when you visit their page or an exploit in winrar that when you open a zip file executes code)
From the attackers perspective, you got in, nice. Mind you you got in through means that have nothing to do with windows (and that’s true most times, especially on desktops). but now? what?
You hacked into the machine for a reason! You might wanna grab the browser cookies (giving you direct access to the accounts that the victim is logged into), grab some files, screenshots, passwords
That’s where the AV kicks in. After the initial exploit the malware behaves like a normal program. But not completely. Assuming that the AV hasn’t seen the same exact malware before (which would an insta kick ban) it’s going to see a random process accessing files in chrome’s directory. HUH. ISNT THAT SOMETHING. quarantined.
Wanna start listening to each and every keystroke? quarantined
Meanwhile the way that the exe ended up in your system was not through an installer, you don’t provide an uninstaller and it was downloaded from www.xXxveryNicEsiteyou.got. HUUUUUUUH
the whole process is a bit simplified of course, but it captures the general idea
So why does linux not have an AV?
FUCK IF I KNOW! It would be very, VERY useful. Writing malware that bypasses AV is an art of its own. Can be done for sure, but it’s an extra step and it’s not fun
background: used to get paid to do shit like that (legally, pentest) and it’s a fun hobby (writing code around it, not hacking people)
I’ve never even considered ClamAV.
I have the idea that it’s just a malware signature DB (changing the signature of a binary is almost as simple as recompiling it with a bit different variables)
Am I incorrect? does it have heruistics/active scanning?
It is pretty exclusively a file scanner, but that, combined with Linux’s privilege separation, any decent firewall and not willfully executing untrusted files is enough for most cases, I would say.
what kind of privilege separation? you’re talking about containers/namespaces?
cause as it is linux desktop has 1 unprivileged user and that’s it. from an attackers perspective privilege escalation is irrelevant - you have access to the screen, keyboard, browser, files. there really is nothing left to gain from gaining root
and if you have any reason to gain root, it’s super easy by just replacing sudo with an alias in .bashrc you’ve got the user’s password
We REALLY need sandboxing and soon, that’s why I want to give fedora silverblue a try but my hopes are quite low
btw windows is in a bit of a better place and M1 mac is in much better place
Because… They are? Whenever there is a problem in Windows itself, they release an update to fix that ASAP.
Defender doesn’t just work against viruses that exploit weaknesses in Windows. It also works against viruses in programs the user installs. The purpose of Defender is the same as any other antivirus software, to detect known virus signatures in downloaded software, as well as attempt to detect programs that display virus-like behaviour. It also attempts to ensure that users only install software from sources they trust. For these purposes, Windows Defender is at least as good as most other antivirus software on the market.
I would also generally recommend using an antivirus program on a Linux/OSX machine, unless you really know the risk you’re accepting by not using one. Even then, I recommend occasionally running ClamAV or a Malwarebytes scan. There is a misconception of “there are no viruses for non-Windows platforms”, but the thing is that a lot of viruses these days are cross-platform compatible, and all it takes is one program or dependency becoming an infection vector. Keep yourselves safe, people!
I never got where the misconception of “*nix doesn’t have malware” came from. Maybe from the 2k era where “malware” was anything that was slowing down your PC (I also don’t get why a malware would slow down your PC, unless it’s a ransomware)?
I remember the c99.php shell from way back which is an amazing example of cross-platform (PHP can run anywhere) “virus” and it was considered a golden standard (2010 era?)
I understand that antivirus software is necessary on Windows, but I’ll never understand the existence of Windows Defender. If Microsoft knows enough to prevent a virus that exploits something in windows, why are they putting their effort into an antivirus program, and not fixing ththe problem in Windows? If someone has a good explanation for this, I’d love to hear it.
ok so let’s start with the exploits. Exploit is a bug (problem) in a piece of software that when… umm… “abused” (well the word is just exploited) it allows you to do stuff that you shouldn’t. An exploit could be live from your browser to the program you use to zip files. The top 2 reasons to use an exploit is to either get initial foothold on a machine (e.g. an exploit in a browser that would allow an attacker to execute arbitrary code when you visit their page or an exploit in winrar that when you open a zip file executes code)
From the attackers perspective, you got in, nice. Mind you you got in through means that have nothing to do with windows (and that’s true most times, especially on desktops). but now? what?
You hacked into the machine for a reason! You might wanna grab the browser cookies (giving you direct access to the accounts that the victim is logged into), grab some files, screenshots, passwords
That’s where the AV kicks in. After the initial exploit the malware behaves like a normal program. But not completely. Assuming that the AV hasn’t seen the same exact malware before (which would an insta kick ban) it’s going to see a random process accessing files in chrome’s directory. HUH. ISNT THAT SOMETHING. quarantined.
Wanna start listening to each and every keystroke? quarantined
Meanwhile the way that the exe ended up in your system was not through an installer, you don’t provide an uninstaller and it was downloaded from www.xXxveryNicEsiteyou.got. HUUUUUUUH
the whole process is a bit simplified of course, but it captures the general idea
So why does linux not have an AV? FUCK IF I KNOW! It would be very, VERY useful. Writing malware that bypasses AV is an art of its own. Can be done for sure, but it’s an extra step and it’s not fun
background: used to get paid to do shit like that (legally, pentest) and it’s a fun hobby (writing code around it, not hacking people)
I can recommend running ClamAV, if anyone is looking for a good one that runs on Linux.
I’ve never even considered ClamAV. I have the idea that it’s just a malware signature DB (changing the signature of a binary is almost as simple as recompiling it with a bit different variables)
Am I incorrect? does it have heruistics/active scanning?
It is pretty exclusively a file scanner, but that, combined with Linux’s privilege separation, any decent firewall and not willfully executing untrusted files is enough for most cases, I would say.
what kind of privilege separation? you’re talking about containers/namespaces?
cause as it is linux desktop has 1 unprivileged user and that’s it. from an attackers perspective privilege escalation is irrelevant - you have access to the screen, keyboard, browser, files. there really is nothing left to gain from gaining root
and if you have any reason to gain root, it’s super easy by just replacing sudo with an alias in .bashrc you’ve got the user’s password
We REALLY need sandboxing and soon, that’s why I want to give fedora silverblue a try but my hopes are quite low
btw windows is in a bit of a better place and M1 mac is in much better place
If you want sandboxing, isn’t firejail pretty exactly what you’re looking for?
Because… They are? Whenever there is a problem in Windows itself, they release an update to fix that ASAP.
Defender doesn’t just work against viruses that exploit weaknesses in Windows. It also works against viruses in programs the user installs. The purpose of Defender is the same as any other antivirus software, to detect known virus signatures in downloaded software, as well as attempt to detect programs that display virus-like behaviour. It also attempts to ensure that users only install software from sources they trust. For these purposes, Windows Defender is at least as good as most other antivirus software on the market.
I would also generally recommend using an antivirus program on a Linux/OSX machine, unless you really know the risk you’re accepting by not using one. Even then, I recommend occasionally running ClamAV or a Malwarebytes scan. There is a misconception of “there are no viruses for non-Windows platforms”, but the thing is that a lot of viruses these days are cross-platform compatible, and all it takes is one program or dependency becoming an infection vector. Keep yourselves safe, people!
I never got where the misconception of “*nix doesn’t have malware” came from. Maybe from the 2k era where “malware” was anything that was slowing down your PC (I also don’t get why a malware would slow down your PC, unless it’s a ransomware)?
I remember the c99.php shell from way back which is an amazing example of cross-platform (PHP can run anywhere) “virus” and it was considered a golden standard (2010 era?)
I think it’s born from a misinderstanding of infection statistics, especially back when windows was also more popular on servers.