Migrate from openldap 2.2 to 2.4 issue

[DEVARIEUX Alain <alain.devarieux@sib.fr>]

Hi!

First excuse me for my approximative english.

I'm trying to migrate from an old Redhat server running openldap 2.2 to 
a brand new one using Centos 7 and openldap 2.4.
Using slapcat / sldapadd I can't have my new server running with an olc 
config.

I'd like to know what I'm doing wrong during this process :

# To remove entryUUID lines because they're not usable with openldap 2.4
sed -i -e "/entryUUID/d" /root/myslapcat.ldif

# Running slapadd with a 'cleaned' version of my old slapd.conf
slapadd -f /tpm/oldserver/slapd.conf -F /etc/openldap/slapd.d/ -c -u -o 
schema-check=yes -l /root/myslapcat.ldif

# moving from file configuration to olc :
slaptest -f /tmp/oldserver/slapdb.conf -F /etc/openldap/slapd.d/

# now, I can start the service without problem :
systemctl start slapd

But, when I try to access the diectory, here is the error messages I have :
Oct 29 16:02:57 ldap01-qualif slapd[12351]: conn=1002 fd=11 ACCEPT from 
IP=10.35.100.87:49238 (IP=0.0.0.0:389)
Oct 29 16:02:57 ldap01-qualif slapd[12351]: conn=1002 op=0 BIND 
dn="cn=Manager,dc=mydomain,dc=fr" method=128
Oct 29 16:02:57 ldap01-qualif slapd[12351]: conn=1002 op=0 BIND 
dn="cn=Manager,dc=mydomain,dc=fr" mech=SIMPLE ssf=0
Oct 29 16:02:57 ldap01-qualif slapd[12351]: conn=1002 op=0 RESULT tag=97 
err=0 text=
Oct 29 16:02:57 ldap01-qualif slapd[12351]: conn=1002 op=1 SRCH 
base="dc=mydomain,dc=fr" scope=1 deref=0 filter="(objectClass=*)"
Oct 29 16:02:57 ldap01-qualif slapd[12351]: conn=1002 op=1 SRCH 
attr=objectclass
Oct 29 16:02:57 ldap01-qualif slapd[12351]: bdb(dc=mydomain,dc=fr): 
BDB0060 PANIC: fatal region error detected; run recovery
Oct 29 16:02:57 ldap01-qualif slapd[12351]: bdb(dc=mydomain,dc=fr): 
BDB0060 PANIC: fatal region error detected; run recovery
Oct 29 16:02:57 ldap01-qualif slapd[12351]: bdb(dc=mydomain,dc=fr): 
BDB0060 PANIC: fatal region error detected; run recovery
Oct 29 16:02:57 ldap01-qualif slapd[12351]: bdb(dc=mydomain,dc=fr): 
BDB0060 PANIC: fatal region error detected; run recovery
Oct 29 16:02:57 ldap01-qualif slapd[12351]: conn=1002 op=1 SEARCH RESULT 
tag=101 err=80 nentries=0 text=internal error

But, if a launch slapd like telling it the configuration file to use, 
everything works well (ie : I can bind to server and view all entries)

slapd -u ldap -f /tmp/oldserver/slapd.conf


Am I missing something obvious? I'm new to openldap...

Regards,
--
Alain Devarieux
Pôle Infrastructures
GIP SIB