1===================
2QEMU Storage Daemon
3===================
4
5Synopsis
6--------
7
8**qemu-storage-daemon** [options]
9
10Description
11-----------
12
13qemu-storage-daemon provides disk image functionality from QEMU, qemu-img, and
14qemu-nbd in a long-running process controlled via QMP commands without running
15a virtual machine. It can export disk images, run block job operations, and
16perform other disk-related operations. The daemon is controlled via a QMP
17monitor and initial configuration from the command-line.
18
19The daemon offers the following subset of QEMU features:
20
21* Block nodes
22* Block jobs
23* Block exports
24* Throttle groups
25* Character devices
26* Crypto and secrets
27* QMP
28* IOThreads
29
30Commands can be sent over a QEMU Monitor Protocol (QMP) connection. See the
31:manpage:`qemu-storage-daemon-qmp-ref(7)` manual page for a description of the
32commands.
33
34The daemon runs until it is stopped using the ``quit`` QMP command or
35SIGINT/SIGHUP/SIGTERM.
36
37**Warning:** Never modify images in use by a running virtual machine or any
38other process; this may destroy the image. Also, be aware that querying an
39image that is being modified by another process may encounter inconsistent
40state.
41
42Options
43-------
44
45.. program:: qemu-storage-daemon
46
47Standard options:
48
49.. option:: -h, --help
50
51  Display help and exit
52
53.. option:: -V, --version
54
55  Display version information and exit
56
57.. option:: -T, --trace [[enable=]PATTERN][,events=FILE][,file=FILE]
58
59  .. include:: ../qemu-option-trace.rst.inc
60
61.. option:: --blockdev BLOCKDEVDEF
62
63  is a block node definition. See the :manpage:`qemu(1)` manual page for a
64  description of block node properties and the :manpage:`qemu-block-drivers(7)`
65  manual page for a description of driver-specific parameters.
66
67.. option:: --chardev CHARDEVDEF
68
69  is a character device definition. See the :manpage:`qemu(1)` manual page for
70  a description of character device properties. A common character device
71  definition configures a UNIX domain socket::
72
73  --chardev socket,id=char1,path=/var/run/qsd-qmp.sock,server=on,wait=off
74
75.. option:: --export [type=]nbd,id=<id>,node-name=<node-name>[,name=<export-name>][,writable=on|off][,bitmap=<name>]
76  --export [type=]vhost-user-blk,id=<id>,node-name=<node-name>,addr.type=unix,addr.path=<socket-path>[,writable=on|off][,logical-block-size=<block-size>][,num-queues=<num-queues>]
77  --export [type=]vhost-user-blk,id=<id>,node-name=<node-name>,addr.type=fd,addr.str=<fd>[,writable=on|off][,logical-block-size=<block-size>][,num-queues=<num-queues>]
78  --export [type=]fuse,id=<id>,node-name=<node-name>,mountpoint=<file>[,growable=on|off][,writable=on|off]
79
80  is a block export definition. ``node-name`` is the block node that should be
81  exported. ``writable`` determines whether or not the export allows write
82  requests for modifying data (the default is off).
83
84  The ``nbd`` export type requires ``--nbd-server`` (see below). ``name`` is
85  the NBD export name (if not specified, it defaults to the given
86  ``node-name``). ``bitmap`` is the name of a dirty bitmap reachable from the
87  block node, so the NBD client can use NBD_OPT_SET_META_CONTEXT with the
88  metadata context name "qemu:dirty-bitmap:BITMAP" to inspect the bitmap.
89
90  The ``vhost-user-blk`` export type takes a vhost-user socket address on which
91  it accept incoming connections. Both
92  ``addr.type=unix,addr.path=<socket-path>`` for UNIX domain sockets and
93  ``addr.type=fd,addr.str=<fd>`` for file descriptor passing are supported.
94  ``logical-block-size`` sets the logical block size in bytes (the default is
95  512). ``num-queues`` sets the number of virtqueues (the default is 1).
96
97  The ``fuse`` export type takes a mount point, which must be a regular file,
98  on which to export the given block node. That file will not be changed, it
99  will just appear to have the block node's content while the export is active
100  (very much like mounting a filesystem on a directory does not change what the
101  directory contains, it only shows a different content while the filesystem is
102  mounted). Consequently, applications that have opened the given file before
103  the export became active will continue to see its original content. If
104  ``growable`` is set, writes after the end of the exported file will grow the
105  block node to fit.
106
107.. option:: --monitor MONITORDEF
108
109  is a QMP monitor definition. See the :manpage:`qemu(1)` manual page for
110  a description of QMP monitor properties. A common QMP monitor definition
111  configures a monitor on character device ``char1``::
112
113  --monitor chardev=char1
114
115.. option:: --nbd-server addr.type=inet,addr.host=<host>,addr.port=<port>[,tls-creds=<id>][,tls-authz=<id>][,max-connections=<n>]
116  --nbd-server addr.type=unix,addr.path=<path>[,tls-creds=<id>][,tls-authz=<id>][,max-connections=<n>]
117  --nbd-server addr.type=fd,addr.str=<fd>[,tls-creds=<id>][,tls-authz=<id>][,max-connections=<n>]
118
119  is a server for NBD exports. Both TCP and UNIX domain sockets are supported.
120  A listen socket can be provided via file descriptor passing (see Examples
121  below). TLS encryption can be configured using ``--object`` tls-creds-* and
122  authz-* secrets (see below).
123
124  To configure an NBD server on UNIX domain socket path
125  ``/var/run/qsd-nbd.sock``::
126
127  --nbd-server addr.type=unix,addr.path=/var/run/qsd-nbd.sock
128
129.. option:: --object help
130  --object <type>,help
131  --object <type>[,<property>=<value>...]
132
133  is a QEMU user creatable object definition. List object types with ``help``.
134  List object properties with ``<type>,help``. See the :manpage:`qemu(1)`
135  manual page for a description of the object properties.
136
137.. option:: --pidfile PATH
138
139  is the path to a file where the daemon writes its pid. This allows scripts to
140  stop the daemon by sending a signal::
141
142    $ kill -SIGTERM $(<path/to/qsd.pid)
143
144  A file lock is applied to the file so only one instance of the daemon can run
145  with a given pid file path. The daemon unlinks its pid file when terminating.
146
147  The pid file is written after chardevs, exports, and NBD servers have been
148  created but before accepting connections. The daemon has started successfully
149  when the pid file is written and clients may begin connecting.
150
151Examples
152--------
153Launch the daemon with QMP monitor socket ``qmp.sock`` so clients can execute
154QMP commands::
155
156  $ qemu-storage-daemon \
157      --chardev socket,path=qmp.sock,server=on,wait=off,id=char1 \
158      --monitor chardev=char1
159
160Launch the daemon from Python with a QMP monitor socket using file descriptor
161passing so there is no need to busy wait for the QMP monitor to become
162available::
163
164  #!/usr/bin/env python3
165  import subprocess
166  import socket
167
168  sock_path = '/var/run/qmp.sock'
169
170  with socket.socket(socket.AF_UNIX, socket.SOCK_STREAM) as listen_sock:
171      listen_sock.bind(sock_path)
172      listen_sock.listen()
173
174      fd = listen_sock.fileno()
175
176      subprocess.Popen(
177          ['qemu-storage-daemon',
178           '--chardev', f'socket,fd={fd},server=on,id=char1',
179           '--monitor', 'chardev=char1'],
180          pass_fds=[fd],
181      )
182
183  # listen_sock was automatically closed when leaving the 'with' statement
184  # body. If the daemon process terminated early then the following connect()
185  # will fail with "Connection refused" because no process has the listen
186  # socket open anymore. Launch errors can be detected this way.
187
188  qmp_sock = socket.socket(socket.AF_UNIX, socket.SOCK_STREAM)
189  qmp_sock.connect(sock_path)
190  ...QMP interaction...
191
192The same socket spawning approach also works with the ``--nbd-server
193addr.type=fd,addr.str=<fd>`` and ``--export
194type=vhost-user-blk,addr.type=fd,addr.str=<fd>`` options.
195
196Export raw image file ``disk.img`` over NBD UNIX domain socket ``nbd.sock``::
197
198  $ qemu-storage-daemon \
199      --blockdev driver=file,node-name=disk,filename=disk.img \
200      --nbd-server addr.type=unix,addr.path=nbd.sock \
201      --export type=nbd,id=export,node-name=disk,writable=on
202
203Export a qcow2 image file ``disk.qcow2`` as a vhosts-user-blk device over UNIX
204domain socket ``vhost-user-blk.sock``::
205
206  $ qemu-storage-daemon \
207      --blockdev driver=file,node-name=file,filename=disk.qcow2 \
208      --blockdev driver=qcow2,node-name=qcow2,file=file \
209      --export type=vhost-user-blk,id=export,addr.type=unix,addr.path=vhost-user-blk.sock,node-name=qcow2
210
211Export a qcow2 image file ``disk.qcow2`` via FUSE on itself, so the disk image
212file will then appear as a raw image::
213
214  $ qemu-storage-daemon \
215      --blockdev driver=file,node-name=file,filename=disk.qcow2 \
216      --blockdev driver=qcow2,node-name=qcow2,file=file \
217      --export type=fuse,id=export,node-name=qcow2,mountpoint=disk.qcow2,writable=on
218
219See also
220--------
221
222:manpage:`qemu(1)`, :manpage:`qemu-block-drivers(7)`, :manpage:`qemu-storage-daemon-qmp-ref(7)`
223