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, "/");
20 cc prog.c -o prog -lguestfs
22 cc prog.c -o prog `pkg-config libguestfs --cflags --libs`
26 Libguestfs is a library for accessing and modifying guest disk images.
27 Amongst the things this is good for: making batch configuration
28 changes to guests, getting disk used/free statistics (see also:
29 virt-df), migrating between virtualization systems (see also:
30 virt-p2v), performing partial backups, performing partial guest
31 clones, cloning guests and changing registry/UUID/hostname info, and
34 Libguestfs uses Linux kernel and qemu code, and can access any type of
35 guest filesystem that Linux and qemu can, including but not limited
36 to: ext2/3/4, btrfs, FAT and NTFS, LVM, many different disk partition
37 schemes, qcow, qcow2, vmdk.
39 Libguestfs provides ways to enumerate guest storage (eg. partitions,
40 LVs, what filesystem is in each LV, etc.). It can also run commands
41 in the context of the guest. Also you can access filesystems over
44 Libguestfs is a library that can be linked with C and C++ management
45 programs (or management programs written in OCaml, Perl, Python, Ruby,
46 Java, Haskell or C#). You can also use it from shell scripts or the
49 You don't need to be root to use libguestfs, although obviously you do
50 need enough permissions to access the disk images.
52 Libguestfs is a large API because it can do many things. For a gentle
53 introduction, please read the L</API OVERVIEW> section next.
57 This section provides a gentler overview of the libguestfs API. We
58 also try to group API calls together, where that may not be obvious
59 from reading about the individual calls in the main section of this
64 Before you can use libguestfs calls, you have to create a handle.
65 Then you must add at least one disk image to the handle, followed by
66 launching the handle, then performing whatever operations you want,
67 and finally closing the handle. By convention we use the single
68 letter C<g> for the name of the handle variable, although of course
69 you can use any name you want.
71 The general structure of all libguestfs-using programs looks like
74 guestfs_h *g = guestfs_create ();
76 /* Call guestfs_add_drive additional times if there are
77 * multiple disk images.
79 guestfs_add_drive (g, "guest.img");
81 /* Most manipulation calls won't work until you've launched
82 * the handle 'g'. You have to do this _after_ adding drives
83 * and _before_ other commands.
87 /* Now you can examine what partitions, LVs etc are available.
89 char **partitions = guestfs_list_partitions (g);
90 char **logvols = guestfs_lvs (g);
92 /* To access a filesystem in the image, you must mount it.
94 guestfs_mount (g, "/dev/sda1", "/");
96 /* Now you can perform filesystem actions on the guest
99 guestfs_touch (g, "/hello");
101 /* You only need to call guestfs_sync if you have made
102 * changes to the guest image. (But if you've made changes
103 * then you *must* sync). See also: guestfs_umount and
104 * guestfs_umount_all calls.
108 /* Close the handle 'g'. */
111 The code above doesn't include any error checking. In real code you
112 should check return values carefully for errors. In general all
113 functions that return integers return C<-1> on error, and all
114 functions that return pointers return C<NULL> on error. See section
115 L</ERROR HANDLING> below for how to handle errors, and consult the
116 documentation for each function call below to see precisely how they
117 return error indications.
121 The image filename (C<"guest.img"> in the example above) could be a
122 disk image from a virtual machine, a L<dd(1)> copy of a physical hard
123 disk, an actual block device, or simply an empty file of zeroes that
124 you have created through L<posix_fallocate(3)>. Libguestfs lets you
125 do useful things to all of these.
127 You can add a disk read-only using L</guestfs_add_drive_ro>, in which
128 case libguestfs won't modify the file.
130 Be extremely cautious if the disk image is in use, eg. if it is being
131 used by a virtual machine. Adding it read-write will almost certainly
132 cause disk corruption, but adding it read-only is safe.
134 You must add at least one disk image, and you may add multiple disk
135 images. In the API, the disk images are usually referred to as
136 C</dev/sda> (for the first one you added), C</dev/sdb> (for the second
139 Once L</guestfs_launch> has been called you cannot add any more images.
140 You can call L</guestfs_list_devices> to get a list of the device
141 names, in the order that you added them. See also L</BLOCK DEVICE
146 Before you can read or write files, create directories and so on in a
147 disk image that contains filesystems, you have to mount those
148 filesystems using L</guestfs_mount>. If you already know that a disk
149 image contains (for example) one partition with a filesystem on that
150 partition, then you can mount it directly:
152 guestfs_mount (g, "/dev/sda1", "/");
154 where C</dev/sda1> means literally the first partition (C<1>) of the
155 first disk image that we added (C</dev/sda>). If the disk contains
156 Linux LVM2 logical volumes you could refer to those instead (eg. C</dev/VG/LV>).
158 If you are given a disk image and you don't know what it contains then
159 you have to find out. Libguestfs can do that too: use
160 L</guestfs_list_partitions> and L</guestfs_lvs> to list possible
161 partitions and LVs, and either try mounting each to see what is
162 mountable, or else examine them with L</guestfs_vfs_type> or
163 L</guestfs_file>. Libguestfs also has a set of APIs for inspection of
164 disk images (see L</INSPECTION> below). But you might find it easier
165 to look at higher level programs built on top of libguestfs, in
166 particular L<virt-inspector(1)>.
168 To mount a disk image read-only, use L</guestfs_mount_ro>. There are
169 several other variations of the C<guestfs_mount_*> call.
171 =head2 FILESYSTEM ACCESS AND MODIFICATION
173 The majority of the libguestfs API consists of fairly low-level calls
174 for accessing and modifying the files, directories, symlinks etc on
175 mounted filesystems. There are over a hundred such calls which you
176 can find listed in detail below in this man page, and we don't even
177 pretend to cover them all in this overview.
179 Specify filenames as full paths, starting with C<"/"> and including
182 For example, if you mounted a filesystem at C<"/"> and you want to
183 read the file called C<"etc/passwd"> then you could do:
185 char *data = guestfs_cat (g, "/etc/passwd");
187 This would return C<data> as a newly allocated buffer containing the
188 full content of that file (with some conditions: see also
189 L</DOWNLOADING> below), or C<NULL> if there was an error.
191 As another example, to create a top-level directory on that filesystem
192 called C<"var"> you would do:
194 guestfs_mkdir (g, "/var");
196 To create a symlink you could do:
198 guestfs_ln_s (g, "/etc/init.d/portmap",
199 "/etc/rc3.d/S30portmap");
201 Libguestfs will reject attempts to use relative paths and there is no
202 concept of a current working directory.
204 Libguestfs can return errors in many situations: for example if the
205 filesystem isn't writable, or if a file or directory that you
206 requested doesn't exist. If you are using the C API (documented here)
207 you have to check for those error conditions after each call. (Other
208 language bindings turn these errors into exceptions).
210 File writes are affected by the per-handle umask, set by calling
211 L</guestfs_umask> and defaulting to 022. See L</UMASK>.
215 Libguestfs contains API calls to read, create and modify partition
216 tables on disk images.
218 In the common case where you want to create a single partition
219 covering the whole disk, you should use the L</guestfs_part_disk>
222 const char *parttype = "mbr";
223 if (disk_is_larger_than_2TB)
225 guestfs_part_disk (g, "/dev/sda", parttype);
227 Obviously this effectively wipes anything that was on that disk image
232 Libguestfs provides access to a large part of the LVM2 API, such as
233 L</guestfs_lvcreate> and L</guestfs_vgremove>. It won't make much sense
234 unless you familiarize yourself with the concepts of physical volumes,
235 volume groups and logical volumes.
237 This author strongly recommends reading the LVM HOWTO, online at
238 L<http://tldp.org/HOWTO/LVM-HOWTO/>.
242 Use L</guestfs_cat> to download small, text only files. This call
243 is limited to files which are less than 2 MB and which cannot contain
244 any ASCII NUL (C<\0>) characters. However it has a very simple
247 L</guestfs_read_file> can be used to read files which contain
248 arbitrary 8 bit data, since it returns a (pointer, size) pair.
249 However it is still limited to "small" files, less than 2 MB.
251 L</guestfs_download> can be used to download any file, with no
252 limits on content or size (even files larger than 4 GB).
254 To download multiple files, see L</guestfs_tar_out> and
259 It's often the case that you want to write a file or files to the disk
262 To write a small file with fixed content, use L</guestfs_write>. To
263 create a file of all zeroes, use L</guestfs_truncate_size> (sparse) or
264 L</guestfs_fallocate64> (with all disk blocks allocated). There are a
265 variety of other functions for creating test files, for example
266 L</guestfs_fill> and L</guestfs_fill_pattern>.
268 To upload a single file, use L</guestfs_upload>. This call has no
269 limits on file content or size (even files larger than 4 GB).
271 To upload multiple files, see L</guestfs_tar_in> and L</guestfs_tgz_in>.
273 However the fastest way to upload I<large numbers of arbitrary files>
274 is to turn them into a squashfs or CD ISO (see L<mksquashfs(8)> and
275 L<mkisofs(8)>), then attach this using L</guestfs_add_drive_ro>. If
276 you add the drive in a predictable way (eg. adding it last after all
277 other drives) then you can get the device name from
278 L</guestfs_list_devices> and mount it directly using
279 L</guestfs_mount_ro>. Note that squashfs images are sometimes
280 non-portable between kernel versions, and they don't support labels or
281 UUIDs. If you want to pre-build an image or you need to mount it
282 using a label or UUID, use an ISO image instead.
286 There are various different commands for copying between files and
287 devices and in and out of the guest filesystem. These are summarised
292 =item B<file> to B<file>
294 Use L</guestfs_cp> to copy a single file, or
295 L</guestfs_cp_a> to copy directories recursively.
297 =item B<file or device> to B<file or device>
299 Use L</guestfs_dd> which efficiently uses L<dd(1)>
300 to copy between files and devices in the guest.
302 Example: duplicate the contents of an LV:
304 guestfs_dd (g, "/dev/VG/Original", "/dev/VG/Copy");
306 The destination (C</dev/VG/Copy>) must be at least as large as the
307 source (C</dev/VG/Original>). To copy less than the whole
308 source device, use L</guestfs_copy_size>.
310 =item B<file on the host> to B<file or device>
312 Use L</guestfs_upload>. See L</UPLOADING> above.
314 =item B<file or device> to B<file on the host>
316 Use L</guestfs_download>. See L</DOWNLOADING> above.
322 L</guestfs_ll> is just designed for humans to read (mainly when using
323 the L<guestfish(1)>-equivalent command C<ll>).
325 L</guestfs_ls> is a quick way to get a list of files in a directory
326 from programs, as a flat list of strings.
328 L</guestfs_readdir> is a programmatic way to get a list of files in a
329 directory, plus additional information about each one. It is more
330 equivalent to using the L<readdir(3)> call on a local filesystem.
332 L</guestfs_find> and L</guestfs_find0> can be used to recursively list
335 =head2 RUNNING COMMANDS
337 Although libguestfs is a primarily an API for manipulating files
338 inside guest images, we also provide some limited facilities for
339 running commands inside guests.
341 There are many limitations to this:
347 The kernel version that the command runs under will be different
348 from what it expects.
352 If the command needs to communicate with daemons, then most likely
353 they won't be running.
357 The command will be running in limited memory.
361 The network may not be available unless you enable it
362 (see L</guestfs_set_network>).
366 Only supports Linux guests (not Windows, BSD, etc).
370 Architecture limitations (eg. won't work for a PPC guest on
375 For SELinux guests, you may need to enable SELinux and load policy
376 first. See L</SELINUX> in this manpage.
380 The two main API calls to run commands are L</guestfs_command> and
381 L</guestfs_sh> (there are also variations).
383 The difference is that L</guestfs_sh> runs commands using the shell, so
384 any shell globs, redirections, etc will work.
386 =head2 CONFIGURATION FILES
388 To read and write configuration files in Linux guest filesystems, we
389 strongly recommend using Augeas. For example, Augeas understands how
390 to read and write, say, a Linux shadow password file or X.org
391 configuration file, and so avoids you having to write that code.
393 The main Augeas calls are bound through the C<guestfs_aug_*> APIs. We
394 don't document Augeas itself here because there is excellent
395 documentation on the L<http://augeas.net/> website.
397 If you don't want to use Augeas (you fool!) then try calling
398 L</guestfs_read_lines> to get the file as a list of lines which
399 you can iterate over.
403 We support SELinux guests. To ensure that labeling happens correctly
404 in SELinux guests, you need to enable SELinux and load the guest's
411 Before launching, do:
413 guestfs_set_selinux (g, 1);
417 After mounting the guest's filesystem(s), load the policy. This
418 is best done by running the L<load_policy(8)> command in the
421 guestfs_sh (g, "/usr/sbin/load_policy");
423 (Older versions of C<load_policy> require you to specify the
424 name of the policy file).
428 Optionally, set the security context for the API. The correct
429 security context to use can only be known by inspecting the
430 guest. As an example:
432 guestfs_setcon (g, "unconfined_u:unconfined_r:unconfined_t:s0");
436 This will work for running commands and editing existing files.
438 When new files are created, you may need to label them explicitly,
439 for example by running the external command
440 C<restorecon pathname>.
444 Certain calls are affected by the current file mode creation mask (the
445 "umask"). In particular ones which create files or directories, such
446 as L</guestfs_touch>, L</guestfs_mknod> or L</guestfs_mkdir>. This
447 affects either the default mode that the file is created with or
448 modifies the mode that you supply.
450 The default umask is C<022>, so files are created with modes such as
451 C<0644> and directories with C<0755>.
453 There are two ways to avoid being affected by umask. Either set umask
454 to 0 (call C<guestfs_umask (g, 0)> early after launching). Or call
455 L</guestfs_chmod> after creating each file or directory.
457 For more information about umask, see L<umask(2)>.
459 =head2 ENCRYPTED DISKS
461 Libguestfs allows you to access Linux guests which have been
462 encrypted using whole disk encryption that conforms to the
463 Linux Unified Key Setup (LUKS) standard. This includes
464 nearly all whole disk encryption systems used by modern
467 Use L</guestfs_vfs_type> to identify LUKS-encrypted block
468 devices (it returns the string C<crypto_LUKS>).
470 Then open these devices by calling L</guestfs_luks_open>.
471 Obviously you will require the passphrase!
473 Opening a LUKS device creates a new device mapper device
474 called C</dev/mapper/mapname> (where C<mapname> is the
475 string you supply to L</guestfs_luks_open>).
476 Reads and writes to this mapper device are decrypted from and
477 encrypted to the underlying block device respectively.
479 LVM volume groups on the device can be made visible by calling
480 L</guestfs_vgscan> followed by L</guestfs_vg_activate_all>.
481 The logical volume(s) can now be mounted in the usual way.
483 Use the reverse process to close a LUKS device. Unmount
484 any logical volumes on it, deactivate the volume groups
485 by caling C<guestfs_vg_activate (g, 0, ["/dev/VG"])>.
486 Then close the mapper device by calling
487 L</guestfs_luks_close> on the C</dev/mapper/mapname>
488 device (I<not> the underlying encrypted block device).
492 Libguestfs has APIs for inspecting an unknown disk image to find out
493 if it contains operating systems. (These APIs used to be in a
494 separate Perl-only library called L<Sys::Guestfs::Lib(3)> but since
495 version 1.5.3 the most frequently used part of this library has been
496 rewritten in C and moved into the core code).
498 Add all disks belonging to the unknown virtual machine and call
499 L</guestfs_launch> in the usual way.
501 Then call L</guestfs_inspect_os>. This function uses other libguestfs
502 calls and certain heuristics, and returns a list of operating systems
503 that were found. An empty list means none were found. A single
504 element is the root filesystem of the operating system. For dual- or
505 multi-boot guests, multiple roots can be returned, each one
506 corresponding to a separate operating system. (Multi-boot virtual
507 machines are extremely rare in the world of virtualization, but since
508 this scenario can happen, we have built libguestfs to deal with it.)
510 For each root, you can then call various C<guestfs_inspect_get_*>
511 functions to get additional details about that operating system. For
512 example, call L</guestfs_inspect_get_type> to return the string
513 C<windows> or C<linux> for Windows and Linux-based operating systems
516 Un*x-like and Linux-based operating systems usually consist of several
517 filesystems which are mounted at boot time (for example, a separate
518 boot partition mounted on C</boot>). The inspection rules are able to
519 detect how filesystems correspond to mount points. Call
520 C<guestfs_inspect_get_mountpoints> to get this mapping. It might
521 return a hash table like this example:
524 / => /dev/vg_guest/lv_root
525 /usr => /dev/vg_guest/lv_usr
527 The caller can then make calls to L</guestfs_mount_options> to
528 mount the filesystems as suggested.
530 Be careful to mount filesystems in the right order (eg. C</> before
531 C</usr>). Sorting the keys of the hash by length, shortest first,
534 Inspection currently only works for some common operating systems.
535 Contributors are welcome to send patches for other operating systems
536 that we currently cannot detect.
538 Encrypted disks must be opened before inspection. See
539 L</ENCRYPTED DISKS> for more details. The L</guestfs_inspect_os>
540 function just ignores any encrypted devices.
542 A note on the implementation: The call L</guestfs_inspect_os> performs
543 inspection and caches the results in the guest handle. Subsequent
544 calls to C<guestfs_inspect_get_*> return this cached information, but
545 I<do not> re-read the disks. If you change the content of the guest
546 disks, you can redo inspection by calling L</guestfs_inspect_os>
549 =head2 SPECIAL CONSIDERATIONS FOR WINDOWS GUESTS
551 Libguestfs can mount NTFS partitions. It does this using the
552 L<http://www.ntfs-3g.org/> driver.
554 DOS and Windows still use drive letters, and the filesystems are
555 always treated as case insensitive by Windows itself, and therefore
556 you might find a Windows configuration file referring to a path like
557 C<c:\windows\system32>. When the filesystem is mounted in libguestfs,
558 that directory might be referred to as C</WINDOWS/System32>.
560 Drive letter mappings are outside the scope of libguestfs. You have
561 to use libguestfs to read the appropriate Windows Registry and
562 configuration files, to determine yourself how drives are mapped (see
563 also L<hivex(3)> and L<virt-inspector(1)>).
565 Replacing backslash characters with forward slash characters is also
566 outside the scope of libguestfs, but something that you can easily do.
568 Where we can help is in resolving the case insensitivity of paths.
569 For this, call L</guestfs_case_sensitive_path>.
571 Libguestfs also provides some help for decoding Windows Registry
572 "hive" files, through the library C<hivex> which is part of the
573 libguestfs project although ships as a separate tarball. You have to
574 locate and download the hive file(s) yourself, and then pass them to
575 C<hivex> functions. See also the programs L<hivexml(1)>,
576 L<hivexsh(1)>, L<hivexregedit(1)> and L<virt-win-reg(1)> for more help
579 =head2 USING LIBGUESTFS WITH OTHER PROGRAMMING LANGUAGES
581 Although we don't want to discourage you from using the C API, we will
582 mention here that the same API is also available in other languages.
584 The API is broadly identical in all supported languages. This means
585 that the C call C<guestfs_mount(g,path)> is
586 C<$g-E<gt>mount($path)> in Perl, C<g.mount(path)> in Python,
587 and C<Guestfs.mount g path> in OCaml. In other words, a
588 straightforward, predictable isomorphism between each language.
590 Error messages are automatically transformed
591 into exceptions if the language supports it.
593 We don't try to "object orientify" parts of the API in OO languages,
594 although contributors are welcome to write higher level APIs above
595 what we provide in their favourite languages if they wish.
601 You can use the I<guestfs.h> header file from C++ programs. The C++
602 API is identical to the C API. C++ classes and exceptions are not
607 The C# bindings are highly experimental. Please read the warnings
608 at the top of C<csharp/Libguestfs.cs>.
612 This is the only language binding that is working but incomplete.
613 Only calls which return simple integers have been bound in Haskell,
614 and we are looking for help to complete this binding.
618 Full documentation is contained in the Javadoc which is distributed
623 For documentation see the file C<guestfs.mli>.
627 For documentation see L<Sys::Guestfs(3)>.
631 For documentation do:
639 Use the Guestfs module. There is no Ruby-specific documentation, but
640 you can find examples written in Ruby in the libguestfs source.
642 =item B<shell scripts>
644 For documentation see L<guestfish(1)>.
648 =head2 LIBGUESTFS GOTCHAS
650 L<http://en.wikipedia.org/wiki/Gotcha_(programming)>: "A feature of a
651 system [...] that works in the way it is documented but is
652 counterintuitive and almost invites mistakes."
654 Since we developed libguestfs and the associated tools, there are
655 several things we would have designed differently, but are now stuck
656 with for backwards compatibility or other reasons. If there is ever a
657 libguestfs 2.0 release, you can expect these to change. Beware of
662 =item Autosync / forgetting to sync.
664 When modifying a filesystem from C or another language, you B<must>
665 unmount all filesystems and call L</guestfs_sync> explicitly before
666 you close the libguestfs handle. You can also call:
668 guestfs_set_autosync (g, 1);
670 to have the unmount/sync done automatically for you when the handle 'g'
671 is closed. (This feature is called "autosync", L</guestfs_set_autosync>
674 If you forget to do this, then it is entirely possible that your
675 changes won't be written out, or will be partially written, or (very
676 rarely) that you'll get disk corruption.
678 Note that in L<guestfish(3)> autosync is the default. So quick and
679 dirty guestfish scripts that forget to sync will work just fine, which
680 can make this very puzzling if you are trying to debug a problem.
682 =item Mount option C<-o sync> should not be the default.
684 If you use L</guestfs_mount>, then C<-o sync,noatime> are added
685 implicitly. However C<-o sync> does not add any reliability benefit,
686 but does have a very large performance impact.
688 The work around is to use L</guestfs_mount_options> and set the mount
689 options that you actually want to use.
691 =item Read-only should be the default.
693 In L<guestfish(3)>, I<--ro> should be the default, and you should
694 have to specify I<--rw> if you want to make changes to the image.
696 This would reduce the potential to corrupt live VM images.
698 Note that many filesystems change the disk when you just mount and
699 unmount, even if you didn't perform any writes. You need to use
700 L</guestfs_add_drive_ro> to guarantee that the disk is not changed.
702 =item guestfish command line is hard to use.
704 C<guestfish disk.img> doesn't do what people expect (open C<disk.img>
705 for examination). It tries to run a guestfish command C<disk.img>
706 which doesn't exist, so it fails. In earlier versions of guestfish
707 the error message was also unintuitive, but we have corrected this
708 since. Like the Bourne shell, we should have used C<guestfish -c
709 command> to run commands.
711 =item guestfish megabyte modifiers don't work right on all commands
713 In recent guestfish you can use C<1M> to mean 1 megabyte (and
714 similarly for other modifiers). What guestfish actually does is to
715 multiply the number part by the modifier part and pass the result to
716 the C API. However this doesn't work for a few APIs which aren't
717 expecting bytes, but are already expecting some other unit
720 The most common is L</guestfs_lvcreate>. The guestfish command:
724 does not do what you might expect. Instead because
725 L</guestfs_lvcreate> is already expecting megabytes, this tries to
726 create a 100 I<terabyte> (100 megabytes * megabytes) logical volume.
727 The error message you get from this is also a little obscure.
729 This could be fixed in the generator by specially marking parameters
730 and return values which take bytes or other units.
732 =item Library should return errno with error messages.
734 It would be a nice-to-have to be able to get the original value of
735 'errno' from inside the appliance along error paths (where set).
736 Currently L<guestmount(1)> goes through hoops to try to reverse the
737 error message string into an errno, see the function error() in
740 In libguestfs 1.5.4, the protocol was changed so that the
741 Linux errno is sent back from the daemon.
745 =head2 PROTOCOL LIMITS
747 Internally libguestfs uses a message-based protocol to pass API calls
748 and their responses to and from a small "appliance" (see L</INTERNALS>
749 for plenty more detail about this). The maximum message size used by
750 the protocol is slightly less than 4 MB. For some API calls you may
751 need to be aware of this limit. The API calls which may be affected
752 are individually documented, with a link back to this section of the
755 A simple call such as L</guestfs_cat> returns its result (the file
756 data) in a simple string. Because this string is at some point
757 internally encoded as a message, the maximum size that it can return
758 is slightly under 4 MB. If the requested file is larger than this
759 then you will get an error.
761 In order to transfer large files into and out of the guest filesystem,
762 you need to use particular calls that support this. The sections
763 L</UPLOADING> and L</DOWNLOADING> document how to do this.
765 You might also consider mounting the disk image using our FUSE
766 filesystem support (L<guestmount(1)>).
768 =head2 KEYS AND PASSPHRASES
770 Certain libguestfs calls take a parameter that contains sensitive key
771 material, passed in as a C string.
773 In the future we would hope to change the libguestfs implementation so
774 that keys are L<mlock(2)>-ed into physical RAM, and thus can never end
775 up in swap. However this is I<not> done at the moment, because of the
776 complexity of such an implementation.
778 Therefore you should be aware that any key parameter you pass to
779 libguestfs might end up being written out to the swap partition. If
780 this is a concern, scrub the swap partition or don't use libguestfs on
783 =head1 CONNECTION MANAGEMENT
787 C<guestfs_h> is the opaque type representing a connection handle.
788 Create a handle by calling L</guestfs_create>. Call L</guestfs_close>
789 to free the handle and release all resources used.
791 For information on using multiple handles and threads, see the section
792 L</MULTIPLE HANDLES AND MULTIPLE THREADS> below.
794 =head2 guestfs_create
796 guestfs_h *guestfs_create (void);
798 Create a connection handle.
800 You have to call L</guestfs_add_drive> on the handle at least once.
802 This function returns a non-NULL pointer to a handle on success or
805 After configuring the handle, you have to call L</guestfs_launch>.
807 You may also want to configure error handling for the handle. See
808 L</ERROR HANDLING> section below.
812 void guestfs_close (guestfs_h *g);
814 This closes the connection handle and frees up all resources used.
816 =head1 ERROR HANDLING
818 The convention in all functions that return C<int> is that they return
819 C<-1> to indicate an error. You can get additional information on
820 errors by calling L</guestfs_last_error> and/or by setting up an error
821 handler with L</guestfs_set_error_handler>.
823 The default error handler prints the information string to C<stderr>.
825 Out of memory errors are handled differently. The default action is
826 to call L<abort(3)>. If this is undesirable, then you can set a
827 handler using L</guestfs_set_out_of_memory_handler>.
829 =head2 guestfs_last_error
831 const char *guestfs_last_error (guestfs_h *g);
833 This returns the last error message that happened on C<g>. If
834 there has not been an error since the handle was created, then this
837 The lifetime of the returned string is until the next error occurs, or
838 L</guestfs_close> is called.
840 The error string is not localized (ie. is always in English), because
841 this makes searching for error messages in search engines give the
842 largest number of results.
844 =head2 guestfs_set_error_handler
846 typedef void (*guestfs_error_handler_cb) (guestfs_h *g,
849 void guestfs_set_error_handler (guestfs_h *g,
850 guestfs_error_handler_cb cb,
853 The callback C<cb> will be called if there is an error. The
854 parameters passed to the callback are an opaque data pointer and the
855 error message string.
857 Note that the message string C<msg> is freed as soon as the callback
858 function returns, so if you want to stash it somewhere you must make
861 The default handler prints messages on C<stderr>.
863 If you set C<cb> to C<NULL> then I<no> handler is called.
865 =head2 guestfs_get_error_handler
867 guestfs_error_handler_cb guestfs_get_error_handler (guestfs_h *g,
870 Returns the current error handler callback.
872 =head2 guestfs_set_out_of_memory_handler
874 typedef void (*guestfs_abort_cb) (void);
875 int guestfs_set_out_of_memory_handler (guestfs_h *g,
878 The callback C<cb> will be called if there is an out of memory
879 situation. I<Note this callback must not return>.
881 The default is to call L<abort(3)>.
883 You cannot set C<cb> to C<NULL>. You can't ignore out of memory
886 =head2 guestfs_get_out_of_memory_handler
888 guestfs_abort_fn guestfs_get_out_of_memory_handler (guestfs_h *g);
890 This returns the current out of memory handler.
894 Libguestfs needs a kernel and initrd.img, which it finds by looking
895 along an internal path.
897 By default it looks for these in the directory C<$libdir/guestfs>
898 (eg. C</usr/local/lib/guestfs> or C</usr/lib64/guestfs>).
900 Use L</guestfs_set_path> or set the environment variable
901 L</LIBGUESTFS_PATH> to change the directories that libguestfs will
902 search in. The value is a colon-separated list of paths. The current
903 directory is I<not> searched unless the path contains an empty element
904 or C<.>. For example C<LIBGUESTFS_PATH=:/usr/lib/guestfs> would
905 search the current directory and then C</usr/lib/guestfs>.
907 =head1 HIGH-LEVEL API ACTIONS
911 We guarantee the libguestfs ABI (binary interface), for public,
912 high-level actions as outlined in this section. Although we will
913 deprecate some actions, for example if they get replaced by newer
914 calls, we will keep the old actions forever. This allows you the
915 developer to program in confidence against the libguestfs API.
925 =head2 GROUPS OF FUNCTIONALITY IN THE APPLIANCE
927 Using L</guestfs_available> you can test availability of
928 the following groups of functions. This test queries the
929 appliance to see if the appliance you are currently using
930 supports the functionality.
934 =head2 GUESTFISH supported COMMAND
936 In L<guestfish(3)> there is a handy interactive command
937 C<supported> which prints out the available groups and
938 whether they are supported by this build of libguestfs.
939 Note however that you have to do C<run> first.
941 =head2 SINGLE CALLS AT COMPILE TIME
943 Since version 1.5.8, C<E<lt>guestfs.hE<gt>> defines symbols
944 for each C API function, such as:
946 #define LIBGUESTFS_HAVE_DD 1
948 if L</guestfs_dd> is available.
950 Before version 1.5.8, if you needed to test whether a single
951 libguestfs function is available at compile time, we recommended using
952 build tools such as autoconf or cmake. For example in autotools you
955 AC_CHECK_LIB([guestfs],[guestfs_create])
956 AC_CHECK_FUNCS([guestfs_dd])
958 which would result in C<HAVE_GUESTFS_DD> being either defined
959 or not defined in your program.
961 =head2 SINGLE CALLS AT RUN TIME
963 Testing at compile time doesn't guarantee that a function really
964 exists in the library. The reason is that you might be dynamically
965 linked against a previous I<libguestfs.so> (dynamic library)
966 which doesn't have the call. This situation unfortunately results
967 in a segmentation fault, which is a shortcoming of the C dynamic
968 linking system itself.
970 You can use L<dlopen(3)> to test if a function is available
971 at run time, as in this example program (note that you still
972 need the compile time check as well):
982 #ifdef LIBGUESTFS_HAVE_DD
986 /* Test if the function guestfs_dd is really available. */
987 dl = dlopen (NULL, RTLD_LAZY);
989 fprintf (stderr, "dlopen: %s\n", dlerror ());
992 has_function = dlsym (dl, "guestfs_dd") != NULL;
996 printf ("this libguestfs.so does NOT have guestfs_dd function\n");
998 printf ("this libguestfs.so has guestfs_dd function\n");
999 /* Now it's safe to call
1000 guestfs_dd (g, "foo", "bar");
1004 printf ("guestfs_dd function was not found at compile time\n");
1008 You may think the above is an awful lot of hassle, and it is.
1009 There are other ways outside of the C linking system to ensure
1010 that this kind of incompatibility never arises, such as using
1013 Requires: libguestfs >= 1.0.80
1017 <!-- old anchor for the next section -->
1018 <a name="state_machine_and_low_level_event_api"/>
1024 Internally, libguestfs is implemented by running an appliance (a
1025 special type of small virtual machine) using L<qemu(1)>. Qemu runs as
1026 a child process of the main program.
1032 | | child process / appliance
1033 | | __________________________
1035 +-------------------+ RPC | +-----------------+ |
1036 | libguestfs <--------------------> guestfsd | |
1037 | | | +-----------------+ |
1038 \___________________/ | | Linux kernel | |
1039 | +--^--------------+ |
1040 \_________|________________/
1048 The library, linked to the main program, creates the child process and
1049 hence the appliance in the L</guestfs_launch> function.
1051 Inside the appliance is a Linux kernel and a complete stack of
1052 userspace tools (such as LVM and ext2 programs) and a small
1053 controlling daemon called L</guestfsd>. The library talks to
1054 L</guestfsd> using remote procedure calls (RPC). There is a mostly
1055 one-to-one correspondence between libguestfs API calls and RPC calls
1056 to the daemon. Lastly the disk image(s) are attached to the qemu
1057 process which translates device access by the appliance's Linux kernel
1058 into accesses to the image.
1060 A common misunderstanding is that the appliance "is" the virtual
1061 machine. Although the disk image you are attached to might also be
1062 used by some virtual machine, libguestfs doesn't know or care about
1063 this. (But you will care if both libguestfs's qemu process and your
1064 virtual machine are trying to update the disk image at the same time,
1065 since these usually results in massive disk corruption).
1067 =head1 STATE MACHINE
1069 libguestfs uses a state machine to model the child process:
1080 / | \ \ guestfs_launch
1091 \______/ <------ \________/
1093 The normal transitions are (1) CONFIG (when the handle is created, but
1094 there is no child process), (2) LAUNCHING (when the child process is
1095 booting up), (3) alternating between READY and BUSY as commands are
1096 issued to, and carried out by, the child process.
1098 The guest may be killed by L</guestfs_kill_subprocess>, or may die
1099 asynchronously at any time (eg. due to some internal error), and that
1100 causes the state to transition back to CONFIG.
1102 Configuration commands for qemu such as L</guestfs_add_drive> can only
1103 be issued when in the CONFIG state.
1105 The high-level API offers two calls that go from CONFIG through
1106 LAUNCHING to READY. L</guestfs_launch> blocks until the child process
1107 is READY to accept commands (or until some failure or timeout).
1108 L</guestfs_launch> internally moves the state from CONFIG to LAUNCHING
1109 while it is running.
1111 High-level API actions such as L</guestfs_mount> can only be issued
1112 when in the READY state. These high-level API calls block waiting for
1113 the command to be carried out (ie. the state to transition to BUSY and
1114 then back to READY). But using the low-level event API, you get
1115 non-blocking versions. (But you can still only carry out one
1116 operation per handle at a time - that is a limitation of the
1117 communications protocol we use).
1119 Finally, the child process sends asynchronous messages back to the
1120 main program, such as kernel log messages. Mostly these are ignored
1121 by the high-level API, but using the low-level event API you can
1122 register to receive these messages.
1124 =head2 SETTING CALLBACKS TO HANDLE EVENTS
1126 The child process generates events in some situations. Current events
1127 include: receiving a log message, the child process exits.
1129 Use the C<guestfs_set_*_callback> functions to set a callback for
1130 different types of events.
1132 Only I<one callback of each type> can be registered for each handle.
1133 Calling C<guestfs_set_*_callback> again overwrites the previous
1134 callback of that type. Cancel all callbacks of this type by calling
1135 this function with C<cb> set to C<NULL>.
1137 =head2 guestfs_set_log_message_callback
1139 typedef void (*guestfs_log_message_cb) (guestfs_h *g, void *opaque,
1140 char *buf, int len);
1141 void guestfs_set_log_message_callback (guestfs_h *g,
1142 guestfs_log_message_cb cb,
1145 The callback function C<cb> will be called whenever qemu or the guest
1146 writes anything to the console.
1148 Use this function to capture kernel messages and similar.
1150 Normally there is no log message handler, and log messages are just
1153 =head2 guestfs_set_subprocess_quit_callback
1155 typedef void (*guestfs_subprocess_quit_cb) (guestfs_h *g, void *opaque);
1156 void guestfs_set_subprocess_quit_callback (guestfs_h *g,
1157 guestfs_subprocess_quit_cb cb,
1160 The callback function C<cb> will be called when the child process
1161 quits, either asynchronously or if killed by
1162 L</guestfs_kill_subprocess>. (This corresponds to a transition from
1163 any state to the CONFIG state).
1165 =head2 guestfs_set_launch_done_callback
1167 typedef void (*guestfs_launch_done_cb) (guestfs_h *g, void *opaque);
1168 void guestfs_set_launch_done_callback (guestfs_h *g,
1169 guestfs_launch_done_cb cb,
1172 The callback function C<cb> will be called when the child process
1173 becomes ready first time after it has been launched. (This
1174 corresponds to a transition from LAUNCHING to the READY state).
1176 =head2 guestfs_set_close_callback
1178 typedef void (*guestfs_close_cb) (guestfs_h *g, void *opaque);
1179 void guestfs_set_close_callback (guestfs_h *g,
1180 guestfs_close_cb cb,
1183 The callback function C<cb> will be called while the handle
1184 is being closed (synchronously from L</guestfs_close>).
1186 Note that libguestfs installs an L<atexit(3)> handler to try to
1187 clean up handles that are open when the program exits. This
1188 means that this callback might be called indirectly from
1189 L<exit(3)>, which can cause unexpected problems in higher-level
1190 languages (eg. if your HLL interpreter has already been cleaned
1191 up by the time this is called, and if your callback then jumps
1192 into some HLL function).
1194 =head2 guestfs_set_progress_callback
1196 typedef void (*guestfs_progress_cb) (guestfs_h *g, void *opaque,
1197 int proc_nr, int serial,
1198 uint64_t position, uint64_t total);
1199 void guestfs_set_progress_callback (guestfs_h *g,
1200 guestfs_progress_cb cb,
1203 Some long-running operations can generate progress messages. If
1204 this callback is registered, then it will be called each time a
1205 progress message is generated (usually two seconds after the
1206 operation started, and three times per second thereafter until
1207 it completes, although the frequency may change in future versions).
1209 The callback receives two numbers: C<position> and C<total>.
1210 The units of C<total> are not defined, although for some
1211 operations C<total> may relate in some way to the amount of
1212 data to be transferred (eg. in bytes or megabytes), and
1213 C<position> may be the portion which has been transferred.
1215 The only defined and stable parts of the API are:
1221 The callback can display to the user some type of progress bar or
1222 indicator which shows the ratio of C<position>:C<total>.
1226 0 E<lt>= C<position> E<lt>= C<total>
1230 If any progress notification is sent during a call, then a final
1231 progress notification is always sent when C<position> = C<total>.
1233 This is to simplify caller code, so callers can easily set the
1234 progress indicator to "100%" at the end of the operation, without
1235 requiring special code to detect this case.
1239 The callback also receives the procedure number and serial number of
1240 the call. These are only useful for debugging protocol issues, and
1241 the callback can normally ignore them. The callback may want to
1242 print these numbers in error messages or debugging messages.
1244 =head1 PRIVATE DATA AREA
1246 You can attach named pieces of private data to the libguestfs handle,
1247 and fetch them by name for the lifetime of the handle. This is called
1248 the private data area and is only available from the C API.
1250 To attach a named piece of data, use the following call:
1252 void guestfs_set_private (guestfs_h *g, const char *key, void *data);
1254 C<key> is the name to associate with this data, and C<data> is an
1255 arbitrary pointer (which can be C<NULL>). Any previous item with the
1256 same name is overwritten.
1258 You can use any C<key> you want, but names beginning with an
1259 underscore character are reserved for internal libguestfs purposes
1260 (for implementing language bindings). It is recommended to prefix the
1261 name with some unique string to avoid collisions with other users.
1263 To retrieve the pointer, use:
1265 void *guestfs_get_private (guestfs_h *g, const char *key);
1267 This function returns C<NULL> if either no data is found associated
1268 with C<key>, or if the user previously set the C<key>'s C<data>
1271 Libguestfs does not try to look at or interpret the C<data> pointer in
1272 any way. As far as libguestfs is concerned, it need not be a valid
1273 pointer at all. In particular, libguestfs does I<not> try to free the
1274 data when the handle is closed. If the data must be freed, then the
1275 caller must either free it before calling L</guestfs_close> or must
1276 set up a close callback to do it (see L</guestfs_set_close_callback>,
1277 and note that only one callback can be registered for a handle).
1279 The private data area is implemented using a hash table, and should be
1280 reasonably efficient for moderate numbers of keys.
1282 =head1 BLOCK DEVICE NAMING
1284 In the kernel there is now quite a profusion of schemata for naming
1285 block devices (in this context, by I<block device> I mean a physical
1286 or virtual hard drive). The original Linux IDE driver used names
1287 starting with C</dev/hd*>. SCSI devices have historically used a
1288 different naming scheme, C</dev/sd*>. When the Linux kernel I<libata>
1289 driver became a popular replacement for the old IDE driver
1290 (particularly for SATA devices) those devices also used the
1291 C</dev/sd*> scheme. Additionally we now have virtual machines with
1292 paravirtualized drivers. This has created several different naming
1293 systems, such as C</dev/vd*> for virtio disks and C</dev/xvd*> for Xen
1296 As discussed above, libguestfs uses a qemu appliance running an
1297 embedded Linux kernel to access block devices. We can run a variety
1298 of appliances based on a variety of Linux kernels.
1300 This causes a problem for libguestfs because many API calls use device
1301 or partition names. Working scripts and the recipe (example) scripts
1302 that we make available over the internet could fail if the naming
1305 Therefore libguestfs defines C</dev/sd*> as the I<standard naming
1306 scheme>. Internally C</dev/sd*> names are translated, if necessary,
1307 to other names as required. For example, under RHEL 5 which uses the
1308 C</dev/hd*> scheme, any device parameter C</dev/sda2> is translated to
1309 C</dev/hda2> transparently.
1311 Note that this I<only> applies to parameters. The
1312 L</guestfs_list_devices>, L</guestfs_list_partitions> and similar calls
1313 return the true names of the devices and partitions as known to the
1316 =head2 ALGORITHM FOR BLOCK DEVICE NAME TRANSLATION
1318 Usually this translation is transparent. However in some (very rare)
1319 cases you may need to know the exact algorithm. Such cases include
1320 where you use L</guestfs_config> to add a mixture of virtio and IDE
1321 devices to the qemu-based appliance, so have a mixture of C</dev/sd*>
1322 and C</dev/vd*> devices.
1324 The algorithm is applied only to I<parameters> which are known to be
1325 either device or partition names. Return values from functions such
1326 as L</guestfs_list_devices> are never changed.
1332 Is the string a parameter which is a device or partition name?
1336 Does the string begin with C</dev/sd>?
1340 Does the named device exist? If so, we use that device.
1341 However if I<not> then we continue with this algorithm.
1345 Replace initial C</dev/sd> string with C</dev/hd>.
1347 For example, change C</dev/sda2> to C</dev/hda2>.
1349 If that named device exists, use it. If not, continue.
1353 Replace initial C</dev/sd> string with C</dev/vd>.
1355 If that named device exists, use it. If not, return an error.
1359 =head2 PORTABILITY CONCERNS
1361 Although the standard naming scheme and automatic translation is
1362 useful for simple programs and guestfish scripts, for larger programs
1363 it is best not to rely on this mechanism.
1365 Where possible for maximum future portability programs using
1366 libguestfs should use these future-proof techniques:
1372 Use L</guestfs_list_devices> or L</guestfs_list_partitions> to list
1373 actual device names, and then use those names directly.
1375 Since those device names exist by definition, they will never be
1380 Use higher level ways to identify filesystems, such as LVM names,
1381 UUIDs and filesystem labels.
1387 =head2 COMMUNICATION PROTOCOL
1389 Don't rely on using this protocol directly. This section documents
1390 how it currently works, but it may change at any time.
1392 The protocol used to talk between the library and the daemon running
1393 inside the qemu virtual machine is a simple RPC mechanism built on top
1394 of XDR (RFC 1014, RFC 1832, RFC 4506).
1396 The detailed format of structures is in C<src/guestfs_protocol.x>
1397 (note: this file is automatically generated).
1399 There are two broad cases, ordinary functions that don't have any
1400 C<FileIn> and C<FileOut> parameters, which are handled with very
1401 simple request/reply messages. Then there are functions that have any
1402 C<FileIn> or C<FileOut> parameters, which use the same request and
1403 reply messages, but they may also be followed by files sent using a
1406 =head3 ORDINARY FUNCTIONS (NO FILEIN/FILEOUT PARAMS)
1408 For ordinary functions, the request message is:
1410 total length (header + arguments,
1411 but not including the length word itself)
1412 struct guestfs_message_header (encoded as XDR)
1413 struct guestfs_<foo>_args (encoded as XDR)
1415 The total length field allows the daemon to allocate a fixed size
1416 buffer into which it slurps the rest of the message. As a result, the
1417 total length is limited to C<GUESTFS_MESSAGE_MAX> bytes (currently
1418 4MB), which means the effective size of any request is limited to
1419 somewhere under this size.
1421 Note also that many functions don't take any arguments, in which case
1422 the C<guestfs_I<foo>_args> is completely omitted.
1424 The header contains the procedure number (C<guestfs_proc>) which is
1425 how the receiver knows what type of args structure to expect, or none
1428 The reply message for ordinary functions is:
1430 total length (header + ret,
1431 but not including the length word itself)
1432 struct guestfs_message_header (encoded as XDR)
1433 struct guestfs_<foo>_ret (encoded as XDR)
1435 As above the C<guestfs_I<foo>_ret> structure may be completely omitted
1436 for functions that return no formal return values.
1438 As above the total length of the reply is limited to
1439 C<GUESTFS_MESSAGE_MAX>.
1441 In the case of an error, a flag is set in the header, and the reply
1442 message is slightly changed:
1444 total length (header + error,
1445 but not including the length word itself)
1446 struct guestfs_message_header (encoded as XDR)
1447 struct guestfs_message_error (encoded as XDR)
1449 The C<guestfs_message_error> structure contains the error message as a
1452 =head3 FUNCTIONS THAT HAVE FILEIN PARAMETERS
1454 A C<FileIn> parameter indicates that we transfer a file I<into> the
1455 guest. The normal request message is sent (see above). However this
1456 is followed by a sequence of file chunks.
1458 total length (header + arguments,
1459 but not including the length word itself,
1460 and not including the chunks)
1461 struct guestfs_message_header (encoded as XDR)
1462 struct guestfs_<foo>_args (encoded as XDR)
1463 sequence of chunks for FileIn param #0
1464 sequence of chunks for FileIn param #1 etc.
1466 The "sequence of chunks" is:
1468 length of chunk (not including length word itself)
1469 struct guestfs_chunk (encoded as XDR)
1471 struct guestfs_chunk (encoded as XDR)
1474 struct guestfs_chunk (with data.data_len == 0)
1476 The final chunk has the C<data_len> field set to zero. Additionally a
1477 flag is set in the final chunk to indicate either successful
1478 completion or early cancellation.
1480 At time of writing there are no functions that have more than one
1481 FileIn parameter. However this is (theoretically) supported, by
1482 sending the sequence of chunks for each FileIn parameter one after
1483 another (from left to right).
1485 Both the library (sender) I<and> the daemon (receiver) may cancel the
1486 transfer. The library does this by sending a chunk with a special
1487 flag set to indicate cancellation. When the daemon sees this, it
1488 cancels the whole RPC, does I<not> send any reply, and goes back to
1489 reading the next request.
1491 The daemon may also cancel. It does this by writing a special word
1492 C<GUESTFS_CANCEL_FLAG> to the socket. The library listens for this
1493 during the transfer, and if it gets it, it will cancel the transfer
1494 (it sends a cancel chunk). The special word is chosen so that even if
1495 cancellation happens right at the end of the transfer (after the
1496 library has finished writing and has started listening for the reply),
1497 the "spurious" cancel flag will not be confused with the reply
1500 This protocol allows the transfer of arbitrary sized files (no 32 bit
1501 limit), and also files where the size is not known in advance
1502 (eg. from pipes or sockets). However the chunks are rather small
1503 (C<GUESTFS_MAX_CHUNK_SIZE>), so that neither the library nor the
1504 daemon need to keep much in memory.
1506 =head3 FUNCTIONS THAT HAVE FILEOUT PARAMETERS
1508 The protocol for FileOut parameters is exactly the same as for FileIn
1509 parameters, but with the roles of daemon and library reversed.
1511 total length (header + ret,
1512 but not including the length word itself,
1513 and not including the chunks)
1514 struct guestfs_message_header (encoded as XDR)
1515 struct guestfs_<foo>_ret (encoded as XDR)
1516 sequence of chunks for FileOut param #0
1517 sequence of chunks for FileOut param #1 etc.
1519 =head3 INITIAL MESSAGE
1521 Because the underlying channel (QEmu -net channel) doesn't have any
1522 sort of connection control, when the daemon launches it sends an
1523 initial word (C<GUESTFS_LAUNCH_FLAG>) which indicates that the guest
1524 and daemon is alive. This is what L</guestfs_launch> waits for.
1526 =head1 MULTIPLE HANDLES AND MULTIPLE THREADS
1528 All high-level libguestfs actions are synchronous. If you want
1529 to use libguestfs asynchronously then you must create a thread.
1531 Only use the handle from a single thread. Either use the handle
1532 exclusively from one thread, or provide your own mutex so that two
1533 threads cannot issue calls on the same handle at the same time.
1535 =head1 QEMU WRAPPERS
1537 If you want to compile your own qemu, run qemu from a non-standard
1538 location, or pass extra arguments to qemu, then you can write a
1539 shell-script wrapper around qemu.
1541 There is one important rule to remember: you I<must C<exec qemu>> as
1542 the last command in the shell script (so that qemu replaces the shell
1543 and becomes the direct child of the libguestfs-using program). If you
1544 don't do this, then the qemu process won't be cleaned up correctly.
1546 Here is an example of a wrapper, where I have built my own copy of
1550 qemudir=/home/rjones/d/qemu
1551 exec $qemudir/x86_64-softmmu/qemu-system-x86_64 -L $qemudir/pc-bios "$@"
1553 Save this script as C</tmp/qemu.wrapper> (or wherever), C<chmod +x>,
1554 and then use it by setting the LIBGUESTFS_QEMU environment variable.
1557 LIBGUESTFS_QEMU=/tmp/qemu.wrapper guestfish
1559 Note that libguestfs also calls qemu with the -help and -version
1560 options in order to determine features.
1562 =head1 LIBGUESTFS VERSION NUMBERS
1564 Since April 2010, libguestfs has started to make separate development
1565 and stable releases, along with corresponding branches in our git
1566 repository. These separate releases can be identified by version
1569 even numbers for stable: 1.2.x, 1.4.x, ...
1570 .-------- odd numbers for development: 1.3.x, 1.5.x, ...
1576 | `-------- sub-version
1578 `------ always '1' because we don't change the ABI
1580 Thus "1.3.5" is the 5th update to the development branch "1.3".
1582 As time passes we cherry pick fixes from the development branch and
1583 backport those into the stable branch, the effect being that the
1584 stable branch should get more stable and less buggy over time. So the
1585 stable releases are ideal for people who don't need new features but
1586 would just like the software to work.
1588 Our criteria for backporting changes are:
1594 Documentation changes which don't affect any code are
1595 backported unless the documentation refers to a future feature
1596 which is not in stable.
1600 Bug fixes which are not controversial, fix obvious problems, and
1601 have been well tested are backported.
1605 Simple rearrangements of code which shouldn't affect how it works get
1606 backported. This is so that the code in the two branches doesn't get
1607 too far out of step, allowing us to backport future fixes more easily.
1611 We I<don't> backport new features, new APIs, new tools etc, except in
1612 one exceptional case: the new feature is required in order to
1613 implement an important bug fix.
1617 A new stable branch starts when we think the new features in
1618 development are substantial and compelling enough over the current
1619 stable branch to warrant it. When that happens we create new stable
1620 and development versions 1.N.0 and 1.(N+1).0 [N is even]. The new
1621 dot-oh release won't necessarily be so stable at this point, but by
1622 backporting fixes from development, that branch will stabilize over
1625 =head1 ENVIRONMENT VARIABLES
1629 =item LIBGUESTFS_APPEND
1631 Pass additional options to the guest kernel.
1633 =item LIBGUESTFS_DEBUG
1635 Set C<LIBGUESTFS_DEBUG=1> to enable verbose messages. This
1636 has the same effect as calling C<guestfs_set_verbose (g, 1)>.
1638 =item LIBGUESTFS_MEMSIZE
1640 Set the memory allocated to the qemu process, in megabytes. For
1643 LIBGUESTFS_MEMSIZE=700
1645 =item LIBGUESTFS_PATH
1647 Set the path that libguestfs uses to search for kernel and initrd.img.
1648 See the discussion of paths in section PATH above.
1650 =item LIBGUESTFS_QEMU
1652 Set the default qemu binary that libguestfs uses. If not set, then
1653 the qemu which was found at compile time by the configure script is
1656 See also L</QEMU WRAPPERS> above.
1658 =item LIBGUESTFS_TRACE
1660 Set C<LIBGUESTFS_TRACE=1> to enable command traces. This
1661 has the same effect as calling C<guestfs_set_trace (g, 1)>.
1665 Location of temporary directory, defaults to C</tmp>.
1667 If libguestfs was compiled to use the supermin appliance then each
1668 handle will require rather a large amount of space in this directory
1669 for short periods of time (~ 80 MB). You can use C<$TMPDIR> to
1670 configure another directory to use in case C</tmp> is not large
1682 L<virt-inspector(1)>,
1683 L<virt-list-filesystems(1)>,
1684 L<virt-list-partitions(1)>,
1693 L<http://libguestfs.org/>.
1695 Tools with a similar purpose:
1704 To get a list of bugs against libguestfs use this link:
1706 L<https://bugzilla.redhat.com/buglist.cgi?component=libguestfs&product=Virtualization+Tools>
1708 To report a new bug against libguestfs use this link:
1710 L<https://bugzilla.redhat.com/enter_bug.cgi?component=libguestfs&product=Virtualization+Tools>
1712 When reporting a bug, please check:
1718 That the bug hasn't been reported already.
1722 That you are testing a recent version.
1726 Describe the bug accurately, and give a way to reproduce it.
1730 Run libguestfs-test-tool and paste the B<complete, unedited>
1731 output into the bug report.
1737 Richard W.M. Jones (C<rjones at redhat dot com>)
1741 Copyright (C) 2009-2010 Red Hat Inc.
1742 L<http://libguestfs.org/>
1744 This library is free software; you can redistribute it and/or
1745 modify it under the terms of the GNU Lesser General Public
1746 License as published by the Free Software Foundation; either
1747 version 2 of the License, or (at your option) any later version.
1749 This library is distributed in the hope that it will be useful,
1750 but WITHOUT ANY WARRANTY; without even the implied warranty of
1751 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
1752 Lesser General Public License for more details.
1754 You should have received a copy of the GNU Lesser General Public
1755 License along with this library; if not, write to the Free Software
1756 Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA