[Date Prev][Date Next] [Chronological] [Thread] [Top]

OpenLdap stopped with database of 3 000 000 entries



Hello
   
  I am having problems with openldap when the size of the entries reaches 3  000 000 entries.

  I am using openldap-2.3.14 and db-4.4.16

  The hardware has 4 G of memory running Centos 2.6.9-22 on a 32 bits plateform
I can slapadd the 3 000 000 entries without problem

  The content of the bdb reaches a total of 3.7G
  -rw-------  1 root root 806M Jan  5 12:33 dn2id.bdb
-rw-------  1 root root 2.9G Jan  5 12:33 id2entry.bdb
   
  When i try to search the entries, i can see the slapd process reaching 3.2 G in memory and suddently stopped
  slapd-2.3.14: ch_malloc.c:107: ch_calloc: Assertion `0' failed.
Aborted
   
  
My slapd.conf:
##############
  
database        bdb
suffix          ""
rootdn          "cn=Admin"
rootpw          admin
  directory       //usr/ldap//openldap/var/openldap-data
  # Indices to maintain
loglevel 0
cachesize 500000
#tried cachesize 3500000
#tried cachesize 1000000
#tried cachesize 100000
  index           uniqueID eq
   
  
My DB_CONFIG
############
set_cachesize 1 0 2
#tried set_cachesize 0 20000000 2
#tried set_cachesize 0 512000000 2
  I have tried a lot of different (cachesize/set_cachesize) parameter in slapd.conf / DB_CONFIG configuration but i get the same result.
   
  I know that i am reaching the memory size limit of a linux process.
  
Any idea what is going wrong & what needs to be done ?
   
   
  Thanks in advance
  Jean Claude Martys
Telecom Architecture
  
 
   
   

		
---------------------------------
 Nouveau : téléphonez moins cher avec Yahoo! Messenger ! Découvez les tarifs exceptionnels pour appeler la France et l'international.Téléchargez la version beta.