How to Install OpenVZ On Debian 9

Traducciones al Español
Estamos traduciendo nuestros guías y tutoriales al Español. Es posible que usted esté viendo una traducción generada automáticamente. Estamos trabajando con traductores profesionales para verificar las traducciones de nuestro sitio web. Este proyecto es un trabajo en curso.
Create a Linode account to try this guide with a $ credit.
This credit will be applied to any valid services used during your first  days.

What is OpenVZ?

OpenVZ is a software-based OS virtualization tool enabling the deployment, management, and modification of isolated virtual Linux environments from within a host Linux distribution. An extensive array of prebuilt OS templates in a variety of Linux distributions allow users to rapidly download and deploy virtual environments with ease.

Before You Begin

  1. Working through this tutorial requires a root user account, and is written as if commands are issued as the root user. Readers choosing to use a limited user account will need to prefix commands with sudo where required. If you have yet to create a limited user account, follow the steps in the Setting Up and Securing a Compute Instance guide.

  2. The instructions in this guide were written for and tested on Debian 9 only. They are unlikely to work for other Debian or Ubuntu distributions.

  3. Certain essential modifications to your Debian 9 system are required to run OpenVZ, including the removal and replacement of Systemd with SystemV, and the use of a custom Linux kernel. Before continuing, be certain that all software currently installed on the machine will be compatible with these changes.

    Note
    Although not required, it is recommended to create a separate Ext4 filesystem partition for OpenVZ templates. By default, both the Debian 9 installer and the Linode Manager format newly created partitions with Ext4. For information on how to accomplish this configuration, follow the steps to create a disk in the Managing Disks and Storage on a Linode guide.

Optional: Create A Separate Partition For OpenVZ Templates

If you intend to dedicate an entire Linode VPS to running OpenVZ and no other services, it’s recommended to create separate partitions for the host server and its processes, and any OpenVZ virtual server templates. The following table illustrates the recommended partitioning scheme:

PartitionDescriptionTypical Size
/Root partition4-12 GB
swapPaging partition2 times RAM or RAM + 2GB (depending on available hard drive space)
/vzPartition to host OpenVZ templatesAll remaining hard drive space
  1. Log into your Linode Manager and select your Linode. Power down the machine, and verify the job completed by viewing the Host Job Queue section. Under the Disks tab, click Create a new Disk. Add a label of your choosing, select “ext4” in the Type drop-down menu, and allocate as much space as you can in the Size field. Click Save Changes; an optimal configuration will resemble the image below.

  2. Under the Dashboard tab, click your main Configuration Profile. Under the Block Device Assignment tab, assign your new partition to an open device. Click Save Changes when finished.

  3. Boot the Linode and log in via SSH. Issue the command below to verify that the new disk has been created properly. The output will display your newly created disk.

    fdisk -l
    
  4. Create a mount point for the new device:

     mkdir /vztemp
    
  5. Mount the new disk. Be sure to replace /dev/sdc with your device name:

     mount /dev/sdc /vztemp
    

Remove the Metadata Checksum Feature From Ext4 Volumes

Before OpenVZ can be installed, the system must be configured for compatibility. Debian 9 supports a new checksum feature that is incompatible with custom OpenVZ kernels. Depending on your preference, you may choose to either remove metadata_csum from a mounted partition or reformat the affected partition to a compatible Ext4 volume. Choose either method and follow the instructions in the appropriate section below.

  1. List available disk partitions.

     lsblk
    
  2. Check if metadata_csum is installed in any mounted disk partitions shown in Step 1 (not including the SWAP partition). Follow the format below for each partition, replacing /dev/sda1 with the appropriate volume name. If the below command yields no output for mounted disk volumes, you may skip this section.

     dumpe2fs -h /dev/sda1 2>/dev/null | grep -e metadata_csum
    

Remove Metadata Checksum From Mounted Partitions

  1. Issue the commands below to add code to the fsck file:

     echo "copy_exec /sbin/e2fsck" | sudo tee -a /usr/share/initramfs-tools/hooks/fsck
     echo "copy_exec /sbin/tune2fs" | sudo tee -a /usr/share/initramfs-tools/hooks/fsck
    
  2. Create a new file in the directory designated below and name it tune. Copy and paste the text below into this new file and save:

    File: /etc/initramfs-tools/scripts/local-premount/tune
    1
    2
    3
    4
    5
    6
    7
    8
    9
    
    #!/bin/sh
    
    if [ "$readonly" != "y" ] ;
    then exit 0 ;
    fi
    
    e2fsck -f $Volume
    tune2fs -O -metadata_csum $Volume
    e2fsck -f $Volume
  3. Update file properties and existing initramfs image to load the tune script:

     chmod 755 /etc/initramfs-tools/scripts/local-premount/tune
     update-initramfs -u -k all
    
  4. Reboot your system and run the command below to verify that metadata_csum was disabled from all affected partitions. Again, replace “/dev/sda1” with the correct volume names.

     dumpe2fs -h /dev/sda1 2>/dev/null | grep -e metadata_csum
    

Format a Compatible Ext4 Volume

  1. Choose the Ext4 volume you would like to format and issue the command below, replacing /dev/sda3 with your selected volume. An output of “0” indicates success.

    Important
    Formatting a volume with the mkfs command may result in data loss.
     mkfs -t ext4 -O -metadata_csum /dev/sda3
    

Replace Systemd with SystemV

  1. Install SystemV utilities.

     apt install sysvinit-core sysvinit-utils
    
  2. From the Linode Manager, reboot your machine to release Systemd.

  3. Remove Systemd from your machine:

     apt --auto-remove remove systemd
    
  4. Create file avoid-systemd and paste in the contents below:

    File: /etc/apt/preferences.d/avoid-systemd
    1
    2
    3
    
    Package: *systemd*
    Pin: release *
    Pin-Priority: -1

Add OpenVZ Repository

  1. Create a new repository source file and paste in the contents below:

    File: /etc/apt/sources.list.d/openvz.list
    deb http://download.openvz.org/debian jessie main
    deb http://download.openvz.org/debian wheezy main
  2. Add the repository key to your system:

     wget -qO - http://ftp.openvz.org/debian/archive.key | sudo apt-key add -
    
  3. As of this guide’s publication date, the OpenVZ repository key is invalid, and issuing the apt update command will generate a warning from the system. The command should succeed. If it does not, update the system with the following argument:

     apt --allow-unauthenticated update
    

Install OpenVZ Packages

  1. Install OpenVZ with required packages.

     KPackage="linux-image-openvz-$(dpkg --print-architecture)"
     sudo apt --allow-unauthenticated --install-recommends install $KPackage vzdump ploop initramfs-tools dirmngr
    
  2. The installation should create a new directory, /vz. If this directory does not exist after the installation, create a symbolic link using the command below:

     ln -s /var/lib/vz/ /vz
    
  3. Create file vznet.conf and paste in the line below:

    File: /etc/vz/vznet.conf
    1
    
    EXTERNAL_SCRIPT="/usr/sbin/vznetaddbr"
  4. This step is optional, and will cause OpenVZ virtual instances to stop when the OpenVZ service is stopped. If this behavior is desired, issue the command below.

     echo 'VE_STOP_MODE=stop' | sudo tee -a /etc/vz/vznet.conf
    

Boot Into The OpenVZ Kernel

The system must be configured to boot the OpenVZ kernel each time the server is restarted.

  1. Open the grub.cfg file in less, or your preferred text editor:

     less /boot/grub/grub.cfg
    
  2. Within the grub.cfg file, look for a section resembling the following:

    File: /boot/grub/grub.cfg
     1
     2
     3
     4
     5
     6
     7
     8
     9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    19
    20
    21
    
    . . .
    
    menuentry 'Debian GNU/Linux' --class debian --class gnu-linux --class gnu --class os $menuentry_id_option 'gnulinux-simple-e025e52b-91c4-4f64-962d-79f244caa92a' {
        gfxmode $linux_gfx_mode
        insmod gzio
        if [ x$grub_platform = xxen ]; then insmod xzio; insmod lzopio; fi
        insmod ext2
        set root='hd0'
        if [ x$feature_platform_search_hint = xy ]; then
            search --no-floppy --fs-uuid --set=root --hint-bios=hd0 --hint-efi=hd0 --hint-baremetal=ahci0  e025e52b-91c4-4f64-962d-79f244caa92a
        else
            search --no-floppy --fs-uuid --set=root e025e52b-91c4-4f64-962d-79f244caa92a
        fi
        echo    'Loading Linux 4.9.0-3-amd64 ...'
        linux   /boot/vmlinuz-4.9.0-3-amd64 root=/dev/sda ro console=ttyS0,19200n8 net.ifnames=0
        echo    'Loading initial ramdisk ...'
        initrd  /boot/initrd.img-4.9.0-3-amd64
        }
    submenu 'Advanced options for Debian GNU/Linux' $menuentry_id_option 'gnulinux-advanced-e025e52b-91c4-4f64-962d-79f244caa92a'
    
    . . .
    Copy the text entry preceeding `submenu`, in this example the text would be: **Advanced options for Debian GNU/Linux**.
    
  3. Within the grub.cfg file underneath the “submenu” line, you will see multiple indented “menuentry” sections. These represent the available kernels. From these, you need to locate the newly installed OpenVZ kernel menu entry. It should look similar to the content below. Note that some will be recovery kernels and should be ignored:

    File: /boot/grub/grub.cfg
     1
     2
     3
     4
     5
     6
     7
     8
     9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    19
    20
    
    . . .
    
        menuentry 'Debian GNU/Linux, with Linux 2.6.32-openvz-042stab123.9-amd64' --class debian --class gnu-linux --class gnu --class os $menuentry_id_option 'gnulinux-2.6.32-openvz-042stab123.9-amd64-advanced-e025e52b-91c4-4f64-962d-79f244caa92a' {
                gfxmode $linux_gfx_mode
                insmod gzio
                if [ x$grub_platform = xxen ]; then insmod xzio; insmod lzopio; fi
                insmod ext2
                set root='hd0'
                if [ x$feature_platform_search_hint = xy ]; then
                  search --no-floppy --fs-uuid --set=root --hint-bios=hd0 --hint-efi=hd0 --hint-baremetal=ahci0  e025e52b-91c4-4f64-962d-79f244caa92a
                else
                  search --no-floppy --fs-uuid --set=root e025e52b-91c4-4f64-962d-79f244caa92a
                fi
                echo    'Loading Linux 2.6.32-openvz-042stab123.9-amd64 ...'
                linux   /boot/vmlinuz-2.6.32-openvz-042stab123.9-amd64 root=/dev/sda ro console=ttyS0,19200n8 net.ifnames=0
                echo    'Loading initial ramdisk ...'
                initrd  /boot/initrd.img-2.6.32-openvz-042stab123.9-amd64
        }
    
    . . .
      Again, write down the text directly after "menuentry" in single quotes. Here, the text to copy is **Debian GNU/Linux, with Linux 2.6.32-openvz-042stab123.9-amd64**.
    
  4. Close the grub.cfg file and open /etc/default/grub in your preferred text editor. Locate the line that begins with GRUB_DEFAULT=. Delete the default value for this parameter and enter the text you copied in the previous steps, following the format below. With the example above, the value would be the following:

     GRUB_DEFAULT="Advanced options for Debian GNU/Linux>Debian GNU/Linux, with Linux 2.6.32-openvz-042stab123.9-amd64"
    

    Note that both copied strings are separated with the carrot “>” character.

  5. Save and close the grub file, and issue the below command to reload the grub bootloader with the new kernel value:

     update-grub
    
  6. By default, kernel loading is not handled by Grub, but by the Linode Manager. Login to your Linode Manager and select your Linode. Click on your configuration profile. Under the “Boot Settings” section, select “GRUB 2” from the Kernel dropdown-list (see image below). Save your changes and exit.

    Linode Manager - Select Kernel

  7. Reboot your server and issue the command below to verify the OpenVZ kernel was loaded:

     uname -r
    

    If the OpenVZ kernel was not loaded, it is most likely the grub file that is misconfigured. Check and make certain the correct kernel was chosen and entered correctly.

Download And Deploy An OS Template

  1. Start the OpenVZ service:

     service vz start
     service vz status
    
  2. Register with the official OpenVZ template repository:

     sudo gpg --recv-keys $(echo $(sudo gpg --batch --search-keys security@openvz.org 2>&1 | grep -ie ' key.*created' | sed -e 's|key|@|g' | cut -f 2 -d '@') | cut -f 1 -d ' ' | cut -f 1 -d ',')
    
  3. Edit /etc/vz/vz.conf and change the following line to use simfs instead of ploop:

    File: /etc/vz/vz.conf
    1
    
    VE_LAYOUT=simfs
  4. List available OS templates for download:

     vztmpl-dl --list-remote
    
  5. From the available list of templates, select one to download. Using the format below, issue the following command, replacing centos7-x86_64 with the template you selected:

     vztmpl-dl --gpg-check centos7-x86_64
    
  6. OpenVZ refers to each installed OS template as a “Container.” You must create a Container ID (CTID) for each downloaded template. Issue the below command, replacing [CTID] with any number (101 is recommended) and the CentOS 7 template name with your downloaded template.

     vzctl create [CTID] --ostemplate centos7-x86_64
    
  7. If you set up a separate disk partition for OpenVZ templates, use the command below to create the container within the new disk. Replace –ostemplate with your template name, and –name with a descriptive name of your choice:

    vzctl create [CTID] --ostemplate debian-8.0-x86_64 --layout simfs --name centos7 --private /vztemp/vz/private/$VEID --root /vztemp/vz/root/$VEID --config basic
    
  8. A configuration file will now have been created for your OS template. The location of this file is specified as part of the output of the previous command. Open this file now to make the following changes below. The config file will be named in the [CTID].conf format.

    • Give your virtual environment an IP address. The recommended format is 192.168.0.[CTID]. In this case it would be 192.168.0.101.
    • Provide a nameserver. Google’s nameserver (8.8.8.8) should be sufficient.
    • If you have trouble booting into your virtual environment, you may try changing VE_LAYOUT back to ploop from simfs.

    You may also configure other options at your discretion, such as SWAP and RAM allocation. Save and close when finished.

    File: /etc/vz/conf/101.conf
     1
     2
     3
     4
     5
     6
     7
     8
     9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    19
    20
    21
    22
    23
    24
    25
    
    . . .
    
    # RAM
    PHYSPAGES="0:256M"
    
    # Swap
    SWAPPAGES="0:512M"
    
    # Disk quota parameters (in form of softlimit:hardlimit)
    DISKSPACE="2G:2.2G"
    DISKINODES="131072:144179"
    QUOTATIME="0"
    
    # CPU fair scheduler parameter
    CPUUNITS="1000"
    
    NETFILTER="stateless"
    VE_ROOT="/var/lib/vz/root/$VEID"
    VE_PRIVATE="/var/lib/vz/private/$VEID"
    VE_LAYOUT="simfs"
    OSTEMPLATE="centos7-x86_64"
    ORIGIN_SAMPLE="vswap-256m"
    NAMESERVER="8.8.8.8"
    IP_ADDRESS="192.168.0.101/24"
    HOSTNAME="centos-7"
  9. Boot into your newly created container using the commands below. Replace [CTID] with your container’s CTID number. To exit any container session while leaving the virtual environment running, type exit in the command line.

     vzctl start [CTID]
     vzctl enter [CTID]
    

Configure Internet Access To Containers

Containers have no way to access the internet or be accessed from the internet. The host server must be configured to pass on requests to and from each installed virtual environment.

Configure Access From Container To Internet

Note
You may need to login as root with su - in order to run the iptables-save commands in this section.
  1. On the host server, issue the following command using Iptables. Replace the brackets and contents with the appropriate information. For the container IP address, make sure to list it in CIDR notation. Include IP address and subnet, or xxx.xxx.xxx.xxx/xx, in order to encompass a range of IP addresses that will enable access to any containers added in the future. For example, entering 192.168.0.0/24 will setup routing for IP addresses 192.168.0.0 through 192.168.0.255:

     iptables -t nat -A POSTROUTING -s [container IP] -o eth0 -j SNAT --to [host server IP]
    
  2. If you have iptables-persistent installed, skip this step. Save the new Iptables rules:

     iptables-save > /etc/iptables.conf
    
  3. Configure your firewall to allow forwarded requests. If you are not using iptables-persistent save the rule:

     iptables -A FORWARD -s 192.168.0.0/24 -j ACCEPT
     iptables-save > /etc/iptables.conf
    
  4. You should now have access to the internet from within your container environment. Try updating packages from your container to verify the connection.

Configure Access From Internet To Container

  1. If you need to access a specific service on your container from the internet, you will need to reserve a port on the host machine and route access through it. Issue the following command, replacing any bracketed values with the appropriate information:

     iptables -t nat -A PREROUTING -p tcp -d [host_ip] --dport [host_port_number] -i eth0 -j DNAT --to-destination [container_ip:container_port_number]
    
  2. Save your new rule. Skip this step if you have iptables-persistent installed:

     iptables-save > /etc/iptables.conf
    

Where To Go From Here

After installing OpenVZ, downloading a template, creating a container, and configuring internet access, your virtual environment will function exactly like any normal Linux environment: requiring regular updates, security configuration, etc. Most configuration can be done from the host server via OpenVZ commands.

See the “OpenVZ Basic Operations” link in the External Resources section to familiarize yourself with basic administration commands. Additional user-created templates can also be downloaded, which are not included in the main template listing. You can find these by following the “OpenVZ User Contributed Templates” link.

More Information

You may wish to consult the following resources for additional information on this topic. While these are provided in the hope that they will be useful, please note that we cannot vouch for the accuracy or timeliness of externally hosted materials.

This page was originally published on


Your Feedback Is Important

Let us know if this guide was helpful to you.


Join the conversation.
Read other comments or post your own below. Comments must be respectful, constructive, and relevant to the topic of the guide. Do not post external links or advertisements. Before posting, consider if your comment would be better addressed by contacting our Support team or asking on our Community Site.
The Disqus commenting system for Linode Docs requires the acceptance of Functional Cookies, which allow us to analyze site usage so we can measure and improve performance. To view and create comments for this article, please update your Cookie Preferences on this website and refresh this web page. Please note: You must have JavaScript enabled in your browser.