[Date Prev][Date Next]
- To: openldap <openldap-software@OpenLDAP.org>
- Subject: Threads, redux
- From: Ben Beuchler <email@example.com>
- Date: Fri, 2 Sep 2005 17:34:36 -0500
- Content-disposition: inline
- Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com; h=received:message-id:date:from:reply-to:to:subject:mime-version:content-type:content-transfer-encoding:content-disposition; b=lU6B5u4JEC5xtas00xjErw6RWn3tWI1GtjtkPg13lV2EXzIYkK+dN5A82u/pergEQoPS38+X76dw/LfXd9lPE3fMGINYkvBf04yYqYJBwq9G4VUd2u7I/Guw7Ykcfy8kpyW25pAQsgMi/QATZzU3r+YqjiOplxEtoPGFCp1hVhI=
I'm trying to decide if my slapd's hunger for threads is acceptable or
a sign of some underlying pathology. I work at a small college with
around 1000 users. We're using LDAP to provide network authentication
for ~200 OS X workstations and 1000 email accounts (providing both
mailbox information to the smtp server and login iformation to the
POP/IMAP server). When I left "threads" at the default I would see
errors indicating it was deferring requests for "too many executing"
and was also deferring binds.
Leaving the thread limit anywhere below 100 would eventually produce
those errors. After several hours of heavy use, it would sit at
around 65 - 85 threads.
So. Does this imply that there's something screwed up with my slapd? Or is
that a reasonable number for an installation my size?