guestfs-faq - libguestfs Frequently Asked Questions (FAQ)
libguestfs is a way to create, access and modify disk images. You can look inside disk images, modify the files they contain, create them from scratch, resize them, and much more. It's especially useful from scripts and programs and from the command line.
libguestfs is a C library (hence "lib-"), and a set of tools built on this library, and a set of bindings in many different programming languages.
For more information about what libguestfs can do read the introduction on the home page (http://libguestfs.org).
Virt tools (website: http://virt-tools.org) are a whole set of virtualization management tools aimed at system administrators. Some of them come from libguestfs, some from libvirt and many others from other open source projects. So virt tools is a superset of libguestfs. However libguestfs comes with many important tools. See http://libguestfs.org for a full list.
libvirt is not a requirement for libguestfs.
libguestfs works with any disk image, including ones created in VMware, KVM, qemu, VirtualBox, Xen, and many other hypervisors, and ones which you have created from scratch.
Red Hat sponsors (ie. pays for) development of libguestfs and a huge number of other open source projects. But you can run libguestfs and the virt tools on many different Linux distros and Mac OS X. Some virt tools have been ported to Windows.
- vs. kpartx
Libguestfs takes a different approach from kpartx. kpartx needs root, and mounts filesystems on the host kernel (which can be insecure - see "SECURITY" in guestfs(3)). Libguestfs isolates your host kernel from guests, is more flexible, scriptable, supports LVM, doesn't require root, is isolated from other processes, and cleans up after itself. Libguestfs is more than just file access because you can use it to create images from scratch.
- vs. vdfuse
vdfuse is like kpartx but for VirtualBox images. See the kpartx comparison above. You can use libguestfs on the partition files exposed by vdfuse, although it's not necessary since libguestfs can access VirtualBox images directly.
- vs. qemu-nbd
nbd is like kpartx but for qcow2 images. See the kpartx comparison above. You can use libguestfs and qemu-nbd together for access to block devices over the network.
- vs. mounting filesystems in the host
Mounting guest filesystems in the host is insecure and should be avoided completely for untrusted guests. Use libguestfs to provide a layer of protection against filesystem exploits. See also guestmount(1).
- vs. parted
Libguestfs supports LVM. Libguestfs uses parted and provides most parted features through the libguestfs API.
The simplest method is:
Libguestfs development happens along an unstable branch and we periodically create a stable branch which we backport stable patches to. To find out more, read "LIBGUESTFS VERSION NUMBERS" in guestfs(3).
If you are a Red Hat customer using Red Hat Enterprise Linux, please contact Red Hat Support: http://redhat.com/support
There is a mailing list, mainly for development, but users are also welcome to ask questions about libguestfs and the virt tools: https://www.redhat.com/mailman/listinfo/libguestfs
You can also talk to us on IRC channel
#libguestfs on FreeNode. We're not always around, so please stay in the channel after asking your question and someone will get back to you.
For other virt tools (not ones supplied with libguestfs) there is a general virt tools mailing list: https://www.redhat.com/mailman/listinfo/virt-tools-list
Please use the following link to enter a bug in Bugzilla:
Include as much detail as you can and a way to reproduce the problem.
Include the full output of libguestfs-test-tool(1).
This error indicates that qemu failed or the host kernel could not boot. To get further information about the failure, you have to run:
If, after using this, you still don't understand the failure, contact us (see previous section).
See also "LIBGUESTFS GOTCHAS" in guestfs(3) for some "gotchas" with using the libguestfs API.
Typical symptoms of this problem:
- You get an error when you create a file where the filename contains non-ASCII characters, particularly non 8-bit characters from Asian languages (Chinese, Japanese, etc). The filesystem is VFAT.
- When you list a directory from a VFAT filesystem, filenames appear as question marks.
This is a design flaw of the GNU/Linux system.
VFAT stores long filenames as UTF-16 characters. When opening or returning filenames, the Linux kernel has to translate these to some form of 8 bit string. UTF-8 would be the obvious choice, except for Linux users who persist in using non-UTF-8 locales (the user's locale is not known to the kernel because it's a function of libc).
Therefore you have to tell the kernel what translation you want done when you mount the filesystem. The two methods are the
iocharset parameter (which is not relevant to libguestfs) and the
So to use a VFAT filesystem you must add the
utf8 flag when mounting. From guestfish, use:
><fs> mount-options utf8 /dev/sda1 /
or on the guestfish command line:
guestfish [...] -m /dev/sda1:/:utf8
or from the API:
guestfs_mount_options (g, "utf8", "/dev/sda1", "/");
The kernel will then translate filenames to and from UTF-8 strings.
We considered adding this mount option transparently, but unfortunately there are several problems with doing that:
- On some Linux systems, the
utf8mount option doesn't work. We don't precisely understand what systems or why, but this was reliably reported by one user.
- It would prevent you from using the
iocharsetparameter because it is incompatible with
utf8. It is probably not a good idea to use this parameter, but we don't want to prevent it.
The filesystem was not prepared correctly with mkisofs or genisoimage. Make sure the filesystem was created using Joliet and/or Rock Ridge extensions. libguestfs does not require any special mount options to handle the filesystem.
- Fedora ≥ 11, RHEL ≥ 5.3, EPEL 5
yum install '*guestf*'
For the latest builds, see: http://koji.fedoraproject.org/koji/packageinfo?packageID=8391
- Red Hat Enterprise Linux 6
It is part of the default install. On RHEL 6 (only) you have to install
libguestfs-winsupportto get Windows guest support.
- RHEL 6.4
A preview repository will be announced at a later date.
- Debian Squeeze (6)
Use Hilko Bengen's backport repository: http://people.debian.org/~bengen/libguestfs/
- Debian Wheezy and later (7+)
Official Debian packages are available: http://packages.debian.org/search?keywords=libguestfs (thanks Hilko Bengen).
We don't have an Ubuntu maintainer, and the packages supplied by Canonical (which are outside our control) are often broken. Try compiling from source (next section).
Canonical decided to change the permissions on the kernel so that it's not readable except by root. This is completely stupid, but they won't change it (https://bugs.launchpad.net/ubuntu/+source/linux/+bug/759725). So every user should do this:
sudo chmod 0644 /boot/vmlinuz*
- Ubuntu 10.04
- Ubuntu 12.04
libguestfs in this version of Ubuntu works, but you need to update febootstrap and seabios to the latest versions.
You need febootstrap ≥ 3.14-2 from: http://packages.ubuntu.com/precise/febootstrap
After installing or updating febootstrap, rebuild the appliance:
You need seabios ≥ 0.6.2-0ubuntu2.1 or ≥ 0.6.2-0ubuntu3 from: http://packages.ubuntu.com/precise-updates/seabios or http://packages.ubuntu.com/quantal/seabios
Also you need to do (see above):
sudo chmod 0644 /boot/vmlinuz*
Libguestfs was added to Gentoo in 2012-07. Do:
- Other Linux distro
Compile from source (next section).
- Other non-Linux distro
You'll have to compile from source, and port it.
If your Linux distro has a working port of febootstrap (that is, Fedora, Red Hat Enterprise Linux >= 6.3, Debian, Ubuntu and ArchLinux) then you should just be able to compile from source in the usual way. Download the latest tarball from http://libguestfs.org/download, unpack it, and start by reading the README file.
If you don't have febootstrap, you will need to use the "fixed appliance method". See: http://libguestfs.org/download/binaries/appliance/
Patches to port febootstrap to more Linux distros are welcome.
Because of the complexity of building the libguestfs appliance, the source RPMs provided cannot be rebuilt directly using
If you use Koji (which is open source software and may be installed locally), then the SRPMs can be rebuilt in Koji. https://fedoraproject.org/wiki/Koji
If you don't have or want to use Koji, then you have to give libguestfs access to the network so it can download the RPMs for building the appliance. You also need to set an RPM macro to tell libguestfs to use the network. Put the following line into a file called
If you are using mock, do:
mock -D '%libguestfs_buildnet 1' [etc]
That's because it does a lot of things.
By far the most important thing you can do is to install and properly configure Squid. Note that the default configuration that ships with Squid is rubbish, so configuring it is not optional.
A very good place to start with Squid configuration is here: https://fedoraproject.org/wiki/Extras/MockTricks#Using_Squid_to_Speed_Up_Mock_package_downloads
Make sure Squid is running, and that the environment variables
$ftp_proxy are pointing to it.
With Squid running and correctly configured, appliance builds should be reduced to a few minutes.
Note: Most of the information in this section has moved: guestfs-performance(1).
In libguestfs < 1.13.16, the mount command ("guestfs_mount" in guestfs(3)) enabled option
-o sync implicitly. This causes very poor write performance, and was one of the main gotchas for new libguestfs users.
For libguestfs < 1.13.16, replace mount with
mount-options, leaving the first parameter as an empty string.
You can also do this with more recent versions of libguestfs, but if you know that you are using libguestfs ≥ 1.13.16 then it's safe to use plain mount.
If the underlying disk is not fully allocated (eg. sparse raw or qcow2) then writes can be slow because the host operating system has to do costly disk allocations while you are writing. The solution is to use a fully allocated format instead, ie. non-sparse raw, or qcow2 with the
libguestfs caches a large-ish appliance in:
If the environment variable
TMPDIR is defined, then
$TMPDIR/.guestfs-<UID> is used instead.
It is safe to delete this directory when you are not using libguestfs.
If the input to virt-sparsify(1) is raw, then the output will be raw sparse. Make sure you are measuring the output with a tool which understands sparseness such as
du-sh. It can make a huge difference:
$ ls -lh test1.img -rw-rw-r--. 1 rjones rjones 100M Aug 8 08:08 test1.img $ du -sh test1.img 3.6M test1.img
(Compare the apparent size 100M vs the actual size 3.6M)
If all this confuses you, use a non-sparse output by specifying the --convert option, eg:
virt-sparsify --convert qcow2 disk.raw disk.qcow2
We recommend you start by reading the API overview: "API OVERVIEW" in guestfs(3).
Although the API overview covers the C API, it is still worth reading even if you are going to use another programming language, because the API is the same, just with simple logical changes to the names of the calls:
C guestfs_ln_sf (g, target, linkname); Python g.ln_sf (target, linkname); OCaml g#ln_sf target linkname; Perl $g->ln_sf (target, linkname); Shell (guestfish) ln-sf target linkname PHP guestfs_ln_sf ($g, $target, $linkname);
Once you're familiar with the API overview, you should look at this list of starting points for other language bindings: "USING LIBGUESTFS WITH OTHER PROGRAMMING LANGUAGES" in guestfs(3).
In general, yes. However this is not legal advice. You should read the license that comes with libguestfs, and if you have specific questions about your obligations when distributing libguestfs, contact a lawyer. In the source tree the license is in the file
COPYING.LIB (LGPLv2+ for the library and bindings) and
COPYING (GPLv2+ for the standalone programs).
There are two
LIBGUESTFS_* environment variables you can set in order to get more information from libguestfs.
Set this to 1 and libguestfs will print out each command / API call in a format which is similar to guestfish commands.
Set this to 1 in order to enable massive amounts of debug messages. If you think there is some problem inside the libguestfs appliance, then you should use this option.
To set these from the shell, do this before running the program:
export LIBGUESTFS_TRACE=1 export LIBGUESTFS_DEBUG=1
For csh/tcsh the equivalent commands would be:
setenv LIBGUESTFS_TRACE 1 setenv LIBGUESTFS_DEBUG 1
For further information, see: "ENVIRONMENT VARIABLES" in guestfs(3).
You can use the same environment variables above. Alternatively use the guestfish options -x (to trace commands) or -v (to get the full debug output), or both.
For further information, see: guestfish(1).
For best results, call these functions as early as possible, just after creating the guestfs handle if you can, and definitely before calling launch.
Use the event API. For examples, see: "SETTING CALLBACKS TO HANDLE EVENTS" in guestfs(3).
Enable debugging and then read this documentation on the appliance boot process: "INTERNALS" in guestfs(3).
Enable debugging and look at the full output. If you cannot work out what is going on, file a bug report, including the complete output of libguestfs-test-tool(1).
We offer a command called guestmount(1) which lets you mount guest filesystems on the host. This is implemented as a FUSE module. Why don't we just implement the whole of libguestfs using this mechanism, instead of having the large and rather complicated API?
The reasons are twofold. Firstly, libguestfs offers API calls for doing things like creating and deleting partitions and logical volumes, which don't fit into a filesystem model very easily. Or rather, you could fit them in: for example, creating a partition could be mapped to
mkdir /fs/hda1 but then you'd have to specify some method to choose the size of the partition (maybe
echo 100M > /fs/hda1/.size), and the partition type, start and end sectors etc., but once you've done that the filesystem-based API starts to look more complicated than the call-based API we currently have.
The second reason is for efficiency. FUSE itself is reasonably efficient, but it does make lots of small, independent calls into the FUSE module. In guestmount these have to be translated into messages to the libguestfs appliance which has a big overhead (in time and round trips). For example, reading a file in 64 KB chunks is inefficient because each chunk would turn into a single round trip. In the libguestfs API it is much more efficient to download an entire file or directory through one of the streaming calls like
The problems are similar to the problems with FUSE.
GVFS is a better abstraction than POSIX/FUSE. There is an FTP backend for GVFS, which is encouraging because FTP is conceptually similar to the libguestfs API. However the GVFS FTP backend makes multiple simultaneous connections in order to keep interactivity, which we can't easily do with libguestfs.
Usually this is not a good idea. The question is answered in more detail in this mailing list posting: https://www.redhat.com/archives/libguestfs/2010-August/msg00024.html
A lot of people are confused by the two superficially similar tools we provide:
$ guestfish --ro -a guest.img ><fs> run ><fs> fsck /dev/sda1 $ virt-rescue --ro guest.img ><rescue> /sbin/fsck /dev/sda1
And the related question which then arises is why you can't type in full shell commands with all the --options in guestfish (but you can in virt-rescue(1)).
guestfish(1) is a program providing structured access to the guestfs(3) API. It happens to be a nice interactive shell too, but its primary purpose is structured access from shell scripts. Think of it more like a language binding, like Python and other bindings, but for shell. The key differentiating factor of guestfish (and the libguestfs API in general) is the ability to automate changes.
virt-rescue(1) is a free-for-all freeform way to boot the libguestfs appliance and make arbitrary changes to your VM. It's not structured, you can't automate it, but for making quick ad-hoc fixes to your guests, it can be quite useful.
But, libguestfs also has a "backdoor" into the appliance allowing you to send arbitrary shell commands. It's not as flexible as virt-rescue, because you can't interact with the shell commands, but here it is anyway:
><fs> debug sh "cmd arg1 arg2 ..."
Note that you should not rely on this. It could be removed or changed in future. If your program needs some operation, please add it to the libguestfs API instead.
What's the deal with
guestfish -i? Why does virt-cat only work on a real VM image, but virt-df works on any disk image? What does "no root device found in this operating system image" mean?
These questions are all related at a fundamental level which may not be immediately obvious.
At the guestfs(3) API level, a "disk image" is just a pile of partitions and filesystems.
In contrast, when the virtual machine boots, it mounts those filesystems into a consistent hierarchy such as:
/ (/dev/sda2) | +-- /boot (/dev/sda1) | +-- /home (/dev/vg_external/Homes) | +-- /usr (/dev/vg_os/lv_usr) | +-- /var (/dev/vg_os/lv_var)
(or drive letters on Windows).
The API first of all sees the disk image at the "pile of filesystems" level. But it also has a way to inspect the disk image to see if it contains an operating system, and how the disks are mounted when the operating system boots: "INSPECTION" in guestfs(3).
Users expect some tools (like virt-cat(1)) to work with VM paths:
virt-cat fedora.img /var/log/messages
How does virt-cat know that
/var is a separate partition? The trick is that virt-cat performs inspection on the disk image, and uses that to translate the path correctly.
Some tools (including virt-cat(1), virt-edit(1), virt-ls(1)) use inspection to map VM paths. Other tools, such as virt-df(1) and virt-filesystems(1) operate entirely at the raw "big pile of filesystems" level of the libguestfs API, and don't use inspection.
guestfish(1) is in an interesting middle ground. If you use the -a and -m command line options, then you have to tell guestfish exactly how to add disk images and where to mount partitions. This is the raw API level.
If you use the -i option, libguestfs performs inspection and mounts the filesystems for you.
no root device found in this operating system image is related to this. It means inspection was unable to locate an operating system within the disk image you gave it. You might see this from programs like virt-cat if you try to run them on something which is just a disk image, not a virtual machine disk image.
Richard W.M. Jones (
rjones at redhat dot com)
Copyright (C) 2012 Red Hat Inc.