I've done a slapcat of the master, removed one postalAddress, slapadd and restart replica : the entry is relicated with success (then it locks to another one and I can't modify LDAP directory feeding process). Why slapadd succeds whereas replication with schema checking disabled fails ?
As documented, slapadd is not as rigorous in its data checking as ldapadd or replication.
Do you have any suggestions to make replication works without removing duplicate postalAddress entries ?
No. You need to fix your broken data. --Quanah -- Quanah Gibson-Mount Principal Software Engineer Zimbra, Inc -------------------- Zimbra :: the leader in open source messaging and collaboration