[EXTERNAL] Re: Weird issue with file permissions

[Ken Brown]

On 7/2/2022 3:37 PM, Lavrentiev, Anton (NIH/NLM/NCBI) [C] wrote:
>> what your test program was actually doing.  But you seem to be assuming that
>> calling fchmod on a socket descriptor should affect the permissions on the
>> socket file (assuming the socket is bound).  Is that documented anywhere?  POSIX
>> says that the behavior of fchmod on a socket descriptor is unspecified
> 
> The socket file descriptor for a bound UNIX sockets refers to an object in a filesystem
> (it's practically a file), which the bind() system call creates.  The access to the socket
> is controlled by the permission bits, when someone actually tries to connect to it,
> so permissions should be working for these objects (otherwise, there's no other way!)
> 
> And fchmod() for a bound Unix socket works on Linux and many other Unix flavors, actually.

That's not what I'm seeing when I run your test program on Linux:

$ ./sun
fstat mode = 140666
stat mode = 140777

$ ls -l .socket
srwxrwxrwx. 1 kbrown kbrown 0 Jul  2 17:47 .socket=

So calling fchmod on the socket descriptor did not change the permissions of the 
file to which the socket was bound.

And on freeBSD, calling fchmod on a socket descriptor is apparently an error:

   https://www.freebsd.org/cgi/man.cgi?query=fchmod&sektion=2&n=1

Ken