Installing Kolab on a system with 1GB and no swap

[Leo Walkling]

Hello,
I want to install Kolab 2.4, but when I run setup-kolab, the LDAP server 
setup fails. The error message says that it could not allocate 2.8GB of 
cache (see message below).
Can I reduce the cache memory requirements of the 389 directory server 
or the number of threads it uses?
I think I have to do that before setup-kolab calls setup-ds-admin.pl. 
That is because setup-ds-admin.pl creates, starts and initializes the 
server in one pass, leaving me no time to modify the 'dse.ldif'.

Rationale:
I'm on a Centos 6 system with only 1GB memory and no swap, so the 
current memory requirements of 389 are just too high for me.
The latter is because it is running under OpenVZ.

Regards
Leo Walkling



Could not import LDIF file '/tmp/ldiftghAi0.ldif'.  Error: 512.  Output: 
importing data ...
[23/Sep/2012:07:50:46 +0200] - WARNING: Import is running with 
nsslapd-db-private-import-mem on; No other process is allowed to access 
the database
[23/Sep/2012:07:50:46 +0200] - check_and_set_import_cache: pagesize: 
4096, pages: 50025989, procpages: 50178
[23/Sep/2012:07:50:46 +0200] - Import allocates 80041580KB import cache.
[23/Sep/2012:07:50:46 +0200] - Setting ncache to: 20 to keep each chunk 
below 4Gbytes
[23/Sep/2012:07:50:46 +0200] - libdb: malloc: 2853570916: Cannot 
allocate memory
[23/Sep/2012:07:50:46 +0200] - libdb: Unable to allocate memory for 
mutexes from the region
[23/Sep/2012:07:50:46 +0200] - libdb: PANIC: Cannot allocate memory
[23/Sep/2012:07:50:46 +0200] - libdb: process-private: unable to find 
environment
[23/Sep/2012:07:50:46 +0200] - Unable to open new DB_ENV for 
import/export! -30974
[23/Sep/2012:07:50:46 +0200] - dblayer_instance_start fail: 
DB_RUNRECOVERY: Fatal error, run database recovery (-30974)