hdb_search: does not match filter

Hello, 
I'm running openldap 2.3.43 on CentOS.
I'm tring to setup a working master-slave sync system.
I populate master server using slapadd -q -l from a ~160k entry ldif file. Every time I (re)start master server it takes long time to become available (about 20 minutes), in the logs I read these entries:

Nov 10 13:56:17 ldap01 slapd[2191]: hdb_search: 1 does not match filter
Nov 10 13:56:17 ldap01 slapd[2191]: entry_decode: ""
Nov 10 13:56:17 ldap01 slapd[2191]: <= entry_decode()
Nov 10 13:56:17 ldap01 slapd[2191]: hdb_search: 2 does not match filter
Nov 10 13:56:17 ldap01 slapd[2191]: entry_decode: ""
Nov 10 13:56:17 ldap01 slapd[2191]: <= entry_decode()
Nov 10 13:56:17 ldap01 slapd[2191]: hdb_search: 3 does not match filter
Nov 10 13:56:17 ldap01 slapd[2191]: entry_decode: ""
...
...
...
Nov 10 14:17:58 ldap01 slapd[32246]: hdb_search: 162959 does not match filter
Nov 10 14:17:58 ldap01 slapd[32246]: entry_decode: ""
Nov 10 14:17:58 ldap01 slapd[32246]: <= entry_decode()
Nov 10 14:17:58 ldap01 slapd[32246]: send_ldap_result: conn=-1 op=0 p=0
Nov 10 14:17:58 ldap01 slapd[32246]: slapd starting

If I try to comment the syncprov section:

overlay syncprov
syncprov-checkpoint 100 10
syncprov-sessionlog 100

in slapd.conf the server startup is very quick (2-3 seconds).

Is there something wrong in what I'm doing? Is this very long startup normal?

Thanks, 
G.