Robert Petkus wrote:
You may want to take a look at back-hdb and set an appropriately large
value for idlcachesize -- 3x what one wold normally use with back-bdb
according to the man pages.
I don't think the choice of slapd backend has anything to do with the
original question.
slapd will not automatically read all entries into memory on startup. If
you want that to happen, you should issue a subtree ldapsearch command
(using the rootDN or some other identity with unlimited search
privilege) on the database after slapd starts.
Of course the ldapsearch will just cause every entry to be read. You
must set the entry cache size to a number >= the number of entries in
the database if you want them all to stay cached in memory.
Robert Petkus
William.Zhang wrote:
Dear :
I am wondering if there is anyway to config the openldap (or berkeley
db) so that it can read all entries in memory when startup for
performance ??
thanks in advance!
-----------------
Regards.
?????????????????????????????????????
?b?j(????????????????????????????????????
--
-- Howard Chu
Chief Architect, Symas Corp. http://www.symas.com
Director, Highland Sun http://highlandsun.com/hyc
OpenLDAP Core Team http://www.openldap.org/project/
