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