URGENT: CPU and RAM spikes on clean install

[markcarsonboxz]

I recently performed a clean re-image and install of CentOS 5.5 and Plesk 9.5.2

Each time I login and browse the Plesk Control Panel, the CPU and RAM usage spikes. Using Virtuosso, the list of processes contain:

13542 0.3 0.0 /usr/bin/sw-engine-cgi -c /usr/local/psa/admin/conf/php.ini -d auto_prepend_file=auth.php3 -u
13549 7.5 0.2 /usr/bin/sw-engine-cgi -c /usr/local/psa/admin/conf/php.ini -d auto_prepend_file=auth.php3 -u
13550 0.3 0.0 /usr/bin/sw-engine-cgi -c /usr/local/psa/admin/conf/php.ini -d auto_prepend_file=auth.php3 -u
13553 0.0 0.0 /usr/bin/sw-engine-cgi -c /usr/local/psa/admin/conf/php.ini -d auto_prepend_file=auth.php3 -u

Each of these have been as high as 25%, killing the VPS.

Any reason why this is occurring? (threads from old posts indicate a problem/bug in Plesk, but this was v8.3)

 

[markcarsonboxz]

Looking through some logs, I found the following:

/sw_cp_server/error_log:

2010-07-21 00:05:57: (mod_fastcgi.c.2582) FastCGI-stderr: PHP Warning: unlink(/usr/local/psa/tmp/components_dirty_flag) [<a href='function.unlink'>function.unlink</a>]: No such file or directory in /usr/local/psa/admin/auto_prepend/auth.php3 on line 514

/httpd/error_log:

[Wed Jul 21 09:15:55 2010] [warn] RSA server certificate CommonName (CN) `plesk' does NOT match server name!?
[Wed Jul 21 09:15:55 2010] [warn] RSA server certificate CommonName (CN) `plesk' does NOT match server name!?
[Wed Jul 21 09:15:55 2010] [warn] Init: SSL server IP/port conflict: atmail.webmail:443 (/etc/httpd/conf.d/zzz_atmail_vhost.conf:48) vs. horde.webmail:443 (/etc/httpd/conf.d/zzz_horde_vhost.conf:41)
[Wed Jul 21 09:15:55 2010] [warn] Init: SSL server IP/port conflict: default-217-160-239-233:443 (/etc/httpd/conf.d/zz010_psa_httpd.conf:78) vs. horde.webmail:443 (/etc/httpd/conf.d/zzz_horde_vhost.conf:41)
[Wed Jul 21 09:15:55 2010] [warn] Init: You should not use name-based virtual hosts in conjunction with SSL!!

/messages
2010-07-21T09:15:53.379111+01:00 s15376792 rsyslogd: [origin software="rsyslogd" swVersion="3.22.1" x-pid="7696" x-info="http://www.rsyslog.com"] (re)start
2010-07-21T09:15:53.377615+01:00 s15376792 rsyslogd: WARNING: rsyslogd is running in compatibility mode. Automatically generated config directives may interfer with your rsyslog.conf settings. We suggest upgrading your config and adding -c3 as the first rsyslogd option.
2010-07-21T09:15:53.378721+01:00 s15376792 rsyslogd: Warning: backward compatibility layer added to following directive to rsyslog.conf: ModLoad imuxsock

 

[IgorG]

http://forum.parallels.com/showthread.php?t=99861

 

[markcarsonboxz]

Thanks for the prompt reply. I did look at this post, but thought that the issue was too specific to that post.
I got a bit p**sed off and completed a reimage.
If this happens, I'll know where to look

 

[markcarsonboxz]

Reimaged the server. Same problem with no updates applied - using version 9.0.1.

List of installation tasks completed:

Update existing php packages from 5.1.6 to 5.2.13
wget -q -O - http://www.atomicorp.com/installers/atomic.sh | sh
yum --disablerepo=centos* update php
Install php-mcrypt (resolving dependencies manually)
yum install libmcrypt
yum install libmcrypt-devel
ldconfig -p | grep libmcrypt.so.4
ls -la /usr/lib64/libmcrypt.so.4
yum install php-mcrypt.x86_64
Restart httpd process
/etc/init.d/httpd restart
Install other php extensions; required for later application installation
yum --disablerepo=centos* install php-tidy.x86_64

Install PCRE with Unicode Support
wget http://chrisjean.com/wp-content/uploads/2009/01/pcre-66-27x86_64.rpm
rpm -Uvh ~/pcre-66-27x86_64.rpm

Required at least PHP 5.2, with tidy and mcrypt.
Followed instructions from 1&1 to update php to 5.2.13 here
Researched and installed mcrypt from same location
Hopefully PCRE update hasn't broke Plesk!

The following httpd error appears to be fixed:

/httpd/error_log
[Wed Jul 21 12:23:25 2010] [error] [client 127.0.0.1] Directory index forbidden by Options directive: /var/www/html/

/sw_cp_server/error_log:
2010-07-21 13:56:53: (mod_fastcgi.c.1755) connect failed: Connection refused on unix:/usr/local/psa/tmp/sw-engine-8880.sock-0
2010-07-21 13:56:53: (mod_fastcgi.c.2904) backend died, we disable it for a 5 seconds and send the request to another backend instead: reconnects: 0 load: 1
2010-07-21 13:56:53: (mod_fastcgi.c.2677) child signaled: 9
Downloading the file products.inf3: 2010-07-21 14:12:01: (log.c.135) server stopped
2010-07-21 14:12:22: (log.c.75) server started

mysqld.log:
100721 14:12:22 mysqld started
100721 14:12:22 InnoDB: Started; log sequence number 0 763813
100721 14:12:22 [Note] /usr/libexec/mysqld: ready for connections.
Version: '5.0.45' socket: '/var/lib/mysql/mysql.sock' port: 0 Source distribution

I have not completed any other configuration or installations tasks.

I would appreciate any comments or help with this.

 

[markcarsonboxz]

More Information:
Kernel: 2.6.9-023stab051.3-smp (CentOS? Really?)
I attach a file containing all installed packages.