1 Libguestfs is a library for accessing and modifying guest disk images.
2 Amongst the things this is good for: making batch configuration
3 changes to guests, getting disk used/free statistics (see also:
4 virt-df), migrating between virtualization systems (see also:
5 virt-p2v), performing partial backups, performing partial guest
6 clones, cloning guests and changing registry/UUID/hostname info, and
9 Libguestfs uses Linux kernel and qemu code, and can access any type of
10 guest filesystem that Linux and qemu can, including but not limited
11 to: ext2/3/4, btrfs, FAT and NTFS, LVM, many different disk partition
12 schemes, qcow, qcow2, vmdk.
14 Libguestfs provides ways to enumerate guest storage (eg. partitions,
15 LVs, what filesystem is in each LV, etc.). It can also run commands
16 in the context of the guest. Also you can access filesystems over
19 Libguestfs is a library that can be linked with C and C++ management
20 programs (or management programs written in OCaml, Perl, Python, Ruby,
21 Java, Haskell or C#). You can also use it from shell scripts or the
24 Libguestfs was written by Richard W.M. Jones (rjones@redhat.com) and
25 hacked on by lots of other people. For discussion, development,
26 patches, etc. please use the mailing list:
28 http://www.redhat.com/mailman/listinfo/libguestfs
32 ----------------------------------------------------------------------
34 http://libguestfs.org/
38 ----------------------------------------------------------------------
40 - recent QEMU >= 0.10 with vmchannel support
41 http://lists.gnu.org/archive/html/qemu-devel/2009-02/msg01042.html
49 - XDR, rpcgen (on Linux these are provided by glibc)
51 - squashfs-tools (mksquashfs only)
53 - genisoimage / mkisofs
55 - (Optional) hivex >= 1.2.1 to build Windows Registry support
57 - (Optional) FUSE to build the FUSE module
59 - (Optional) Augeas (http://augeas.net/)
61 - perldoc (pod2man, pod2text) to generate the manual pages and
64 - (Optional) Readline to have nicer command-line editing in guestfish.
66 - (Optional) xmllint to validate virt-inspector RELAX NG schema
68 - (Optional) OCaml + OCaml library xml-light if you want to rebuild
69 the generated files, and also to build the OCaml bindings
70 (http://tech.motion-twin.com/xmllight.html)
72 - (Optional) local Fedora mirror
74 - (Optional) Perl if you want to build the perl bindings
76 - (Optional) Python if you want to build the python bindings
78 - (Optional) Ruby, rake if you want to build the ruby bindings
80 - (Optional) Java, JNI, jpackage-utils if you want to build the java
83 - (Optional) GHC if you want to build the Haskell bindings
85 - (Optional) Perl XML::XPath, Sys::Virt modules (for libvirt support
88 - (Optional, but highly recommended) perl-libintl for translating perl code.
90 Running ./configure will check you have all the requirements installed
95 ----------------------------------------------------------------------
97 Then make the daemon, library and root filesystem:
99 ./configure [--with-mirror=URI]
102 Use the optional --with-mirror parameter to specify the URI of a local
103 Fedora mirror. See the discussion of the MIRROR parameter in the
104 febootstrap(8) manpage.
106 Finally run the tests:
110 If everything works, you can install the library and tools by running
111 this command as root:
117 ----------------------------------------------------------------------
119 We provide packages for Fedora >= 11 in Fedora. Use those, or build
120 from our source RPMs - it's far simpler that way.
122 You can compile libguestfs on Fedora 10 but you cannot use it with the
123 version of qemu in Fedora 10. You need to compile your own qemu, see
124 section 'qemu' below.
127 RHEL / EPEL / CentOS etc
128 ----------------------------------------------------------------------
130 We provide packages in EPEL which cover RHEL/CentOS >= 5. Use those
131 or build from our source RPMs.
135 ----------------------------------------------------------------------
137 libguestfs is now built as a package in Debian by Guido Gunther and
138 the other Debian libvirt maintainers. See:
140 http://wiki.debian.org/Teams/DebianLibvirtTeam#Packages
142 You can build for Debian in two different ways, either building a
143 Fedora-based appliance using febootstrap, yum, rpm, fakeroot,
144 fakechroot (all packaged in Debian). However the recommended way is
145 to build a Debian-based appliance using debootstrap and debirf.
147 Both ways are supported by the configure script.
151 ----------------------------------------------------------------------
153 By far the most common problem is with broken or incompatible
156 Different versions of qemu have problems booting the appliance for
157 different reasons. This varies between versions of qemu, and Linux
158 distributions which add their own patches.
160 If you find a problem, you could try using your own qemu built from
161 source (qemu is very easy to build from source), with a 'qemu
162 wrapper'. Qemu wrappers are described in the guestfs(3) manpage.
166 ----------------------------------------------------------------------
168 By default the configure script will look for qemu-kvm (KVM support).
169 You will need a reasonably recent processor for this to work. KVM is
170 much faster than using plain Qemu.
172 You may also need to enable KVM support for non-root users, by following
175 http://www.linux-kvm.org/page/FAQ#How_can_I_use_kvm_with_a_non-privileged_user.3F
177 On some systems, this will work too:
181 On some systems, the chmod will not survive a reboot, and you will
182 need to make edits to the udev configuration.
186 ----------------------------------------------------------------------
188 Previous versions of libguestfs required something called "vmchannel".
189 Vmchannel is a special device given to virtual machines which allows
190 them to communicate in some way with the host, often (but not always)
191 without using a traditional network device. In reality, there is no
192 one thing called "vmchannel". This idea has been reimplemented
193 several times under the name vmchannel, and other hypervisors have
194 their own incompatible implementation(s) too.
196 In libguestfs <= 1.0.71, we required a specific vmchannel which is
197 properly known as "guestfwd" and has been upstream in qemu since here:
199 http://lists.gnu.org/archive/html/qemu-devel/2009-02/msg01042.html
201 In libguestfs >= 1.0.71 we don't require any vmchannel implementation,
202 as long as qemu has been compiled with support for SLIRP (user mode
203 networking, or "-net user"), which is almost always the case.
205 However we still offer the ability to use vmchannel, and in future we
206 may add support for other types of qemu, which is useful in a few
207 cases, specifically where qemu packagers decide to compile out support
208 for SLIRP (qemu packagers: please don't do this).
212 ----------------------------------------------------------------------
214 If you configure with --enable-supermin then we will build a supermin
215 appliance (supermin = super-minimized). This is a very specialized
216 appliance which is built on-the-fly at runtime (specifically, when you
217 call guestfs_launch).
219 The normal appliance is a self-contained Linux operating system, based
220 on the Fedora/RHEL/CentOS Linux distro. So it contains a complete
221 copy of all the libraries and programs needed, like kernel, libc,
222 bash, coreutils etc etc.
224 The supermin appliance removes the kernel and all the executable
225 libraries and programs from the appliance. That just leaves a
226 skeleton of config files and some data files, which is obviously
227 massively smaller than the normal appliance. At runtime we rebuild
228 the appliance on-the-fly from the libraries and programs on the host
229 (eg. pulling in the real /lib/libc.so, the real /bin/bash etc.)
231 Although this process of rebuilding the appliance each time sounds
232 slow, it turns out to be faster than using the prebuilt appliance.
233 (Most of the saving comes from not compressing the appliance - it
234 transpires that decompressing the appliance is the slowest part of the
235 whole boot sequence). On my machine, a new appliance can be built in
236 under a fifth of a second, and the boot time is several seconds
239 The big advantage of the supermin appliance for distributions like
240 Fedora is that it gets security fixes automatically from the host, so
241 there is no need to rebuild the whole of libguestfs for a security
242 update in some underlying library.
244 There are several DISADVANTAGES:
246 It won't work at all except in very narrow, controlled cases like the
247 Fedora packaging case. We control the dependencies of the libguestfs
248 RPM tightly to ensure that the required binaries are actually present
251 Furthermore there are certain unlikely changes in the packages on the
252 host which could break a supermin appliance, eg. an updated library
253 which depends on an additional data file.
255 Also supermin appliances are subjected to changes in the host kernel
256 which might break compatibility with qemu -- these are, of course,
257 real bugs in any case.
259 Lastly, supermin appliances really can't be moved between branches of
260 distributions (eg. built on Fedora 12 and moved to Fedora 10) because
261 they are not self-contained and they rely on certain libraries being
262 around. You shouldn't do this anyway.
264 Use supermin appliances with caution.
267 Notes on cross-architecture support
268 ----------------------------------------------------------------------
270 At the moment we basically don't support cross-architecture or
271 32-on-64. This limits what is possible for some guests. Filesystem
272 operations and FUSE will work fine, but running commands in guests may
275 To enable this requires work for cross-architecture and 32-on-64
276 support in febootstrap, fakeroot and fakechroot.
278 The daemon/ directory contains its own configure script. This is so
279 that in future we will be able to cross-compile the daemon.
283 ----------------------------------------------------------------------
285 On my machines I can usually rebuild the appliance in around 3
286 minutes. If it takes much longer for you, use a local Fedora mirror
289 To use squid to cache yum downloads, read this first:
290 https://lists.dulug.duke.edu/pipermail/yum/2006-August/009041.html
291 (In brief, because yum chooses random mirrors each time, squid doesn't
292 work very well with default yum configuration. To get around this,
293 choose a Fedora mirror which is close to you, set this with
294 './configure --with-mirror=[...]', and then proxy the whole lot
295 through squid by setting http_proxy environment variable).
297 You will also need to substantially increase the squid configuration
299 http://fedoraproject.org/wiki/Using_Mock_to_test_package_builds#Using_Squid_to_Speed_Up_Mock_package_downloads
302 Porting to other Linux distros / non-Linux
303 ----------------------------------------------------------------------
305 libguestfs itself should be fairly portable to other Linux
306 distributions. Non-Linux ports are trickier, but we will accept
307 patches if they aren't too invasive.
309 The main porting issues are with the dependencies needed to build the
310 appliance. You will need to find or port the following packages
316 - rpm-python http://www.rpm.org/
317 - yum http://yum.baseurl.org/
318 - febootstrap http://people.redhat.com/~rjones/febootstrap/
321 Copyright and license information
322 ----------------------------------------------------------------------
324 Copyright (C) 2009-2010 Red Hat Inc.
326 The library is distributed under the LGPLv2+. The programs are
327 distributed under the GPLv2+. Please see the files COPYING and
328 COPYING.LIB for full license information.