If you have the Brave Browser installed on your Windows devices, then you may also have Brave VPN services installed on the machine. Brave installs these services without user consent on Windows devices.
Brave Firewall + VPN is an extra service that Brave users may subscribe to for a monthly fee. Launched in mid-2022, it is a cooperation between Brave Software, maker of Brave Browser, and Guardian, the company that operates the VPN and the firewall solution. The firewall and VPN solution is available for $9.99 per month.
I don’t use Windows but if you install a program that requires a service on Linux, the service will be written to your system’s services daemon awaiting your activation. I don’t see what the issue with that is.
What’s to stop the installer on Linux from configuring the service such that the service always runs on boot? e.g.
systemctl enable malware.service.You still need to manually enable the service. The configuration of the service has zero effect on its activation or lifecycle.
Huh? Any script can create a service, enable it and then start it. What would make you think the brave package (or just the application itself) can’t do this?
Not possible to start or enable a created service without user intervention. You don’t know what you are talking about.
deleted by creator
Bruh you just ran the command to enable the ‘written’ service. Comprehension is a problem in this community.
deleted by creator
OK… challenge accepted. Maybe you don’t know about systemd user services.
Content of
mytrojan.sh:#!/usr/bin/env bash echo "Writing the service unit file" cat > ~/.config/systemd/user/my_test_service.service << EOF [Unit] Description=Script Daemon For Test User Services [Service] Type=simple User= #Group= ExecStart=/home/user/bin/myscript.sh Restart=on-failure StandardOutput=file:%h/log_file [Install] WantedBy=default.target EOF echo "Reloading systemd for the user" systemctl --user daemon-reload || exit 1 echo "Enabling and starting the service" systemctl --user enable --now my_test_service.serviceContent of
myscript.sh:$ cat ~/bin/myscript.sh #!/usr/bin/env bash while true do now=$(date) me=$(whoami) echo "User $me at $now" sleep 10 doneNow run the script (
mytrojan.sh) and check service status after that:$ ./mytrojan.sh Writing the service unit file Reloading systemd for the user Enabling and starting the service $ systemctl --user status my_test_service.service ● my_test_service.service - Script Daemon For Test User Services Loaded: loaded (/home/user/.config/systemd/user/my_test_service.service; enabled; vendor preset: ena> Active: active (running) since Thu 2023-10-19 12:15:21 EEST; 6s ago Main PID: 1666383 (myscript.sh) Tasks: 2 (limit: 18757) Memory: 556.0K CPU: 4ms CGroup: /user.slice/user-1000.slice/user@1000.service/app.slice/my_test_service.service ├─1666383 /bin/bash /home/user/bin/myscript.sh └─1666387 sleep 10 Oct 19 12:15:21 tesla systemd[1866318]: Started Script Daemon For Test User ServicesYou failed. This requires the user to run a script aka manual intervention.
I thought that you only were ignorant, but no, you’re more than that!
Maybe am ignorant but at least I understand the questions before I answer them.
Systemd “enabled” services are literal symlinks… whenever a target runs, it tries to start also all the service files on its “wants” directory.
You can literally enable any service for next boot by making a symlink in
/etc/systemd/system/multi-user.target.wants/(or whichever other target you want it to run on) as root (and installation scripts are run as root).ln -s /usr/lib/systemd/system/whatever.service /etc/systemd/system/multi-user.target.wants/whatever.serviceThis is actually very close (just tested and confirmed it). I somehow stand corrected about requiring manual enablement but this is just using the package manager to do the dirty work for you.
However the program itself cannot write into those directories without root permissions. You still have to allow your package manager to do this with root permissions as mentioned.
Linux doesn’t have “installers” as Linux uses package managers. The only way you can get malware is if you manually add a bad repo.
So it doesn’t really matter in the long run
deleted by creator
Are you really serious making this claim? lol.
Yes, prove me wrong. As long as your running a up to date system there shouldn’t be anything that could be easily compromised.
I’ve been using Linux (and UNIX) professionally since the kernel version started with a “1.” I have no need to try to prove anything to you. Linux has installers other than just those invoked by a package manager, and it is laughable that you claim otherwise.