[Date Prev][Date Next] [Chronological] [Thread] [Top]

Re: Too many filehandles

Replying to myself,

Increased 'threads' from 64 to 256, added a new index, then for each 5
minutes, ran this:

#lsof  | grep slapd | wc
    799    7257  115089
# lsof  | grep slapd | wc
    855    7783  122269
# lsof  | grep slapd | wc
    965    8761  139745

by the final lsof run we're seeing repeated 'too many file handles'
entries in /var/log/messages.

One line of thought is why slapd is using tcpwrapper support when it's
run the standard redhat way via init script.

Other environmental variables I missed on my original post:

# Allow LDAPv2 client connections.  This is NOT the default.
allow bind_v2

# Note syslogd should be logging with a 'dash' notation so that it's
# logging async 
loglevel    256

# We need lots of threads, this may or maybe not enough
threads     512

# We don't want to leave Exim hanging around, but we do want to stay
# alive long enough for the SOAP service
idletimeout 62

sizelimit 250000

index objectClass                       eq,pres
index ou,cn,mail,surname,givenname      eq,pres,sub
index uidNumber,gidNumber,loginShell    eq,pres
index uid,memberUid                     eq,pres,sub
index nisMapName,nisMapEntry            eq,pres,sub
index entryUUID,entryCSN  eq
index owner,dc pres,eq

# $OpenLDAP: pkg/ldap/servers/slapd/DB_CONFIG,v 1.1 2004/06/18 02:49:08
kurt Exp $
# Example DB_CONFIG file for use with slapd(8) BDB/HDB databases.
# See Sleepycat Berkeley DB documentation
#   <http://www.sleepycat.com/docs/ref/env/db_config.html>
# for detail description of DB_CONFIG syntax and semantics.
# Hints can also be found in the OpenLDAP Software FAQ
#   <http://www.openldap.org/faq/index.cgi?file=2>

# 2.5GB split between 3 physical cache
set_cachesize 2 524288000 3

# Data Directory
#set_data_dir db

# Transaction Log settings
set_lg_regionmax 262144
set_lg_bsize 2097152
set_lg_dir logs

set_lk_max_locks   10000
set_lk_max_lockers 10000
set_lk_max_objects 10000

# When using (and only when using) slapadd(8) or slapindex(8),
# the following flags may be useful:
set_flags DB_TXN_NOSYNC

^-------- these were commented out earlier, so I've uncommented them. No
difference here.

Attachment: signature.asc
Description: This is a digitally signed message part