From: Robert Schueler (schueler@emt.iis.fhg.de)
Date: Thu Aug 05 2004 - 08:34:57 EDT
Hello,
many thanks to John L. Lanier. He also point me to the variables of the
proc-subsystem.
I increased the value of per_proc_data_sizeup up to my amount of
physical memory, like recommended. [1]
default value: per_proc_data_size = 134217728
current value: per_proc_data_size = 536870912
After reboot concurrency-tests succeed, without restricting the thread
pool.
Robert
[1] 4.4.6.3
http://h30097.www3.hp.com/docs/base_doc/DOCUMENTATION/V51B_HTML/ARH9GCTE/TITLE.HTM
On Wed, 2004-08-04 at 17:07 +0200, Robert Schueler wrote:
> Admins,
>
> we are running into a problem with the current openldap-2.1 and
> openldap-2.2 series in our cluster-environment. Both cluster consist of
> two DS20E with 512MB and 1024MB of RAM and have installed Tru64 UNIX
> 5.1B with latest patch_kit.
>
> Everytime I run the test-suite from openldap-sources test008-concurrency
> failed. The matter should be the following log-entry:
>
> ch_malloc of 16777216 bytes failed
>
> If I restrict the maximum size of thread pool (slapd.conf) to 7
> concurrency-test succeed. With 8 and above it will fail. Default size is
> set to 16.
>
> Is somebody aware of an internal kernel-limit or something else, which
> we have reached?
>
> Any help is welcome,
>
> Robert.
-- Fraunhofer-Institut für Digitale Medientechnologie http://www.idmt.fraunhofer.de/ Informationen zur Klangfeldsynthese http://www.iosono-sound.de/
This archive was generated by hypermail 2.1.7 : Sat Apr 12 2008 - 10:50:06 EDT