----- "Aaron Richton" <richton(a)nbcs.rutgers.edu&gt; wrote: I was about to reply the same, but you anticipated me :) I've tried the above, and it works as expected as soon as the "relay" statement is omitted. In fact, it requires the superior database to already exist. Probably, that test should either be relaxed or moved to db_open(). With respect to Guillaume's second question, I see the issue. In principle, what he needs to do is something like rwm-map attribute telephoneNumber homePhone rwm-map attribute * telephoneNumber so that real homePhone is mapped to virtual telephoneNumber, and real telephoneNumber is hidden. Unfortunately, this seems to result in a "double mapping" for telephoneNumber. I need to look at the logic of mapping, since what Guillaume wants to do seems to make sense, so it should be allowed. As per the reason of hiding everything not working, it's related to the fact that hiding everything does not allow "objectClass" to be returned, which causes the search filter to fail because the objectClass is not present. Unfortunately, the objectClass attribute cannot be mapped, so it cannot be explicitly preserved by adding rwm-map attribute objectClass * I recommend he files an ITS for each of those two issues. Ing. Pierangelo Masarati OpenLDAP Core Team SysNet s.r.l. via Dossi, 8 - 27100 Pavia - ITALIA http://www.sys-net.it ----------------------------------- Office: +39 02 23998309 Mobile: +39 333 4963172 Fax: +39 0382 476497 Email: ando(a)sys-net.it -----------------------------------