On Fri, Oct 03, 2008 at 05:25:51PM -0500, Nicolas Williams wrote:
> I'm investigating a possible bug in unresolvable SID ID mapping.
You can observe the bug I'm investigating like so:
% idmap show -c sid:S-1-5-1-2-3-4-1
sid:S-1-5-1-2-3-4-1 -> uid:2162761759
% idmap show -c sid:S-1-5-1-2-3-4-1
sid:S-1-5-1-2-3-4-1 -> uid:2162761760
% idmap show -c sid:S-1-5-1-2-3-4-1
sid:S-1-5-1-2-3-4-1 -> uid:2162761761
%
Whereas:
% idmap show -c usid:S-1-5-1-2-3-4-2
usid:S-1-5-1-2-3-4-2 -> uid:2162761764
% idmap show -c usid:S-1-5-1-2-3-4-2
usid:S-1-5-1-2-3-4-2 -> uid:2162761764
% idmap show -c usid:S-1-5-1-2-3-4-2
usid:S-1-5-1-2-3-4-2 -> uid:2162761764
%
Basically, if the caller doesn't know the type of a given unresolvable
SID (and why should they?) then you get a different UID/GID every time.
Actually, that happens even if you claim it's a group ('gsid').
When that happens only the last UID/GID returned then resolves back to
the original SID.
Nico
--
_______________________________________________
cifs-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/cifs-discuss
