I've configured a second host to replicate the main LDAP server via syncrepl in the slapd.conf:

syncrepl rid=666 
        provider=ldaps://my-main-server.com 
        type=refreshAndPersist 
        searchBase="dc=Staff,dc=my-main-server,dc=com" 
        filter="(objectClass=*)" 
        scope=sub 
        schemachecking=off 
        bindmethod=simple 
        binddn="cn=repadmin,dc=my-main-server,dc=com" 
        credentials=mypassword 

When I restart slapd, it writes to /var/log/debug

Jun 11 15:48:33 cluster-mn-04 slapd[29441]: @(#) $OpenLDAP: slapd 2.4.9 (Mar 31 2009 07:18:37) $ ^Ibuildd@yellow:/build/buildd/openldap2.3-2.4.9/debian/build/servers/slapd 
Jun 11 15:48:34 cluster-mn-04 slapd[29442]: slapd starting 
Jun 11 15:48:34 cluster-mn-04 slapd[29442]: null_callback : error code 0x14 
Jun 11 15:48:34 cluster-mn-04 slapd[29442]: syncrepl_entry: rid=666 be_modify failed (20) 
Jun 11 15:48:34 cluster-mn-04 slapd[29442]: do_syncrepl: rid=666 quitting 

I've looked into the sources for the return code and found only

#define LDAP_TYPE_OR_VALUE_EXISTS       0x14

in include/ldap.h. Anyway, I don't quite get what the error message means.

Can you help me debugging this problem and figure out why the LDAP replication doesn't work? I've managed to put a "manual" copy via slapcat and slapadd into the database, but I'd like to sync automatically.

UPDATE: "Solved" by removing /var/lib/ldap/* and re-importing the database with slapadd.