[Date Prev][Date Next]
[Chronological]
[Thread]
[Top]
Splitting data for replication
- To: OpenLDAP mail-list <openldap-software@OpenLDAP.org>
- Subject: Splitting data for replication
- From: Jorge Ortiz Claver <jortiz@idsk.com>
- Date: Mon, 28 Apr 2003 21:41:30 +0200
- User-agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.0.1) Gecko/20020823 Netscape/7.0
Hi,
I have a ldap database I'd like to replicate to different slaves. Data
is organized as follows:
o=mycompany
|
-----------------------------------------
| |
|
ou=domain1 ou=domain2 ou=domain3
Replication would be made at domains level, replicating domain1data to
one server, domain2 to another one and so on. I know I can't have all
the data in the same database so I need to chop the data out into
different databases.
Here is my questions: what is the better way to distribute the data
among different databases but keep a "virtual" view of the whole tree
for clients? I mean I don't want the clients to send three or four
queries in order to search for a user. I'd like to make organization
transparent to clients.
Another requirement is (I guess that's not too complex) I need to
maintain some domains data out of the main server so for example, I'd
need to maintain domain3 data in a different server and replicate it to
the main server.
Any idea, suggestion or advice?
Thanks in advance
Jorge Ortiz