Ron Artstein wrote:
What the FMActually, I don't think it's the fine manual's fault at all. Mount is mount, permissions are permissions.
needed was a line saying that in order to run a file it has to have
execute permissions *and be on a device mounted with the exec flag*.
What I think should happen is that, when a partition is mounted "noexec", it should do just that - no exec. Even if the filesystem itself has the x flag on, it should not appear to be on.
Had that been the situation: Try to run the program Get "permission denied". Do "ls" - see "-rw-rw-r--". Do "chmod a+x a.out". Try to run the program. Get "permission denied" again. Do "ls" - see "-rw-rw-r--". Realize that something is preventing the "x" flag from being set. Remeber "noexec" mount option.
If i would continue your line of thought, why don't just display to the user *her* effective permissions to the file?
An "ls" will display just "-r--". Who cares that some group has write permission? Or that the owner has execute permission. Why even show who the file owner is? you don't really need it.
Nah, that's not the Unix Way (tm).
Besides, even if the user has a file system mounted as noexec, she should be able to change permissions as she sees fit (e.g. chmod ug+x myscript.pl). If she chose to remount later, with different flags, then those changes could mean something.
Sometimes, diagnosing things need a leap of thought. It will always require that. :)
Shachar.
-------------------------------------------------------------------------- Haifa Linux Club Mailing List (http://www.haifux.org) To unsub send an empty message to [EMAIL PROTECTED]
