nbdfuse - present a network block device in a FUSE filesystem
nbdfuse [-o FUSE-OPTION] [-P PIDFILE] [-r] MOUNTPOINT[/FILENAME] URI
nbdfuse MOUNTPOINT[/FILENAME] --command CMD [ARGS ...] nbdfuse MOUNTPOINT[/FILENAME] --socket-activation CMD [ARGS ...] nbdfuse MOUNTPOINT[/FILENAME] --fd N nbdfuse MOUNTPOINT[/FILENAME] --tcp HOST PORT nbdfuse MOUNTPOINT[/FILENAME] --unix SOCKET nbdfuse MOUNTPOINT[/FILENAME] --vsock CID PORT
nbdfuse presents a Network Block Device as a local file inside a FUSE filesystem.
The FUSE filesystem is mounted at MOUNTPOINT and contains a single virtual file called FILENAME (defaulting to nbd). Reads and writes to the virtual file or device are turned into reads and writes to the NBD device.
In nbdfuse ≥ 1.6 you can also create a "naked" mountpoint by mounting over any regular file called MOUNTPOINT (the existing contents of the file do not matter).
The NBD server itself can be local or remote. The server can be specified as an NBD URI (like
nbd://localhost) or in various other ways (see "MODES").
fusermount -u MOUNTPOINT to unmount the filesystem after you have used it.
If there is a remote NBD server running on
example.com at the default NBD port number (10809) then you can turn it into a local file by doing:
$ mkdir dir $ nbdfuse dir nbd://example.com & $ ls -l dir/ total 0 -rw-rw-rw-. 1 nbd nbd 1073741824 Jan 1 10:10 nbd
The file is called dir/nbd and you can read and write to it as if it is a normal file. Note that writes to the file will write to the remote NBD server. After using it, unmount it:
$ fusermount -u dir $ rmdir dir
nbdkit(1) has an -s option allowing it to serve over stdin/stdout. You can combine this with nbdfuse as follows:
$ mkdir dir $ nbdfuse dir/ramdisk --command nbdkit -s memory 1G & $ ls -l dir/ total 0 -rw-rw-rw-. 1 nbd nbd 1073741824 Jan 1 10:10 ramdisk $ dd if=/dev/urandom bs=1M count=100 of=mp/ramdisk conv=notrunc,nocreat 100+0 records in 100+0 records out 104857600 bytes (105 MB, 100 MiB) copied, 2.08319 s, 50.3 MB/s
When you have finished with the RAM disk, you can unmount it as below which will cause nbdkit to exit and the RAM disk contents to be discarded:
$ fusermount -u dir $ rmdir dir
qemu-nbd(8) cannot serve over stdin/stdout, but it can use systemd socket activation. You can combine this with nbdfuse and use it to open any file format which qemu understands:
$ mkdir dir $ nbdfuse dir/file.raw \ --socket-activation qemu-nbd -f qcow2 file.qcow2 & $ ls -l dir/ total 0 -rw-rw-rw-. 1 nbd nbd 1073741824 Jan 1 10:10 file.raw
File dir/file.raw is in raw format, backed by file.qcow2. Any changes made to dir/file.raw are reflected into the qcow2 file. To unmount the file do:
$ fusermount -u dir $ rmdir dir
nbdkit(1) is able to both access and transparently uncompress remote disk images on web servers, so you can convert them into virtual files:
$ mkdir dir $ nbdfuse dir/disk.iso \ --command nbdkit -s curl --filter=xz \ http://builder.libguestfs.org/fedora-30.xz & $ ls -l dir/ total 0 -rw-rw-rw-. 1 nbd nbd 6442450944 Jan 1 10:10 disk.iso $ file dir/disk.iso dir/disk.iso: DOS/MBR boot sector $ qemu-system-x86_64 -m 4G \ -drive file=dir/disk.iso,format=raw,if=virtio,snapshot=on $ fusermount -u dir
In this example we have used the virtual file to boot qemu, but qemu can much more efficiently access NBD servers directly so in the real world that would be the preferred method.
Display brief command line help and exit.
Display FUSE options and exit. See -o below.
Pass extra options to FUSE. To get a list of all the extra options supported by FUSE, use --fuse-help.
Some potentially useful FUSE options:
Allow other users to see the filesystem. This option has no effect unless you enable it globally in /etc/fuse.conf.
Allow the kernel to cache files (reduces the number of reads that have to go through the libnbd(3) API). This is generally a good idea if you can afford the extra memory usage.
Use these options to map UIDs and GIDs.
When nbdfuse is ready to serve, write the nbdfuse process ID (PID) to PIDFILE. This can be used in scripts to wait until nbdfuse is ready. Note you mustn't try to kill nbdfuse. Use
fusermount -u to unmount the mountpoint which will cause nbdfuse to exit cleanly.
Access the network block device read-only. The virtual file will have read-only permissions, and any writes will return errors.
Display the package name and version and exit.
Modes are used to select the NBD server. The default mode uses an NBD URI (see nbd_connect_uri(3) and https://github.com/NetworkBlockDevice/nbd/blob/master/doc/uri.md). For example this specifies a TLS-encrypted connection to
10809, with export name
nbdfuse dir nbds://example.com/disk
Other modes are:
Select command mode. In this mode an NBD server can be run directly from the command line with nbdfuse communicating with the server over the server’s stdin/stdout. Normally you would use this with
nbdkit -s. See "EXAMPLES" above and nbd_connect_command(3).
Select file descriptor mode. In this mode a connected socket is passed to nbdfuse. nbdfuse connects to the socket on the numbered file descriptor. See also nbd_connect_socket(3).
Select systemd socket activation mode. This is similar to --command, but is used for servers like qemu-nbd(8) which support systemd socket activation. See "EXAMPLES" above and nbd_connect_systemd_socket_activation(3).
Select TCP mode. Connect to an NBD server on a host and port over an unencrypted TCP socket. See also nbd_connect_tcp(3).
Select Unix mode. Connect to an NBD server on a Unix domain socket. See also nbd_connect_unix(3).
Select vsock mode. Connect to an NBD server on a
AF_VSOCK socket. See also nbd_connect_vsock(3).
It is tempting (and possible) to loop mount the file. However this will be very slow and may sometimes deadlock. Better alternatives are to use nbd-client(8) or qemu-nbd(8), or more securely libguestfs(3), guestfish(1) or guestmount(1) which can all access NBD servers.
You can use this to access NBD servers, but it is usually better (and definitely much faster) to use libnbd(3) directly instead. To access NBD servers from the command line, look at nbdsh(1). To copy to and from an NBD server use nbdcopy(1).
This program is similar in concept to nbd-client(8) (which turns NBD into /dev/nbdX device nodes), except:
nbd-client is faster because it uses a special kernel module
nbd-client requires root, but nbdfuse can be used by any user
nbdfuse virtual files can be mounted anywhere in the filesystem
nbdfuse uses libnbd to talk to the NBD server
nbdfuse requires FUSE support in the kernel
qemu-nbd(8) can also attach itself to /dev/nbdX device nodes. The differences from nbdfuse are similar to the list above.
libnbd(3), nbdcopy(1), nbdinfo(1), nbdsh(1), fusermount(1), mount.fuse(8), nbd_connect_uri(3), nbd_connect_command(3), nbd_connect_socket(3), nbd_connect_systemd_socket_activation(3), nbd_connect_tcp(3), nbd_connect_unix(3), nbd_connect_vsock(3), libguestfs(3), guestfish(1), guestmount(1), nbdkit(1), nbdkit-loop(1), qemu-nbd(8), nbd-client(8).
Richard W.M. Jones
Copyright (C) 2019 Red Hat Inc.
This library is free software; you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version.
This library is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License along with this library; if not, write to the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA