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
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
|
====================
libvirt Installation
====================
.. contents::
Installing from distribution repositories
-----------------------------------------
This is the recommended option to install libvirt. Libvirt is present in the
package repositories of all major distributions. Installing a package from the
package manager ensures that it's properly compiled, installed, started, and
updated during the lifecycle of the distribution.
For users who wish to use the most recent version, certain distributions also
allow installing the most recent versions of virtualization packages:
**Fedora**
Refer to https://fedoraproject.org/wiki/Virtualization_Preview_Repository
**Gentoo**
The ``app-emulation/libvirt`` is regularly updated, but newest versions are
usually marked as testing by the ``~*`` keyword.
**openSUSE**
Refer to https://build.opensuse.org/package/show/Virtualization/libvirt
Preparing sources
-----------------
Libvirt can be built both from release tarballs and from a git checkout using
the same steps once the source code is prepared. Note that the build system
requires that the build directory is separate from the top level source
directory.
By default further steps will build libvirt inside a subdirectory of the source
tree named ``build``.
Refer to the `downloads page <downloads.html>`__ for official tarballs and the
git repository.
Unpacking a source tarball
~~~~~~~~~~~~~~~~~~~~~~~~~~
Download a source tarball of the version you want to compile and unpack it
using the following commands:
::
$ xz -dc libvirt-x.x.x.tar.xz | tar xvf -
$ cd libvirt-x.x.x
Git checkout
~~~~~~~~~~~~
A git checkout/clone is already in correct state for next steps. Just change
your working directory to the checkout.
Configuring the project
-----------------------
The libvirt build process uses the **Meson** build system. To configure for a
build use the following command. Note that the ``build`` argument is the name
of the build directory which will be created.
::
$ meson build [options]
To get the complete list of the options run the following command:
::
$ meson configure
Be aware that by default the build is configured with a local ``prefix`` path
which will not interoperate with OS vendor provided binaries, since the UNIX
socket paths will all be different. To produce a build that is compatible with
normal OS vendor prefixes, use
::
$ meson build -Dsystem=true
Explicitly enabling required functionality
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
By default each module of functionality of libvirtd is optionally enabled,
meaning it will be enabled if the build environment contains the required
dependencies.
To ensure that your build contains the required functionality it's recommended
to explicitly enable given modules, in which case the configure step will end
with an error if dependencies are not present. **Example:** to build the
libvirt project with support for the **qemu** driver use the following options:
::
$ meson build -Dsystem=true -Ddriver_qemu=enabled
Notes:
~~~~~~
By default when the ``meson`` is run from within a GIT checkout, it will turn
on -Werror for builds. This can be disabled with --werror=false, but this is
not recommended.
Please ensure that you have the appropriate minimal ``meson`` version installed
in your build environment. The minimal version for a specific package can be
checked in the top level ``meson.build`` file in the ``meson_version`` field.
Compiling the sources
---------------------
To build the configured project run (note that ``-C build`` is a path to the
build directory):
::
$ ninja -C build
The ``build`` directory now contains the built binaries.
Additionally you can also run the test suite:
::
$ ninja -C build test
Running compiled binaries from build directory
----------------------------------------------
For testing or development purposes it's usually not necessary to install the
built binaries into your system. Instead simply run libvirt directly from the
source tree. For example to run a privileged libvirtd instance
::
$ su -
# service libvirtd stop (or systemctl stop libvirtd.service)
# /home/to/your/checkout/build/src/libvirtd
It is also possible to run virsh directly from the build tree using the
./run script (which sets some environment variables):
::
$ pwd
/home/to/your/checkout/build
$ ./run ./tools/virsh ....
**Note:** The libvirt project provides `multiple daemons <daemons.html>`__ and
the above steps may replace only some of them with the custom compiled instances.
In most cases this should work but keep that fact in mind.
Installing compiled binaries
----------------------------
**Important:** Manual installation of libvirt is generally not recommended and
you should prefer installation from your operating system's package repository
or from manually built packages which are then installed using the package
manager. Overwriting an installation of libvirt from the package manager by a
manually compiled installation may not work properly.
Installing the compiled binaries into the appropriate location (based on
how the build was configured) is done by the following command:
::
$ sudo ninja -C build install
Note the use of **sudo** with the *ninja install* command. Using
sudo is only required when installing to a location your user does not
have write access to. Installing to a system location is a good example
of this.
If you are installing to a location that your user *does* have write
access to, then you can instead run the *ninja install* command without
putting **sudo** before it.
After installation you you **may** have to run ``ldconfig`` or a similar
utility to update your list of installed shared libs, or adjust the paths where
the system looks for binaries and shared libraries.
The libvirt project provides `multiple daemons <daemons.html>`__ based on your
configuration. You have to ensure that you start the appropriate processes for
the freshly installed libvirt to be usable (e.g. even monolithic ``libvirtd``
requires in most configurations that ``virtlogd`` is started).
|
