Full Disaster Recovery

Valid for versions 94 through the latest version

Version:

82

94


Overview

Warning:
  • Only experienced system administrators should use the procedures in this document.
  • cPanel, L.L.C. cannot support failed hard drives or hard drive recovery. We are not responsible for any data loss.

In some circumstances, you may need to completely restore your server from a crashed or failed hard drive.

Prepare to restore your server

If you have recent remote backups and can’t boot the failed hard drive, you can proceed to rebuilding your server.

Warning:

If you don’t have recent remote backups, we strongly recommend that you rework your backup strategy to store regularly-scheduled backups in a remote location.

If you don’t have recent remote backups but can boot the failed hard drive, perform one of the following actions:

  • If you can access WHM from the failed hard drive, create a backup with WHM’s Backup Configuration interface (WHM >> Home >> Backup >> Backup Configuration).
  • If you can’t access WHM from the failed hard drive, run the /usr/local/cpanel/scripts/pkgacct script.

After you complete the backup procedure, transfer the backup files to a remote location.

Rebuild the server

Some data centers and hosting providers offer system restoration and imaging. Contact your hosting provider for more information.

Important:
  • The replacement cPanel & WHM server should contain identical software to the original server. This includes MySQL® or MariaDB®, PHP, PHP extensions, Apache®, Apache extensions, and other software options. If you do not use the same version of MySQL or MariaDB on the new server, you may encounter serious errors.
  • If you wish to update any of the software on the server, we strongly recommend that you perform the transfer or restoration first and then perform any updates.

Manually provision the new hard drive

Important:
  • If you use an ISO, get it from the operating system’s official mirrors. Third-party drive images may cause unexpected problems.
  • If you use CentOS, we strongly recommend that you use the CentOS Minimal ISO.

To manually provision the new hard drive, perform the following steps:

  1. Install a new hard drive as the primary hard drive and make the old hard drive the secondary hard drive.
  2. Install the operating system. For more information, read our Installation Guide documentation.
  3. Install cPanel & WHM with the following commands:
    1
    2
    3
    
    cd /home
    curl -o latest -L https://securedownloads.cpanel.net/latest
    sh latest

Request a new hard drive and image

Make sure to request that the data center installs your old hard drive as a secondary drive. After they finish this, perform one of the following actions to update cPanel & WHM to the latest version:

  • Run the /usr/local/cpanel/scripts/upcp script.
  • Use WHM’s Upgrade to Latest Version interface (WHM >> Home >> cPanel >> Upgrade to Latest Version).

Recover your files

If you don’t have a recent remote backup and can’t boot the failed hard drive on your server, you can proceed to restoring your files.

In all other scenarios, perform the following steps as the root user to recover your files:

  1. Mount the secondary drive in a chroot environment. To do this, run the following commands, where sdb1 represents the device name of your secondary drive:
    1
    2
    
    mkdir /mnt/chroot/
    mount /dev/sdb1 /mnt/chroot/
    Important:

    You must mount the drive and partitions before you can bind mount the filesystems.

  2. Mount the proc, dev, and sys filesystems in a chroot environment. To do this, run the following commands:
    1
    2
    3
    
    mount --bind /proc /mnt/chroot/proc
    mount --rbind /dev /mnt/chroot/dev
    mount --bind /sys /mnt/chroot/sys
    Note:

    We use the rbind command instead of the bind command for the dev directory in order to ensure that we mount the /dev/pts partition properly.

  3. Run the grep chroot /etc/mtab command to verify that the system properly mounted each filesystem. Your results should resemble the following output:
    1
    2
    3
    4
    
    /dev/sdb1 /mnt/chroot none rw,bind 0 0
    /mnt/chroot/proc /proc none rw,bind 0 0
    /mnt/chroot/dev /dev none rw,bind 0 0
    /mnt/chroot/sys /sys none rw,bind 0 0
  4. Run the screen command to start a screen session. This lets you use the chroot environment in a session that you can reconnect to if you lose your connection. Disconnections from sessions with chroot environments can cause problems for services such as MySQL, which may experience InnoDB issues.
  5. Run the following commands to let the system perform as if you booted into the crashed drive:
    1
    2
    
    chroot /mnt/chroot /bin/bash -l
    export PS1="{chrooted}$PS1"

    Your results should resemble the following output:

    {chrooted}bash-4.1#
  6. Run the source /etc/environment && source /etc/profile command to recover the cPanel & WHM environment.
  7. To start required services from the old hard drive, run the necessary commands to start any services that the /usr/local/cpanel/scripts/pkgacct script will require. For example, you will need the MySQL service for Webmail databases, or the PostgreSQL® service if you use that database service.
    • If your server runs CentOS 6, run the following command:
      service mysql start
    • If your server runs CentOS 7 or 8, run the following command:
      test -f /etc/sysconfig/mysql && source /etc/sysconfig/mysql; /usr/sbin/mysqld --daemonize --user=mysql --pid-file=/var/run/mysqld/mysqld.pid ${MYSQLD_OPTS}
  8. To start dnsadmin, run the following command:
    /usr/local/cpanel/libexec/dnsadmin-startup --start
  9. To back up accounts from the old hard drive, run the following commands. These commands store the backups in the /mnt/chroot/home directory and also stores the feature lists and packages settings:
    1
    2
    
    cd /var/cpanel/users
    for i in `ls -1 *`; do /scripts/pkgacct --skipdnszones $i; done
    Warning:

    We strongly recommend that you do not use the rsync command to back up accounts. This may cause issues with services such as MySQL.

  10. To back up service configurations from the old hard drive, run the following commands:
    1
    2
    
    cd /usr/local/cpanel
    bin/cpconftool --modules=cpanel::smtp::exim,cpanel::system::backups,cpanel::system::mysql,cpanel::system::whmconf,cpanel::easy::apache,cpanel::ui::themes --backup

    The results will resemble the following output:

    1
    2
    
    Backup Successful
    /home/whm-config-backup-configuration__to__backup-10.550000-1452006507.tar.gz
  11. Transfer backup files to remote storage. Use the scp command or the rsync command to copy the files to a remote storage location.
    Important:

    You can directly transfer the files from the old hard drive to the new hard drive. However, we strongly recommend that you transfer them to a remote location first. This protects your backup files from hardware issues that may have caused the hard drive to fail.

  12. Run the following commands to stop all services in the chroot environment, unmount the filesystems, and exit the chroot environment:
    1
    2
    3
    
    service mysql stop
    umount {/proc,/dev,/sys}
    exit
  13. To close the screen session, run the exit command.

Restore your files

To restore the content from the backup files to your new hard drive in the server, perform the following steps as the root user:

  1. Run the scp command or the rsync command to copy the files from the remote location to the new hard drive.
  2. To restore feature lists to the server, run the following commands:
    1
    2
    
    cd /backups
    tar -xzvf _var_cpanel.tar.gz var/cpanel/features tar -xzvf _var_cpanel.tar.gz var/cpanel/features
  3. To restore accounts to the server, run the following commands, where BACKUPDIRECTORY/TYPE/DATE/ represents your backup directory:
    1
    2
    
    cd BACKUPDIRECTORY/TYPE/DATE/accounts/
    for archive in `ls`; do /scripts/restorepkg $archive; done
    More:

    For more information, read our The restorepkg Script documentation.

  4. Run the /usr/local/cpanel/bin/cpconftool script to restore the configuration settings for Apache, backups, cPanel themes, Exim, MySQL, and WHM to the server.
  5. If you run CloudLinux, restore CloudLinux settings to the server.

Test the server

Test the websites, applications, and services on the new server. Make certain that you have successfully restored the server’s data and configuration.

Reconnect to a screen session

To reconnect to a screen session, perform the following steps as the root user:

  1. Run the screen -ls command to list your active sessions. Your results should resemble the following output:
    1
    2
    3
    4
    
    There are screens on:
        12565.screen1   (Detached)
        12568.screen2   (Detached)
    2 Sockets in /var/run/screen/S-root.
  2. Run the screen -x screenname command to reconnect to the session, where screenname is the session’s name.