1 # libguestfs generated file
2 # WARNING: THIS FILE IS GENERATED BY 'src/generator.ml'.
3 # ANY CHANGES YOU MAKE TO THIS FILE WILL BE LOST.
5 # Copyright (C) 2009 Red Hat Inc.
7 # This library is free software; you can redistribute it and/or
8 # modify it under the terms of the GNU Lesser General Public
9 # License as published by the Free Software Foundation; either
10 # version 2 of the License, or (at your option) any later version.
12 # This library is distributed in the hope that it will be useful,
13 # but WITHOUT ANY WARRANTY; without even the implied warranty of
14 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
15 # Lesser General Public License for more details.
17 # You should have received a copy of the GNU Lesser General Public
18 # License along with this library; if not, write to the Free Software
19 # Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
25 Sys::Guestfs - Perl bindings for libguestfs
31 my $h = Sys::Guestfs->new ();
32 $h->add_drive ('guest.img');
35 $h->mount ('/dev/sda1', '/');
41 The C<Sys::Guestfs> module provides a Perl XS binding to the
42 libguestfs API for examining and modifying virtual machine
45 Amongst the things this is good for: making batch configuration
46 changes to guests, getting disk used/free statistics (see also:
47 virt-df), migrating between virtualization systems (see also:
48 virt-p2v), performing partial backups, performing partial guest
49 clones, cloning guests and changing registry/UUID/hostname info, and
52 Libguestfs uses Linux kernel and qemu code, and can access any type of
53 guest filesystem that Linux and qemu can, including but not limited
54 to: ext2/3/4, btrfs, FAT and NTFS, LVM, many different disk partition
55 schemes, qcow, qcow2, vmdk.
57 Libguestfs provides ways to enumerate guest storage (eg. partitions,
58 LVs, what filesystem is in each LV, etc.). It can also run commands
59 in the context of the guest. Also you can access filesystems over FTP.
63 All errors turn into calls to C<croak> (see L<Carp(3)>).
77 XSLoader::load ('Sys::Guestfs');
79 =item $h = Sys::Guestfs->new ();
81 Create a new guestfs handle.
87 my $class = ref ($proto) || $proto;
89 my $self = Sys::Guestfs::_create ();
94 =item $h->add_cdrom ($filename);
96 This function adds a virtual CD-ROM disk image to the guest.
98 This is equivalent to the qemu parameter C<-cdrom filename>.
100 Note that this call checks for the existence of C<filename>. This
101 stops you from specifying other types of drive which are supported
102 by qemu such as C<nbd:> and C<http:> URLs. To specify those, use
103 the general C<$h-E<gt>config> call instead.
105 =item $h->add_drive ($filename);
107 This function adds a virtual machine disk image C<filename> to the
108 guest. The first time you call this function, the disk appears as IDE
109 disk 0 (C</dev/sda>) in the guest, the second time as C</dev/sdb>, and
112 You don't necessarily need to be root when using libguestfs. However
113 you obviously do need sufficient permissions to access the filename
114 for whatever operations you want to perform (ie. read access if you
115 just want to read the image or write access if you want to modify the
118 This is equivalent to the qemu parameter C<-drive file=filename,cache=off>.
120 Note that this call checks for the existence of C<filename>. This
121 stops you from specifying other types of drive which are supported
122 by qemu such as C<nbd:> and C<http:> URLs. To specify those, use
123 the general C<$h-E<gt>config> call instead.
125 =item $h->add_drive_ro ($filename);
127 This adds a drive in snapshot mode, making it effectively
130 Note that writes to the device are allowed, and will be seen for
131 the duration of the guestfs handle, but they are written
132 to a temporary file which is discarded as soon as the guestfs
133 handle is closed. We don't currently have any method to enable
134 changes to be committed, although qemu can support this.
136 This is equivalent to the qemu parameter
137 C<-drive file=filename,snapshot=on>.
139 Note that this call checks for the existence of C<filename>. This
140 stops you from specifying other types of drive which are supported
141 by qemu such as C<nbd:> and C<http:> URLs. To specify those, use
142 the general C<$h-E<gt>config> call instead.
144 =item $h->aug_close ();
146 Close the current Augeas handle and free up any resources
147 used by it. After calling this, you have to call
148 C<$h-E<gt>aug_init> again before you can use any other
151 =item ($nrnodes, $created) = $h->aug_defnode ($name, $expr, $val);
153 Defines a variable C<name> whose value is the result of
156 If C<expr> evaluates to an empty nodeset, a node is created,
157 equivalent to calling C<$h-E<gt>aug_set> C<expr>, C<value>.
158 C<name> will be the nodeset containing that single node.
160 On success this returns a pair containing the
161 number of nodes in the nodeset, and a boolean flag
162 if a node was created.
164 =item $nrnodes = $h->aug_defvar ($name, $expr);
166 Defines an Augeas variable C<name> whose value is the result
167 of evaluating C<expr>. If C<expr> is NULL, then C<name> is
170 On success this returns the number of nodes in C<expr>, or
171 C<0> if C<expr> evaluates to something which is not a nodeset.
173 =item $val = $h->aug_get ($path);
175 Look up the value associated with C<path>. If C<path>
176 matches exactly one node, the C<value> is returned.
178 =item $h->aug_init ($root, $flags);
180 Create a new Augeas handle for editing configuration files.
181 If there was any previous Augeas handle associated with this
182 guestfs session, then it is closed.
184 You must call this before using any other C<$h-E<gt>aug_*>
187 C<root> is the filesystem root. C<root> must not be NULL,
190 The flags are the same as the flags defined in
191 E<lt>augeas.hE<gt>, the logical I<or> of the following
196 =item C<AUG_SAVE_BACKUP> = 1
198 Keep the original file with a C<.augsave> extension.
200 =item C<AUG_SAVE_NEWFILE> = 2
202 Save changes into a file with extension C<.augnew>, and
203 do not overwrite original. Overrides C<AUG_SAVE_BACKUP>.
205 =item C<AUG_TYPE_CHECK> = 4
207 Typecheck lenses (can be expensive).
209 =item C<AUG_NO_STDINC> = 8
211 Do not use standard load path for modules.
213 =item C<AUG_SAVE_NOOP> = 16
215 Make save a no-op, just record what would have been changed.
217 =item C<AUG_NO_LOAD> = 32
219 Do not load the tree in C<$h-E<gt>aug_init>.
223 To close the handle, you can call C<$h-E<gt>aug_close>.
225 To find out more about Augeas, see L<http://augeas.net/>.
227 =item $h->aug_insert ($path, $label, $before);
229 Create a new sibling C<label> for C<path>, inserting it into
230 the tree before or after C<path> (depending on the boolean
233 C<path> must match exactly one existing node in the tree, and
234 C<label> must be a label, ie. not contain C</>, C<*> or end
235 with a bracketed index C<[N]>.
237 =item $h->aug_load ();
239 Load files into the tree.
241 See C<aug_load> in the Augeas documentation for the full gory
244 =item @matches = $h->aug_ls ($path);
246 This is just a shortcut for listing C<$h-E<gt>aug_match>
247 C<path/*> and sorting the resulting nodes into alphabetical order.
249 =item @matches = $h->aug_match ($path);
251 Returns a list of paths which match the path expression C<path>.
252 The returned paths are sufficiently qualified so that they match
253 exactly one node in the current tree.
255 =item $h->aug_mv ($src, $dest);
257 Move the node C<src> to C<dest>. C<src> must match exactly
258 one node. C<dest> is overwritten if it exists.
260 =item $nrnodes = $h->aug_rm ($path);
262 Remove C<path> and all of its children.
264 On success this returns the number of entries which were removed.
266 =item $h->aug_save ();
268 This writes all pending changes to disk.
270 The flags which were passed to C<$h-E<gt>aug_init> affect exactly
273 =item $h->aug_set ($path, $val);
275 Set the value associated with C<path> to C<value>.
277 =item $h->blockdev_flushbufs ($device);
279 This tells the kernel to flush internal buffers associated
282 This uses the L<blockdev(8)> command.
284 =item $blocksize = $h->blockdev_getbsz ($device);
286 This returns the block size of a device.
288 (Note this is different from both I<size in blocks> and
289 I<filesystem block size>).
291 This uses the L<blockdev(8)> command.
293 =item $ro = $h->blockdev_getro ($device);
295 Returns a boolean indicating if the block device is read-only
296 (true if read-only, false if not).
298 This uses the L<blockdev(8)> command.
300 =item $sizeinbytes = $h->blockdev_getsize64 ($device);
302 This returns the size of the device in bytes.
304 See also C<$h-E<gt>blockdev_getsz>.
306 This uses the L<blockdev(8)> command.
308 =item $sectorsize = $h->blockdev_getss ($device);
310 This returns the size of sectors on a block device.
311 Usually 512, but can be larger for modern devices.
313 (Note, this is not the size in sectors, use C<$h-E<gt>blockdev_getsz>
316 This uses the L<blockdev(8)> command.
318 =item $sizeinsectors = $h->blockdev_getsz ($device);
320 This returns the size of the device in units of 512-byte sectors
321 (even if the sectorsize isn't 512 bytes ... weird).
323 See also C<$h-E<gt>blockdev_getss> for the real sector size of
324 the device, and C<$h-E<gt>blockdev_getsize64> for the more
325 useful I<size in bytes>.
327 This uses the L<blockdev(8)> command.
329 =item $h->blockdev_rereadpt ($device);
331 Reread the partition table on C<device>.
333 This uses the L<blockdev(8)> command.
335 =item $h->blockdev_setbsz ($device, $blocksize);
337 This sets the block size of a device.
339 (Note this is different from both I<size in blocks> and
340 I<filesystem block size>).
342 This uses the L<blockdev(8)> command.
344 =item $h->blockdev_setro ($device);
346 Sets the block device named C<device> to read-only.
348 This uses the L<blockdev(8)> command.
350 =item $h->blockdev_setrw ($device);
352 Sets the block device named C<device> to read-write.
354 This uses the L<blockdev(8)> command.
356 =item $content = $h->cat ($path);
358 Return the contents of the file named C<path>.
360 Note that this function cannot correctly handle binary files
361 (specifically, files containing C<\0> character which is treated
362 as end of string). For those you need to use the C<$h-E<gt>download>
363 function which has a more complex interface.
365 Because of the message protocol, there is a transfer limit
366 of somewhere between 2MB and 4MB. To transfer large files you should use
369 =item $checksum = $h->checksum ($csumtype, $path);
371 This call computes the MD5, SHAx or CRC checksum of the
374 The type of checksum to compute is given by the C<csumtype>
375 parameter which must have one of the following values:
381 Compute the cyclic redundancy check (CRC) specified by POSIX
382 for the C<cksum> command.
386 Compute the MD5 hash (using the C<md5sum> program).
390 Compute the SHA1 hash (using the C<sha1sum> program).
394 Compute the SHA224 hash (using the C<sha224sum> program).
398 Compute the SHA256 hash (using the C<sha256sum> program).
402 Compute the SHA384 hash (using the C<sha384sum> program).
406 Compute the SHA512 hash (using the C<sha512sum> program).
410 The checksum is returned as a printable string.
412 =item $h->chmod ($mode, $path);
414 Change the mode (permissions) of C<path> to C<mode>. Only
415 numeric modes are supported.
417 =item $h->chown ($owner, $group, $path);
419 Change the file owner to C<owner> and group to C<group>.
421 Only numeric uid and gid are supported. If you want to use
422 names, you will need to locate and parse the password file
423 yourself (Augeas support makes this relatively easy).
425 =item $output = $h->command (\@arguments);
427 This call runs a command from the guest filesystem. The
428 filesystem must be mounted, and must contain a compatible
429 operating system (ie. something Linux, with the same
430 or compatible processor architecture).
432 The single parameter is an argv-style list of arguments.
433 The first element is the name of the program to run.
434 Subsequent elements are parameters. The list must be
435 non-empty (ie. must contain a program name). Note that
436 the command runs directly, and is I<not> invoked via
437 the shell (see C<$h-E<gt>sh>).
439 The return value is anything printed to I<stdout> by
442 If the command returns a non-zero exit status, then
443 this function returns an error message. The error message
444 string is the content of I<stderr> from the command.
446 The C<$PATH> environment variable will contain at least
447 C</usr/bin> and C</bin>. If you require a program from
448 another location, you should provide the full path in the
451 Shared libraries and data files required by the program
452 must be available on filesystems which are mounted in the
453 correct places. It is the caller's responsibility to ensure
454 all filesystems that are needed are mounted at the right
457 Because of the message protocol, there is a transfer limit
458 of somewhere between 2MB and 4MB. To transfer large files you should use
461 =item @lines = $h->command_lines (\@arguments);
463 This is the same as C<$h-E<gt>command>, but splits the
464 result into a list of lines.
466 See also: C<$h-E<gt>sh_lines>
468 Because of the message protocol, there is a transfer limit
469 of somewhere between 2MB and 4MB. To transfer large files you should use
472 =item $h->config ($qemuparam, $qemuvalue);
474 This can be used to add arbitrary qemu command line parameters
475 of the form C<-param value>. Actually it's not quite arbitrary - we
476 prevent you from setting some parameters which would interfere with
477 parameters that we use.
479 The first character of C<param> string must be a C<-> (dash).
481 C<value> can be NULL.
483 =item $h->cp ($src, $dest);
485 This copies a file from C<src> to C<dest> where C<dest> is
486 either a destination filename or destination directory.
488 =item $h->cp_a ($src, $dest);
490 This copies a file or directory from C<src> to C<dest>
491 recursively using the C<cp -a> command.
493 =item $result = $h->debug ($subcmd, \@extraargs);
495 The C<$h-E<gt>debug> command exposes some internals of
496 C<guestfsd> (the guestfs daemon) that runs inside the
499 There is no comprehensive help for this command. You have
500 to look at the file C<daemon/debug.c> in the libguestfs source
501 to find out what you can do.
503 =item $output = $h->df ();
505 This command runs the C<df> command to report disk space used.
507 This command is mostly useful for interactive sessions. It
508 is I<not> intended that you try to parse the output string.
509 Use C<statvfs> from programs.
511 =item $output = $h->df_h ();
513 This command runs the C<df -h> command to report disk space used
514 in human-readable format.
516 This command is mostly useful for interactive sessions. It
517 is I<not> intended that you try to parse the output string.
518 Use C<statvfs> from programs.
520 =item $kmsgs = $h->dmesg ();
522 This returns the kernel messages (C<dmesg> output) from
523 the guest kernel. This is sometimes useful for extended
524 debugging of problems.
526 Another way to get the same information is to enable
527 verbose messages with C<$h-E<gt>set_verbose> or by setting
528 the environment variable C<LIBGUESTFS_DEBUG=1> before
531 =item $h->download ($remotefilename, $filename);
533 Download file C<remotefilename> and save it as C<filename>
534 on the local machine.
536 C<filename> can also be a named pipe.
538 See also C<$h-E<gt>upload>, C<$h-E<gt>cat>.
540 =item $h->drop_caches ($whattodrop);
542 This instructs the guest kernel to drop its page cache,
543 and/or dentries and inode caches. The parameter C<whattodrop>
544 tells the kernel what precisely to drop, see
545 L<http://linux-mm.org/Drop_Caches>
547 Setting C<whattodrop> to 3 should drop everything.
549 This automatically calls L<sync(2)> before the operation,
550 so that the maximum guest memory is freed.
552 =item $sizekb = $h->du ($path);
554 This command runs the C<du -s> command to estimate file space
557 C<path> can be a file or a directory. If C<path> is a directory
558 then the estimate includes the contents of the directory and all
559 subdirectories (recursively).
561 The result is the estimated size in I<kilobytes>
562 (ie. units of 1024 bytes).
564 =item $h->e2fsck_f ($device);
566 This runs C<e2fsck -p -f device>, ie. runs the ext2/ext3
567 filesystem checker on C<device>, noninteractively (C<-p>),
568 even if the filesystem appears to be clean (C<-f>).
570 This command is only needed because of C<$h-E<gt>resize2fs>
571 (q.v.). Normally you should use C<$h-E<gt>fsck>.
573 =item $h->end_busy ();
575 This sets the state to C<READY>, or if in C<CONFIG> then it leaves the
576 state as is. This is only used when implementing
577 actions using the low-level API.
579 For more information on states, see L<guestfs(3)>.
581 =item $equality = $h->equal ($file1, $file2);
583 This compares the two files C<file1> and C<file2> and returns
584 true if their content is exactly equal, or false otherwise.
586 The external L<cmp(1)> program is used for the comparison.
588 =item $existsflag = $h->exists ($path);
590 This returns C<true> if and only if there is a file, directory
591 (or anything) with the given C<path> name.
593 See also C<$h-E<gt>is_file>, C<$h-E<gt>is_dir>, C<$h-E<gt>stat>.
595 =item $description = $h->file ($path);
597 This call uses the standard L<file(1)> command to determine
598 the type or contents of the file. This also works on devices,
599 for example to find out whether a partition contains a filesystem.
601 The exact command which runs is C<file -bsL path>. Note in
602 particular that the filename is not prepended to the output
605 =item @names = $h->find ($directory);
607 This command lists out all files and directories, recursively,
608 starting at C<directory>. It is essentially equivalent to
609 running the shell command C<find directory -print> but some
610 post-processing happens on the output, described below.
612 This returns a list of strings I<without any prefix>. Thus
613 if the directory structure was:
619 then the returned list from C<$h-E<gt>find> C</tmp> would be
627 If C<directory> is not a directory, then this command returns
630 The returned list is sorted.
632 =item $status = $h->fsck ($fstype, $device);
634 This runs the filesystem checker (fsck) on C<device> which
635 should have filesystem type C<fstype>.
637 The returned integer is the status. See L<fsck(8)> for the
638 list of status codes from C<fsck>.
646 Multiple status codes can be summed together.
650 A non-zero return code can mean "success", for example if
651 errors have been corrected on the filesystem.
655 Checking or repairing NTFS volumes is not supported
660 This command is entirely equivalent to running C<fsck -a -t fstype device>.
662 =item $append = $h->get_append ();
664 Return the additional kernel options which are added to the
665 guest kernel command line.
667 If C<NULL> then no options are added.
669 =item $autosync = $h->get_autosync ();
671 Get the autosync flag.
673 =item $label = $h->get_e2label ($device);
675 This returns the ext2/3/4 filesystem label of the filesystem on
678 =item $uuid = $h->get_e2uuid ($device);
680 This returns the ext2/3/4 filesystem UUID of the filesystem on
683 =item $path = $h->get_path ();
685 Return the current search path.
687 This is always non-NULL. If it wasn't set already, then this will
688 return the default path.
690 =item $qemu = $h->get_qemu ();
692 Return the current qemu binary.
694 This is always non-NULL. If it wasn't set already, then this will
695 return the default qemu binary name.
697 =item $state = $h->get_state ();
699 This returns the current state as an opaque integer. This is
700 only useful for printing debug and internal error messages.
702 For more information on states, see L<guestfs(3)>.
704 =item $verbose = $h->get_verbose ();
706 This returns the verbose messages flag.
708 =item @paths = $h->glob_expand ($pattern);
710 This command searches for all the pathnames matching
711 C<pattern> according to the wildcard expansion rules
714 If no paths match, then this returns an empty list
715 (note: not an error).
717 It is just a wrapper around the C L<glob(3)> function
718 with flags C<GLOB_MARK|GLOB_BRACE>.
719 See that manual page for more details.
721 =item $h->grub_install ($root, $device);
723 This command installs GRUB (the Grand Unified Bootloader) on
724 C<device>, with the root directory being C<root>.
726 =item @lines = $h->head ($path);
728 This command returns up to the first 10 lines of a file as
731 Because of the message protocol, there is a transfer limit
732 of somewhere between 2MB and 4MB. To transfer large files you should use
735 =item @lines = $h->head_n ($nrlines, $path);
737 If the parameter C<nrlines> is a positive number, this returns the first
738 C<nrlines> lines of the file C<path>.
740 If the parameter C<nrlines> is a negative number, this returns lines
741 from the file C<path>, excluding the last C<nrlines> lines.
743 If the parameter C<nrlines> is zero, this returns an empty list.
745 Because of the message protocol, there is a transfer limit
746 of somewhere between 2MB and 4MB. To transfer large files you should use
749 =item $dump = $h->hexdump ($path);
751 This runs C<hexdump -C> on the given C<path>. The result is
752 the human-readable, canonical hex dump of the file.
754 Because of the message protocol, there is a transfer limit
755 of somewhere between 2MB and 4MB. To transfer large files you should use
758 =item @filenames = $h->initrd_list ($path);
760 This command lists out files contained in an initrd.
762 The files are listed without any initial C</> character. The
763 files are listed in the order they appear (not necessarily
764 alphabetical). Directory names are listed as separate items.
766 Old Linux kernels (2.4 and earlier) used a compressed ext2
767 filesystem as initrd. We I<only> support the newer initramfs
768 format (compressed cpio files).
770 =item $busy = $h->is_busy ();
772 This returns true iff this handle is busy processing a command
773 (in the C<BUSY> state).
775 For more information on states, see L<guestfs(3)>.
777 =item $config = $h->is_config ();
779 This returns true iff this handle is being configured
780 (in the C<CONFIG> state).
782 For more information on states, see L<guestfs(3)>.
784 =item $dirflag = $h->is_dir ($path);
786 This returns C<true> if and only if there is a directory
787 with the given C<path> name. Note that it returns false for
788 other objects like files.
790 See also C<$h-E<gt>stat>.
792 =item $fileflag = $h->is_file ($path);
794 This returns C<true> if and only if there is a file
795 with the given C<path> name. Note that it returns false for
796 other objects like directories.
798 See also C<$h-E<gt>stat>.
800 =item $launching = $h->is_launching ();
802 This returns true iff this handle is launching the subprocess
803 (in the C<LAUNCHING> state).
805 For more information on states, see L<guestfs(3)>.
807 =item $ready = $h->is_ready ();
809 This returns true iff this handle is ready to accept commands
810 (in the C<READY> state).
812 For more information on states, see L<guestfs(3)>.
814 =item $h->kill_subprocess ();
816 This kills the qemu subprocess. You should never need to call this.
820 Internally libguestfs is implemented by running a virtual machine
823 You should call this after configuring the handle
824 (eg. adding drives) but before performing any actions.
826 =item @devices = $h->list_devices ();
828 List all the block devices.
830 The full block device names are returned, eg. C</dev/sda>
832 =item @partitions = $h->list_partitions ();
834 List all the partitions detected on all block devices.
836 The full partition device names are returned, eg. C</dev/sda1>
838 This does not return logical volumes. For that you will need to
841 =item $listing = $h->ll ($directory);
843 List the files in C<directory> (relative to the root directory,
844 there is no cwd) in the format of 'ls -la'.
846 This command is mostly useful for interactive sessions. It
847 is I<not> intended that you try to parse the output string.
849 =item @listing = $h->ls ($directory);
851 List the files in C<directory> (relative to the root directory,
852 there is no cwd). The '.' and '..' entries are not returned, but
853 hidden files are shown.
855 This command is mostly useful for interactive sessions. Programs
856 should probably use C<$h-E<gt>readdir> instead.
858 =item %statbuf = $h->lstat ($path);
860 Returns file information for the given C<path>.
862 This is the same as C<$h-E<gt>stat> except that if C<path>
863 is a symbolic link, then the link is stat-ed, not the file it
866 This is the same as the C<lstat(2)> system call.
868 =item $h->lvcreate ($logvol, $volgroup, $mbytes);
870 This creates an LVM volume group called C<logvol>
871 on the volume group C<volgroup>, with C<size> megabytes.
873 =item $h->lvm_remove_all ();
875 This command removes all LVM logical volumes, volume groups
876 and physical volumes.
878 B<This command is dangerous. Without careful use you
879 can easily destroy all your data>.
881 =item $h->lvremove ($device);
883 Remove an LVM logical volume C<device>, where C<device> is
884 the path to the LV, such as C</dev/VG/LV>.
886 You can also remove all LVs in a volume group by specifying
887 the VG name, C</dev/VG>.
889 =item $h->lvresize ($device, $mbytes);
891 This resizes (expands or shrinks) an existing LVM logical
892 volume to C<mbytes>. When reducing, data in the reduced part
895 =item @logvols = $h->lvs ();
897 List all the logical volumes detected. This is the equivalent
898 of the L<lvs(8)> command.
900 This returns a list of the logical volume device names
901 (eg. C</dev/VolGroup00/LogVol00>).
903 See also C<$h-E<gt>lvs_full>.
905 =item @logvols = $h->lvs_full ();
907 List all the logical volumes detected. This is the equivalent
908 of the L<lvs(8)> command. The "full" version includes all fields.
910 =item $h->mkdir ($path);
912 Create a directory named C<path>.
914 =item $h->mkdir_p ($path);
916 Create a directory named C<path>, creating any parent directories
917 as necessary. This is like the C<mkdir -p> shell command.
919 =item $dir = $h->mkdtemp ($template);
921 This command creates a temporary directory. The
922 C<template> parameter should be a full pathname for the
923 temporary directory name with the final six characters being
926 For example: "/tmp/myprogXXXXXX" or "/Temp/myprogXXXXXX",
927 the second one being suitable for Windows filesystems.
929 The name of the temporary directory that was created
932 The temporary directory is created with mode 0700
933 and is owned by root.
935 The caller is responsible for deleting the temporary
936 directory and its contents after use.
938 See also: L<mkdtemp(3)>
940 =item $h->mkfs ($fstype, $device);
942 This creates a filesystem on C<device> (usually a partition
943 or LVM logical volume). The filesystem type is C<fstype>, for
946 =item $h->mkswap ($device);
948 Create a swap partition on C<device>.
950 =item $h->mkswap_L ($label, $device);
952 Create a swap partition on C<device> with label C<label>.
954 =item $h->mkswap_U ($uuid, $device);
956 Create a swap partition on C<device> with UUID C<uuid>.
958 =item $h->mount ($device, $mountpoint);
960 Mount a guest disk at a position in the filesystem. Block devices
961 are named C</dev/sda>, C</dev/sdb> and so on, as they were added to
962 the guest. If those block devices contain partitions, they will have
963 the usual names (eg. C</dev/sda1>). Also LVM C</dev/VG/LV>-style
966 The rules are the same as for L<mount(2)>: A filesystem must
967 first be mounted on C</> before others can be mounted. Other
968 filesystems can only be mounted on directories which already
971 The mounted filesystem is writable, if we have sufficient permissions
972 on the underlying device.
974 The filesystem options C<sync> and C<noatime> are set with this
975 call, in order to improve reliability.
977 =item $h->mount_loop ($file, $mountpoint);
979 This command lets you mount C<file> (a filesystem image
980 in a file) on a mount point. It is entirely equivalent to
981 the command C<mount -o loop file mountpoint>.
983 =item $h->mount_options ($options, $device, $mountpoint);
985 This is the same as the C<$h-E<gt>mount> command, but it
986 allows you to set the mount options as for the
987 L<mount(8)> I<-o> flag.
989 =item $h->mount_ro ($device, $mountpoint);
991 This is the same as the C<$h-E<gt>mount> command, but it
992 mounts the filesystem with the read-only (I<-o ro>) flag.
994 =item $h->mount_vfs ($options, $vfstype, $device, $mountpoint);
996 This is the same as the C<$h-E<gt>mount> command, but it
997 allows you to set both the mount options and the vfstype
998 as for the L<mount(8)> I<-o> and I<-t> flags.
1000 =item @devices = $h->mounts ();
1002 This returns the list of currently mounted filesystems. It returns
1003 the list of devices (eg. C</dev/sda1>, C</dev/VG/LV>).
1005 Some internal mounts are not shown.
1007 =item $h->mv ($src, $dest);
1009 This moves a file from C<src> to C<dest> where C<dest> is
1010 either a destination filename or destination directory.
1012 =item $status = $h->ntfs_3g_probe ($rw, $device);
1014 This command runs the L<ntfs-3g.probe(8)> command which probes
1015 an NTFS C<device> for mountability. (Not all NTFS volumes can
1016 be mounted read-write, and some cannot be mounted at all).
1018 C<rw> is a boolean flag. Set it to true if you want to test
1019 if the volume can be mounted read-write. Set it to false if
1020 you want to test if the volume can be mounted read-only.
1022 The return value is an integer which C<0> if the operation
1023 would succeed, or some non-zero value documented in the
1024 L<ntfs-3g.probe(8)> manual page.
1026 =item $h->ping_daemon ();
1028 This is a test probe into the guestfs daemon running inside
1029 the qemu subprocess. Calling this function checks that the
1030 daemon responds to the ping message, without affecting the daemon
1031 or attached block device(s) in any other way.
1033 =item $h->pvcreate ($device);
1035 This creates an LVM physical volume on the named C<device>,
1036 where C<device> should usually be a partition name such
1039 =item $h->pvremove ($device);
1041 This wipes a physical volume C<device> so that LVM will no longer
1044 The implementation uses the C<pvremove> command which refuses to
1045 wipe physical volumes that contain any volume groups, so you have
1046 to remove those first.
1048 =item $h->pvresize ($device);
1050 This resizes (expands or shrinks) an existing LVM physical
1051 volume to match the new size of the underlying device.
1053 =item @physvols = $h->pvs ();
1055 List all the physical volumes detected. This is the equivalent
1056 of the L<pvs(8)> command.
1058 This returns a list of just the device names that contain
1059 PVs (eg. C</dev/sda2>).
1061 See also C<$h-E<gt>pvs_full>.
1063 =item @physvols = $h->pvs_full ();
1065 List all the physical volumes detected. This is the equivalent
1066 of the L<pvs(8)> command. The "full" version includes all fields.
1068 =item @lines = $h->read_lines ($path);
1070 Return the contents of the file named C<path>.
1072 The file contents are returned as a list of lines. Trailing
1073 C<LF> and C<CRLF> character sequences are I<not> returned.
1075 Note that this function cannot correctly handle binary files
1076 (specifically, files containing C<\0> character which is treated
1077 as end of line). For those you need to use the C<$h-E<gt>read_file>
1078 function which has a more complex interface.
1080 =item $h->resize2fs ($device);
1082 This resizes an ext2 or ext3 filesystem to match the size of
1083 the underlying device.
1085 I<Note:> It is sometimes required that you run C<$h-E<gt>e2fsck_f>
1086 on the C<device> before calling this command. For unknown reasons
1087 C<resize2fs> sometimes gives an error about this and sometimes not.
1088 In any case, it is always safe to call C<$h-E<gt>e2fsck_f> before
1089 calling this function.
1091 =item $h->rm ($path);
1093 Remove the single file C<path>.
1095 =item $h->rm_rf ($path);
1097 Remove the file or directory C<path>, recursively removing the
1098 contents if its a directory. This is like the C<rm -rf> shell
1101 =item $h->rmdir ($path);
1103 Remove the single directory C<path>.
1105 =item $h->scrub_device ($device);
1107 This command writes patterns over C<device> to make data retrieval
1110 It is an interface to the L<scrub(1)> program. See that
1111 manual page for more details.
1113 B<This command is dangerous. Without careful use you
1114 can easily destroy all your data>.
1116 =item $h->scrub_file ($file);
1118 This command writes patterns over a file to make data retrieval
1121 The file is I<removed> after scrubbing.
1123 It is an interface to the L<scrub(1)> program. See that
1124 manual page for more details.
1126 =item $h->scrub_freespace ($dir);
1128 This command creates the directory C<dir> and then fills it
1129 with files until the filesystem is full, and scrubs the files
1130 as for C<$h-E<gt>scrub_file>, and deletes them.
1131 The intention is to scrub any free space on the partition
1134 It is an interface to the L<scrub(1)> program. See that
1135 manual page for more details.
1137 =item $h->set_append ($append);
1139 This function is used to add additional options to the
1140 guest kernel command line.
1142 The default is C<NULL> unless overridden by setting
1143 C<LIBGUESTFS_APPEND> environment variable.
1145 Setting C<append> to C<NULL> means I<no> additional options
1146 are passed (libguestfs always adds a few of its own).
1148 =item $h->set_autosync ($autosync);
1150 If C<autosync> is true, this enables autosync. Libguestfs will make a
1151 best effort attempt to run C<$h-E<gt>umount_all> followed by
1152 C<$h-E<gt>sync> when the handle is closed
1153 (also if the program exits without closing handles).
1155 This is disabled by default (except in guestfish where it is
1156 enabled by default).
1158 =item $h->set_busy ();
1160 This sets the state to C<BUSY>. This is only used when implementing
1161 actions using the low-level API.
1163 For more information on states, see L<guestfs(3)>.
1165 =item $h->set_e2label ($device, $label);
1167 This sets the ext2/3/4 filesystem label of the filesystem on
1168 C<device> to C<label>. Filesystem labels are limited to
1171 You can use either C<$h-E<gt>tune2fs_l> or C<$h-E<gt>get_e2label>
1172 to return the existing label on a filesystem.
1174 =item $h->set_e2uuid ($device, $uuid);
1176 This sets the ext2/3/4 filesystem UUID of the filesystem on
1177 C<device> to C<uuid>. The format of the UUID and alternatives
1178 such as C<clear>, C<random> and C<time> are described in the
1179 L<tune2fs(8)> manpage.
1181 You can use either C<$h-E<gt>tune2fs_l> or C<$h-E<gt>get_e2uuid>
1182 to return the existing UUID of a filesystem.
1184 =item $h->set_path ($path);
1186 Set the path that libguestfs searches for kernel and initrd.img.
1188 The default is C<$libdir/guestfs> unless overridden by setting
1189 C<LIBGUESTFS_PATH> environment variable.
1191 Setting C<path> to C<NULL> restores the default path.
1193 =item $h->set_qemu ($qemu);
1195 Set the qemu binary that we will use.
1197 The default is chosen when the library was compiled by the
1200 You can also override this by setting the C<LIBGUESTFS_QEMU>
1201 environment variable.
1203 Setting C<qemu> to C<NULL> restores the default qemu binary.
1205 =item $h->set_ready ();
1207 This sets the state to C<READY>. This is only used when implementing
1208 actions using the low-level API.
1210 For more information on states, see L<guestfs(3)>.
1212 =item $h->set_verbose ($verbose);
1214 If C<verbose> is true, this turns on verbose messages (to C<stderr>).
1216 Verbose messages are disabled unless the environment variable
1217 C<LIBGUESTFS_DEBUG> is defined and set to C<1>.
1219 =item $h->sfdisk ($device, $cyls, $heads, $sectors, \@lines);
1221 This is a direct interface to the L<sfdisk(8)> program for creating
1222 partitions on block devices.
1224 C<device> should be a block device, for example C</dev/sda>.
1226 C<cyls>, C<heads> and C<sectors> are the number of cylinders, heads
1227 and sectors on the device, which are passed directly to sfdisk as
1228 the I<-C>, I<-H> and I<-S> parameters. If you pass C<0> for any
1229 of these, then the corresponding parameter is omitted. Usually for
1230 'large' disks, you can just pass C<0> for these, but for small
1231 (floppy-sized) disks, sfdisk (or rather, the kernel) cannot work
1232 out the right geometry and you will need to tell it.
1234 C<lines> is a list of lines that we feed to C<sfdisk>. For more
1235 information refer to the L<sfdisk(8)> manpage.
1237 To create a single partition occupying the whole disk, you would
1238 pass C<lines> as a single element list, when the single element being
1239 the string C<,> (comma).
1241 See also: C<$h-E<gt>sfdisk_l>, C<$h-E<gt>sfdisk_N>
1243 B<This command is dangerous. Without careful use you
1244 can easily destroy all your data>.
1246 =item $h->sfdisk_N ($device, $partnum, $cyls, $heads, $sectors, $line);
1248 This runs L<sfdisk(8)> option to modify just the single
1249 partition C<n> (note: C<n> counts from 1).
1251 For other parameters, see C<$h-E<gt>sfdisk>. You should usually
1252 pass C<0> for the cyls/heads/sectors parameters.
1254 B<This command is dangerous. Without careful use you
1255 can easily destroy all your data>.
1257 =item $partitions = $h->sfdisk_disk_geometry ($device);
1259 This displays the disk geometry of C<device> read from the
1260 partition table. Especially in the case where the underlying
1261 block device has been resized, this can be different from the
1262 kernel's idea of the geometry (see C<$h-E<gt>sfdisk_kernel_geometry>).
1264 The result is in human-readable format, and not designed to
1267 =item $partitions = $h->sfdisk_kernel_geometry ($device);
1269 This displays the kernel's idea of the geometry of C<device>.
1271 The result is in human-readable format, and not designed to
1274 =item $partitions = $h->sfdisk_l ($device);
1276 This displays the partition table on C<device>, in the
1277 human-readable output of the L<sfdisk(8)> command. It is
1278 not intended to be parsed.
1280 =item $output = $h->sh ($command);
1282 This call runs a command from the guest filesystem via the
1285 This is like C<$h-E<gt>command>, but passes the command to:
1287 /bin/sh -c "command"
1289 Depending on the guest's shell, this usually results in
1290 wildcards being expanded, shell expressions being interpolated
1293 All the provisos about C<$h-E<gt>command> apply to this call.
1295 =item @lines = $h->sh_lines ($command);
1297 This is the same as C<$h-E<gt>sh>, but splits the result
1298 into a list of lines.
1300 See also: C<$h-E<gt>command_lines>
1302 =item $h->sleep ($secs);
1304 Sleep for C<secs> seconds.
1306 =item %statbuf = $h->stat ($path);
1308 Returns file information for the given C<path>.
1310 This is the same as the C<stat(2)> system call.
1312 =item %statbuf = $h->statvfs ($path);
1314 Returns file system statistics for any mounted file system.
1315 C<path> should be a file or directory in the mounted file system
1316 (typically it is the mount point itself, but it doesn't need to be).
1318 This is the same as the C<statvfs(2)> system call.
1320 =item @stringsout = $h->strings ($path);
1322 This runs the L<strings(1)> command on a file and returns
1323 the list of printable strings found.
1325 Because of the message protocol, there is a transfer limit
1326 of somewhere between 2MB and 4MB. To transfer large files you should use
1329 =item @stringsout = $h->strings_e ($encoding, $path);
1331 This is like the C<$h-E<gt>strings> command, but allows you to
1332 specify the encoding.
1334 See the L<strings(1)> manpage for the full list of encodings.
1336 Commonly useful encodings are C<l> (lower case L) which will
1337 show strings inside Windows/x86 files.
1339 The returned strings are transcoded to UTF-8.
1341 Because of the message protocol, there is a transfer limit
1342 of somewhere between 2MB and 4MB. To transfer large files you should use
1347 This syncs the disk, so that any writes are flushed through to the
1348 underlying disk image.
1350 You should always call this if you have modified a disk image, before
1353 =item @lines = $h->tail ($path);
1355 This command returns up to the last 10 lines of a file as
1358 Because of the message protocol, there is a transfer limit
1359 of somewhere between 2MB and 4MB. To transfer large files you should use
1362 =item @lines = $h->tail_n ($nrlines, $path);
1364 If the parameter C<nrlines> is a positive number, this returns the last
1365 C<nrlines> lines of the file C<path>.
1367 If the parameter C<nrlines> is a negative number, this returns lines
1368 from the file C<path>, starting with the C<-nrlines>th line.
1370 If the parameter C<nrlines> is zero, this returns an empty list.
1372 Because of the message protocol, there is a transfer limit
1373 of somewhere between 2MB and 4MB. To transfer large files you should use
1376 =item $h->tar_in ($tarfile, $directory);
1378 This command uploads and unpacks local file C<tarfile> (an
1379 I<uncompressed> tar file) into C<directory>.
1381 To upload a compressed tarball, use C<$h-E<gt>tgz_in>.
1383 =item $h->tar_out ($directory, $tarfile);
1385 This command packs the contents of C<directory> and downloads
1386 it to local file C<tarfile>.
1388 To download a compressed tarball, use C<$h-E<gt>tgz_out>.
1390 =item $h->tgz_in ($tarball, $directory);
1392 This command uploads and unpacks local file C<tarball> (a
1393 I<gzip compressed> tar file) into C<directory>.
1395 To upload an uncompressed tarball, use C<$h-E<gt>tar_in>.
1397 =item $h->tgz_out ($directory, $tarball);
1399 This command packs the contents of C<directory> and downloads
1400 it to local file C<tarball>.
1402 To download an uncompressed tarball, use C<$h-E<gt>tar_out>.
1404 =item $h->touch ($path);
1406 Touch acts like the L<touch(1)> command. It can be used to
1407 update the timestamps on a file, or, if the file does not exist,
1408 to create a new zero-length file.
1410 =item %superblock = $h->tune2fs_l ($device);
1412 This returns the contents of the ext2, ext3 or ext4 filesystem
1413 superblock on C<device>.
1415 It is the same as running C<tune2fs -l device>. See L<tune2fs(8)>
1416 manpage for more details. The list of fields returned isn't
1417 clearly defined, and depends on both the version of C<tune2fs>
1418 that libguestfs was built against, and the filesystem itself.
1420 =item $h->umount ($pathordevice);
1422 This unmounts the given filesystem. The filesystem may be
1423 specified either by its mountpoint (path) or the device which
1424 contains the filesystem.
1426 =item $h->umount_all ();
1428 This unmounts all mounted filesystems.
1430 Some internal mounts are not unmounted by this call.
1432 =item $h->upload ($filename, $remotefilename);
1434 Upload local file C<filename> to C<remotefilename> on the
1437 C<filename> can also be a named pipe.
1439 See also C<$h-E<gt>download>.
1441 =item $h->vg_activate ($activate, \@volgroups);
1443 This command activates or (if C<activate> is false) deactivates
1444 all logical volumes in the listed volume groups C<volgroups>.
1445 If activated, then they are made known to the
1446 kernel, ie. they appear as C</dev/mapper> devices. If deactivated,
1447 then those devices disappear.
1449 This command is the same as running C<vgchange -a y|n volgroups...>
1451 Note that if C<volgroups> is an empty list then B<all> volume groups
1452 are activated or deactivated.
1454 =item $h->vg_activate_all ($activate);
1456 This command activates or (if C<activate> is false) deactivates
1457 all logical volumes in all volume groups.
1458 If activated, then they are made known to the
1459 kernel, ie. they appear as C</dev/mapper> devices. If deactivated,
1460 then those devices disappear.
1462 This command is the same as running C<vgchange -a y|n>
1464 =item $h->vgcreate ($volgroup, \@physvols);
1466 This creates an LVM volume group called C<volgroup>
1467 from the non-empty list of physical volumes C<physvols>.
1469 =item $h->vgremove ($vgname);
1471 Remove an LVM volume group C<vgname>, (for example C<VG>).
1473 This also forcibly removes all logical volumes in the volume
1476 =item @volgroups = $h->vgs ();
1478 List all the volumes groups detected. This is the equivalent
1479 of the L<vgs(8)> command.
1481 This returns a list of just the volume group names that were
1482 detected (eg. C<VolGroup00>).
1484 See also C<$h-E<gt>vgs_full>.
1486 =item @volgroups = $h->vgs_full ();
1488 List all the volumes groups detected. This is the equivalent
1489 of the L<vgs(8)> command. The "full" version includes all fields.
1491 =item $h->wait_ready ();
1493 Internally libguestfs is implemented by running a virtual machine
1496 You should call this after C<$h-E<gt>launch> to wait for the launch
1499 =item $chars = $h->wc_c ($path);
1501 This command counts the characters in a file, using the
1502 C<wc -c> external command.
1504 =item $lines = $h->wc_l ($path);
1506 This command counts the lines in a file, using the
1507 C<wc -l> external command.
1509 =item $words = $h->wc_w ($path);
1511 This command counts the words in a file, using the
1512 C<wc -w> external command.
1514 =item $h->write_file ($path, $content, $size);
1516 This call creates a file called C<path>. The contents of the
1517 file is the string C<content> (which can contain any 8 bit data),
1518 with length C<size>.
1520 As a special case, if C<size> is C<0>
1521 then the length is calculated using C<strlen> (so in this case
1522 the content cannot contain embedded ASCII NULs).
1524 I<NB.> Owing to a bug, writing content containing ASCII NUL
1525 characters does I<not> work, even if the length is specified.
1526 We hope to resolve this bug in a future version. In the meantime
1527 use C<$h-E<gt>upload>.
1529 Because of the message protocol, there is a transfer limit
1530 of somewhere between 2MB and 4MB. To transfer large files you should use
1533 =item $h->zero ($device);
1535 This command writes zeroes over the first few blocks of C<device>.
1537 How many blocks are zeroed isn't specified (but it's I<not> enough
1538 to securely wipe the device). It should be sufficient to remove
1539 any partition tables, filesystem superblocks and so on.
1541 See also: C<$h-E<gt>scrub_device>.
1543 =item $h->zerofree ($device);
1545 This runs the I<zerofree> program on C<device>. This program
1546 claims to zero unused inodes and disk blocks on an ext2/3
1547 filesystem, thus making it possible to compress the filesystem
1550 You should B<not> run this program if the filesystem is
1553 It is possible that using this program can damage the filesystem
1554 or data on the filesystem.
1564 Copyright (C) 2009 Red Hat Inc.
1568 Please see the file COPYING.LIB for the full license.
1572 L<guestfs(3)>, L<guestfish(1)>.