• NAME
• DESCRIPTION
  • Easy mounting at boot time
  • Mounting using systemd mount points
• LOADING THE LINUX KERNEL MODULE
  • RHEL and nbd.ko
• ATTACHING NBD DEVICES TO A VIRTUAL MACHINE
  • Using libvirt XML
  • Using qemu directly
• SEE ALSO
• AUTHORS
• COPYRIGHT
• LICENSE

NAME

nbdkit-client - how to mount NBD filesystems on a client machine

DESCRIPTION

For NBD exports that contain filesystems there are several approaches to mounting them on a physical machine.

For virtual machines, see the section "ATTACHING NBD DEVICES TO A VIRTUAL MACHINE" at the end.

Easy mounting at boot time

For simple setups the following method is the easiest way to get an NBD filesystem to mount at boot. Create or edit /etc/rc.local or /etc/rc.d/rc.local:

 #!/bin/sh -
 nm-online
 modprobe nbd
 nbd-client server /dev/nbd0
 mount /dev/nbd0 /mnt

Mounting using systemd mount points

You can use systemd mount points to mount NBD filesystems at boot and/or on demand.

Set up an nbdtab(5) mapping. If /etc/nbdtab doesn't exist, then create it first. Add this line:

 nbd0 server / bs=512,persist

As a workaround for https://github.com/NetworkBlockDevice/nbd/issues/91 you must currently modify the nbd@.service file:

 # cp /usr/lib/systemd/system/nbd@.service /etc/systemd/system/
 # vi /etc/systemd/system/nbd@.service

and edit or create these settings in the [Service] section:

 [Service]
 Type=oneshot
 RemainAfterExit=yes
 ExecStart=/usr/sbin/nbd-client %i
 ExecStop=/usr/sbin/nbd-client -d /dev/%i

Finally create a systemd mount file called /etc/systemd/system/mnt.mount:

 [Unit]
 Requires=nbd@nbd0.service
 [Mount]
 What=/dev/nbd0
 Where=/mnt
 Type=ext4

You can either reboot now or do:

 # systemctl start mnt.mount

Other systemd services which need this mount point can depend on this mount unit.

LOADING THE LINUX KERNEL MODULE

The native Linux NBD client is a kernel module called nbd.ko. It is not always loaded on demand. To ensure it is loaded you may need to do:

 # echo nbd > /etc/modules-load.d/nbd.conf

This will not take effect until you reboot, so to load it right away do:

 # modprobe nbd

RHEL and nbd.ko

Red Hat Enterprise Linux 8 enabled the nbd.ko Linux kernel module but only for Unix domain sockets (ie. local connections). This means you cannot connect to an NBD server over a TCP network. This also affects Linux distributions derived from RHEL like CentOS, Alma and others.

This does not affect use of nbdkit as an NBD server, only the Linux kernel as an NBD client. Userspace Linux clients such as libnbd(3) tools will work.

ATTACHING NBD DEVICES TO A VIRTUAL MACHINE

Notice in these cases that the virtual machine does not use the NBD protocol directly. Instead, the virtual machine sees a local disk. Thus there is no need to enable an NBD client or kernel module inside the virtual machine. Behind the scenes the hypervisor (eg. Qemu) converts the local disk into an NBD connection.

Using libvirt XML

Use the virsh(1) edit subcommand to modify the libvirt XML of a virtual machine:

 # virsh edit guest-name

The <disk> element should be placed in the <devices> section of the XML, after any other <disk> elements. For more information about libvirt XML see https://libvirt.org/formatdomain.html

For NBD devices served over a Unix domain socket (nbdkit -U option) add:

 <disk device="disk" type="network">
   <source protocol="nbd">
     <host transport="unix" socket="/path/to/unix.sock"/>
   </source>
   <target dev="vdb" bus="virtio"/>
   <driver name="qemu" type="raw"/>
 </disk>

If using a TCP socket (nbdkit -p option):

 <disk device="disk" type="network">
   <source protocol="nbd">
     <host name="localhost" port="10809"/>
   </source>
   <target dev="vdb" bus="virtio"/>
   <driver name="qemu" type="raw"/>
 </disk>

Using qemu directly

Qemu can open NBD URIs. To get nbdkit to show the URI it is serving use the --print-uri option.

For example:

 $ nbdkit -f -U - --print-uri memory 1G
 nbd+unix://?socket=/tmp/nbdkitTV6kS8/socket
 Shell-quoted URI: "nbd+unix://?socket=/tmp/nbdkitTV6kS8/socket"
 Command to query the NBD endpoint:
   nbdinfo "nbd+unix://?socket=/tmp/nbdkitTV6kS8/socket"

 $ qemu-system-x86_64 [...] \
     -drive file="nbd+unix://?socket=/tmp/nbdkitTV6kS8/socket",format=raw,if=virtio

SEE ALSO

nbdkit(1), nbdkit-loop(1), nbdkit-service(1), nbd-client(8), nbdtab(5), systemd(1), systemd.mount(5), virsh(1), https://libvirt.org/formatdomain.html.

AUTHORS

Richard W.M. Jones

COPYRIGHT

Copyright Red Hat

LICENSE

Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

• Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.

• Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.

• Neither the name of Red Hat nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission.

THIS SOFTWARE IS PROVIDED BY RED HAT AND CONTRIBUTORS ''AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL RED HAT OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.