Web Servers on Unix and Linux - Server lagging. Seems like PHP errors

This is Interesting: Free IT Magazines  
Home > Archive > Web Servers on Unix and Linux > January 2006 > Server lagging. Seems like PHP errors





You are viewing an archived Text-only version of the thread. To view this thread in it's original format and/or if you want to reply to this thread please [click here]

Author Server lagging. Seems like PHP errors
madray@gmail.com

2006-01-13, 9:54 pm

Good Day!

My friend's server is lagging. I tried to find out the reason of lags,
but reached a deadblock...

server configuration:
Pentium4 2.8GHz SMP, 2Gb RAM

Red Hat
kernel - 2.4.31-ow1
apache - 1.3.34
php - 4.4.0 (compiled for mysql: 4.0.23, static libmysql). Bug #34156
(http://bugs.php.net/34156) proof.
mysql - 4.1.15

apache config:

Timeout 50
KeepAlive Off
MaxKeepAliveRequests 100
KeepAliveTimeout 5
MinSpareServers 5
MaxSpareServers 50
StartServers 8
MaxClients 270
MaxRequestsPerChild 100

libphp4 module is from apache2.

mysql:

max_connections 270
wait_timeout 50

php:

max_execution_time = 30
max_input_time = 60
memory_limit = 32M
zlib.output_compression = Off
error_reporting = E_ALL
log_errors = on
error_log = /var/logs/php/errlog.txt (But php doesn't log anything!)

zend_optimizer is working.


top:

14:40:44 up 23 days, 21:31, 1 user, load average: 33.27, 24.44,
23.16
447 processes: 440 sleeping, 7 running, 0 zombie, 0 stopped
CPU0 states: 84.20% user 12.18% system 0.0% nice 0.0% iowait
2.7% idle
CPU1 states: 86.23% user 10.9% system 0.0% nice 0.0% iowait
2.14% idle
Mem: 2069856k av, 1642928k used, 426928k free, 0k shrd,
20660k buff
68720k active, 361656k inactive
Swap: 2096440k av, 1058424k used, 1038016k free
134220k cached

PID USER PRI NI SIZE RSS SHARE STAT %CPU %MEM TIME CPU
COMMAND
17378 root 20 0 12552 11M 11976 R 99.9 0.5 1:14 0 httpd
27037 mysql 9 0 1204M 230M 35344 S 16.2 11.4 7:23 1
mysqld

ps -aux|grep httpd|wc -l
253


httpd's errorlog has strange lines:

Allowed memory size of 20971520 bytes exhausted (tried to allocate 0
bytes)

[notice] child 9951 exit signal Segmentation fault (11)

[warn] child process 13361 did not exit, sending another SIGHUP
(SIGHTERM and then SIGHKILL)
<.same type of errors.>
[notice] SIGHUP received. Attemping to restart.


When httpd's childs dying count of mysql's processes is abnormally
growing up (10-15 up to limit).
Acording to 'show processlist' all 250 processes locked.

Also, memory size of 20MB in errorlog is strange for me. Httpd's cpu
consuming is strange too.


gdb didn't give something ubnormal. httpd is working fine and then
catching SIGHTERM.
DumpCores doesn't working.

Tried to change httpd's options, but it didn't give something.


I think, it is php. All scripts are safe. They working fine on another
server.

How do you think will php 4.4.1 help?

Thx for any idea how to "repair" this server. And sorry for my bad
English, please

Sponsored Links






Free braindumps | Software forum | Database administration forum

Copyright 2003 - 2008 webservertalk.com