|
Thanks for your nice help. I got what the problem now…. Greetings .... BipinDas From: Prasanna Ram
Venkatachalam [mailto:vpram86@xxxxxxxxx] Your server is using prefork only. If thats the case, at max
load there might be enormous processes getting spawned and serving requests!
The current values in prefork are not sufficient to handle this much
connections (2000)! The MaxClients is set to 256 which is much lesser compared
to 2000.I dont know how system will behave if simultaneous 2000 connections are
served! Definitely creating process is an overhead and consumes much of system
resources. I dont use prefork much in these cases :( May be this will help
On Wed, May 20, 2009 at 12:50 PM, BipinDas-Gmail <bipinkdas@xxxxxxxxx> wrote: Hi, Here follows the version and
build parameters of my Server. I am not much expertise in Apache, I think I am
using PREFORK. But both worker and prefork is
enabled here. Server version: Apache/2.2.3 Server built: Nov
12 2008 10:40:14 Server's Module Magic Number:
20051115:3 Server loaded: APR 1.2.7,
APR-Util 1.2.7 Compiled using: APR 1.2.7,
APR-Util 1.2.7 Architecture:
64-bit Server
MPM: Prefork
threaded: no
forked: yes (variable process count) Server compiled with.... -D
APACHE_MPM_DIR="server/mpm/prefork" -D APR_HAS_SENDFILE -D APR_HAS_MMAP -D APR_HAVE_IPV6
(IPv4-mapped addresses enabled) -D
APR_USE_SYSVSEM_SERIALIZE -D
APR_USE_PTHREAD_SERIALIZE -D
SINGLE_LISTEN_UNSERIALIZED_ACCEPT -D APR_HAS_OTHER_CHILD -D
AP_HAVE_RELIABLE_PIPED_LOGS -D
DYNAMIC_MODULE_LIMIT=128 -D
HTTPD_ROOT="/etc/httpd" -D
SUEXEC_BIN="/usr/sbin/suexec" -D
DEFAULT_PIDLOG="logs/httpd.pid" -D
DEFAULT_SCOREBOARD="logs/apache_runtime_status" -D
DEFAULT_LOCKFILE="logs/accept.lock" -D
DEFAULT_ERRORLOG="logs/error_log" -D
AP_TYPES_CONFIG_FILE="conf/mime.types" -D
SERVER_CONFIG_FILE="conf/httpd.conf" Thanks for your time. Greetings .... BipinDas From: Prasanna Ram Venkatachalam [mailto:vpram86@xxxxxxxxx]
First we need to figure out which model your Apache is using. If it
uses worker this setting is also fine! Please remember, in worker ServerLimit
is the maximum processes that can run simultaneously and ThreadsPerChild is the
threads running inside each process! Since in worker model, each thread
will service a request, we got a MAX of ServerLimit x ThreadsPerChild
threads having connections and servicing requests. So we can put any
combination of these to achieve our goal. 50 and 41 is just an example i gave!
:) <Note: if your apache is runing on windows, These two models will not be
there. Apache in Windows uses WINNT model and UNIX flavors use either of
prefork or worker) On Wed, May 20, 2009 at 12:28 PM, BipinDas-Gmail <bipinkdas@xxxxxxxxx>
wrote:
|
|