--On Monday, January 03, 2011 9:46 PM +0100 Steeg Carson steeg.carson@googlemail.com wrote:

2011/1/3 Quanah Gibson-Mount quanah@zimbra.com:

...

--On Monday, January 03, 2011 9:22 PM +0100 Steeg Carson steeg.carson@googlemail.com wrote:

...

I also used strace to catch the problem, but i did not find where the slapd write to disk. It seems, it comes via set up the futex write locks?  I'm not a kernel programmer and the code from openldap is hard to read. But strace I use for several years to find errors, it was also the first choice for me, to find the error. It is not a normal file that will be written...

The DB is always read off disk into the BDB backing cache the very first time slapd is run (assuming back-bdb or back-hdb are your backends).  I don't see any bug/issue here.

But does the slapd process during this time _write_ (heavy the whole time) to the disk?

Did you configure you BDB backing cache to be in memory, or on disk? The default is for it to be on the disk. If you want it to be in memory, set an shm_key value in the database configuration.

--Quanah

--

Quanah Gibson-Mount Sr. Member of Technical Staff Zimbra, Inc A Division of VMware, Inc. -------------------- Zimbra :: the leader in open source messaging and collaboration