A Linux Server is not Windows XP and if reboot fail you usually still connect by SSH and do something. This commands will show you how to remotely hard reboot machine. Hard reboot mean that shutdown scripts will not run and machine reboot immediately without syncing hard disk drives, shutdown applications etc, it’s more like hitting the reset button.

echo 1 > /proc/sys/kernel/sysrq

echo b > /proc/sysrq-trigger

This commands enable sysrq and after this calls fast reboot. If you want to force shutdown machine try this.

Force shutdown

echo 1 > /proc/sys/kernel/sysrq

echo o > /proc/sysrq-trigger

This came handy, when I had a server that had some IO error and it can no longer read from disk, only few cached binaries into memory kept it running (kernel, SSHD, bash), I could still access the machine via SSH but can no longer do anything, forcing the reboot as mentioned above was my only resort, and it worked like charm…

No related posts.

Every so often, MySQL
tables have a way of corrupting themselves. MySQL offers a quick and
painless method of repairing those tables.

Login to your server via SSH and change directories to the database that is having problems.

cd /usr/local/mysql/var/[DBNAME]/

You will need to replace [DBNAME] with the actual name of your database.

Stop the MySQL server:

/etc/rc.d/init.d/mysql stop

NOTE: Accounts created before July 7th, 2003 will want to use:

/etc/rc.d/init.d/mysqld stop

To check the tables:

myisamchk *.MYI

To repair tables:

myisamchk -r *.MYI

Restart MySQL:

/etc/rc.d/init.d/mysql start

NOTE: Accounts created before July 7th, 2003 will want to use:

/etc/rc.d/init.d/mysqld start

Alternatively, if you do not want to shut down MySQL, you can use mysqlcheck.

mysqlcheck [DBNAME]

To repair the database tables:

mysqlcheck -r [DBNAME]

You will need to replace [DBNAME] with the actual name of your database.

You can find additional documentation here:

http://dev.mysql.com/doc/mysql/en/Table_maintenance.html

http://dev.mysql.com/doc/mysql/en/Using_mysqlcheck.html

Related posts:

1. Linux Server Type: Restoring MySQL SimLink. If you’re unable to restart mysqld and troubleshooting the cause:...

This simple step by step guide will show you how to mount a new hard drive or partitions onto your Linux server

Initially you wanna check the hard drives on your system

fdisk -l

so now you now whats being seen by the system

The mounting point path must already be created with proper permissions. So a more likely flow of commands would be below:

How To Unmount A Drive In Linux

This command is very easy to type wrong. It is NOT unmount. Take another closer look if thats what you saw at first. It is umount — no n here!

No related posts.

Just go to the directory , like cd / and try this

du -cksh *

No related posts.

The cause of the EasyApache failure, according to the log, was a failure of the package manager, in this case yum.  This was caused by a problem with domain name resolution, which was preventing yum from accessing its mirrors.  Here is what I found:

[root@server ~]# yum check-update
Could not retrieve mirrorlist http://mirrorlist.centos.org/?release=5&arch=x86_64&repo=os error was
[Errno 4] IOError: <urlopen error (-2, ‘Name or service not known’)>
Error: Cannot retrieve repository metadata (repomd.xml) for repository: base. Please verify its path and try again
[root@server ~]# host mirrorlist.centos.org
[root@server ~]#

Note that no result was returned for mirrorlist.centos.org.  To correct this, I added the following to the top of the /etc/resolv.conf file:

nameserver 4.2.2.1

This is one of the root nameservers.  After I made this change, mirrorlist.centos.org was resolvable, and yum worked correctly:

[root@server ~]# host mirrorlist.centos.org
mirrorlist.centos.org has address 72.21.40.11
[root@server ~]# yum check-update
base                      100% |=========================| 1.1 kB    00:00    
updates                   100% |=========================|  951 B    00:00    
addons                    100% |=========================|  951 B    00:00    
extras                    100% |=========================| 1.1 kB    00:00    
Excluding Packages in global exclude list
Finished

wget.x86_64                              1.10.2-7.el5           base           
[root@server ~]#

Adding a root nameserver to /etc/resolv.conf is meant to be a temporary measure, so you will need to take a look at the other nameservers listed in the file and make sure they are valid recursive nameservers that will do host look-ups for domains that they do not host.

EasyApache should work properly now.

No related posts.

/sbin/ifconfig | grep -i hwaddr

The sequence of hexadecimal digits that appears to the right of eth0 HWAddr (e.g. 08:00:27:ED:DA:8b) is
your network card’s MAC Address

No related posts.

Below you can find some steps to do so manually

Log into your server and create a cPanel backup for the account

/scripts/pkgacct <username> <location> backup

<username> refers to the username you want to backup.

<location> is the path to where the backup will be placed (in your case you want to put /backup/cpbackup here).

So backup tells pkgacct that you want to create a backup file.

If you leave this off it creates a cpmove archive. Not sure what the difference is between a cpmove and a backup file, but this tells pkgacct you just want to create a <username>.tar.gz file in <location>.

And after the backup has been completed you can simple move it by pushing it via SCP

here is the tutorial  http://www3.wiredgorilla.com/content/view/322/1/

No related posts.

The reason you’re having trouble is because CentOS 5 uses bind-9.3 and CentOS 4 uses bind-9.2 – The changes between these versions is significant enough that cPanel won’t work with it properly.

For instance by default there is no /etc/named.conf file created when bind-9.3 is installed and the default named.conf file for bind-9.3 is significantly different than in older versions.

Until the code is updated to work with bind-9.3, I have found the workaround to be to remove bind-9.3 packages and install bind-9.2 packages as follows:
 

Log into your servers via SSH and firstly we BIND

yum remove bind-*

yum remove bind

Then we ll install the old BIND rpm’s from a mirror closed to you

rpm -Uvh –nodeps http://ftp.monash.edu.au/pub/linux/CentOS/4.6/os/x86_64/CentOS/RPMS/bind-libs-9.2.4-28.el4.x86_64.rpm
rpm -Uvh –nodeps http://ftp.monash.edu.au/pub/linux/CentOS/4.6/os/x86_64/CentOS/RPMS/bind-utils-9.2.4-28.el4.x86_64.rpm
rpm -Uvh –nodeps http://ftp.monash.edu.au/pub/linux/CentOS/4.6/os/x86_64/CentOS/RPMS/bind-9.2.4-28.el4.x86_64.rpm
rpm -Uvh –nodeps http://ftp.monash.edu.au/pub/linux/CentOS/4.6/os/x86_64/CentOS/RPMS/caching-nameserver-7.3-3.noarch.rpm

simlink this
ln -s ../../lib/libcrypto.so.0.9.8b /usr/lib/libcrypto.so.4

service named start

The –nodeps flag is tagged on because of dependency "libcrypto.so.4" for the bind packages.
IMPORTANT!

Run those cPanel SSH comands

/scripts/fixndc
/scripts/rebuildnamedconf

You’ll also need to edit /etc/yum.conf and add a line in the [main] section
 that starts "exclude=" or append to this line if it already exists "bind* caching-nameserver".

nano  /etc/yum.conf

This prevents yum from updating bind-9.2 to bind-9.3 and in the process breaking cPanel.
If you fail to do this when cPanel update is run BIND will get updated and the DNS functionality of cPanel will break.
I hope this helps shed some light on this anoying issue.

PS: To find the dowload mirrors appropiate to your location please go here

http://www.centos.org/modules/tinycontent/index.php?id=13 

No related posts.

uname -a

and to upgrade your kernel

yum update \kernel*

check if your kernel is added to grub

cat /boot/grub/grub.conf

and then reboot

shutdown -rf now

No related posts.

mkdir $HOME/.ssh
chmod -R og= $HOME/.ssh
cd .ssh
nano authorized_keys2

Add your key you created with PuttyGen , make absolutly sure you r key is in one line ONLY , otherwise it wont work!

Adjust your  sshd_config

nano /etc/ssh/sshd_config

to

RSAAuthentication yes
PubkeyAuthentication yes
AuthorizedKeysFile      .ssh/authorized_keys2

No related posts.

Firstly login to your Windows server via Remote Desktop, launch IE, and go to the following address:

http://download.r1soft.com

Click on the Windows Stable link under CDP Agent, and download the latest Windows Agent. Once it is downloaded, run it.

You will see the following screen:

{gallery}winagent1{/gallery}

This is just a typical Windows installation program; follow through the prompts and install it. You will then be shown the following screen:

{gallery}winagent2{/gallery}

 Click on Finish, and the Windows Agent Key Utility will load. Click on Next, accept the license agreement, and you will be presented with the following screen:

{gallery}winagent3{/gallery}

Click on CDP Server Allow, and then click on the "Add Server Key" button (padlock with the little green +).  Leave the protocol as http and the port as 80 (unless your setup is different; if so, adjust accordingly)and enter the host name of your central backup server. Click OK, and the utility will add the backup server key to the server to be backed up.

One last thing to do; open up TCP port 1167 on the server firewall.
This can be done through the Windows Firewall or your hosting control panel (e.g. Plesk).

Everything is now completed on the server to be backed up; now you
just need to add it in the CDP interface. Login as administrator and
click on the New Host button.

{gallery}winagent4{/gallery}

Now fill out the form like this:

Host Name or IP: The IP address/hostname of the server you just installed the agent on.
Host Description: A description of the server you installed the agent on
Host Type: Select "Windows"
Tick the Use Default Network Settings box

Leave both "Licensed Addon Module" checkboxes blank.

Now select the Volume you want the server to place the backups in, check the "Create Disk Safe Next" box, and click on OK.

These settings are usually fine for a standard CDP server setup;
however, keep in mind that your server setup may mean that you need to
use different settings from the ones listed here.

The next page contains Disk Safe configuration settings.

{gallery}agentcompression{/gallery}

The first,
Compression Level, I would recommend setting to at least 1.
Uncompressed backups can use up all the available disk space on your
backup server quickly; setting the Compression Level to 1 can reduce
backup size by up to 50% without too significant a hit on CPU
consumption. Leaving the Defragment setting at 5% is fine.

I would not recommend encryption unless the data you are backing up sensitive data. Now click on OK.

You should now be directed to the host summary page. CDP
automatically attempts to find the host; click on History at the top
and see if it was successful or not.

{gallery}hosthistory{/gallery}

If it was, you can now start the
first backup and setup a recurring backup schedule to your liking in
the Backup and Restore tab in the host’s page.

{gallery}backupandrestore{/gallery}

Here is a brief overview of setting up a recurring backup schedule:

1. Login to CDP as admin
2. Click on the Host
3. Click on the "Backup and Restore" option
4. Click on "Schedule Backup"

Now fill out the settings to your liking. If you have a separate backup drive, I recommend skipping it. 

If you now need to setup a granular MySQL backup using the MySQL addon, click on the MySQL button.

{gallery}agentmysql{/gallery}

Fill out the description and the root username and password of your MySQL server. I recommend changing the Connection Type to Socket File – the default socket file location on CentOS servers is:

/var/lib/mysql/mysql.sock

CDP will now backup your MySQL databases table by table in line with your backup schedule. 

All done, you have setup a Windows server to be backed up by R1soft

No related posts.

First, login to the new server as root. Then browse to the following page:

http://download.r1soft.com/

Click on the Linux Stable link under CDP Agent, and copy the link location of the latest CentOS agent.

Now, go back to the server, and execute the following command:

For 64bit CentOS : wget http://download.r1soft.com/d/linux-agent/1.46.5-x86_64/linux-agent-64-1.46.5-centos.run

For 32bit CentOS : wget http://download.r1soft.com/d/linux-agent/1.46.5-x86/linux-agent-32-1.46.5-centos.run

for newer kernels go to the Pre
Release
section

wget http://download.r1soft.com/d/linux-agent/1.47.0-x86_64/linux-agent-64-1.47.0-centos.run

The link will look different depending on the release you download. Once this has finished, execute the file:

sh linux-agent-64-1.46.5-centos.run

or

sh linux-agent-32-1.46.5-centos.run

If you are installing a custom kernel then you need to install the headers first

yum install kernel-devel

A graphical wizard will appear. Just go through here accepting the defaults; they usually work fine. Once this is done, we need to create the server key. Browse to the following directory:

cd /etc/buagent/server.allow/

Now create a file called the IP of your main backup server.

nano 111.222.333.444 (Your backup server IP)

Browse to your main backup server, login as administrator, click on Options, then click on Server Key. You need to copy this key into the file you just created on the server to be backed up. Once this is done, save the file. The path to the file should look like this:

/etc/buagent/server.allow/SERVER_IP

One last thing to do; open up TCP port 1167 on the server firewall. This can easily be done through a iptables frontend like csf, or just use iptables via the command line.

Everything is now completed on the server to be backed up; now you just need to add it in the CDP interface. Login as administrator and click on the New Host button.

{gallery}winagent4{/gallery}

Now fill out the form like this:

Host Name or IP: The IP address/hostname of the server you just installed the agent on.

Host Description: A description of the server you installed the agent on

Host Type: Select “Linux”

Tick the Use Default Network Settings box

In the “Licensed Add-on Modules” section, you need to tick the Control Panel option if you want CDP to integrate with a hosting

control panel (like cPanel or Plesk), and you need to tick the MySQL module if you want CDP to perform a proper granular (table by

table) backup of the databases on the server.

Now select the Volume you want the server to place the backups in, check the “Create Disk Safe Next” box, and click on OK.

These settings are usually fine for a standard CDP server setup; however, keep in mind that your server setup may mean that you need to use different settings from the ones listed here.

The next page contains Disk Safe configuration settings.

{gallery}agentcompression{/gallery}

The first, Compression Level, I would recommend setting to at least 1. Uncompressed backups can use up all the available disk space on your backup server quickly; setting the Compression Level to 1 can reduce backup size by up to 50% without too significant a hit on CPU consumption. Leaving the Defragment setting at 5% is fine.

I would not recommend encryption unless the data you are backing up sensitive data. Now click on OK.

You should now be directed to the host summary page. CDP automatically attempts to find the host; click on History at the top and see if it was succesful or not.

{gallery}hosthistory{/gallery}

If it was, you can now start the first backup and setup a recurring backup schedule to your liking in the Backup and Restore tab in the host’s page.

{gallery}backupandrestore{/gallery}

Here is a brief overview of setting up a recurring backup schedule:

1. Login to CDP as admin

2. Click on the Host

3. Click on the “Backup and Restore” option

4. Click on “Schedule Backup”

Now fill out the settings to your liking. Do backup the swap partition (CDP will skip it anyway), disable the tmp partition. If you have a second hard drive for backups (for example, for cPanel backups), do not include it.

If you now need to setup a granular MySQL backup using the MySQL addon, click on the MySQL button.

{gallery}agentmysql{/gallery}

Fill out the description and the root username and password of your
MySQL server. I recommend changing the Connection Type to Socket File -
the default socket file location on CentOS servers is:

/var/lib/mysql/mysql.sock

CDP will now backup your MySQL databases table by table in line with your backup schedule.

All done, you have setup a Linux server to be backed up by R1soft

No related posts.

Follow these steps to prepare a Linux or Windows server backup system,

prodced by R1Soft,   for a bare metal restore.

1. Put the R1Soft Live Rescue CD into the machine you are going to restore
to, and boot.

Press enter at the splash screen.

2. Once you are logged in, type

netconfig

and follow the prompts to configure
the network settings. Start SSH if you need to access the machine remotely.

3. Type ping google.com to test the network connectivity.

4. To simplify things, its usually a good idea to check if there are any
existing partitions on the drives you are restoring to. To do so, type

fdisk -l

to obtain a list of hard drives. Take note of the drive labels e.g. /dev/sda.

Then type

cfdisk /disk/label

to access the partition table of that specific
disk. From here you can delete all existing partitions until there is only free
space, and then quit the program.

5. The machine itself is now ready to be restored to;

Login to the R1Soft
Backup Server web interface, click on the host you want to restore, click on the
Backup & Restore tab,  and then click on Bare Metal Restore.

Now just follow
the prompts to restore the machine.

6. Once the restore is completed, you will need to unplug the USB CD drive
and make sure the boot order in the BIOS is set to the main hard
drive.

No related posts.

This tutorial will guide you through setting up and configuring Cacti on a Linux server. 

This tutorial assumes that you are running a vanilla installation of CentOS 5.1 with a typical LAMP setup. The main requirements for a Cacti installation are Apache, MySQL, PHP (with php-snmp and php-mysql), and net-snmp. 

First of all, login to your server via SSH. One of the core requirements of Cacti is RRDTool, which unfortunately isn’t in the CentOS repositories. However, there are a number of RPMs maintained for RHEL, which you can find here:

http://dag.wieers.com/rpm/packages/rrdtool/ 

I used the i386 packages for RHEL5, which at the time of writing were as follows:

rrdtool-1.2.23-1.el5.rf.i386.rpm
rrdtool-devel-1.2.23-1.el5.rf.i386.rpm
perl-rrdtool-1.2.23-1.el5.rf.i386.rpm

wget these to your server, and then run the following rpm command to install them:

rpm -ivh  rrdtool-1.2.23-1.el5.rf.i386.rpm rrdtool-devel-1.2.23-1.el5.rf.i386.rpm perl-rrdtool-1.2.23-1.el5.rf.i386.rpm

If you haven’t done so already, install net-snmp:

yum install net-snmp

At this point we will add a simple snmp configuration file for Cacti. Navigate to /etc/snmp and edit snmpd.conf. Delete everything in there and insert the following:

rocommunity public
syslocation locatationofserver
syscontact admin@server.com
dontPrintUnits true
includeAllDisks

This simple configuration is enough to get Cacti up and running. Save and exit the file, then restart snmpd:

service snmpd restart 

Now setup a user account for cacti. Depending on your system setup you will have to do this through SSH, or you can set up an account through cPanel/Plesk/Webmin. Once this is done, create a new MySQL database and assign a user to it. 

Now go to the cacti website and download the latest release from here:

http://cacti.net/download_cacti.php 

Extract it, and upload it via FTP. Alternatively, wget the archive using SSH, like so:

wget http://www.cacti.net/downloads/cacti-0.8.7b.tar.gz

Now edit the file config.php in the include directory. Change the database connections settings to match the database you created, and save it. Now just browse to the cacti directory in your web browser and follow the installer.

The final step to take is to setup the cronjob to run the poller (which processes all the Cacti stats). You can do this through your control panel or through SSH by editing the /etc/crontab file. Here is what mine looks like:

*/5 * * * * cacti php /var/www/html/poller.php > /dev/null 2>&1

This runs the poller as the user cacti every 5 minutes, which is the standard setting.

Now login to Cacti. You will be prompted to change your administrator password when you first login. Once this is done, you will be presented with the Cacti console.

Congratulations, you have succesfully installed Cacti! You can now add devices and configure graphs using the control panel. 

No related posts.

Webmin, a free and powerful control panel, it easy to install and allows you to carry out administrative tasks in an easy to use browser based frontend. This tutorial will guide you through installing Webmin.

This tutorial assumes that you have a vanilla server CentOS5 installation.

1. Login to your server via SSH and download the latest Webmin release, like so:

wget http://prdownloads.sourceforge.net/webadmin/webmin-1.410-1.noarch.rpm

2. Install the package:

rpm -ivh webmin-1.410-1.noarch.rpm

Thats it! If all goes well, Webmin will start up on port 10000, where you can login as root using your root password. All in all, the process should take you no more than 5 minutes. 

{gallery}webmin{/gallery} 

No related posts.

2. Notify your DataCenter to setup Reverse DNS for the IP addresses used above
and for your name servers. (Normally you put in a ticket to request RDNS for “NS1.yourdomain.com = IP#1 NS2.yourdomain.com = IP#2″ and they will complete this within a day)

Please note it can take up to 36 hours for your name servers to propogate through the web. Once they have propogated procede to step #3.

1. Login to WHM as root

2. Click “Edit Setup” from the left hand menu. (x2 theme)

3. Type: ns1.yourdomain.com in the Primary Nameserver field.

4. Press Assign IP Address

5. Then Press Add an A Entry for this nameserver

6. Type ns2.yourdomain.com in the Secondary Nameserver field.

7. Click Assign IP Address

8. Then Click Add an A Entry for this nameserver

9. Click Manage Nameserver IPs from the left menu in WHM.

10. If you see any nameservers that do not belong to you, remove them now.

11. Click Initial NameServer Setup from the left menu in WHM
Run this for the “initial name server setup”

12. Login to SSH to restart bind. (Be sure it restarts, so don’t do this in WHM)

13. Type: service named stop

14. Type: service named start

15. While in SSH do the following to verify installation has gone well.

16. Type: pico /etc/wwwacct.conf
Scroll to the bottom and insure that your name servers are listed like this:
NS ns1.yourdomain.com
NS2 ns2.yourdomain.com

17. While still in SSH do the following to verify installation has gone correctly.
Type: pico /etc/resolv.conf
It should look like this:

domain yourdomain.com
search yourdomain.com
#nameserver 127.0.0.1
nameserver ip address #1 here
nameserver ip address #2 here

18. Check one more thing in SSH to make sure Name Server Configuration went GOOD.
Type: pico /etc/nameserverips
It should look like this:
ip address 1=ns1.yourdomain.com
ip address 2=ns2.yourdomain.com

There may be more IP Addresses that are equal to 0 this is ok.

No related posts.

Log into root via SSH and disable selinux

nano /etc/selinux/config

and change

SELINUX=enforcing

to

SELINUX=disabled

You now need to either reboot for this to take effect, or disable it using this command:

setenforce 0

Then check if it is disabled:

cat /selinux/enforce

If this returns 0, SELinux is disabled. 

Also you need to stop your firewall IP tables

service iptables stop

iptables -L -n

Ok, so now after you have ordered a cPanel licence you need to initiate the install with those commands

mkdir /home/cpins
cd /home/cpins
wget http://layer1.cpanel.net/latest
sh latest  

The install can take quiet a while and is depending on hardware configuration and network speed

cPanel requires a fresh/clean server!
If you are serving websites off this server (and are
not already running cPanel) this installer will
overwrite all of your config files.  You should hit
Ctrl+C NOW!!!

Now is the time to go get another cup of coffee or two

 After the initial setup , it is always a good idea to firstly install your IPs for your name servers

In SSH via root

Edit the file

pico /etc/nameserverips

The format of the file is IP address = name server.  For example:

111.222.333.444=ns1.yourserver.com

Just put the IP address that you want to add in the first part and the name of the new name server in the second part.

To reboot or add IPs

/etc/init.d/ipaliases restart 

So now just logg into WHM

http://yournewserverip:2086/ 

After accepting the licence aggreement on the first page , you get redirected to 

Step 2 Basic cPanel/WHM Setup

There you set your  Server Contact E-Mail Address and other information

Once you are satisfied with the configuration, click the Next Step Button.

Quotas are now being setup in the right frame. You can continue at any time any
quota setup will finish in the background.

If you wish to enable the nameserver, you can do so in the right frame.
Otherwise, just click the Next Step button. 

 {mospagebreak}

Lets say that we are activating name servers and those would be the results then

Name Server Activated
Ensuring caching-nameserver is installed
Loading "installonlyn" plugin
Setting up Install Process
Setting up repositories
Reading repository metadata in from local files
Excluding Packages in global exclude list
Finished
Parsing package install arguments
Nothing to do
Activating name server monitoring (chkservd)
Setting up rndc configuration
Checking in /etc/named.conf to rcs system
Changing ownership of /etc/named.conf: named:named
Restarting Bind
Starting named: [
OK
]
Restarting Nameserver
Starting named: [
OK
]
Restarting chkservd
Stopping chkservd: [
OK
]
Starting chkservd: [
OK
]

Please complete the resolver
configuration in the right frame, and then click the Next Step button.

Please set the mysql
password in the right frame, and click Next Step.

And thats it

Inital Setup is now
complete, click below to enter your Web Host Manager®

So the first thing after the initial setup should be that you add your second IP onto the server, so you name servers will function properly!

 go to IP Functions > Add New IP >

and add your second IP there ! 

OK, so now you can set the second name server in SSH

pico /etc/nameserverips 

After this is done assign the IP adress and add an A entry for this name server in

Main >> Server
Configuration >> Basic cPanel/WHM
Setup

Additionally I would strongly suggest to always do a yum upgrade to the newest kernel !

 uname -a

yum update \kernel*

Make sure the kernel is in grub

cat /boot/grub/grub.conf

and then reboot

 shutdown -rf now

{mospagebreak}

OK, so after this initiall setup and reboot , lets go and secure this box now

Chirpy from ConfigServer has got some great free tools to help you with that

1/ ConfigServer Security & Firewall (csf) 

A Stateful Packet Inspection (SPI) firewall, Login/Intrusion Detection and
Security application for Linux servers.

http://www.configserver.com/cp/csf.html 

2/ ConfigServer Explorer (cse)

This is an exclusive!  and free! add-on product for cPanel/WHM. The product
provides you with a full featured Filesystem explorer and a Virtual Console to
use within your web browser in WHM. It gives you root access from the top level
of your disks and allows you to enter non-interactive commands and see the
output.

http://www.configserver.com/cp/cse.html

3/ ConfigServer Mail Queues (cmq)

This is an exclusive!  and free! add-on product for cPanel/WHM. The product
provides you with a full featured interface to the cPanel exim email queues from
within WHM.

http://www.configserver.com/cp/cmq.html

4/ ConfigServer Mail Manage (cmm)

This is an exclusive!  and free! add-on product for cPanel/WHM. The product
provides you with an interface to the cPanel user accounts email configuration
without having to login to their accounts. It is domain based rather than
account based and allows you to do all the following from within WHM:

http://www.configserver.com/cp/cmm.html

Additionally you really want to protect your tmp folder as well

 /scripts/securetmp

and follow the prompts. 

OK, so now lets log back in WHM to finish the initial securing

 {mospagebreak}

Go to

Main >> DNS Functions >> Add an A Entry for your Hostname and add the A entry!

Go to

 Main >> Server
Configuration >> Tweak Settings

 Those are the settings I got and they suit my fine although I guess it depends totaly on the individuals needs

{mospagebreak}

OK , so now to Main >> Security >> Security Center

To be continued  

No related posts.

Here’s what made it work for me (run as
root):


1) First remove all gems

gem uninstall rails
gem uninstall mongrel
gem uninstalled fastthread

2)
Now run cPanel’s Ruby installation script

/scripts/installruby  
gem install rails 
gem install mongrel 
gem install fastthread

3)
Now second command as required by cPanel

/usr/local/cpanel/bin/ror_setup

4)
It seems cPanel’s "fast head" forgets to install, fast Thread

gem install fastthread

5)
App’s don’t run still? – The issue in my case was outdated GEM! Mongrel did
not want to run with the old version. So run:

gem update --system

6)
Restart cPanel (not required but I like to do it)

/etc/init.d/cpanel restart

And
there you have it. Works like a charm on my Centos 5

No related posts.

The vast majority of affected systems are initially accessed using SSH with no indications of brute force or exploitation of the underlying service. Despite non-trivial passwords, intermediary users and nonstandard ports, the attacker is able to gain access to the affected servers with no password failures. The majority of the affect servers come from a single undisclosed data-center. All affected systems have password-based authentication enabled. Based upon these findings, it’s believed that the attacker has gained access to a database of root login credentials for a large group of Linux servers.

Once access is gained, the attacker downloads and compiles Stealth Zapper 1.0, which is used to clean all traces of the attackers access and movements through the system. The attacker then downloads a script used to gather information from Apache and compiles a list of statistics for each site hosted on the server. This information is then sent to an undisclosed location for the attacker to view. Once the information is sent successfully, the attacker downloads an agent binary built from the Boxer 0.99 BETA 3 root-kit. This binary is secured with encrypted keys to only allow access from the attackers Boxer installation. This agent binary is built with several additional scripts developed by the attacker to load a web server into memory and inject the random JavaScript into the HTML code after Apache has served the file, but before it has traveled through the TCP transport back to the web site visitor. The attacker will first run the agent binary to load it into memory. This activates the root-kit, which will then go on to copy itself to the seven binary locations below which will keep the agent running at all times, including after a reboot.

/sbin/ifconfig

/sbin/fsck

/sbin/route

/bin/basename

/bin/cat

/bin/mount

/bin/touch

The rootkit renames these system binaries by adding a random set of characters to the end of the file name. Additionally, a 0 byte file with a different set of random characters is created based upon the target binary’s name similar to the following:

/sbin/routewWmVTnBL6szkobbNZ9Iz

/sbin/routeGnAxnt168fMJAxHiru22

These files are hidden on the live filesystem of an affected system. In order to view these files, the system must be rebooted into a safe environment such as a system rescue CD.
The JavaScript being loaded by this web server is directing users to another server that scans the web site user for a number of known vulnerabilities. These vulnerabilities are then used to add the web site user to a bot net. More information about the JavaScript hacks can be found at: http://www.finjan.com/Pressrelease.aspx?id=1820&PressLan=1819&lan=3.
If you feel your server is compromised, you can run the tests below to confirm.
The easiest test is to attempt to create a directory with a numerical name:

mkdir 1

If your server is compromised, this will result in the error below:

[root\@cpanel ~]# mkdir 1
mkdir: cannot create directory `1′: No such file or directory

This isn’t always the case in older variants of the rootkit. To be certain your server isn’t compromised, it’s best to sniff packets for a brief 3-5 minute period. You can do this using the command below:

tcpdump -nAs 2048 src port 80 | grep "[a-zA-Z]\{5\}\.js’"

If this reports packets being sent that match the regex above, then the server is most likely compromised. Additional detection methods require an in-depth knowledge of kernel debugging.
Cleaning the Random JavaScript Toolkit requires the server to be booted into a safe environment and the removal of all infected binaries. Since it is believed that the attacker has access to the database of login credentials, the only way to prevent being hacked again is changing the password and not releasing it to anyone. The preferred method however is to move to SSH Keys and remove password authentication altogether. It is recommended that you contact your data-center, NOC, or a qualified administrator to have the server properly cleaned , the OS reinstalled and secured.

More information on this issue as well as discussions can be found at the following URLs:

1. http://forums.cpanel.net

2. http://www.webhostingtalk.com/showthread.php?t=651748
3. http://www.finjan.com/Pressrelease.aspx?id=1820&PressLan=1819&lan=3

No related posts.

Very annoying , isnt it

So we did a yum upgrade and a force upcp and tried to recompile again with this new result

So it appeared easyapache had  problems with the krb5 packages

Tried to find the related packages

rpm -qa | grep krb5

and got those results

krb5-libs-1.5-29
pam_krb5-2.2.14-1
krb5-devel-1.6.1-17.el5
krb5-devel-1.5-29
krb5-libs-1.5-29
pam_krb5-2.2.11-1
krb5-workstation-1.5-29
krb5-libs-1.6.1-17.el5
pam_krb5-2.2.11-1
krb5-devel-1.5-29
krb5-libs-1.6.1-17.el5
pam_krb5-2.2.14-1

So I deleted the conflicting packages including the dependances ( –allmatches will show you the info , for example :  rpm -e –allmatches krb5-libs-1.5-29 )

 rpm -e –allmatches krb5-libs-1.5-29 krb5-devel-1.5-29.i386 krb5-devel-1.5-29.x86_646 openssl-devel-0.9.8b-8.3.el5_0.2.x86_64

So by deleting hopefully eapache will start
I run another  force upcp and the new attempt of recompiling Apache to 2.2.8 with easyapache 3 in WHM was successful

Additional info:
To find your easyapache error logs go here

cd /usr/local/cpanel/logs/easy/apache/

I guess you could call this, trials and tribulations of a cPanel Junkie

No related posts.