5 guestfs - Library for accessing and modifying virtual machine images
11 guestfs_h *g = guestfs_create ();
12 guestfs_add_drive (g, "guest.img");
14 guestfs_mount (g, "/dev/sda1", "/");
15 guestfs_touch (g, "/hello");
16 guestfs_umount (g, "/");
19 cc prog.c -o prog -lguestfs
21 cc prog.c -o prog `pkg-config libguestfs --cflags --libs`
25 Libguestfs is a library for accessing and modifying guest disk images.
26 Amongst the things this is good for: making batch configuration
27 changes to guests, getting disk used/free statistics (see also:
28 virt-df), migrating between virtualization systems (see also:
29 virt-p2v), performing partial backups, performing partial guest
30 clones, cloning guests and changing registry/UUID/hostname info, and
33 Libguestfs uses Linux kernel and qemu code, and can access any type of
34 guest filesystem that Linux and qemu can, including but not limited
35 to: ext2/3/4, btrfs, FAT and NTFS, LVM, many different disk partition
36 schemes, qcow, qcow2, vmdk.
38 Libguestfs provides ways to enumerate guest storage (eg. partitions,
39 LVs, what filesystem is in each LV, etc.). It can also run commands
40 in the context of the guest. Also you can access filesystems over
43 Libguestfs is a library that can be linked with C and C++ management
44 programs (or management programs written in OCaml, Perl, Python, Ruby,
45 Java, PHP, Haskell or C#). You can also use it from shell scripts or the
48 You don't need to be root to use libguestfs, although obviously you do
49 need enough permissions to access the disk images.
51 Libguestfs is a large API because it can do many things. For a gentle
52 introduction, please read the L</API OVERVIEW> section next.
54 There are also some example programs in the L<guestfs-examples(3)>
59 This section provides a gentler overview of the libguestfs API. We
60 also try to group API calls together, where that may not be obvious
61 from reading about the individual calls in the main section of this
66 Before you can use libguestfs calls, you have to create a handle.
67 Then you must add at least one disk image to the handle, followed by
68 launching the handle, then performing whatever operations you want,
69 and finally closing the handle. By convention we use the single
70 letter C<g> for the name of the handle variable, although of course
71 you can use any name you want.
73 The general structure of all libguestfs-using programs looks like
76 guestfs_h *g = guestfs_create ();
78 /* Call guestfs_add_drive additional times if there are
79 * multiple disk images.
81 guestfs_add_drive (g, "guest.img");
83 /* Most manipulation calls won't work until you've launched
84 * the handle 'g'. You have to do this _after_ adding drives
85 * and _before_ other commands.
89 /* Now you can examine what partitions, LVs etc are available.
91 char **partitions = guestfs_list_partitions (g);
92 char **logvols = guestfs_lvs (g);
94 /* To access a filesystem in the image, you must mount it.
96 guestfs_mount (g, "/dev/sda1", "/");
98 /* Now you can perform filesystem actions on the guest
101 guestfs_touch (g, "/hello");
103 /* This is only needed for libguestfs < 1.5.24. Since then
104 * it is done automatically when you close the handle. See
105 * discussion of autosync in this page.
109 /* Close the handle 'g'. */
112 The code above doesn't include any error checking. In real code you
113 should check return values carefully for errors. In general all
114 functions that return integers return C<-1> on error, and all
115 functions that return pointers return C<NULL> on error. See section
116 L</ERROR HANDLING> below for how to handle errors, and consult the
117 documentation for each function call below to see precisely how they
118 return error indications. See L<guestfs-examples(3)> for fully worked
123 The image filename (C<"guest.img"> in the example above) could be a
124 disk image from a virtual machine, a L<dd(1)> copy of a physical hard
125 disk, an actual block device, or simply an empty file of zeroes that
126 you have created through L<posix_fallocate(3)>. Libguestfs lets you
127 do useful things to all of these.
129 The call you should use in modern code for adding drives is
130 L</guestfs_add_drive_opts>. To add a disk image, allowing writes, and
131 specifying that the format is raw, do:
133 guestfs_add_drive_opts (g, filename,
134 GUESTFS_ADD_DRIVE_OPTS_FORMAT, "raw",
137 You can add a disk read-only using:
139 guestfs_add_drive_opts (g, filename,
140 GUESTFS_ADD_DRIVE_OPTS_FORMAT, "raw",
141 GUESTFS_ADD_DRIVE_OPTS_READONLY, 1,
144 or by calling the older function L</guestfs_add_drive_ro>. In either
145 case libguestfs won't modify the file.
147 Be extremely cautious if the disk image is in use, eg. if it is being
148 used by a virtual machine. Adding it read-write will almost certainly
149 cause disk corruption, but adding it read-only is safe.
151 You must add at least one disk image, and you may add multiple disk
152 images. In the API, the disk images are usually referred to as
153 C</dev/sda> (for the first one you added), C</dev/sdb> (for the second
156 Once L</guestfs_launch> has been called you cannot add any more images.
157 You can call L</guestfs_list_devices> to get a list of the device
158 names, in the order that you added them. See also L</BLOCK DEVICE
163 Before you can read or write files, create directories and so on in a
164 disk image that contains filesystems, you have to mount those
165 filesystems using L</guestfs_mount_options> or L</guestfs_mount_ro>.
166 If you already know that a disk image contains (for example) one
167 partition with a filesystem on that partition, then you can mount it
170 guestfs_mount_options (g, "", "/dev/sda1", "/");
172 where C</dev/sda1> means literally the first partition (C<1>) of the
173 first disk image that we added (C</dev/sda>). If the disk contains
174 Linux LVM2 logical volumes you could refer to those instead
175 (eg. C</dev/VG/LV>). Note that these are libguestfs virtual devices,
176 and are nothing to do with host devices.
178 If you are given a disk image and you don't know what it contains then
179 you have to find out. Libguestfs can do that too: use
180 L</guestfs_list_partitions> and L</guestfs_lvs> to list possible
181 partitions and LVs, and either try mounting each to see what is
182 mountable, or else examine them with L</guestfs_vfs_type> or
183 L</guestfs_file>. To list just filesystems, use
184 L</guestfs_list_filesystems>.
186 Libguestfs also has a set of APIs for inspection of unknown disk
187 images (see L</INSPECTION> below). But you might find it easier to
188 look at higher level programs built on top of libguestfs, in
189 particular L<virt-inspector(1)>.
191 To mount a filesystem read-only, use L</guestfs_mount_ro>. There are
192 several other variations of the C<guestfs_mount_*> call.
194 =head2 FILESYSTEM ACCESS AND MODIFICATION
196 The majority of the libguestfs API consists of fairly low-level calls
197 for accessing and modifying the files, directories, symlinks etc on
198 mounted filesystems. There are over a hundred such calls which you
199 can find listed in detail below in this man page, and we don't even
200 pretend to cover them all in this overview.
202 Specify filenames as full paths, starting with C<"/"> and including
205 For example, if you mounted a filesystem at C<"/"> and you want to
206 read the file called C<"etc/passwd"> then you could do:
208 char *data = guestfs_cat (g, "/etc/passwd");
210 This would return C<data> as a newly allocated buffer containing the
211 full content of that file (with some conditions: see also
212 L</DOWNLOADING> below), or C<NULL> if there was an error.
214 As another example, to create a top-level directory on that filesystem
215 called C<"var"> you would do:
217 guestfs_mkdir (g, "/var");
219 To create a symlink you could do:
221 guestfs_ln_s (g, "/etc/init.d/portmap",
222 "/etc/rc3.d/S30portmap");
224 Libguestfs will reject attempts to use relative paths and there is no
225 concept of a current working directory.
227 Libguestfs can return errors in many situations: for example if the
228 filesystem isn't writable, or if a file or directory that you
229 requested doesn't exist. If you are using the C API (documented here)
230 you have to check for those error conditions after each call. (Other
231 language bindings turn these errors into exceptions).
233 File writes are affected by the per-handle umask, set by calling
234 L</guestfs_umask> and defaulting to 022. See L</UMASK>.
238 Libguestfs contains API calls to read, create and modify partition
239 tables on disk images.
241 In the common case where you want to create a single partition
242 covering the whole disk, you should use the L</guestfs_part_disk>
245 const char *parttype = "mbr";
246 if (disk_is_larger_than_2TB)
248 guestfs_part_disk (g, "/dev/sda", parttype);
250 Obviously this effectively wipes anything that was on that disk image
255 Libguestfs provides access to a large part of the LVM2 API, such as
256 L</guestfs_lvcreate> and L</guestfs_vgremove>. It won't make much sense
257 unless you familiarize yourself with the concepts of physical volumes,
258 volume groups and logical volumes.
260 This author strongly recommends reading the LVM HOWTO, online at
261 L<http://tldp.org/HOWTO/LVM-HOWTO/>.
265 Use L</guestfs_cat> to download small, text only files. This call
266 is limited to files which are less than 2 MB and which cannot contain
267 any ASCII NUL (C<\0>) characters. However it has a very simple
270 L</guestfs_read_file> can be used to read files which contain
271 arbitrary 8 bit data, since it returns a (pointer, size) pair.
272 However it is still limited to "small" files, less than 2 MB.
274 L</guestfs_download> can be used to download any file, with no
275 limits on content or size (even files larger than 4 GB).
277 To download multiple files, see L</guestfs_tar_out> and
282 It's often the case that you want to write a file or files to the disk
285 To write a small file with fixed content, use L</guestfs_write>. To
286 create a file of all zeroes, use L</guestfs_truncate_size> (sparse) or
287 L</guestfs_fallocate64> (with all disk blocks allocated). There are a
288 variety of other functions for creating test files, for example
289 L</guestfs_fill> and L</guestfs_fill_pattern>.
291 To upload a single file, use L</guestfs_upload>. This call has no
292 limits on file content or size (even files larger than 4 GB).
294 To upload multiple files, see L</guestfs_tar_in> and L</guestfs_tgz_in>.
296 However the fastest way to upload I<large numbers of arbitrary files>
297 is to turn them into a squashfs or CD ISO (see L<mksquashfs(8)> and
298 L<mkisofs(8)>), then attach this using L</guestfs_add_drive_ro>. If
299 you add the drive in a predictable way (eg. adding it last after all
300 other drives) then you can get the device name from
301 L</guestfs_list_devices> and mount it directly using
302 L</guestfs_mount_ro>. Note that squashfs images are sometimes
303 non-portable between kernel versions, and they don't support labels or
304 UUIDs. If you want to pre-build an image or you need to mount it
305 using a label or UUID, use an ISO image instead.
309 There are various different commands for copying between files and
310 devices and in and out of the guest filesystem. These are summarised
315 =item B<file> to B<file>
317 Use L</guestfs_cp> to copy a single file, or
318 L</guestfs_cp_a> to copy directories recursively.
320 =item B<file or device> to B<file or device>
322 Use L</guestfs_dd> which efficiently uses L<dd(1)>
323 to copy between files and devices in the guest.
325 Example: duplicate the contents of an LV:
327 guestfs_dd (g, "/dev/VG/Original", "/dev/VG/Copy");
329 The destination (C</dev/VG/Copy>) must be at least as large as the
330 source (C</dev/VG/Original>). To copy less than the whole
331 source device, use L</guestfs_copy_size>.
333 =item B<file on the host> to B<file or device>
335 Use L</guestfs_upload>. See L</UPLOADING> above.
337 =item B<file or device> to B<file on the host>
339 Use L</guestfs_download>. See L</DOWNLOADING> above.
343 =head2 UPLOADING AND DOWNLOADING TO PIPES AND FILE DESCRIPTORS
345 Calls like L</guestfs_upload>, L</guestfs_download>,
346 L</guestfs_tar_in>, L</guestfs_tar_out> etc appear to only take
347 filenames as arguments, so it appears you can only upload and download
348 to files. However many Un*x-like hosts let you use the special device
349 files C</dev/stdin>, C</dev/stdout>, C</dev/stderr> and C</dev/fd/N>
350 to read and write from stdin, stdout, stderr, and arbitrary file
353 For example, L<virt-cat(1)> writes its output to stdout by
356 guestfs_download (filename, "/dev/stdout");
358 and you can write tar output to a pipe C<fd> by doing:
361 snprintf (devfd, sizeof devfd, "/dev/fd/%d", fd);
362 guestfs_tar_out ("/", devfd);
366 L</guestfs_ll> is just designed for humans to read (mainly when using
367 the L<guestfish(1)>-equivalent command C<ll>).
369 L</guestfs_ls> is a quick way to get a list of files in a directory
370 from programs, as a flat list of strings.
372 L</guestfs_readdir> is a programmatic way to get a list of files in a
373 directory, plus additional information about each one. It is more
374 equivalent to using the L<readdir(3)> call on a local filesystem.
376 L</guestfs_find> and L</guestfs_find0> can be used to recursively list
379 =head2 RUNNING COMMANDS
381 Although libguestfs is primarily an API for manipulating files
382 inside guest images, we also provide some limited facilities for
383 running commands inside guests.
385 There are many limitations to this:
391 The kernel version that the command runs under will be different
392 from what it expects.
396 If the command needs to communicate with daemons, then most likely
397 they won't be running.
401 The command will be running in limited memory.
405 The network may not be available unless you enable it
406 (see L</guestfs_set_network>).
410 Only supports Linux guests (not Windows, BSD, etc).
414 Architecture limitations (eg. won't work for a PPC guest on
419 For SELinux guests, you may need to enable SELinux and load policy
420 first. See L</SELINUX> in this manpage.
424 I<Security:> It is not safe to run commands from untrusted, possibly
425 malicious guests. These commands may attempt to exploit your program
426 by sending unexpected output. They could also try to exploit the
427 Linux kernel or qemu provided by the libguestfs appliance. They could
428 use the network provided by the libguestfs appliance to bypass
429 ordinary network partitions and firewalls. They could use the
430 elevated privileges or different SELinux context of your program
433 A secure alternative is to use libguestfs to install a "firstboot"
434 script (a script which runs when the guest next boots normally), and
435 to have this script run the commands you want in the normal context of
436 the running guest, network security and so on. For information about
437 other security issues, see L</SECURITY>.
441 The two main API calls to run commands are L</guestfs_command> and
442 L</guestfs_sh> (there are also variations).
444 The difference is that L</guestfs_sh> runs commands using the shell, so
445 any shell globs, redirections, etc will work.
447 =head2 CONFIGURATION FILES
449 To read and write configuration files in Linux guest filesystems, we
450 strongly recommend using Augeas. For example, Augeas understands how
451 to read and write, say, a Linux shadow password file or X.org
452 configuration file, and so avoids you having to write that code.
454 The main Augeas calls are bound through the C<guestfs_aug_*> APIs. We
455 don't document Augeas itself here because there is excellent
456 documentation on the L<http://augeas.net/> website.
458 If you don't want to use Augeas (you fool!) then try calling
459 L</guestfs_read_lines> to get the file as a list of lines which
460 you can iterate over.
464 We support SELinux guests. To ensure that labeling happens correctly
465 in SELinux guests, you need to enable SELinux and load the guest's
472 Before launching, do:
474 guestfs_set_selinux (g, 1);
478 After mounting the guest's filesystem(s), load the policy. This
479 is best done by running the L<load_policy(8)> command in the
482 guestfs_sh (g, "/usr/sbin/load_policy");
484 (Older versions of C<load_policy> require you to specify the
485 name of the policy file).
489 Optionally, set the security context for the API. The correct
490 security context to use can only be known by inspecting the
491 guest. As an example:
493 guestfs_setcon (g, "unconfined_u:unconfined_r:unconfined_t:s0");
497 This will work for running commands and editing existing files.
499 When new files are created, you may need to label them explicitly,
500 for example by running the external command
501 C<restorecon pathname>.
505 Certain calls are affected by the current file mode creation mask (the
506 "umask"). In particular ones which create files or directories, such
507 as L</guestfs_touch>, L</guestfs_mknod> or L</guestfs_mkdir>. This
508 affects either the default mode that the file is created with or
509 modifies the mode that you supply.
511 The default umask is C<022>, so files are created with modes such as
512 C<0644> and directories with C<0755>.
514 There are two ways to avoid being affected by umask. Either set umask
515 to 0 (call C<guestfs_umask (g, 0)> early after launching). Or call
516 L</guestfs_chmod> after creating each file or directory.
518 For more information about umask, see L<umask(2)>.
520 =head2 ENCRYPTED DISKS
522 Libguestfs allows you to access Linux guests which have been
523 encrypted using whole disk encryption that conforms to the
524 Linux Unified Key Setup (LUKS) standard. This includes
525 nearly all whole disk encryption systems used by modern
528 Use L</guestfs_vfs_type> to identify LUKS-encrypted block
529 devices (it returns the string C<crypto_LUKS>).
531 Then open these devices by calling L</guestfs_luks_open>.
532 Obviously you will require the passphrase!
534 Opening a LUKS device creates a new device mapper device
535 called C</dev/mapper/mapname> (where C<mapname> is the
536 string you supply to L</guestfs_luks_open>).
537 Reads and writes to this mapper device are decrypted from and
538 encrypted to the underlying block device respectively.
540 LVM volume groups on the device can be made visible by calling
541 L</guestfs_vgscan> followed by L</guestfs_vg_activate_all>.
542 The logical volume(s) can now be mounted in the usual way.
544 Use the reverse process to close a LUKS device. Unmount
545 any logical volumes on it, deactivate the volume groups
546 by caling C<guestfs_vg_activate (g, 0, ["/dev/VG"])>.
547 Then close the mapper device by calling
548 L</guestfs_luks_close> on the C</dev/mapper/mapname>
549 device (I<not> the underlying encrypted block device).
553 Libguestfs has APIs for inspecting an unknown disk image to find out
554 if it contains operating systems. (These APIs used to be in a
555 separate Perl-only library called L<Sys::Guestfs::Lib(3)> but since
556 version 1.5.3 the most frequently used part of this library has been
557 rewritten in C and moved into the core code).
559 Add all disks belonging to the unknown virtual machine and call
560 L</guestfs_launch> in the usual way.
562 Then call L</guestfs_inspect_os>. This function uses other libguestfs
563 calls and certain heuristics, and returns a list of operating systems
564 that were found. An empty list means none were found. A single
565 element is the root filesystem of the operating system. For dual- or
566 multi-boot guests, multiple roots can be returned, each one
567 corresponding to a separate operating system. (Multi-boot virtual
568 machines are extremely rare in the world of virtualization, but since
569 this scenario can happen, we have built libguestfs to deal with it.)
571 For each root, you can then call various C<guestfs_inspect_get_*>
572 functions to get additional details about that operating system. For
573 example, call L</guestfs_inspect_get_type> to return the string
574 C<windows> or C<linux> for Windows and Linux-based operating systems
577 Un*x-like and Linux-based operating systems usually consist of several
578 filesystems which are mounted at boot time (for example, a separate
579 boot partition mounted on C</boot>). The inspection rules are able to
580 detect how filesystems correspond to mount points. Call
581 C<guestfs_inspect_get_mountpoints> to get this mapping. It might
582 return a hash table like this example:
585 / => /dev/vg_guest/lv_root
586 /usr => /dev/vg_guest/lv_usr
588 The caller can then make calls to L</guestfs_mount_options> to
589 mount the filesystems as suggested.
591 Be careful to mount filesystems in the right order (eg. C</> before
592 C</usr>). Sorting the keys of the hash by length, shortest first,
595 Inspection currently only works for some common operating systems.
596 Contributors are welcome to send patches for other operating systems
597 that we currently cannot detect.
599 Encrypted disks must be opened before inspection. See
600 L</ENCRYPTED DISKS> for more details. The L</guestfs_inspect_os>
601 function just ignores any encrypted devices.
603 A note on the implementation: The call L</guestfs_inspect_os> performs
604 inspection and caches the results in the guest handle. Subsequent
605 calls to C<guestfs_inspect_get_*> return this cached information, but
606 I<do not> re-read the disks. If you change the content of the guest
607 disks, you can redo inspection by calling L</guestfs_inspect_os>
608 again. (L</guestfs_inspect_list_applications> works a little
609 differently from the other calls and does read the disks. See
610 documentation for that function for details).
612 =head2 SPECIAL CONSIDERATIONS FOR WINDOWS GUESTS
614 Libguestfs can mount NTFS partitions. It does this using the
615 L<http://www.ntfs-3g.org/> driver.
617 =head3 DRIVE LETTERS AND PATHS
619 DOS and Windows still use drive letters, and the filesystems are
620 always treated as case insensitive by Windows itself, and therefore
621 you might find a Windows configuration file referring to a path like
622 C<c:\windows\system32>. When the filesystem is mounted in libguestfs,
623 that directory might be referred to as C</WINDOWS/System32>.
625 Drive letter mappings are outside the scope of libguestfs. You have
626 to use libguestfs to read the appropriate Windows Registry and
627 configuration files, to determine yourself how drives are mapped (see
628 also L<hivex(3)> and L<virt-inspector(1)>).
630 Replacing backslash characters with forward slash characters is also
631 outside the scope of libguestfs, but something that you can easily do.
633 Where we can help is in resolving the case insensitivity of paths.
634 For this, call L</guestfs_case_sensitive_path>.
636 =head3 ACCESSING THE WINDOWS REGISTRY
638 Libguestfs also provides some help for decoding Windows Registry
639 "hive" files, through the library C<hivex> which is part of the
640 libguestfs project although ships as a separate tarball. You have to
641 locate and download the hive file(s) yourself, and then pass them to
642 C<hivex> functions. See also the programs L<hivexml(1)>,
643 L<hivexsh(1)>, L<hivexregedit(1)> and L<virt-win-reg(1)> for more help
646 =head3 SYMLINKS ON NTFS-3G FILESYSTEMS
648 Ntfs-3g tries to rewrite "Junction Points" and NTFS "symbolic links"
649 to provide something which looks like a Linux symlink. The way it
650 tries to do the rewriting is described here:
652 L<http://www.tuxera.com/community/ntfs-3g-advanced/junction-points-and-symbolic-links/>
654 The essential problem is that ntfs-3g simply does not have enough
655 information to do a correct job. NTFS links can contain drive letters
656 and references to external device GUIDs that ntfs-3g has no way of
657 resolving. It is almost certainly the case that libguestfs callers
658 should ignore what ntfs-3g does (ie. don't use L</guestfs_readlink> on
661 Instead if you encounter a symbolic link on an ntfs-3g filesystem, use
662 L</guestfs_lgetxattr> to read the C<system.ntfs_reparse_data> extended
663 attribute, and read the raw reparse data from that (you can find the
664 format documented in various places around the web).
666 =head3 EXTENDED ATTRIBUTES ON NTFS-3G FILESYSTEMS
668 There are other useful extended attributes that can be read from
669 ntfs-3g filesystems (using L</guestfs_getxattr>). See:
671 L<http://www.tuxera.com/community/ntfs-3g-advanced/extended-attributes/>
673 =head2 USING LIBGUESTFS WITH OTHER PROGRAMMING LANGUAGES
675 Although we don't want to discourage you from using the C API, we will
676 mention here that the same API is also available in other languages.
678 The API is broadly identical in all supported languages. This means
679 that the C call C<guestfs_mount(g,path)> is
680 C<$g-E<gt>mount($path)> in Perl, C<g.mount(path)> in Python,
681 and C<Guestfs.mount g path> in OCaml. In other words, a
682 straightforward, predictable isomorphism between each language.
684 Error messages are automatically transformed
685 into exceptions if the language supports it.
687 We don't try to "object orientify" parts of the API in OO languages,
688 although contributors are welcome to write higher level APIs above
689 what we provide in their favourite languages if they wish.
695 You can use the I<guestfs.h> header file from C++ programs. The C++
696 API is identical to the C API. C++ classes and exceptions are not
701 The C# bindings are highly experimental. Please read the warnings
702 at the top of C<csharp/Libguestfs.cs>.
706 This is the only language binding that is working but incomplete.
707 Only calls which return simple integers have been bound in Haskell,
708 and we are looking for help to complete this binding.
712 Full documentation is contained in the Javadoc which is distributed
717 For documentation see L<guestfs-ocaml(3)>.
721 For documentation see L<Sys::Guestfs(3)>.
725 For documentation see C<README-PHP> supplied with libguestfs
726 sources or in the php-libguestfs package for your distribution.
728 The PHP binding only works correctly on 64 bit machines.
732 For documentation see L<guestfs-python(3)>.
736 For documentation see L<guestfs-ruby(3)>.
738 =item B<shell scripts>
740 For documentation see L<guestfish(1)>.
744 =head2 LIBGUESTFS GOTCHAS
746 L<http://en.wikipedia.org/wiki/Gotcha_(programming)>: "A feature of a
747 system [...] that works in the way it is documented but is
748 counterintuitive and almost invites mistakes."
750 Since we developed libguestfs and the associated tools, there are
751 several things we would have designed differently, but are now stuck
752 with for backwards compatibility or other reasons. If there is ever a
753 libguestfs 2.0 release, you can expect these to change. Beware of
758 =item Autosync / forgetting to sync.
760 When modifying a filesystem from C or another language, you B<must>
761 unmount all filesystems and call L</guestfs_sync> explicitly before
762 you close the libguestfs handle. You can also call:
764 guestfs_set_autosync (g, 1);
766 to have the unmount/sync done automatically for you when the handle 'g'
767 is closed. (This feature is called "autosync", L</guestfs_set_autosync>
770 If you forget to do this, then it is entirely possible that your
771 changes won't be written out, or will be partially written, or (very
772 rarely) that you'll get disk corruption.
774 Note that in L<guestfish(3)> autosync is the default. So quick and
775 dirty guestfish scripts that forget to sync will work just fine, which
776 can make this very puzzling if you are trying to debug a problem.
778 Update: Autosync is enabled by default for all API users starting from
781 =item Mount option C<-o sync> should not be the default.
783 If you use L</guestfs_mount>, then C<-o sync,noatime> are added
784 implicitly. However C<-o sync> does not add any reliability benefit,
785 but does have a very large performance impact.
787 The work around is to use L</guestfs_mount_options> and set the mount
788 options that you actually want to use.
790 =item Read-only should be the default.
792 In L<guestfish(3)>, I<--ro> should be the default, and you should
793 have to specify I<--rw> if you want to make changes to the image.
795 This would reduce the potential to corrupt live VM images.
797 Note that many filesystems change the disk when you just mount and
798 unmount, even if you didn't perform any writes. You need to use
799 L</guestfs_add_drive_ro> to guarantee that the disk is not changed.
801 =item guestfish command line is hard to use.
803 C<guestfish disk.img> doesn't do what people expect (open C<disk.img>
804 for examination). It tries to run a guestfish command C<disk.img>
805 which doesn't exist, so it fails. In earlier versions of guestfish
806 the error message was also unintuitive, but we have corrected this
807 since. Like the Bourne shell, we should have used C<guestfish -c
808 command> to run commands.
810 =item guestfish megabyte modifiers don't work right on all commands
812 In recent guestfish you can use C<1M> to mean 1 megabyte (and
813 similarly for other modifiers). What guestfish actually does is to
814 multiply the number part by the modifier part and pass the result to
815 the C API. However this doesn't work for a few APIs which aren't
816 expecting bytes, but are already expecting some other unit
819 The most common is L</guestfs_lvcreate>. The guestfish command:
823 does not do what you might expect. Instead because
824 L</guestfs_lvcreate> is already expecting megabytes, this tries to
825 create a 100 I<terabyte> (100 megabytes * megabytes) logical volume.
826 The error message you get from this is also a little obscure.
828 This could be fixed in the generator by specially marking parameters
829 and return values which take bytes or other units.
831 =item Ambiguity between devices and paths
833 There is a subtle ambiguity in the API between a device name
834 (eg. C</dev/sdb2>) and a similar pathname. A file might just happen
835 to be called C<sdb2> in the directory C</dev> (consider some non-Unix
838 In the current API we usually resolve this ambiguity by having two
839 separate calls, for example L</guestfs_checksum> and
840 L</guestfs_checksum_device>. Some API calls are ambiguous and
841 (incorrectly) resolve the problem by detecting if the path supplied
842 begins with C</dev/>.
844 To avoid both the ambiguity and the need to duplicate some calls, we
845 could make paths/devices into structured names. One way to do this
846 would be to use a notation like grub (C<hd(0,0)>), although nobody
847 really likes this aspect of grub. Another way would be to use a
848 structured type, equivalent to this OCaml type:
850 type path = Path of string | Device of int | Partition of int * int
852 which would allow you to pass arguments like:
855 Device 1 (* /dev/sdb, or perhaps /dev/sda *)
856 Partition (1, 2) (* /dev/sdb2 (or is it /dev/sda2 or /dev/sdb3?) *)
857 Path "/dev/sdb2" (* not a device *)
859 As you can see there are still problems to resolve even with this
860 representation. Also consider how it might work in guestfish.
864 =head2 PROTOCOL LIMITS
866 Internally libguestfs uses a message-based protocol to pass API calls
867 and their responses to and from a small "appliance" (see L</INTERNALS>
868 for plenty more detail about this). The maximum message size used by
869 the protocol is slightly less than 4 MB. For some API calls you may
870 need to be aware of this limit. The API calls which may be affected
871 are individually documented, with a link back to this section of the
874 A simple call such as L</guestfs_cat> returns its result (the file
875 data) in a simple string. Because this string is at some point
876 internally encoded as a message, the maximum size that it can return
877 is slightly under 4 MB. If the requested file is larger than this
878 then you will get an error.
880 In order to transfer large files into and out of the guest filesystem,
881 you need to use particular calls that support this. The sections
882 L</UPLOADING> and L</DOWNLOADING> document how to do this.
884 You might also consider mounting the disk image using our FUSE
885 filesystem support (L<guestmount(1)>).
887 =head2 KEYS AND PASSPHRASES
889 Certain libguestfs calls take a parameter that contains sensitive key
890 material, passed in as a C string.
892 In the future we would hope to change the libguestfs implementation so
893 that keys are L<mlock(2)>-ed into physical RAM, and thus can never end
894 up in swap. However this is I<not> done at the moment, because of the
895 complexity of such an implementation.
897 Therefore you should be aware that any key parameter you pass to
898 libguestfs might end up being written out to the swap partition. If
899 this is a concern, scrub the swap partition or don't use libguestfs on
902 =head2 MULTIPLE HANDLES AND MULTIPLE THREADS
904 All high-level libguestfs actions are synchronous. If you want
905 to use libguestfs asynchronously then you must create a thread.
907 Only use the handle from a single thread. Either use the handle
908 exclusively from one thread, or provide your own mutex so that two
909 threads cannot issue calls on the same handle at the same time.
911 See the graphical program guestfs-browser for one possible
912 architecture for multithreaded programs using libvirt and libguestfs.
916 Libguestfs needs a kernel and initrd.img, which it finds by looking
917 along an internal path.
919 By default it looks for these in the directory C<$libdir/guestfs>
920 (eg. C</usr/local/lib/guestfs> or C</usr/lib64/guestfs>).
922 Use L</guestfs_set_path> or set the environment variable
923 L</LIBGUESTFS_PATH> to change the directories that libguestfs will
924 search in. The value is a colon-separated list of paths. The current
925 directory is I<not> searched unless the path contains an empty element
926 or C<.>. For example C<LIBGUESTFS_PATH=:/usr/lib/guestfs> would
927 search the current directory and then C</usr/lib/guestfs>.
931 If you want to compile your own qemu, run qemu from a non-standard
932 location, or pass extra arguments to qemu, then you can write a
933 shell-script wrapper around qemu.
935 There is one important rule to remember: you I<must C<exec qemu>> as
936 the last command in the shell script (so that qemu replaces the shell
937 and becomes the direct child of the libguestfs-using program). If you
938 don't do this, then the qemu process won't be cleaned up correctly.
940 Here is an example of a wrapper, where I have built my own copy of
944 qemudir=/home/rjones/d/qemu
945 exec $qemudir/x86_64-softmmu/qemu-system-x86_64 -L $qemudir/pc-bios "$@"
947 Save this script as C</tmp/qemu.wrapper> (or wherever), C<chmod +x>,
948 and then use it by setting the LIBGUESTFS_QEMU environment variable.
951 LIBGUESTFS_QEMU=/tmp/qemu.wrapper guestfish
953 Note that libguestfs also calls qemu with the -help and -version
954 options in order to determine features.
958 We guarantee the libguestfs ABI (binary interface), for public,
959 high-level actions as outlined in this section. Although we will
960 deprecate some actions, for example if they get replaced by newer
961 calls, we will keep the old actions forever. This allows you the
962 developer to program in confidence against the libguestfs API.
964 =head2 BLOCK DEVICE NAMING
966 In the kernel there is now quite a profusion of schemata for naming
967 block devices (in this context, by I<block device> I mean a physical
968 or virtual hard drive). The original Linux IDE driver used names
969 starting with C</dev/hd*>. SCSI devices have historically used a
970 different naming scheme, C</dev/sd*>. When the Linux kernel I<libata>
971 driver became a popular replacement for the old IDE driver
972 (particularly for SATA devices) those devices also used the
973 C</dev/sd*> scheme. Additionally we now have virtual machines with
974 paravirtualized drivers. This has created several different naming
975 systems, such as C</dev/vd*> for virtio disks and C</dev/xvd*> for Xen
978 As discussed above, libguestfs uses a qemu appliance running an
979 embedded Linux kernel to access block devices. We can run a variety
980 of appliances based on a variety of Linux kernels.
982 This causes a problem for libguestfs because many API calls use device
983 or partition names. Working scripts and the recipe (example) scripts
984 that we make available over the internet could fail if the naming
987 Therefore libguestfs defines C</dev/sd*> as the I<standard naming
988 scheme>. Internally C</dev/sd*> names are translated, if necessary,
989 to other names as required. For example, under RHEL 5 which uses the
990 C</dev/hd*> scheme, any device parameter C</dev/sda2> is translated to
991 C</dev/hda2> transparently.
993 Note that this I<only> applies to parameters. The
994 L</guestfs_list_devices>, L</guestfs_list_partitions> and similar calls
995 return the true names of the devices and partitions as known to the
998 =head3 ALGORITHM FOR BLOCK DEVICE NAME TRANSLATION
1000 Usually this translation is transparent. However in some (very rare)
1001 cases you may need to know the exact algorithm. Such cases include
1002 where you use L</guestfs_config> to add a mixture of virtio and IDE
1003 devices to the qemu-based appliance, so have a mixture of C</dev/sd*>
1004 and C</dev/vd*> devices.
1006 The algorithm is applied only to I<parameters> which are known to be
1007 either device or partition names. Return values from functions such
1008 as L</guestfs_list_devices> are never changed.
1014 Is the string a parameter which is a device or partition name?
1018 Does the string begin with C</dev/sd>?
1022 Does the named device exist? If so, we use that device.
1023 However if I<not> then we continue with this algorithm.
1027 Replace initial C</dev/sd> string with C</dev/hd>.
1029 For example, change C</dev/sda2> to C</dev/hda2>.
1031 If that named device exists, use it. If not, continue.
1035 Replace initial C</dev/sd> string with C</dev/vd>.
1037 If that named device exists, use it. If not, return an error.
1041 =head3 PORTABILITY CONCERNS WITH BLOCK DEVICE NAMING
1043 Although the standard naming scheme and automatic translation is
1044 useful for simple programs and guestfish scripts, for larger programs
1045 it is best not to rely on this mechanism.
1047 Where possible for maximum future portability programs using
1048 libguestfs should use these future-proof techniques:
1054 Use L</guestfs_list_devices> or L</guestfs_list_partitions> to list
1055 actual device names, and then use those names directly.
1057 Since those device names exist by definition, they will never be
1062 Use higher level ways to identify filesystems, such as LVM names,
1063 UUIDs and filesystem labels.
1069 This section discusses security implications of using libguestfs,
1070 particularly with untrusted or malicious guests or disk images.
1072 =head2 GENERAL SECURITY CONSIDERATIONS
1074 Be careful with any files or data that you download from a guest (by
1075 "download" we mean not just the L</guestfs_download> command but any
1076 command that reads files, filenames, directories or anything else from
1077 a disk image). An attacker could manipulate the data to fool your
1078 program into doing the wrong thing. Consider cases such as:
1084 the data (file etc) not being present
1088 being present but empty
1092 being much larger than normal
1096 containing arbitrary 8 bit data
1100 being in an unexpected character encoding
1104 containing homoglyphs.
1108 =head2 SECURITY OF MOUNTING FILESYSTEMS
1110 When you mount a filesystem under Linux, mistakes in the kernel
1111 filesystem (VFS) module can sometimes be escalated into exploits by
1112 deliberately creating a malicious, malformed filesystem. These
1113 exploits are very severe for two reasons. Firstly there are very many
1114 filesystem drivers in the kernel, and many of them are infrequently
1115 used and not much developer attention has been paid to the code.
1116 Linux userspace helps potential crackers by detecting the filesystem
1117 type and automatically choosing the right VFS driver, even if that
1118 filesystem type is obscure or unexpected for the administrator.
1119 Secondly, a kernel-level exploit is like a local root exploit (worse
1120 in some ways), giving immediate and total access to the system right
1121 down to the hardware level.
1123 That explains why you should never mount a filesystem from an
1124 untrusted guest on your host kernel. How about libguestfs? We run a
1125 Linux kernel inside a qemu virtual machine, usually running as a
1126 non-root user. The attacker would need to write a filesystem which
1127 first exploited the kernel, and then exploited either qemu
1128 virtualization (eg. a faulty qemu driver) or the libguestfs protocol,
1129 and finally to be as serious as the host kernel exploit it would need
1130 to escalate its privileges to root. This multi-step escalation,
1131 performed by a static piece of data, is thought to be extremely hard
1132 to do, although we never say 'never' about security issues.
1134 In any case callers can reduce the attack surface by forcing the
1135 filesystem type when mounting (use L</guestfs_mount_vfs>).
1137 =head2 PROTOCOL SECURITY
1139 The protocol is designed to be secure, being based on RFC 4506 (XDR)
1140 with a defined upper message size. However a program that uses
1141 libguestfs must also take care - for example you can write a program
1142 that downloads a binary from a disk image and executes it locally, and
1143 no amount of protocol security will save you from the consequences.
1145 =head2 INSPECTION SECURITY
1147 Parts of the inspection API (see L</INSPECTION>) return untrusted
1148 strings directly from the guest, and these could contain any 8 bit
1149 data. Callers should be careful to escape these before printing them
1150 to a structured file (for example, use HTML escaping if creating a web
1153 Guest configuration may be altered in unusual ways by the
1154 administrator of the virtual machine, and may not reflect reality
1155 (particularly for untrusted or actively malicious guests). For
1156 example we parse the hostname from configuration files like
1157 C</etc/sysconfig/network> that we find in the guest, but the guest
1158 administrator can easily manipulate these files to provide the wrong
1161 The inspection API parses guest configuration using two external
1162 libraries: Augeas (Linux configuration) and hivex (Windows Registry).
1163 Both are designed to be robust in the face of malicious data, although
1164 denial of service attacks are still possible, for example with
1165 oversized configuration files.
1167 =head2 RUNNING UNTRUSTED GUEST COMMANDS
1169 Be very cautious about running commands from the guest. By running a
1170 command in the guest, you are giving CPU time to a binary that you do
1171 not control, under the same user account as the library, albeit
1172 wrapped in qemu virtualization. More information and alternatives can
1173 be found in the section L</RUNNING COMMANDS>.
1175 =head2 CVE-2010-3851
1177 https://bugzilla.redhat.com/642934
1179 This security bug concerns the automatic disk format detection that
1180 qemu does on disk images.
1182 A raw disk image is just the raw bytes, there is no header. Other
1183 disk images like qcow2 contain a special header. Qemu deals with this
1184 by looking for one of the known headers, and if none is found then
1185 assuming the disk image must be raw.
1187 This allows a guest which has been given a raw disk image to write
1188 some other header. At next boot (or when the disk image is accessed
1189 by libguestfs) qemu would do autodetection and think the disk image
1190 format was, say, qcow2 based on the header written by the guest.
1192 This in itself would not be a problem, but qcow2 offers many features,
1193 one of which is to allow a disk image to refer to another image
1194 (called the "backing disk"). It does this by placing the path to the
1195 backing disk into the qcow2 header. This path is not validated and
1196 could point to any host file (eg. "/etc/passwd"). The backing disk is
1197 then exposed through "holes" in the qcow2 disk image, which of course
1198 is completely under the control of the attacker.
1200 In libguestfs this is rather hard to exploit except under two
1207 You have enabled the network or have opened the disk in write mode.
1211 You are also running untrusted code from the guest (see
1212 L</RUNNING COMMANDS>).
1216 The way to avoid this is to specify the expected disk format when
1217 adding disks (the optional C<format> option to
1218 L</guestfs_add_drive_opts>). You should always do this if the disk is
1219 raw format, and it's a good idea for other cases too.
1221 For disks added from libvirt using calls like L</guestfs_add_domain>,
1222 the format is fetched from libvirt and passed through.
1224 For libguestfs tools, use the I<--format> command line parameter as
1227 =head1 CONNECTION MANAGEMENT
1231 C<guestfs_h> is the opaque type representing a connection handle.
1232 Create a handle by calling L</guestfs_create>. Call L</guestfs_close>
1233 to free the handle and release all resources used.
1235 For information on using multiple handles and threads, see the section
1236 L</MULTIPLE HANDLES AND MULTIPLE THREADS> below.
1238 =head2 guestfs_create
1240 guestfs_h *guestfs_create (void);
1242 Create a connection handle.
1244 You have to call L</guestfs_add_drive_opts> (or one of the equivalent
1245 calls) on the handle at least once.
1247 This function returns a non-NULL pointer to a handle on success or
1250 After configuring the handle, you have to call L</guestfs_launch>.
1252 You may also want to configure error handling for the handle. See
1253 L</ERROR HANDLING> section below.
1255 =head2 guestfs_close
1257 void guestfs_close (guestfs_h *g);
1259 This closes the connection handle and frees up all resources used.
1261 =head1 ERROR HANDLING
1263 API functions can return errors. For example, almost all functions
1264 that return C<int> will return C<-1> to indicate an error.
1266 Additional information is available for errors: an error message
1267 string and optionally an error number (errno) if the thing that failed
1270 You can get at the additional information about the last error on the
1271 handle by calling L</guestfs_last_error>, L</guestfs_last_errno>,
1272 and/or by setting up an error handler with
1273 L</guestfs_set_error_handler>.
1275 When the handle is created, a default error handler is installed which
1276 prints the error message string to C<stderr>. For small short-running
1277 command line programs it is sufficient to do:
1279 if (guestfs_launch (g) == -1)
1280 exit (EXIT_FAILURE);
1282 since the default error handler will ensure that an error message has
1283 been printed to C<stderr> before the program exits.
1285 For other programs the caller will almost certainly want to install an
1286 alternate error handler or do error handling in-line like this:
1288 g = guestfs_create ();
1290 /* This disables the default behaviour of printing errors
1292 guestfs_set_error_handler (g, NULL, NULL);
1294 if (guestfs_launch (g) == -1) {
1295 /* Examine the error message and print it etc. */
1296 char *msg = guestfs_last_error (g);
1297 int errnum = guestfs_last_errno (g);
1298 fprintf (stderr, "%s\n", msg);
1302 Out of memory errors are handled differently. The default action is
1303 to call L<abort(3)>. If this is undesirable, then you can set a
1304 handler using L</guestfs_set_out_of_memory_handler>.
1306 L</guestfs_create> returns C<NULL> if the handle cannot be created,
1307 and because there is no handle if this happens there is no way to get
1308 additional error information. However L</guestfs_create> is supposed
1309 to be a lightweight operation which can only fail because of
1310 insufficient memory (it returns NULL in this case).
1312 =head2 guestfs_last_error
1314 const char *guestfs_last_error (guestfs_h *g);
1316 This returns the last error message that happened on C<g>. If
1317 there has not been an error since the handle was created, then this
1320 The lifetime of the returned string is until the next error occurs, or
1321 L</guestfs_close> is called.
1323 =head2 guestfs_last_errno
1325 int guestfs_last_errno (guestfs_h *g);
1327 This returns the last error number (errno) that happened on C<g>.
1329 If successful, an errno integer not equal to zero is returned.
1331 If no error, this returns 0. This call can return 0 in three
1338 There has not been any error on the handle.
1342 There has been an error but the errno was meaningless. This
1343 corresponds to the case where the error did not come from a
1344 failed system call, but for some other reason.
1348 There was an error from a failed system call, but for some
1349 reason the errno was not captured and returned. This usually
1350 indicates a bug in libguestfs.
1354 Libguestfs tries to convert the errno from inside the applicance into
1355 a corresponding errno for the caller (not entirely trivial: the
1356 appliance might be running a completely different operating system
1357 from the library and error numbers are not standardized across
1358 Un*xen). If this could not be done, then the error is translated to
1359 C<EINVAL>. In practice this should only happen in very rare
1362 =head2 guestfs_set_error_handler
1364 typedef void (*guestfs_error_handler_cb) (guestfs_h *g,
1367 void guestfs_set_error_handler (guestfs_h *g,
1368 guestfs_error_handler_cb cb,
1371 The callback C<cb> will be called if there is an error. The
1372 parameters passed to the callback are an opaque data pointer and the
1373 error message string.
1375 C<errno> is not passed to the callback. To get that the callback must
1376 call L</guestfs_last_errno>.
1378 Note that the message string C<msg> is freed as soon as the callback
1379 function returns, so if you want to stash it somewhere you must make
1382 The default handler prints messages on C<stderr>.
1384 If you set C<cb> to C<NULL> then I<no> handler is called.
1386 =head2 guestfs_get_error_handler
1388 guestfs_error_handler_cb guestfs_get_error_handler (guestfs_h *g,
1391 Returns the current error handler callback.
1393 =head2 guestfs_set_out_of_memory_handler
1395 typedef void (*guestfs_abort_cb) (void);
1396 int guestfs_set_out_of_memory_handler (guestfs_h *g,
1399 The callback C<cb> will be called if there is an out of memory
1400 situation. I<Note this callback must not return>.
1402 The default is to call L<abort(3)>.
1404 You cannot set C<cb> to C<NULL>. You can't ignore out of memory
1407 =head2 guestfs_get_out_of_memory_handler
1409 guestfs_abort_fn guestfs_get_out_of_memory_handler (guestfs_h *g);
1411 This returns the current out of memory handler.
1423 =head2 GROUPS OF FUNCTIONALITY IN THE APPLIANCE
1425 Using L</guestfs_available> you can test availability of
1426 the following groups of functions. This test queries the
1427 appliance to see if the appliance you are currently using
1428 supports the functionality.
1432 =head2 GUESTFISH supported COMMAND
1434 In L<guestfish(3)> there is a handy interactive command
1435 C<supported> which prints out the available groups and
1436 whether they are supported by this build of libguestfs.
1437 Note however that you have to do C<run> first.
1439 =head2 SINGLE CALLS AT COMPILE TIME
1441 Since version 1.5.8, C<E<lt>guestfs.hE<gt>> defines symbols
1442 for each C API function, such as:
1444 #define LIBGUESTFS_HAVE_DD 1
1446 if L</guestfs_dd> is available.
1448 Before version 1.5.8, if you needed to test whether a single
1449 libguestfs function is available at compile time, we recommended using
1450 build tools such as autoconf or cmake. For example in autotools you
1453 AC_CHECK_LIB([guestfs],[guestfs_create])
1454 AC_CHECK_FUNCS([guestfs_dd])
1456 which would result in C<HAVE_GUESTFS_DD> being either defined
1457 or not defined in your program.
1459 =head2 SINGLE CALLS AT RUN TIME
1461 Testing at compile time doesn't guarantee that a function really
1462 exists in the library. The reason is that you might be dynamically
1463 linked against a previous I<libguestfs.so> (dynamic library)
1464 which doesn't have the call. This situation unfortunately results
1465 in a segmentation fault, which is a shortcoming of the C dynamic
1466 linking system itself.
1468 You can use L<dlopen(3)> to test if a function is available
1469 at run time, as in this example program (note that you still
1470 need the compile time check as well):
1476 #include <guestfs.h>
1480 #ifdef LIBGUESTFS_HAVE_DD
1484 /* Test if the function guestfs_dd is really available. */
1485 dl = dlopen (NULL, RTLD_LAZY);
1487 fprintf (stderr, "dlopen: %s\n", dlerror ());
1488 exit (EXIT_FAILURE);
1490 has_function = dlsym (dl, "guestfs_dd") != NULL;
1494 printf ("this libguestfs.so does NOT have guestfs_dd function\n");
1496 printf ("this libguestfs.so has guestfs_dd function\n");
1497 /* Now it's safe to call
1498 guestfs_dd (g, "foo", "bar");
1502 printf ("guestfs_dd function was not found at compile time\n");
1506 You may think the above is an awful lot of hassle, and it is.
1507 There are other ways outside of the C linking system to ensure
1508 that this kind of incompatibility never arises, such as using
1511 Requires: libguestfs >= 1.0.80
1513 =head1 CALLS WITH OPTIONAL ARGUMENTS
1515 A recent feature of the API is the introduction of calls which take
1516 optional arguments. In C these are declared 3 ways. The main way is
1517 as a call which takes variable arguments (ie. C<...>), as in this
1520 int guestfs_add_drive_opts (guestfs_h *g, const char *filename, ...);
1522 Call this with a list of optional arguments, terminated by C<-1>.
1523 So to call with no optional arguments specified:
1525 guestfs_add_drive_opts (g, filename, -1);
1527 With a single optional argument:
1529 guestfs_add_drive_opts (g, filename,
1530 GUESTFS_ADD_DRIVE_OPTS_FORMAT, "qcow2",
1535 guestfs_add_drive_opts (g, filename,
1536 GUESTFS_ADD_DRIVE_OPTS_FORMAT, "qcow2",
1537 GUESTFS_ADD_DRIVE_OPTS_READONLY, 1,
1540 and so forth. Don't forget the terminating C<-1> otherwise
1541 Bad Things will happen!
1543 =head2 USING va_list FOR OPTIONAL ARGUMENTS
1545 The second variant has the same name with the suffix C<_va>, which
1546 works the same way but takes a C<va_list>. See the C manual for
1547 details. For the example function, this is declared:
1549 int guestfs_add_drive_opts_va (guestfs_h *g, const char *filename,
1552 =head2 CONSTRUCTING OPTIONAL ARGUMENTS
1554 The third variant is useful where you need to construct these
1555 calls. You pass in a structure where you fill in the optional
1556 fields. The structure has a bitmask as the first element which
1557 you must set to indicate which fields you have filled in. For
1558 our example function the structure and call are declared:
1560 struct guestfs_add_drive_opts_argv {
1566 int guestfs_add_drive_opts_argv (guestfs_h *g, const char *filename,
1567 const struct guestfs_add_drive_opts_argv *optargs);
1569 You could call it like this:
1571 struct guestfs_add_drive_opts_argv optargs = {
1572 .bitmask = GUESTFS_ADD_DRIVE_OPTS_READONLY_BITMASK |
1573 GUESTFS_ADD_DRIVE_OPTS_FORMAT_BITMASK,
1578 guestfs_add_drive_opts_argv (g, filename, &optargs);
1586 The C<_BITMASK> suffix on each option name when specifying the
1591 You do not need to fill in all fields of the structure.
1595 There must be a one-to-one correspondence between fields of the
1596 structure that are filled in, and bits set in the bitmask.
1600 =head2 OPTIONAL ARGUMENTS IN OTHER LANGUAGES
1602 In other languages, optional arguments are expressed in the
1603 way that is natural for that language. We refer you to the
1604 language-specific documentation for more details on that.
1606 For guestfish, see L<guestfish(1)/OPTIONAL ARGUMENTS>.
1608 =head2 SETTING CALLBACKS TO HANDLE EVENTS
1610 The child process generates events in some situations. Current events
1611 include: receiving a log message, the child process exits.
1613 Use the C<guestfs_set_*_callback> functions to set a callback for
1614 different types of events.
1616 Only I<one callback of each type> can be registered for each handle.
1617 Calling C<guestfs_set_*_callback> again overwrites the previous
1618 callback of that type. Cancel all callbacks of this type by calling
1619 this function with C<cb> set to C<NULL>.
1621 =head2 guestfs_set_log_message_callback
1623 typedef void (*guestfs_log_message_cb) (guestfs_h *g, void *opaque,
1624 char *buf, int len);
1625 void guestfs_set_log_message_callback (guestfs_h *g,
1626 guestfs_log_message_cb cb,
1629 The callback function C<cb> will be called whenever qemu or the guest
1630 writes anything to the console.
1632 Use this function to capture kernel messages and similar.
1634 Normally there is no log message handler, and log messages are just
1637 =head2 guestfs_set_subprocess_quit_callback
1639 typedef void (*guestfs_subprocess_quit_cb) (guestfs_h *g, void *opaque);
1640 void guestfs_set_subprocess_quit_callback (guestfs_h *g,
1641 guestfs_subprocess_quit_cb cb,
1644 The callback function C<cb> will be called when the child process
1645 quits, either asynchronously or if killed by
1646 L</guestfs_kill_subprocess>. (This corresponds to a transition from
1647 any state to the CONFIG state).
1649 =head2 guestfs_set_launch_done_callback
1651 typedef void (*guestfs_launch_done_cb) (guestfs_h *g, void *opaque);
1652 void guestfs_set_launch_done_callback (guestfs_h *g,
1653 guestfs_launch_done_cb cb,
1656 The callback function C<cb> will be called when the child process
1657 becomes ready first time after it has been launched. (This
1658 corresponds to a transition from LAUNCHING to the READY state).
1660 =head2 guestfs_set_close_callback
1662 typedef void (*guestfs_close_cb) (guestfs_h *g, void *opaque);
1663 void guestfs_set_close_callback (guestfs_h *g,
1664 guestfs_close_cb cb,
1667 The callback function C<cb> will be called while the handle
1668 is being closed (synchronously from L</guestfs_close>).
1670 Note that libguestfs installs an L<atexit(3)> handler to try to
1671 clean up handles that are open when the program exits. This
1672 means that this callback might be called indirectly from
1673 L<exit(3)>, which can cause unexpected problems in higher-level
1674 languages (eg. if your HLL interpreter has already been cleaned
1675 up by the time this is called, and if your callback then jumps
1676 into some HLL function).
1678 =head2 guestfs_set_progress_callback
1680 typedef void (*guestfs_progress_cb) (guestfs_h *g, void *opaque,
1681 int proc_nr, int serial,
1682 uint64_t position, uint64_t total);
1683 void guestfs_set_progress_callback (guestfs_h *g,
1684 guestfs_progress_cb cb,
1687 Some long-running operations can generate progress messages. If
1688 this callback is registered, then it will be called each time a
1689 progress message is generated (usually two seconds after the
1690 operation started, and three times per second thereafter until
1691 it completes, although the frequency may change in future versions).
1693 The callback receives two numbers: C<position> and C<total>.
1694 The units of C<total> are not defined, although for some
1695 operations C<total> may relate in some way to the amount of
1696 data to be transferred (eg. in bytes or megabytes), and
1697 C<position> may be the portion which has been transferred.
1699 The only defined and stable parts of the API are:
1705 The callback can display to the user some type of progress bar or
1706 indicator which shows the ratio of C<position>:C<total>.
1710 0 E<lt>= C<position> E<lt>= C<total>
1714 If any progress notification is sent during a call, then a final
1715 progress notification is always sent when C<position> = C<total>.
1717 This is to simplify caller code, so callers can easily set the
1718 progress indicator to "100%" at the end of the operation, without
1719 requiring special code to detect this case.
1723 The callback also receives the procedure number and serial number of
1724 the call. These are only useful for debugging protocol issues, and
1725 the callback can normally ignore them. The callback may want to
1726 print these numbers in error messages or debugging messages.
1728 =head1 PRIVATE DATA AREA
1730 You can attach named pieces of private data to the libguestfs handle,
1731 and fetch them by name for the lifetime of the handle. This is called
1732 the private data area and is only available from the C API.
1734 To attach a named piece of data, use the following call:
1736 void guestfs_set_private (guestfs_h *g, const char *key, void *data);
1738 C<key> is the name to associate with this data, and C<data> is an
1739 arbitrary pointer (which can be C<NULL>). Any previous item with the
1740 same name is overwritten.
1742 You can use any C<key> you want, but names beginning with an
1743 underscore character are reserved for internal libguestfs purposes
1744 (for implementing language bindings). It is recommended to prefix the
1745 name with some unique string to avoid collisions with other users.
1747 To retrieve the pointer, use:
1749 void *guestfs_get_private (guestfs_h *g, const char *key);
1751 This function returns C<NULL> if either no data is found associated
1752 with C<key>, or if the user previously set the C<key>'s C<data>
1755 Libguestfs does not try to look at or interpret the C<data> pointer in
1756 any way. As far as libguestfs is concerned, it need not be a valid
1757 pointer at all. In particular, libguestfs does I<not> try to free the
1758 data when the handle is closed. If the data must be freed, then the
1759 caller must either free it before calling L</guestfs_close> or must
1760 set up a close callback to do it (see L</guestfs_set_close_callback>,
1761 and note that only one callback can be registered for a handle).
1763 The private data area is implemented using a hash table, and should be
1764 reasonably efficient for moderate numbers of keys.
1768 <!-- old anchor for the next section -->
1769 <a name="state_machine_and_low_level_event_api"/>
1775 Internally, libguestfs is implemented by running an appliance (a
1776 special type of small virtual machine) using L<qemu(1)>. Qemu runs as
1777 a child process of the main program.
1783 | | child process / appliance
1784 | | __________________________
1786 +-------------------+ RPC | +-----------------+ |
1787 | libguestfs <--------------------> guestfsd | |
1788 | | | +-----------------+ |
1789 \___________________/ | | Linux kernel | |
1790 | +--^--------------+ |
1791 \_________|________________/
1799 The library, linked to the main program, creates the child process and
1800 hence the appliance in the L</guestfs_launch> function.
1802 Inside the appliance is a Linux kernel and a complete stack of
1803 userspace tools (such as LVM and ext2 programs) and a small
1804 controlling daemon called L</guestfsd>. The library talks to
1805 L</guestfsd> using remote procedure calls (RPC). There is a mostly
1806 one-to-one correspondence between libguestfs API calls and RPC calls
1807 to the daemon. Lastly the disk image(s) are attached to the qemu
1808 process which translates device access by the appliance's Linux kernel
1809 into accesses to the image.
1811 A common misunderstanding is that the appliance "is" the virtual
1812 machine. Although the disk image you are attached to might also be
1813 used by some virtual machine, libguestfs doesn't know or care about
1814 this. (But you will care if both libguestfs's qemu process and your
1815 virtual machine are trying to update the disk image at the same time,
1816 since these usually results in massive disk corruption).
1818 =head1 STATE MACHINE
1820 libguestfs uses a state machine to model the child process:
1831 / | \ \ guestfs_launch
1842 \______/ <------ \________/
1844 The normal transitions are (1) CONFIG (when the handle is created, but
1845 there is no child process), (2) LAUNCHING (when the child process is
1846 booting up), (3) alternating between READY and BUSY as commands are
1847 issued to, and carried out by, the child process.
1849 The guest may be killed by L</guestfs_kill_subprocess>, or may die
1850 asynchronously at any time (eg. due to some internal error), and that
1851 causes the state to transition back to CONFIG.
1853 Configuration commands for qemu such as L</guestfs_add_drive> can only
1854 be issued when in the CONFIG state.
1856 The API offers one call that goes from CONFIG through LAUNCHING to
1857 READY. L</guestfs_launch> blocks until the child process is READY to
1858 accept commands (or until some failure or timeout).
1859 L</guestfs_launch> internally moves the state from CONFIG to LAUNCHING
1860 while it is running.
1862 API actions such as L</guestfs_mount> can only be issued when in the
1863 READY state. These API calls block waiting for the command to be
1864 carried out (ie. the state to transition to BUSY and then back to
1865 READY). There are no non-blocking versions, and no way to issue more
1866 than one command per handle at the same time.
1868 Finally, the child process sends asynchronous messages back to the
1869 main program, such as kernel log messages. You can register a
1870 callback to receive these messages.
1874 =head2 COMMUNICATION PROTOCOL
1876 Don't rely on using this protocol directly. This section documents
1877 how it currently works, but it may change at any time.
1879 The protocol used to talk between the library and the daemon running
1880 inside the qemu virtual machine is a simple RPC mechanism built on top
1881 of XDR (RFC 1014, RFC 1832, RFC 4506).
1883 The detailed format of structures is in C<src/guestfs_protocol.x>
1884 (note: this file is automatically generated).
1886 There are two broad cases, ordinary functions that don't have any
1887 C<FileIn> and C<FileOut> parameters, which are handled with very
1888 simple request/reply messages. Then there are functions that have any
1889 C<FileIn> or C<FileOut> parameters, which use the same request and
1890 reply messages, but they may also be followed by files sent using a
1893 =head3 ORDINARY FUNCTIONS (NO FILEIN/FILEOUT PARAMS)
1895 For ordinary functions, the request message is:
1897 total length (header + arguments,
1898 but not including the length word itself)
1899 struct guestfs_message_header (encoded as XDR)
1900 struct guestfs_<foo>_args (encoded as XDR)
1902 The total length field allows the daemon to allocate a fixed size
1903 buffer into which it slurps the rest of the message. As a result, the
1904 total length is limited to C<GUESTFS_MESSAGE_MAX> bytes (currently
1905 4MB), which means the effective size of any request is limited to
1906 somewhere under this size.
1908 Note also that many functions don't take any arguments, in which case
1909 the C<guestfs_I<foo>_args> is completely omitted.
1911 The header contains the procedure number (C<guestfs_proc>) which is
1912 how the receiver knows what type of args structure to expect, or none
1915 For functions that take optional arguments, the optional arguments are
1916 encoded in the C<guestfs_I<foo>_args> structure in the same way as
1917 ordinary arguments. A bitmask in the header indicates which optional
1918 arguments are meaningful. The bitmask is also checked to see if it
1919 contains bits set which the daemon does not know about (eg. if more
1920 optional arguments were added in a later version of the library), and
1921 this causes the call to be rejected.
1923 The reply message for ordinary functions is:
1925 total length (header + ret,
1926 but not including the length word itself)
1927 struct guestfs_message_header (encoded as XDR)
1928 struct guestfs_<foo>_ret (encoded as XDR)
1930 As above the C<guestfs_I<foo>_ret> structure may be completely omitted
1931 for functions that return no formal return values.
1933 As above the total length of the reply is limited to
1934 C<GUESTFS_MESSAGE_MAX>.
1936 In the case of an error, a flag is set in the header, and the reply
1937 message is slightly changed:
1939 total length (header + error,
1940 but not including the length word itself)
1941 struct guestfs_message_header (encoded as XDR)
1942 struct guestfs_message_error (encoded as XDR)
1944 The C<guestfs_message_error> structure contains the error message as a
1947 =head3 FUNCTIONS THAT HAVE FILEIN PARAMETERS
1949 A C<FileIn> parameter indicates that we transfer a file I<into> the
1950 guest. The normal request message is sent (see above). However this
1951 is followed by a sequence of file chunks.
1953 total length (header + arguments,
1954 but not including the length word itself,
1955 and not including the chunks)
1956 struct guestfs_message_header (encoded as XDR)
1957 struct guestfs_<foo>_args (encoded as XDR)
1958 sequence of chunks for FileIn param #0
1959 sequence of chunks for FileIn param #1 etc.
1961 The "sequence of chunks" is:
1963 length of chunk (not including length word itself)
1964 struct guestfs_chunk (encoded as XDR)
1966 struct guestfs_chunk (encoded as XDR)
1969 struct guestfs_chunk (with data.data_len == 0)
1971 The final chunk has the C<data_len> field set to zero. Additionally a
1972 flag is set in the final chunk to indicate either successful
1973 completion or early cancellation.
1975 At time of writing there are no functions that have more than one
1976 FileIn parameter. However this is (theoretically) supported, by
1977 sending the sequence of chunks for each FileIn parameter one after
1978 another (from left to right).
1980 Both the library (sender) I<and> the daemon (receiver) may cancel the
1981 transfer. The library does this by sending a chunk with a special
1982 flag set to indicate cancellation. When the daemon sees this, it
1983 cancels the whole RPC, does I<not> send any reply, and goes back to
1984 reading the next request.
1986 The daemon may also cancel. It does this by writing a special word
1987 C<GUESTFS_CANCEL_FLAG> to the socket. The library listens for this
1988 during the transfer, and if it gets it, it will cancel the transfer
1989 (it sends a cancel chunk). The special word is chosen so that even if
1990 cancellation happens right at the end of the transfer (after the
1991 library has finished writing and has started listening for the reply),
1992 the "spurious" cancel flag will not be confused with the reply
1995 This protocol allows the transfer of arbitrary sized files (no 32 bit
1996 limit), and also files where the size is not known in advance
1997 (eg. from pipes or sockets). However the chunks are rather small
1998 (C<GUESTFS_MAX_CHUNK_SIZE>), so that neither the library nor the
1999 daemon need to keep much in memory.
2001 =head3 FUNCTIONS THAT HAVE FILEOUT PARAMETERS
2003 The protocol for FileOut parameters is exactly the same as for FileIn
2004 parameters, but with the roles of daemon and library reversed.
2006 total length (header + ret,
2007 but not including the length word itself,
2008 and not including the chunks)
2009 struct guestfs_message_header (encoded as XDR)
2010 struct guestfs_<foo>_ret (encoded as XDR)
2011 sequence of chunks for FileOut param #0
2012 sequence of chunks for FileOut param #1 etc.
2014 =head3 INITIAL MESSAGE
2016 When the daemon launches it sends an initial word
2017 (C<GUESTFS_LAUNCH_FLAG>) which indicates that the guest and daemon is
2018 alive. This is what L</guestfs_launch> waits for.
2020 =head3 PROGRESS NOTIFICATION MESSAGES
2022 The daemon may send progress notification messages at any time. These
2023 are distinguished by the normal length word being replaced by
2024 C<GUESTFS_PROGRESS_FLAG>, followed by a fixed size progress message.
2026 The library turns them into progress callbacks (see
2027 C<guestfs_set_progress_callback>) if there is a callback registered,
2028 or discards them if not.
2030 The daemon self-limits the frequency of progress messages it sends
2031 (see C<daemon/proto.c:notify_progress>). Not all calls generate
2034 =head1 LIBGUESTFS VERSION NUMBERS
2036 Since April 2010, libguestfs has started to make separate development
2037 and stable releases, along with corresponding branches in our git
2038 repository. These separate releases can be identified by version
2041 even numbers for stable: 1.2.x, 1.4.x, ...
2042 .-------- odd numbers for development: 1.3.x, 1.5.x, ...
2048 | `-------- sub-version
2050 `------ always '1' because we don't change the ABI
2052 Thus "1.3.5" is the 5th update to the development branch "1.3".
2054 As time passes we cherry pick fixes from the development branch and
2055 backport those into the stable branch, the effect being that the
2056 stable branch should get more stable and less buggy over time. So the
2057 stable releases are ideal for people who don't need new features but
2058 would just like the software to work.
2060 Our criteria for backporting changes are:
2066 Documentation changes which don't affect any code are
2067 backported unless the documentation refers to a future feature
2068 which is not in stable.
2072 Bug fixes which are not controversial, fix obvious problems, and
2073 have been well tested are backported.
2077 Simple rearrangements of code which shouldn't affect how it works get
2078 backported. This is so that the code in the two branches doesn't get
2079 too far out of step, allowing us to backport future fixes more easily.
2083 We I<don't> backport new features, new APIs, new tools etc, except in
2084 one exceptional case: the new feature is required in order to
2085 implement an important bug fix.
2089 A new stable branch starts when we think the new features in
2090 development are substantial and compelling enough over the current
2091 stable branch to warrant it. When that happens we create new stable
2092 and development versions 1.N.0 and 1.(N+1).0 [N is even]. The new
2093 dot-oh release won't necessarily be so stable at this point, but by
2094 backporting fixes from development, that branch will stabilize over
2097 =head1 EXTENDING LIBGUESTFS
2099 =head2 ADDING A NEW API ACTION
2101 Large amounts of boilerplate code in libguestfs (RPC, bindings,
2102 documentation) are generated, and this makes it easy to extend the
2105 To add a new API action there are two changes:
2111 You need to add a description of the call (name, parameters, return
2112 type, tests, documentation) to C<generator/generator_actions.ml>.
2114 There are two sorts of API action, depending on whether the call goes
2115 through to the daemon in the appliance, or is serviced entirely by the
2116 library (see L</ARCHITECTURE> above). L</guestfs_sync> is an example
2117 of the former, since the sync is done in the appliance.
2118 L</guestfs_set_trace> is an example of the latter, since a trace flag
2119 is maintained in the handle and all tracing is done on the library
2122 Most new actions are of the first type, and get added to the
2123 C<daemon_functions> list. Each function has a unique procedure number
2124 used in the RPC protocol which is assigned to that action when we
2125 publish libguestfs and cannot be reused. Take the latest procedure
2126 number and increment it.
2128 For library-only actions of the second type, add to the
2129 C<non_daemon_functions> list. Since these functions are serviced by
2130 the library and do not travel over the RPC mechanism to the daemon,
2131 these functions do not need a procedure number, and so the procedure
2132 number is set to C<-1>.
2136 Implement the action (in C):
2138 For daemon actions, implement the function C<do_E<lt>nameE<gt>> in the
2139 C<daemon/> directory.
2141 For library actions, implement the function C<guestfs__E<lt>nameE<gt>>
2142 (note: double underscore) in the C<src/> directory.
2144 In either case, use another function as an example of what to do.
2148 After making these changes, use C<make> to compile.
2150 Note that you don't need to implement the RPC, language bindings,
2151 manual pages or anything else. It's all automatically generated from
2152 the OCaml description.
2154 =head2 ADDING TESTS FOR AN API ACTION
2156 You can supply zero or as many tests as you want per API call. The
2157 tests can either be added as part of the API description
2158 (C<generator/generator_actions.ml>), or in some rarer cases you may
2159 want to drop a script into C<regressions/>. Note that adding a script
2160 to C<regressions/> is slower, so if possible use the first method.
2162 The following describes the test environment used when you add an API
2163 test in C<generator_actions.ml>.
2165 The test environment has 4 block devices:
2169 =item C</dev/sda> 500MB
2171 General block device for testing.
2173 =item C</dev/sdb> 50MB
2175 C</dev/sdb1> is an ext2 filesystem used for testing
2176 filesystem write operations.
2178 =item C</dev/sdc> 10MB
2180 Used in a few tests where two block devices are needed.
2184 ISO with fixed content (see C<images/test.iso>).
2188 To be able to run the tests in a reasonable amount of time, the
2189 libguestfs appliance and block devices are reused between tests. So
2190 don't try testing L</guestfs_kill_subprocess> :-x
2192 Each test starts with an initial scenario, selected using one of the
2193 C<Init*> expressions, described in C<generator/generator_types.ml>.
2194 These initialize the disks mentioned above in a particular way as
2195 documented in C<generator_types.ml>. You should not assume anything
2196 about the previous contents of other disks that are not initialized.
2198 You can add a prerequisite clause to any individual test. This is a
2199 run-time check, which, if it fails, causes the test to be skipped.
2200 Useful if testing a command which might not work on all variations of
2201 libguestfs builds. A test that has prerequisite of C<Always> means to
2202 run unconditionally.
2204 In addition, packagers can skip individual tests by setting
2205 environment variables before running C<make check>.
2207 SKIP_TEST_<CMD>_<NUM>=1
2209 eg: C<SKIP_TEST_COMMAND_3=1> skips test #3 of L</guestfs_command>.
2215 eg: C<SKIP_TEST_ZEROFREE=1> skips all L</guestfs_zerofree> tests.
2217 Packagers can run only certain tests by setting for example:
2219 TEST_ONLY="vfs_type zerofree"
2221 See C<capitests/tests.c> for more details of how these environment
2224 =head2 DEBUGGING NEW API ACTIONS
2226 Test new actions work before submitting them.
2228 You can use guestfish to try out new commands.
2230 Debugging the daemon is a problem because it runs inside a minimal
2231 environment. However you can fprintf messages in the daemon to
2232 stderr, and they will show up if you use C<guestfish -v>.
2234 =head2 FORMATTING CODE AND OTHER CONVENTIONS
2236 Our C source code generally adheres to some basic code-formatting
2237 conventions. The existing code base is not totally consistent on this
2238 front, but we do prefer that contributed code be formatted similarly.
2239 In short, use spaces-not-TABs for indentation, use 2 spaces for each
2240 indentation level, and other than that, follow the K&R style.
2242 If you use Emacs, add the following to one of one of your start-up files
2243 (e.g., ~/.emacs), to help ensure that you get indentation right:
2245 ;;; In libguestfs, indent with spaces everywhere (not TABs).
2246 ;;; Exceptions: Makefile and ChangeLog modes.
2247 (add-hook 'find-file-hook
2248 '(lambda () (if (and buffer-file-name
2249 (string-match "/libguestfs\\>"
2251 (not (string-equal mode-name "Change Log"))
2252 (not (string-equal mode-name "Makefile")))
2253 (setq indent-tabs-mode nil))))
2255 ;;; When editing C sources in libguestfs, use this style.
2256 (defun libguestfs-c-mode ()
2257 "C mode with adjusted defaults for use with libguestfs."
2260 (setq c-indent-level 2)
2261 (setq c-basic-offset 2))
2262 (add-hook 'c-mode-hook
2263 '(lambda () (if (string-match "/libguestfs\\>"
2265 (libguestfs-c-mode))))
2267 Enable warnings when compiling (and fix any problems this
2270 ./configure --enable-gcc-warnings
2274 make syntax-check # checks the syntax of the C code
2275 make check # runs the test suite
2277 =head2 DAEMON CUSTOM PRINTF FORMATTERS
2279 In the daemon code we have created custom printf formatters C<%Q> and
2280 C<%R>, which are used to do shell quoting.
2286 Simple shell quoted string. Any spaces or other shell characters are
2291 Same as C<%Q> except the string is treated as a path which is prefixed
2298 asprintf (&cmd, "cat %R", path);
2300 would produce C<cat /sysroot/some\ path\ with\ spaces>
2302 I<Note:> Do I<not> use these when you are passing parameters to the
2303 C<command{,r,v,rv}()> functions. These parameters do NOT need to be
2304 quoted because they are not passed via the shell (instead, straight to
2305 exec). You probably want to use the C<sysroot_path()> function
2308 =head2 SUBMITTING YOUR NEW API ACTIONS
2310 Submit patches to the mailing list:
2311 L<http://www.redhat.com/mailman/listinfo/libguestfs>
2312 and CC to L<rjones@redhat.com>.
2314 =head2 INTERNATIONALIZATION (I18N) SUPPORT
2316 We support i18n (gettext anyhow) in the library.
2318 However many messages come from the daemon, and we don't translate
2319 those at the moment. One reason is that the appliance generally has
2320 all locale files removed from it, because they take up a lot of space.
2321 So we'd have to readd some of those, as well as copying our PO files
2324 Debugging messages are never translated, since they are intended for
2327 =head2 SOURCE CODE SUBDIRECTORIES
2333 The libguestfs appliance, build scripts and so on.
2337 Automated tests of the C API.
2341 The L<virt-cat(1)>, L<virt-filesystems(1)> and L<virt-ls(1)> commands
2346 Outside contributions, experimental parts.
2350 The daemon that runs inside the libguestfs appliance and carries out
2355 L<virt-df(1)> command and documentation.
2363 L<guestfish(1)>, the command-line shell.
2367 L<guestmount(1)>, FUSE (userspace filesystem) built on top of libguestfs.
2371 The crucially important generator, used to automatically generate
2372 large amounts of boilerplate C code for things like RPC and bindings.
2376 Files used by the test suite.
2378 Some "phony" guest images which we test against.
2382 L<virt-inspector(1)>, the virtual machine image inspector.
2386 M4 macros used by autoconf.
2390 Translations of simple gettext strings.
2394 The build infrastructure and PO files for translations of manpages and
2395 POD files. Eventually this will be combined with the C<po> directory,
2396 but that is rather complicated.
2398 =item C<regressions>
2404 L<virt-rescue(1)> command and documentation.
2408 Source code to the C library.
2412 Command line tools written in Perl (L<virt-resize(1)> and many others).
2416 Test tool for end users to test if their qemu/kernel combination
2417 will work with libguestfs.
2439 =head1 ENVIRONMENT VARIABLES
2443 =item LIBGUESTFS_APPEND
2445 Pass additional options to the guest kernel.
2447 =item LIBGUESTFS_DEBUG
2449 Set C<LIBGUESTFS_DEBUG=1> to enable verbose messages. This
2450 has the same effect as calling C<guestfs_set_verbose (g, 1)>.
2452 =item LIBGUESTFS_MEMSIZE
2454 Set the memory allocated to the qemu process, in megabytes. For
2457 LIBGUESTFS_MEMSIZE=700
2459 =item LIBGUESTFS_PATH
2461 Set the path that libguestfs uses to search for kernel and initrd.img.
2462 See the discussion of paths in section PATH above.
2464 =item LIBGUESTFS_QEMU
2466 Set the default qemu binary that libguestfs uses. If not set, then
2467 the qemu which was found at compile time by the configure script is
2470 See also L</QEMU WRAPPERS> above.
2472 =item LIBGUESTFS_TRACE
2474 Set C<LIBGUESTFS_TRACE=1> to enable command traces. This
2475 has the same effect as calling C<guestfs_set_trace (g, 1)>.
2479 Location of temporary directory, defaults to C</tmp>.
2481 If libguestfs was compiled to use the supermin appliance then the
2482 real appliance is cached in this directory, shared between all
2483 handles belonging to the same EUID. You can use C<$TMPDIR> to
2484 configure another directory to use in case C</tmp> is not large
2491 L<guestfs-examples(3)>,
2492 L<guestfs-ocaml(3)>,
2493 L<guestfs-python(3)>,
2500 L<virt-filesystems(1)>,
2501 L<virt-inspector(1)>,
2502 L<virt-list-filesystems(1)>,
2503 L<virt-list-partitions(1)>,
2512 L<http://libguestfs.org/>.
2514 Tools with a similar purpose:
2523 To get a list of bugs against libguestfs use this link:
2525 L<https://bugzilla.redhat.com/buglist.cgi?component=libguestfs&product=Virtualization+Tools>
2527 To report a new bug against libguestfs use this link:
2529 L<https://bugzilla.redhat.com/enter_bug.cgi?component=libguestfs&product=Virtualization+Tools>
2531 When reporting a bug, please check:
2537 That the bug hasn't been reported already.
2541 That you are testing a recent version.
2545 Describe the bug accurately, and give a way to reproduce it.
2549 Run libguestfs-test-tool and paste the B<complete, unedited>
2550 output into the bug report.
2556 Richard W.M. Jones (C<rjones at redhat dot com>)
2560 Copyright (C) 2009-2010 Red Hat Inc.
2561 L<http://libguestfs.org/>
2563 This library is free software; you can redistribute it and/or
2564 modify it under the terms of the GNU Lesser General Public
2565 License as published by the Free Software Foundation; either
2566 version 2 of the License, or (at your option) any later version.
2568 This library is distributed in the hope that it will be useful,
2569 but WITHOUT ANY WARRANTY; without even the implied warranty of
2570 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
2571 Lesser General Public License for more details.
2573 You should have received a copy of the GNU Lesser General Public
2574 License along with this library; if not, write to the Free Software
2575 Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA