In the Linux kernel, the following vulnerability has been resolved:
exec: Fix ToCToU between perm check and set-uid/gid usage
When opening a file for exec via do_filp_open(), permission checking is
done against the file's metadata at that moment, and on success, a file
pointer is passed back. Much later in the execve() code path, the file
metadata (specifically mode, uid, and gid) is used to determine if/how
to set the uid and gid. However, those values may have changed since the
permissions check, meaning the execution may gain unintended privileges.
For example, if a file could change permissions from executable and not
set-id:
---------x 1 root root 16048 Aug 7 13:16 target
to set-id and non-executable:
---S------ 1 root root 16048 Aug 7 13:16 target
it is possible to gain root privileges when execution should have been
disallowed.
While this race condition is rare in real-world scenarios, it has been
observed (and proven exploitable) when package managers are updating
the setuid bits of installed programs. Such files start with being
world-executable but then are adjusted to be group-exec with a set-uid
bit. For example, "chmod o-x,u+s target" makes "target" executable only
by uid "root" and gid "cdrom", while also becoming setuid-root:
-rwxr-xr-x 1 root cdrom 16048 Aug 7 13:16 target
becomes:
-rwsr-xr-- 1 root cdrom 16048 Aug 7 13:16 target
But racing the chmod means users without group "cdrom" membership can
get the permission to execute "target" just before the chmod, and when
the chmod finishes, the exec reaches brpm_fill_uid(), and performs the
setuid to root, violating the expressed authorization of "only cdrom
group members can setuid to root".
Re-check that we still have execute permissions in case the metadata
has changed. It would be better to keep a copy from the perm-check time,
but until we can do that refactoring, the least-bad option is to do a
full inode_permission() call (under inode lock). It is understood that
this is safe against dead-locks, but hardly optimal.
References
In the Linux kernel, the following vulnerability has been resolved:
exec: Fix ToCToU between perm check and set-uid/gid usage
When opening a file for exec via do_filp_open(), permission checking is
done against the file's metadata at that moment, and on success, a file
pointer is passed back. Much later in the execve() code path, the file
metadata (specifically mode, uid, and gid) is used to determine if/how
to set the uid and gid. However, those values may have changed since the
permissions check, meaning the execution may gain unintended privileges.
For example, if a file could change permissions from executable and not
set-id:
---------x 1 root root 16048 Aug 7 13:16 target
to set-id and non-executable:
---S------ 1 root root 16048 Aug 7 13:16 target
it is possible to gain root privileges when execution should have been
disallowed.
While this race condition is rare in real-world scenarios, it has been
observed (and proven exploitable) when package managers are updating
the setuid bits of installed programs. Such files start with being
world-executable but then are adjusted to be group-exec with a set-uid
bit. For example, "chmod o-x,u+s target" makes "target" executable only
by uid "root" and gid "cdrom", while also becoming setuid-root:
-rwxr-xr-x 1 root cdrom 16048 Aug 7 13:16 target
becomes:
-rwsr-xr-- 1 root cdrom 16048 Aug 7 13:16 target
But racing the chmod means users without group "cdrom" membership can
get the permission to execute "target" just before the chmod, and when
the chmod finishes, the exec reaches brpm_fill_uid(), and performs the
setuid to root, violating the expressed authorization of "only cdrom
group members can setuid to root".
Re-check that we still have execute permissions in case the metadata
has changed. It would be better to keep a copy from the perm-check time,
but until we can do that refactoring, the least-bad option is to do a
full inode_permission() call (under inode lock). It is understood that
this is safe against dead-locks, but hardly optimal.
References