With the increase popularity of the linux desktop and the steamdeck, will new viruses and malwares be developed for linux systems? should we better use an antivirus?
Most modern antivirus software is a virus in and of itself.
We always say that McAfee uses all the system resources so the virus doesn’t have anything to work with.
At home I have MS defender turned on by default on my windows machine. I was copying the contents of one nvme to another the other day and noticed I was only getting 60MiB/sec. I looked at task manager, realized why, turned off proactive scanning of files, and watched it jump to over 2GiB/sec. Really nailed that point home.
I don’t use antivirus on Windows, let alone Linux. Just be mindful of what you’re downloading and you’ll be fine.
Windows has done of the best antivirus built-in
Same here. Ever since I remember I don’t have AV, just staying out of sketchy sites is enough. Most of malware is targeted at the least aware and cautious users.
No, but I follow good security practices instead:
- sandboxing apps where possible using flatpak/bubblewrap or bubblejail which makes bubblewrap convenient.
- opensnitch to block internet access for certain apps or restricting them to specific domains/ip’s. E.g if i install a chatgpt client, it should only contact openai/chatgpt, if it later pushed an update stealing my credentials, this wouldn’t let it.
- i use ublock, google safebrowsing on firefox (because checks are done locally in a privacy respecting way), and pihole with hagezi’s lists.
I also use fedora which has good defaults, in that it uses selinux/wayland out of the box.
when im on windows i use a similar set of tools, sandboxie plus for sandboxing, and a combination of simplewall + portmaster as an opensnitch alternative, the latter provides more fine grained control and simplewall just gives the allow/block popup for apps.
You can also do fun stuff like running the app as a separate user, and using sudo+xhost to access it. Most Linux firewalls allow you to set a rule with uid-affinity, so you could restrict the additional user to a whitelist of IP’s.
Antivirus is a technical attempt at solving a stupid user problem. It does not actually prevent any problems and causes many of its own.
- run only what you need
- get what you need only from trusted sources
- keep what you need up to date
- configure what you need conservatively
- admin/root account only for admin stuff. Don’t use root as your general login.
Frankly, some phishing attempts, especially at work, are pretty good in my opinion.
I don’t because no antivirus can protect you from yourself, I learned that the hard way while I was still using Windows many years ago.
I had antivirus and antimalware on Windows and I kept them updated and ran scans regularly, almost religiously, than I got a malware that antivirus couldn’t detect (they take a while to be updated with new viruses/malware that constantly come out) and the only way to remove it was a blank new install following a specific procedure to clean it all.
That day I understood that no matter how attentive you are about your antivirus, you’re never really protected until it’s you who learns what to do and what not while accessing the web, so I did a bit of research about how to better configure my PC, how to better recognize phishing/scamming, using adblocker, don’t download random software, stuff like that.
I basically learned “how to behave”. I stopped using antivirus/antimalware on Windows and I didn’t catch a virus since, then I switched to Linux and I still apply the same principles of “good behavior”.
That’s not to say I’m immune to viruses, I’m certainly not, but my mindset now is that if I ever catch a virus, it will be my fault for doing something I shouldn’t have done, and I’ll do my best to learn from it instead of relying on software to do that for me.
Run sandboxed software when possible and avoid doing unsafe things.
There is no useful AV for Linux desktops.
- Virus scanners only detect a fraction of the harmful programmes.
- Virus scanners can often be tricked.
- Virus scanners often have security vulnerabilities themselves, which are usually quite serious, since such programmes embed themselves quite deeply in the operating system.
- Virus scanners cause many users to become careless because they rely too much on such tools.
Therefore, from my point of view, the disadvantages outweigh the advantages. Therefore, I do not have such a tool permanently installed, neither under Linux nor under Windows. However, every 6 months I scan my Windows installation with a USB-bootable virus scanner. No actually harmful programme has been found for years.
In my opinion, the following things are much more important than any security software.
- Install updates as soon as possible. Under Windows, you can use tools like Chocolatey for this.
- Only install software from trustworthy sources.
- Only install software that you really need.
- Only use root or administrator rights if you have to. For everything else, the rights of the user account are sufficient.
- Create backups regularly.
- Think before you act.
Especially the last point is a problem for many users. I can’t tell you how many times I’ve witnessed someone receiving an alleged invoice from mobile provider A by email and opening it, even though they had a contract with provider B.
Haven’t really felt the need to. On Linux ad blocking + common sense has worked out fine. When I was still using Windows I just relied on Windows Defender since around the Windows 8.1 days, but either way my time downloading .exe files from sketchy sites is long behind me.
I don’t, but that’s because of how I have things setup around the network. While most people here say it’s because they don’t need it, I am in a position that I need SOMETHING simply because others in the household could bring in malware and rather than trust them to make smart decisions, I proactively monitor all network activity for anything unusual. That being said, I have clamAv installed and run a weekly scan, but my real “antivirus solutions” are as follows: A syslog server that’s connected to grafana/MySQL and alerts me based on very specific criteria. Along this, I’ve got my network firewall configured to block all “untrustworthy countries” in and out 100%, as well as use an IDS/IPS (also connected to syslog for alerts). Lastly, an internal DNS which grabs from like 20 sources that include some reliable lists with malware domains and such, and a custom list of my own that I add to as useful security news feeds hit my RSS feed with urls in their blog posts.
Actually got a list of other things going on in the network to make it even more secure, but just wanted to list the main things that’ll give you a step up in the anti-malware front.
None of this is buletproof without proper care for how you use the Internet, though.
- Check for router/modem/firewall updates weekly if they can’t be auto updated
- never click any links in an email even if you feel you know you trust it (exception to this would be something you KNOW is coming into your inbox, such as an account registration verification)
- avoid tiny urls or suspicious looking urls when possible
- don’t open ports unless you really really really know what you’re doing. If you absolutely need to open a port, then for the love of god define the source IP address/CIDR. Opening ports to the world includes opening them to cyber criminals
- turn off upnp, I don’t care what that game or service you use says, it doesn’t need it
You get the idea. My message got way too long and turned into a rant lol.
Nope… Raw dawg that shit online.
I wonder how people get phished on IG and FB. Like the get account taken over and blackmailed over a social media account… That shit baffles me. I’m pretty sure I know how they get phished but I’m just shocked it happens to smart people.
Smart people are not immune to moments of panic or laziness or cockiness. I don’t know about you, but I don’t always check email headers even tho that’s the closest to best way to verify the identity of the sender. And if that link verifiably goes to a website I trust, and I was expecting them to reach out, and I just have to login to check my orders and… wait, why does the url have a “redir=” parameter? Oh fuck oh god oh fuck why does the login page say “amzaon.com” instead of “amazon.com” like in the email’s link??? FUCK DAMMIT SHIT
And if that link verifiably goes to a website I trust, and I was expecting them to reach out, and I just have to login to check my orders and… wait, why does the url have a “redir=” parameter? Oh fuck oh god oh fuck why does the login page say “amzaon.com” instead of “amazon.com” like in the email’s link??? FUCK DAMMIT SHIT
This is definitely a situation where having a password manager with auto-filling is nice. When you save your login for
amazon.comit ties it to the URL as well. So if you end up going toamzaon.comby any means and don’t manage to catch it, your password manager won’t fill in your details because it doesn’t recognize the domain.Of course, this won’t stop you from say, using one of the “Login with Google/Apple/Amazon/etc” buttons on some dodgy website, and granting it access to your account (because you’d be redirected to
google.com/apple.com/amazon.com) but it’s at least an layer of “Wait, something isn’t right here” when the auto-fill doesn’t trigger.Password managers are an absolute must-have in this day and age. That and MFA. And making as few accounts as humanly possible.
But, the more general concepts I’m trying to get at are that pobody’s nerfect, you don’t know what you don’t know you don’t know, and we’re all just apes prone to lapses in judgment at innoportune times.
you don’t know what you don’t know you don’t know, and we’re all just apes prone to lapses in judgment at innoportune times.
Oh for sure, I 100% agree! My reply was more of an educational “Hey, in case you’ve run into this before, this is a great way to prevent it from occurring again” sort of deal. No one is born with all-encompassing knowledge of the world and everything/anything they could ever interact with, and subsequently no one should be faulted for running into something like phishing scams where they’re designed to exploit someone’s potential lack of knowledge or even as you mentioned, a lapse in judgment.
I normally am good about avoiding phishing scams but almost fell victim to one because a close trusted friend of mine had their account compromised, and sent a link to something on Steam that seemed in line with what they’d normally bring up with me - and it was exactly the fact that my password manager didn’t prompt me to fill in my Steam login details on that fake page that prevented me from trying to login.
(Well that and I do have Steam Guard/MFA enabled, but still)
No. Not needed for the most part if you aren’t downloading dumb shit on the internet and keeping things up to date.
Nope.
Why
I don’t have wine installed.
Even then, the good old “Reinstall if things feel sluggish” always helps.
You dont need an antivirus if you don’t run software from unknown sources, and you keep your machine up to date.
For the most part, if you do all that and you’re still attacked, they would probably get past an antivirus, anyways. I’ve never had an antivirus catch anything after decades of running Windows.
A bit thing to note for people running home servers is to watch what they expose to the internet. Insecure software and insecure configurations are huge targets for botnets these days.
deleted by creator
Linux still can get viruses. Just not as likely
The pattern of using managed repositories instead of downloading random shit off the internet prevents a lot of it outright.
Just have to remember which repos are “managed” and which are not. Installing stuff from PyPi or NPM might actually require you to read through quite a lot of code before installing. I don’t think most people are that diligent.
yup, that’s why i avoid it like the plague.
It’s .deb’s and .rpm’s all the way down.
And sometimes flatpaks. And sometimes AppImages.
But never pips, gems or any of that sort of …
Yup, and much of stuff is just common apps. A browser based exploit may not care which OS you’re running Firefox/Chrome under (or Edge, I suppose if you’re one of the dozen running that in Linux). The log4j vulnerability was first seen in Minecraft hosts and clients, and that equally affects those running on Linux. Steam has also had vulnerabilities in the past, as have various document editors.
That said, I really can’t say there’s any consumer-level antivirus that I’d trust to be effective on my desktop (especially without impacting performance) even on Windows, let alone Linux. If you really know what your box is doing and it doesn’t change much you might consider SELinux rules etc, but those are a significant amount of work even for servers (which tend to behave more predictably).
Not using software from untrustworthy sources or visiting sketchy sites, plus some monitoring of your network traffic may be a more effective solution.
This (Also is there a service that audits code’s maliciousness for me? I sometimes use some unfamous stuffs from github for gaming, customizing etc. As long as I don’t give them a root permission I should be ok?)
There are tools that do this but it’s mostly for known issues or vulnerabilities (i.e. bad practice in coding or done after the fact). Some stuff runs code on at sandbox looking for sketchy behavior, but hackers may work around that by having malware that only activates in CD l specific circumstances or times.
There can also be “privilege escalation” exploits which can allow a rogue process to obtain root/admin, and even without the process would have full access to whatever the user does in terms of network/device/hardware/process access (or sudo, as applicable).
Not being able to run as root doesn’t help all that much if the process is able to access sensitive saved data, files and/or the internet
deleted by creator
I asked 😺👍
deleted by creator
I didn’t start using Linux in this decade, the last, or the one before that
Did I ask?
The duality of the man, so old but so edgy.
deleted by creator
No one cares
deleted by creator
I usually haven’t, but I installed Clam about a month ago on my desktop, ran a full and complete scan, then left it running scheduled scans. Hasn’t found anything, and I get a lot of software from outside of my package manager, and use wine for a lot of it, so I’d say my risk/exposure is higher than most.
I think it’s fine to go without AV on a linux desktop, but I like the peace of mind. There will definitely be more things targeting linux sytems as/if more market share is acquired, but in terms of security it’s more important that you harden the system than run an AV.
