[Kolab-devel] Re: OpenLDAP performance issues

[Martin Konold]

Am Freitag, 20. Mai 2005 10:39 schrieb Dieter Kluenter:

Hi,

> 1. provide a DB_CONFIG with sufficient cachesize, as a guessing some
>    10 MB should be sufficient, but could be increased.

Where can I lookup the syntax of DB_CONFIG? Does this syntax change for minor 
updates/releases of bdb?

> 2. change to hdb backend, as this is optimized for search operations,
>    but lack performance on heavy write operations.

Are the configuration options compatible to bdb? Sofar we did most QA with bdb 
and I want to avoid to change the backend if possible.

> 3. in slapd.conf increase the value for cachesize to 2,000

Done!

> 4. add idlecachesize with a value of 6,000 to slapd.conf

I chose 10000 in order to be on the save side as I assume that in the worst 
case I am wasting some memory.

> 5. checkpointing the logfile is only done after write operations, so
>    10 minutes are OK, I would probabely increase the byte value to 512.

Done!

> 6. reduce idletimeout to 10 seconds, but mention in the docs that this
>    value could be increased if problems occur with data transport.

Done!

> 7. configure clients to only search with scope one and set searchbase
>    with appropriate DN's.

Hmm,... I expected that the clients do that already.

> 8. add 'database monitor' for ease of monitoring statistics.

Please point me to some specification/documentation. Are there any drawbacks?

> By the way, the 'stop' in access rules is just redundant.

IMHO imho it makes it more readable. Are there any disadvantages?

Regards,
-- martin

-- 
"I am committed to helping Ohio deliver its electoral votes to the
President next year."  -- 2004, Wally O'Dell - CEO of Diebold, Inc. 
e r f r a k o n - Stuttgart, Germany
Erlewein, Frank, Konold & Partner - Beratende Ingenieure und Physiker