For a variety of reasons I would like to run the back-end database beneath slapd on a different machine than slapd.
This appears to be difficult using the Berkeley DB.
Can anyone offer suggestions on how to run slapd and the database --- any database --- on separate machines?
Thanks in advance for any advice you can offer.
Kelly, Terence P wrote:
For a variety of reasons I would like to run the back-end database beneath slapd on a different machine than slapd.
This appears to be difficult using the Berkeley DB.
Can anyone offer suggestions on how to run slapd and the database --- any database --- on separate machines?
Stated like this, it sounds a bit odd. You could run the real slapd where the database is going to be physically stored, and a proxy (back-ldap, back-meta) on the machine the clients are supposed to contact.
p.
Ing. Pierangelo Masarati OpenLDAP Core Team
SysNet s.r.l. via Dossi, 8 - 27100 Pavia - ITALIA http://www.sys-net.it --------------------------------------- Office: +39 02 23998309 Mobile: +39 333 4963172 Email: pierangelo.masarati@sys-net.it ---------------------------------------
"Kelly, Terence P" terence.p.kelly@hp.com writes:
For a variety of reasons I would like to run the back-end database beneath slapd on a different machine than slapd.
This appears to be difficult using the Berkeley DB.
Can anyone offer suggestions on how to run slapd and the database --- any database --- on separate machines?
Thanks in advance for any advice you can offer.
If the remote database is a ldap server, you may use back_ldap. If the remote database is a RDBM you may use a combination of back_sql, back_relay and probably back_ldap or back_meta. If the remote database is of any other type you may use back_perl. But note that the performance would be rather poor compared to back_bdb or back_hdb.
-Dieter
Kelly, Terence P wrote:
For a variety of reasons I would like to run the back-end database beneath slapd on a different machine than slapd.
This appears to be difficult using the Berkeley DB.
Makes no sense at all, given that BerkeleyDB is an *embedded* database. I.e., it doesn't run in a separate process, it is designed to run in the same process as the application that uses it.
Can anyone offer suggestions on how to run slapd and the database --- any database --- on separate machines?
Why? I think if you provide more details about your motivation it would be easier to suggest usable alternatives.
openldap-software@openldap.org
-
Dieter Kluenter -
Howard Chu -
Kelly, Terence P -
Pierangelo Masarati