News for hackers. Stuff that matters.
Re:Attacking Unix Systems via Cups, Part I by stevekemp on Thursday September 26, 2024 @10:36PM (#41665733)
Binding to ports under 1024 traditionally requires root privileges. (These days of course that isn't quite as true as it used to be.)
Re:Attacking Unix Systems via Cups, Part I by amluto on Friday September 27, 2024 @12:05AM (#41666142)
Why on Earth do ordinary systems need CUPS binding to a port at all?
Re:Attacking Unix Systems via Cups, Part I by a96 on Friday September 27, 2024 @03:39AM (#41667402)
They don't need. Want, for listening to network printers announcing themselves. It's a very bad system, even then.
Re:Attacking Unix Systems via Cups, Part I by amluto on Friday September 27, 2024 @10:03AM (#41670631)
Have you ever seen a network printer announce itself by talking to a print daemon on port 631?

I’ve seen network printers announce themselves over DNS-SD/mDNS and over NetBIOS, AppleTalk, etc. All of those are a layer beneath the print daemon.

Re:Attacking Unix Systems via Cups, Part I by yrro on Friday September 27, 2024 @12:02PM (#41672206)
They don't. cups-browsed is a legacy component that isn't needed on ordinary systems, which outsource printer discovery to an mDNS service such as avahi.
Re:Attacking Unix Systems via Cups, Part I by ronjakoi on Friday September 27, 2024 @01:14AM (#41666497)
It's common practice to open the socket to start listening on the <1024 port, then drop the root privileges and continue as a different user.
Re:Attacking Unix Systems via Cups, Part I by cpach on Friday September 27, 2024 @08:17AM (#41669537)
On a modern Linux system, it’s better to use the CAP_NET_BIND_SERVICE capability instead. Then you don’t need to start the process as root at all.