blob: 55c6be391d15c60fa88933517b3dbc46a30d63b1 [file] [log] [blame]
bellardacd935e2004-11-15 22:57:26 +00001@example
2@c man begin SYNOPSIS
3usage: qemu-img command [command options]
4@c man end
5@end example
6
Kevin Wolf48467322012-08-16 10:56:35 +02007@c man begin DESCRIPTION
8qemu-img allows you to create, convert and modify images offline. It can handle
9all image formats supported by QEMU.
10
11@b{Warning:} Never use qemu-img to modify images in use by a running virtual
12machine or any other process; this may destroy the image. Also, be aware that
13querying an image that is being modified by another process may encounter
14inconsistent state.
15@c man end
16
bellardacd935e2004-11-15 22:57:26 +000017@c man begin OPTIONS
18
19The following commands are supported:
Stuart Brady153859b2009-06-07 00:42:17 +010020
21@include qemu-img-cmds.texi
bellardacd935e2004-11-15 22:57:26 +000022
23Command parameters:
24@table @var
25@item filename
26 is a disk image filename
ths5fafdf22007-09-16 21:08:06 +000027@item fmt
Kevin Wolff932c042009-10-28 12:49:15 +010028is the disk image format. It is guessed automatically in most cases. See below
29for a description of the supported disk formats.
bellardacd935e2004-11-15 22:57:26 +000030
Kashyap Chamarthye5357562012-10-18 11:25:34 +053031@item --backing-chain
32will enumerate information about backing files in a disk image chain. Refer
33below for further description.
34
ths5fafdf22007-09-16 21:08:06 +000035@item size
Kevin Wolfeff44262009-06-04 15:39:39 +020036is the disk image size in bytes. Optional suffixes @code{k} or @code{K}
37(kilobyte, 1024) @code{M} (megabyte, 1024k) and @code{G} (gigabyte, 1024M)
38and T (terabyte, 1024G) are supported. @code{b} is ignored.
bellardacd935e2004-11-15 22:57:26 +000039
40@item output_filename
ths5fafdf22007-09-16 21:08:06 +000041is the destination disk image filename
bellardacd935e2004-11-15 22:57:26 +000042
43@item output_fmt
44 is the destination format
Kevin Wolfeff44262009-06-04 15:39:39 +020045@item options
46is a comma separated list of format specific options in a
47name=value format. Use @code{-o ?} for an overview of the options supported
Kevin Wolf3e032362009-10-28 12:49:17 +010048by the used format or see the format descriptions below for details.
Wenchao Xiaef806542013-12-04 17:10:57 +080049@item snapshot_param
50is param used for internal snapshot, format is
51'snapshot.id=[ID],snapshot.name=[NAME]' or '[ID_OR_NAME]'
52@item snapshot_id_or_name
53is deprecated, use snapshot_param instead
bellardacd935e2004-11-15 22:57:26 +000054
55@item -c
56indicates that target image must be compressed (qcow format only)
blueswir1d2c639d2009-01-24 18:19:25 +000057@item -h
58with or without a command shows help and lists the supported formats
Jes Sorensenaaf55b42011-07-19 15:01:34 +020059@item -p
Kevin Wolf0e3bd992014-01-20 15:12:16 +010060display progress bar (compare, convert and rebase commands only).
61If the @var{-p} option is not used for a command that supports it, the
62progress is reported when the process receives a @code{SIGUSR1} signal.
Miroslav Rezaninaf382d432013-02-13 09:09:40 +010063@item -q
64Quiet mode - do not print any output (except errors). There's no progress bar
65in case both @var{-q} and @var{-p} options are used.
Kevin Wolfa22f1232011-08-26 15:27:13 +020066@item -S @var{size}
67indicates the consecutive number of bytes that must contain only zeros
68for qemu-img to create a sparse image during conversion. This value is rounded
69down to the nearest 512 bytes. You may use the common size suffixes like
70@code{k} for kilobytes.
Kevin Wolf3763f262011-12-07 13:57:13 +010071@item -t @var{cache}
72specifies the cache mode that should be used with the (destination) file. See
73the documentation of the emulator's @code{-drive cache=...} option for allowed
74values.
Max Reitz40055952014-07-22 22:58:42 +020075@item -T @var{src_cache}
Stefan Hajnoczibb87fdf2014-09-02 11:01:02 +010076specifies the cache mode that should be used with the source file(s). See
77the documentation of the emulator's @code{-drive cache=...} option for allowed
78values.
blueswir1d2c639d2009-01-24 18:19:25 +000079@end table
80
81Parameters to snapshot subcommand:
82
83@table @option
84
85@item snapshot
86is the name of the snapshot to create, apply or delete
87@item -a
88applies a snapshot (revert disk to saved state)
89@item -c
90creates a snapshot
91@item -d
92deletes a snapshot
93@item -l
94lists all snapshots in the given image
bellardacd935e2004-11-15 22:57:26 +000095@end table
96
Miroslav Rezaninad14ed182013-02-13 09:09:41 +010097Parameters to compare subcommand:
98
99@table @option
100
101@item -f
102First image format
103@item -F
104Second image format
105@item -s
Daniel P. Berrangeb6af0972015-08-26 12:17:13 +0100106Strict mode - fail on different image size or sector allocation
Miroslav Rezaninad14ed182013-02-13 09:09:41 +0100107@end table
108
Alexandre Derumierb2e10492013-09-02 19:07:24 +0100109Parameters to convert subcommand:
110
111@table @option
112
113@item -n
114Skip the creation of the target volume
115@end table
116
bellardacd935e2004-11-15 22:57:26 +0000117Command description:
118
119@table @option
Max Reitz40055952014-07-22 22:58:42 +0200120@item check [-f @var{fmt}] [--output=@var{ofmt}] [-r [leaks | all]] [-T @var{src_cache}] @var{filename}
Kevin Wolfe6184692011-01-17 15:35:28 +0100121
Federico Simoncelli8599ea42013-01-28 06:59:47 -0500122Perform a consistency check on the disk image @var{filename}. The command can
123output in the format @var{ofmt} which is either @code{human} or @code{json}.
Kevin Wolfe6184692011-01-17 15:35:28 +0100124
Kevin Wolf4534ff52012-05-11 16:07:02 +0200125If @code{-r} is specified, qemu-img tries to repair any inconsistencies found
126during the check. @code{-r leaks} repairs only cluster leaks, whereas
127@code{-r all} fixes all kinds of errors, with a higher risk of choosing the
Stefan Weil0546b8c2012-08-10 22:03:25 +0200128wrong fix or hiding corruption that has already occurred.
Kevin Wolf4534ff52012-05-11 16:07:02 +0200129
Kevin Wolfe6184692011-01-17 15:35:28 +0100130Only the formats @code{qcow2}, @code{qed} and @code{vdi} support
131consistency checks.
132
Max Reitzd6635c42014-06-02 22:15:21 +0200133In case the image does not have any inconsistencies, check exits with @code{0}.
134Other exit codes indicate the kind of inconsistency found or if another error
135occurred. The following table summarizes all exit codes of the check subcommand:
136
137@table @option
138
139@item 0
140Check completed, the image is (now) consistent
141@item 1
142Check not completed because of internal errors
143@item 2
144Check completed, image is corrupted
145@item 3
146Check completed, image has leaked clusters, but is not corrupted
147@item 63
148Checks are not supported by the image format
149
150@end table
151
152If @code{-r} is specified, exit codes representing the image state refer to the
153state after (the attempt at) repairing it. That is, a successful @code{-r all}
154will yield the exit code 0, independently of the image state before.
155
Kevin Wolf8063d0f2009-10-28 12:49:16 +0100156@item create [-f @var{fmt}] [-o @var{options}] @var{filename} [@var{size}]
bellardacd935e2004-11-15 22:57:26 +0000157
158Create the new disk image @var{filename} of size @var{size} and format
Kevin Wolf8063d0f2009-10-28 12:49:16 +0100159@var{fmt}. Depending on the file format, you can add one or more @var{options}
160that enable additional features of this format.
bellardacd935e2004-11-15 22:57:26 +0000161
Kevin Wolf8063d0f2009-10-28 12:49:16 +0100162If the option @var{backing_file} is specified, then the image will record
163only the differences from @var{backing_file}. No size needs to be specified in
164this case. @var{backing_file} will never be modified unless you use the
165@code{commit} monitor command (or qemu-img commit).
bellardacd935e2004-11-15 22:57:26 +0000166
Kevin Wolfeff44262009-06-04 15:39:39 +0200167The size can also be specified using the @var{size} option with @code{-o},
168it doesn't need to be specified separately in this case.
169
Max Reitz1b22bff2014-10-24 15:57:40 +0200170@item commit [-q] [-f @var{fmt}] [-t @var{cache}] [-b @var{base}] [-d] [-p] @var{filename}
bellardacd935e2004-11-15 22:57:26 +0000171
Jeff Cody37222902014-01-24 09:02:37 -0500172Commit the changes recorded in @var{filename} in its base image or backing file.
173If the backing file is smaller than the snapshot, then the backing file will be
174resized to be the same size as the snapshot. If the snapshot is smaller than
175the backing file, the backing file will not be truncated. If you want the
176backing file to match the size of the smaller snapshot, you can safely truncate
177it yourself once the commit operation successfully completes.
bellardacd935e2004-11-15 22:57:26 +0000178
Max Reitz9a86fe42014-10-24 15:57:38 +0200179The image @var{filename} is emptied after the operation has succeeded. If you do
180not need @var{filename} afterwards and intend to drop it, you may skip emptying
181@var{filename} by specifying the @code{-d} flag.
182
Max Reitz1b22bff2014-10-24 15:57:40 +0200183If the backing chain of the given image file @var{filename} has more than one
184layer, the backing file into which the changes will be committed may be
185specified as @var{base} (which has to be part of @var{filename}'s backing
186chain). If @var{base} is not specified, the immediate backing file of the top
187image (which is @var{filename}) will be used. For reasons of consistency,
188explicitly specifying @var{base} will always imply @code{-d} (since emptying an
189image after committing to an indirect backing file would lead to different data
190being read from the image due to content in the intermediate backing chain
191overruling the commit target).
192
Max Reitz40055952014-07-22 22:58:42 +0200193@item compare [-f @var{fmt}] [-F @var{fmt}] [-T @var{src_cache}] [-p] [-s] [-q] @var{filename1} @var{filename2}
Miroslav Rezaninad14ed182013-02-13 09:09:41 +0100194
195Check if two images have the same content. You can compare images with
196different format or settings.
197
198The format is probed unless you specify it by @var{-f} (used for
199@var{filename1}) and/or @var{-F} (used for @var{filename2}) option.
200
201By default, images with different size are considered identical if the larger
202image contains only unallocated and/or zeroed sectors in the area after the end
203of the other image. In addition, if any sector is not allocated in one image
204and contains only zero bytes in the second one, it is evaluated as equal. You
205can use Strict mode by specifying the @var{-s} option. When compare runs in
206Strict mode, it fails in case image size differs or a sector is allocated in
207one image and is not allocated in the second one.
208
209By default, compare prints out a result message. This message displays
210information that both images are same or the position of the first different
211byte. In addition, result message can report different image size in case
212Strict mode is used.
213
214Compare exits with @code{0} in case the images are equal and with @code{1}
215in case the images differ. Other exit codes mean an error occurred during
216execution and standard error output should contain an error message.
217The following table sumarizes all exit codes of the compare subcommand:
218
219@table @option
220
221@item 0
222Images are identical
223@item 1
224Images differ
225@item 2
226Error on opening an image
227@item 3
228Error on checking a sector allocation
229@item 4
230Error on reading data
231
232@end table
233
Max Reitz40055952014-07-22 22:58:42 +0200234@item convert [-c] [-p] [-n] [-f @var{fmt}] [-t @var{cache}] [-T @var{src_cache}] [-O @var{output_fmt}] [-o @var{options}] [-s @var{snapshot_id_or_name}] [-l @var{snapshot_param}] [-S @var{sparse_size}] @var{filename} [@var{filename2} [...]] @var{output_filename}
bellardacd935e2004-11-15 22:57:26 +0000235
Wenchao Xiaef806542013-12-04 17:10:57 +0800236Convert the disk image @var{filename} or a snapshot @var{snapshot_param}(@var{snapshot_id_or_name} is deprecated)
237to disk image @var{output_filename} using format @var{output_fmt}. It can be optionally compressed (@code{-c}
Kevin Wolfeff44262009-06-04 15:39:39 +0200238option) or use any format specific options like encryption (@code{-o} option).
bellardacd935e2004-11-15 22:57:26 +0000239
Kevin Wolf8063d0f2009-10-28 12:49:16 +0100240Only the formats @code{qcow} and @code{qcow2} support compression. The
bellardacd935e2004-11-15 22:57:26 +0000241compression is read-only. It means that if a compressed sector is
242rewritten, then it is rewritten as uncompressed data.
243
bellardacd935e2004-11-15 22:57:26 +0000244Image conversion is also useful to get smaller image when using a
Stefan Hajnoczi550830f2014-09-16 15:24:24 +0100245growable format such as @code{qcow}: the empty sectors are detected and
246suppressed from the destination image.
bellardacd935e2004-11-15 22:57:26 +0000247
Peter Lieven11b66992013-10-24 12:07:05 +0200248@var{sparse_size} indicates the consecutive number of bytes (defaults to 4k)
249that must contain only zeros for qemu-img to create a sparse image during
250conversion. If @var{sparse_size} is 0, the source will not be scanned for
251unallocated or zero sectors, and the destination image will always be
252fully allocated.
253
Kevin Wolf8063d0f2009-10-28 12:49:16 +0100254You can use the @var{backing_file} option to force the output image to be
255created as a copy on write image of the specified base image; the
256@var{backing_file} should have the same content as the input's base image,
257however the path, image format, etc may differ.
258
Alexandre Derumierb2e10492013-09-02 19:07:24 +0100259If the @code{-n} option is specified, the target volume creation will be
260skipped. This is useful for formats such as @code{rbd} if the target
261volume has already been created with site specific options that cannot
262be supplied through qemu-img.
263
Kashyap Chamarthye5357562012-10-18 11:25:34 +0530264@item info [-f @var{fmt}] [--output=@var{ofmt}] [--backing-chain] @var{filename}
bellardacd935e2004-11-15 22:57:26 +0000265
266Give information about the disk image @var{filename}. Use it in
267particular to know the size reserved on disk which can be different
bellard19d36792006-08-07 21:34:34 +0000268from the displayed size. If VM snapshots are stored in the disk image,
Benoît Canetc054b3f2012-09-05 13:09:02 +0200269they are displayed too. The command can output in the format @var{ofmt}
270which is either @code{human} or @code{json}.
blueswir1d2c639d2009-01-24 18:19:25 +0000271
Kashyap Chamarthye5357562012-10-18 11:25:34 +0530272If a disk image has a backing file chain, information about each disk image in
273the chain can be recursively enumerated by using the option @code{--backing-chain}.
274
275For instance, if you have an image chain like:
276
277@example
278base.qcow2 <- snap1.qcow2 <- snap2.qcow2
279@end example
280
281To enumerate information about each disk image in the above chain, starting from top to base, do:
282
283@example
284qemu-img info --backing-chain snap2.qcow2
285@end example
286
Paolo Bonzinifacd6e22013-09-04 19:00:34 +0200287@item map [-f @var{fmt}] [--output=@var{ofmt}] @var{filename}
288
289Dump the metadata of image @var{filename} and its backing file chain.
290In particular, this commands dumps the allocation state of every sector
291of @var{filename}, together with the topmost file that allocates it in
292the backing file chain.
293
294Two option formats are possible. The default format (@code{human})
295only dumps known-nonzero areas of the file. Known-zero parts of the
296file are omitted altogether, and likewise for parts that are not allocated
297throughout the chain. @command{qemu-img} output will identify a file
298from where the data can be read, and the offset in the file. Each line
299will include four fields, the first three of which are hexadecimal
300numbers. For example the first line of:
301@example
302Offset Length Mapped to File
3030 0x20000 0x50000 /tmp/overlay.qcow2
3040x100000 0x10000 0x95380000 /tmp/backing.qcow2
305@end example
306@noindent
307means that 0x20000 (131072) bytes starting at offset 0 in the image are
308available in /tmp/overlay.qcow2 (opened in @code{raw} format) starting
309at offset 0x50000 (327680). Data that is compressed, encrypted, or
310otherwise not available in raw format will cause an error if @code{human}
311format is in use. Note that file names can include newlines, thus it is
312not safe to parse this output format in scripts.
313
314The alternative format @code{json} will return an array of dictionaries
315in JSON format. It will include similar information in
316the @code{start}, @code{length}, @code{offset} fields;
317it will also include other more specific information:
318@itemize @minus
319@item
320whether the sectors contain actual data or not (boolean field @code{data};
321if false, the sectors are either unallocated or stored as optimized
322all-zero clusters);
323
324@item
325whether the data is known to read as zero (boolean field @code{zero});
326
327@item
328in order to make the output shorter, the target file is expressed as
329a @code{depth}; for example, a depth of 2 refers to the backing file
330of the backing file of @var{filename}.
331@end itemize
332
333In JSON format, the @code{offset} field is optional; it is absent in
334cases where @code{human} format would omit the entry or exit with an error.
335If @code{data} is false and the @code{offset} field is present, the
336corresponding sectors in the file are not yet in use, but they are
337preallocated.
338
339For more information, consult @file{include/block/block.h} in QEMU's
340source code.
341
blueswir1d2c639d2009-01-24 18:19:25 +0000342@item snapshot [-l | -a @var{snapshot} | -c @var{snapshot} | -d @var{snapshot} ] @var{filename}
343
344List, apply, create or delete snapshots in image @var{filename}.
Stefan Hajnocziae6b0ed2010-04-24 09:12:12 +0100345
Max Reitz40055952014-07-22 22:58:42 +0200346@item rebase [-f @var{fmt}] [-t @var{cache}] [-T @var{src_cache}] [-p] [-u] -b @var{backing_file} [-F @var{backing_fmt}] @var{filename}
Kevin Wolfe6184692011-01-17 15:35:28 +0100347
348Changes the backing file of an image. Only the formats @code{qcow2} and
349@code{qed} support changing the backing file.
350
351The backing file is changed to @var{backing_file} and (if the image format of
352@var{filename} supports this) the backing file format is changed to
Alex Bligha6166732012-10-16 13:46:18 +0100353@var{backing_fmt}. If @var{backing_file} is specified as ``'' (the empty
354string), then the image is rebased onto no backing file (i.e. it will exist
355independently of any backing file).
Kevin Wolfe6184692011-01-17 15:35:28 +0100356
Max Reitz40055952014-07-22 22:58:42 +0200357@var{cache} specifies the cache mode to be used for @var{filename}, whereas
Stefan Hajnoczi3ba67962014-09-02 11:01:03 +0100358@var{src_cache} specifies the cache mode for reading backing files.
Max Reitz40055952014-07-22 22:58:42 +0200359
Kevin Wolfe6184692011-01-17 15:35:28 +0100360There are two different modes in which @code{rebase} can operate:
361@table @option
362@item Safe mode
363This is the default mode and performs a real rebase operation. The new backing
364file may differ from the old one and qemu-img rebase will take care of keeping
365the guest-visible content of @var{filename} unchanged.
366
367In order to achieve this, any clusters that differ between @var{backing_file}
368and the old backing file of @var{filename} are merged into @var{filename}
369before actually changing the backing file.
370
371Note that the safe mode is an expensive operation, comparable to converting
372an image. It only works if the old backing file still exists.
373
374@item Unsafe mode
375qemu-img uses the unsafe mode if @code{-u} is specified. In this mode, only the
376backing file name and format of @var{filename} is changed without any checks
377on the file contents. The user must take care of specifying the correct new
378backing file, or the guest-visible content of the image will be corrupted.
379
380This mode is useful for renaming or moving the backing file to somewhere else.
381It can be used without an accessible old backing file, i.e. you can use it to
382fix an image whose backing file has already been moved/renamed.
383@end table
384
Richard W.M. Jones9fda6ab2012-05-21 14:58:05 +0100385You can use @code{rebase} to perform a ``diff'' operation on two
386disk images. This can be useful when you have copied or cloned
387a guest, and you want to get back to a thin image on top of a
388template or base image.
389
390Say that @code{base.img} has been cloned as @code{modified.img} by
391copying it, and that the @code{modified.img} guest has run so there
392are now some changes compared to @code{base.img}. To construct a thin
393image called @code{diff.qcow2} that contains just the differences, do:
394
395@example
396qemu-img create -f qcow2 -b modified.img diff.qcow2
397qemu-img rebase -b base.img diff.qcow2
398@end example
399
400At this point, @code{modified.img} can be discarded, since
401@code{base.img + diff.qcow2} contains the same information.
402
Stefan Hajnocziae6b0ed2010-04-24 09:12:12 +0100403@item resize @var{filename} [+ | -]@var{size}
404
405Change the disk image as if it had been created with @var{size}.
406
407Before using this command to shrink a disk image, you MUST use file system and
408partitioning tools inside the VM to reduce allocated file systems and partition
409sizes accordingly. Failure to do so will result in data loss!
410
411After using this command to grow a disk image, you must use file system and
412partitioning tools inside the VM to actually begin using the new space on the
413device.
Max Reitz6f176b42013-09-03 10:09:50 +0200414
Max Reitz76a3a342014-10-27 11:12:51 +0100415@item amend [-p] [-f @var{fmt}] [-t @var{cache}] -o @var{options} @var{filename}
Max Reitz6f176b42013-09-03 10:09:50 +0200416
417Amends the image format specific @var{options} for the image file
418@var{filename}. Not all file formats support this operation.
bellardacd935e2004-11-15 22:57:26 +0000419@end table
Kevin Wolfd3067b02012-11-21 14:21:47 +0100420@c man end
bellardacd935e2004-11-15 22:57:26 +0000421
Kevin Wolfd3067b02012-11-21 14:21:47 +0100422@ignore
423@c man begin NOTES
Kevin Wolff932c042009-10-28 12:49:15 +0100424Supported image file formats:
425
426@table @option
427@item raw
428
429Raw disk image format (default). This format has the advantage of
430being simple and easily exportable to all other emulators. If your
431file system supports @emph{holes} (for example in ext2 or ext3 on
432Linux or NTFS on Windows), then only the written sectors will reserve
433space. Use @code{qemu-img info} to know the real size used by the
434image or @code{ls -ls} on Unix/Linux.
435
Hu Tao06247422014-09-10 17:05:48 +0800436Supported options:
437@table @code
438@item preallocation
439Preallocation mode (allowed values: @code{off}, @code{falloc}, @code{full}).
440@code{falloc} mode preallocates space for image by calling posix_fallocate().
441@code{full} mode preallocates space for image by writing zeros to underlying
442storage.
443@end table
444
Kevin Wolff932c042009-10-28 12:49:15 +0100445@item qcow2
446QEMU image format, the most versatile format. Use it to have smaller
447images (useful if your filesystem does not supports holes, for example
448on Windows), optional AES encryption, zlib based compression and
449support of multiple VM snapshots.
Kevin Wolf8063d0f2009-10-28 12:49:16 +0100450
Kevin Wolf3e032362009-10-28 12:49:17 +0100451Supported options:
452@table @code
Kevin Wolfd3067b02012-11-21 14:21:47 +0100453@item compat
Stefan Hajnoczi7fa9e1f2014-01-06 12:39:01 +0800454Determines the qcow2 version to use. @code{compat=0.10} uses the
455traditional image format that can be read by any QEMU since 0.10.
Kevin Wolfd3067b02012-11-21 14:21:47 +0100456@code{compat=1.1} enables image format extensions that only QEMU 1.1 and
Stefan Hajnoczi7fa9e1f2014-01-06 12:39:01 +0800457newer understand (this is the default). Amongst others, this includes zero
458clusters, which allow efficient copy-on-read for sparse images.
Kevin Wolfd3067b02012-11-21 14:21:47 +0100459
Kevin Wolf3e032362009-10-28 12:49:17 +0100460@item backing_file
461File name of a base image (see @option{create} subcommand)
462@item backing_fmt
463Image format of the base image
464@item encryption
Daniel P. Berrange136cd192014-01-22 15:47:10 +0000465If this option is set to @code{on}, the image is encrypted with 128-bit AES-CBC.
Kevin Wolf3e032362009-10-28 12:49:17 +0100466
Daniel P. Berrange136cd192014-01-22 15:47:10 +0000467The use of encryption in qcow and qcow2 images is considered to be flawed by
468modern cryptography standards, suffering from a number of design problems:
469
470@itemize @minus
471@item The AES-CBC cipher is used with predictable initialization vectors based
472on the sector number. This makes it vulnerable to chosen plaintext attacks
473which can reveal the existence of encrypted data.
474@item The user passphrase is directly used as the encryption key. A poorly
475chosen or short passphrase will compromise the security of the encryption.
476@item In the event of the passphrase being compromised there is no way to
477change the passphrase to protect data in any qcow images. The files must
478be cloned, using a different encryption passphrase in the new file. The
479original file must then be securely erased using a program like shred,
480though even this is ineffective with many modern storage technologies.
481@end itemize
482
483Use of qcow / qcow2 encryption is thus strongly discouraged. Users are
484recommended to use an alternative encryption technology such as the
485Linux dm-crypt / LUKS system.
Kevin Wolf3e032362009-10-28 12:49:17 +0100486
487@item cluster_size
488Changes the qcow2 cluster size (must be between 512 and 2M). Smaller cluster
489sizes can improve the image file size whereas larger cluster sizes generally
490provide better performance.
491
492@item preallocation
Hu Tao0e4271b2014-09-10 17:05:49 +0800493Preallocation mode (allowed values: @code{off}, @code{metadata}, @code{falloc},
494@code{full}). An image with preallocated metadata is initially larger but can
495improve performance when the image needs to grow. @code{falloc} and @code{full}
496preallocations are like the same options of @code{raw} format, but sets up
497metadata also.
Kevin Wolf3e032362009-10-28 12:49:17 +0100498
Kevin Wolfd3067b02012-11-21 14:21:47 +0100499@item lazy_refcounts
500If this option is set to @code{on}, reference count updates are postponed with
501the goal of avoiding metadata I/O and improving performance. This is
502particularly interesting with @option{cache=writethrough} which doesn't batch
503metadata updates. The tradeoff is that after a host crash, the reference count
504tables must be rebuilt, i.e. on the next open an (automatic) @code{qemu-img
505check -r all} is required, which may take some time.
506
507This option can only be enabled if @code{compat=1.1} is specified.
508
Chunyan Liu4ab15592014-06-30 14:29:58 +0800509@item nocow
Chunyan Liubc3a7f92014-07-02 12:27:29 +0800510If this option is set to @code{on}, it will turn off COW of the file. It's only
Chunyan Liu4ab15592014-06-30 14:29:58 +0800511valid on btrfs, no effect on other file systems.
512
513Btrfs has low performance when hosting a VM image file, even more when the guest
514on the VM also using btrfs as file system. Turning off COW is a way to mitigate
515this bad performance. Generally there are two ways to turn off COW on btrfs:
516a) Disable it by mounting with nodatacow, then all newly created files will be
517NOCOW. b) For an empty file, add the NOCOW file attribute. That's what this option
518does.
519
520Note: this option is only valid to new or empty files. If there is an existing
521file which is COW and has data blocks already, it couldn't be changed to NOCOW
522by setting @code{nocow=on}. One can issue @code{lsattr filename} to check if
Chunyan Liubc3a7f92014-07-02 12:27:29 +0800523the NOCOW flag is set or not (Capital 'C' is NOCOW flag).
Chunyan Liu4ab15592014-06-30 14:29:58 +0800524
Kevin Wolf3e032362009-10-28 12:49:17 +0100525@end table
526
Kevin Wolfd3067b02012-11-21 14:21:47 +0100527@item Other
528QEMU also supports various other image file formats for compatibility with
Jeff Cody8282db12013-12-17 13:56:06 -0500529older QEMU versions or other hypervisors, including VMDK, VDI, VHD (vpc), VHDX,
530qcow1 and QED. For a full list of supported formats see @code{qemu-img --help}.
Kevin Wolfd3067b02012-11-21 14:21:47 +0100531For a more detailed description of these formats, see the QEMU Emulation User
532Documentation.
Stefan Hajnoczif0858002012-06-13 14:29:15 +0100533
Kevin Wolfd3067b02012-11-21 14:21:47 +0100534The main purpose of the block drivers for these formats is image conversion.
535For running VMs, it is recommended to convert the disk images to either raw or
536qcow2 in order to achieve good performance.
Kevin Wolff932c042009-10-28 12:49:15 +0100537@end table
538
539
bellardacd935e2004-11-15 22:57:26 +0000540@c man end
541
bellardacd935e2004-11-15 22:57:26 +0000542@setfilename qemu-img
543@settitle QEMU disk image utility
544
545@c man begin SEEALSO
546The HTML documentation of QEMU for more precise information and Linux
547user mode emulator invocation.
548@c man end
549
550@c man begin AUTHOR
551Fabrice Bellard
552@c man end
553
554@end ignore