On Friday 22 of January 2010 17:03:51 Jim Meyering wrote: > Can you reproduce the situation in which this new configure-time > option is required on e.g., Fedora 12? So far, I haven't been able to, > since all tty devices properly get the "tty" group, and in that > case, the simple perm-comparison test is adequate.
What I am getting on a sane F-12 installation follows: # ls -l /dev/tty? crw--w----. 1 root root 4, 0 2010-01-22 18:48 /dev/tty0 crw--w----. 1 root root 4, 1 2010-01-22 18:48 /dev/tty1 crw--w----. 1 root tty 4, 2 2010-01-22 18:50 /dev/tty2 crw-------. 1 root root 4, 3 2010-01-22 18:48 /dev/tty3 crw-------. 1 root root 4, 4 2010-01-22 18:48 /dev/tty4 crw-------. 1 root root 4, 5 2010-01-22 18:48 /dev/tty5 crw-------. 1 root root 4, 6 2010-01-22 18:48 /dev/tty6 crw--w----. 1 root tty 4, 7 2010-01-22 18:48 /dev/tty7 crw--w----. 1 root tty 4, 8 2010-01-22 18:48 /dev/tty8 crw--w----. 1 root tty 4, 9 2010-01-22 18:48 /dev/tty9 > I realized that I didn't really remember the details, Nor do I as wasn't anyhow involved in GNU coretuils when the issue was discussed ;-) > and NEWS didn't tell me enough, so started rewriting it: > > who: the "+/-" --mesg (-T) indicator of whether a user/tty is accepting > messages could be incorrectly listed as "+", when in fact, the user was > not accepting messages (mesg no). Before, who would examine only the > permission bits, and not consider the group of the TTY device file. > Thus, when a tty file's group would change somehow e.g., to "root", > that would make it unwritable (via write(1)) by normal users, in spite > of whatever the permission bits might imply. Now, when configured > using the --with-tty-group[=NAME] option, who also compares the group > of the TTY device with NAME (or "tty" if no group name is specified). > How can the group name change (or be set initially) to other than "tty"? > I suspect this is an issue only on older systems. If so, I need to > say that in NEWS, rather than "somehow". It looks like sort of default state to me. To be frank, not sure if intended and/or documented actually :-) > Other than that, everything looks fine (I moved the "const"). I knew I had forgotten something. Sorry about that. It wasn't intentionally. Kamil