At 12:36 AM 3/14/2003 -0500, Christopher Faylor wrote: >On Thu, Mar 13, 2003 at 09:41:55PM -0500, Pierre A. Humblet wrote: >>On Thu, Mar 13, 2003 at 03:20:10PM -0000, Max Bowsher wrote: >>>We've had no proof of advantages (except in one very restricted corner >>>case), and no disproof of disadvantages (i.e. speed penalties). >> >>Running a speed test would be interesting of course, but I point to an >>advantage for most of us: with 32 bit uids the /var/log/lastlog file >>can become enormous. Fortunately (for systems that support sparse >>files), it is sparse. > >I think we'll have to come up with some other way of dealing with >/var/log/lastlog for Win 9x though, won't we?
Chris, 9x is not a real problem because there is no reason to use large uids (mkpasswd won't), although we can't stop users from trying. I am more concerned about old NTs without sparse file support in some corporate environments, where mkpasswd may introduce large uids. I don't fully understand when sparse file support was introduced. By the way, you are right that there is something funny about unlink with FILE_FLAG_DELETE_ON_CLOSE on Win9x, for read-only files. I don't see the problem in 1.3.21, but don't let it creep in an eventual 1.3.22 until further notice. Pierre -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
