This is the mail archive of the
mailing list for the Cygwin project.
Re: id / mkpasswd 
On Fri, Jan 04, 2002 at 11:09:57AM +0100, firstname.lastname@example.org wrote:
>> Nope. The datatype uid_t is equivalent to unsigned short (16bit).
>> The mkpasswd tool doesn't take that into account when generating
>> the uids, unfortunately.
>So, it's just a storage problem of the uid...
>In my memory, the unix datatype uid_t is a long ? no ?
>In the case I've mentioned yesterday, if an other user
>already has the id 43833 (which is probably the case),
>Is there a way to log, in the same time, him on his
>own account and me on mine ?
>I've seen that mkpasswod works with int, the
>value returns by NetUserGetInfo is a DWORD -> ok
>(-long would be better (platform-independent), but...-)
>mkpasswd use the entire information NT returns so :
>Why 32 bits have not been chosen for uid_t too ?
Because we're mean.
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html