Hello, yes "dies" means shutdown
I have made some changes in my config file, but the problem is the same, now I'm with openladap-2.4.11.
Here my config file:
Master A
slapd.conf

# Global Directives:
# Schema and objectClass definitions
include         /etc/ldap/schema/core.schema
include         /etc/ldap/schema/cosine.schema
include         /etc/ldap/schema/nis.schema
include         /etc/ldap/schema/inetorgperson.schema
include         /etc/ldap/schema/samba.schema

pidfile         /var/run/slapd/slapd.pid
argsfile        /var/run/slapd/slapd.args
loglevel       256
sizelimit 500
tool-threads 1

backend bdb

database bdb
suffix  "dc=ar"

overlay syncprov
syncprov-sessionlog 100
syncprov-nopresent TRUE
syncprov-reloadhint TRUE
syncprov-checkpoint 1 1
 
rootdn          "cn=admin,dc=dominio"
rootpw          secret

directory       "/usr/local/var/openldap-data"

dbconfig set_cachesize 0 2097152 0

dbconfig set_lk_max_objects 1500
dbconfig set_lk_max_locks 1500
dbconfig set_lk_max_lockers 1500

index           cn,sn,uid           pres,sub,eq
index           uidNumber,gidNumber             eq
index           sambaSID                        eq
index           sambaPrimaryGroupSID            eq
index           sambaDomainName                 eq
index           objectClass                     pres,eq

lastmod         on
 
access to attrs=userPassword,shadowLastChange
         by dn="cn=admin,dc=ar" write
         by anonymous auth
         by self write
         by * none
 
access to *
         by dn="cn=admin,dc=ar" write
         by * read

access to *
         by dn.base="cn=admin,dc=ar" read
         by * break
 
serverID 3

syncrepl rid=001
         provider=ldap://10.100.130.164
         type=refreshAndPersist
         retry="5 5 300 +"
         searchbase="dc=ar"
         attrs="*"
         bindmethod=simple
         binddn="cn=admin,dc=ar"
         credentials=osde

syncrepl rid=002
         provider=ldap://10.100.130.181
         type=refreshAndPersist
         retry="5 5 300 +"
         searchbase="dc=ar"
         attrs="*"
         bindmethod=simple
         binddn="cn=admin,dc=dominio"
         credentials=secret

mirrormode TRUE

##Accesslog overlay
overlay accesslog
logdb "cn=accesslog"
logops writes
logold (objectclass=auditModify)
logoldattr reqOld
logold (objectclass=auditDelete)
logoldattr reqOld
logsuccess TRUE
logpurge 2+00:00 06:00

##AccessLog DB
database bdb
 
suffix cn=accesslog
rootdn "cn=accesslog"
directory "/usr/local/var/access"
index reqStart eq
index default eq
index entryUUID eq
index entryCSN eq
index objectClass,reqEnd,reqResult,reqDN,reqType
 
access to *
            by dn="cn=admin,dc=dominio" read

database monitor
#*******

2008/7/23 Gavin Henry <ghenry@openldap.org>:
Miguel Jinez wrote:
Hello, I have upgrated openLDAP to 2.4.10, and I can t fix that situation
Someone know what is the problem? or really is a bug?


Hi,

I'm replying to this thread instead of:

 (ITS#5617) N-way multimaster replication problem

Can we see your complete config? Also, "dies", do you mean shutdown?

--
Kind Regards,

Gavin Henry.
OpenLDAP Engineering Team.

E ghenry@OpenLDAP.org

Community developed LDAP software.

http://www.openldap.org/project/