Hi , hope someone knows the answer to this, it causes us massive problems. We have a corporate address book that is created from several address books from our group. All mobile users have a local replica of this database so there are thousands of replicas out there. Every now and then we get a problem with old data replicating back in - deletion stubs have been purged form the server copy and really old dstuff comes back. Users are supposed to keep their own details up to date, and users access level is author, with Create docs, write public docs and Replicate roles.
We want to try to limit potential for this and are thinking of reducing users access to reader. If someone then tries to replicate an old replica to a server copy, will this prevent them, or will it see that the “new” docs were created by a legitimate user and still allow them in? WE thought the first, but seems to do the second!
Subject: Replication of old data and users access levels
We don’t allow anything above Reader access to our Public NAB for just such reasons. FYI, this might help in resolving the problem. Jean-Yves Riverin also posted a utility on openntf.org that might help.
Subject: Replication of old data and users access levels
Firstly:READER can read
Author can read / create
Editor can read / create /modify
Having loads of replicas out there on varing laptops and remote user is going to cause you allot of pain either way. This is because some thing will go wrong and this is normally going to be human error.
The replication can be set to PULL ONLY and this will be over written by someone doing a full replication.
ACL’s can be restricted but a development server can accidentally replicate because of Local DOMAIN server access.
Have you thought of creating a reduced address book for your users? or Directory assistance?