1QEMU-IMG(1) QEMU QEMU-IMG(1)
2
6 qemu-img - QEMU disk image utility
7
9 qemu-img [standard options] command [command options]
10
12 qemu-img allows you to create, convert and modify images offline. It
13 can handle all image formats supported by QEMU.
14 Warning: Never use qemu-img to modify images in use by a running vir‐
16 tual machine or any other process; this may destroy the image. Also, be
17 aware that querying an image that is being modified by another process
18 may encounter inconsistent state.
19
21 Standard options:
22 -h, --help
24 Display this help and exit
25 -V, --version
27 Display version information and exit
28 -T, --trace [[enable=]PATTERN][,events=FILE][,file=FILE]
30 Specify tracing options.
31 [enable=]PATTERN
33 Immediately enable events matching PATTERN (either event
34 name or a globbing pattern). This option is only avail‐
35 able if QEMU has been compiled with the simple, log or
36 ftrace tracing backend. To specify multiple events or
37 patterns, specify the -trace option multiple times.
38 Use -trace help to print a list of names of trace points.
40 events=FILE
42 Immediately enable events listed in FILE. The file must
43 contain one event name (as listed in the trace-events-all
44 file) per line; globbing patterns are accepted too. This
45 option is only available if QEMU has been compiled with
46 the simple, log or ftrace tracing backend.
47 file=FILE
49 Log output traces to FILE. This option is only available
50 if QEMU has been compiled with the simple tracing back‐
51 end.
52 The following commands are supported:
54 amend [--object OBJECTDEF] [--image-opts] [-p] [-q] [-f FMT] [-t CACHE]
56 [--force] -o OPTIONS FILENAME
57 bench [-c COUNT] [-d DEPTH] [-f FMT] [--flush-interval=FLUSH_INTERVAL]
59 [-i AIO] [-n] [--no-drain] [-o OFFSET] [--pattern=PATTERN] [-q] [-s
60 BUFFER_SIZE] [-S STEP_SIZE] [-t CACHE] [-w] [-U] FILENAME
61 bitmap (--merge SOURCE | --add | --remove | --clear | --enable | --dis‐
63 able)... [-b SOURCE_FILE [-F SOURCE_FMT]] [-g GRANULARITY] [--object
64 OBJECTDEF] [--image-opts | -f FMT] FILENAME BITMAP
65 check [--object OBJECTDEF] [--image-opts] [-q] [-f FMT] [--output=OFMT]
67 [-r [leaks | all]] [-T SRC_CACHE] [-U] FILENAME
68 commit [--object OBJECTDEF] [--image-opts] [-q] [-f FMT] [-t CACHE] [-b
70 BASE] [-d] [-p] FILENAME
71 compare [--object OBJECTDEF] [--image-opts] [-f FMT] [-F FMT] [-T
73 SRC_CACHE] [-p] [-q] [-s] [-U] FILENAME1 FILENAME2
74 convert [--object OBJECTDEF] [--image-opts] [--target-image-opts]
76 [--target-is-zero] [--bitmaps] [-U] [-C] [-c] [-p] [-q] [-n] [-f FMT]
77 [-t CACHE] [-T SRC_CACHE] [-O OUTPUT_FMT] [-B BACKING_FILE] [-o
78 OPTIONS] [-l SNAPSHOT_PARAM] [-S SPARSE_SIZE] [-m NUM_COROUTINES] [-W]
79 [--salvage] FILENAME [FILENAME2 [...]] OUTPUT_FILENAME
80 create [--object OBJECTDEF] [-q] [-f FMT] [-b BACKING_FILE] [-F BACK‐
82 ING_FMT] [-u] [-o OPTIONS] FILENAME [SIZE]
83 dd [--image-opts] [-U] [-f FMT] [-O OUTPUT_FMT] [bs=BLOCK_SIZE]
85 [count=BLOCKS] [skip=BLOCKS] if=INPUT of=OUTPUT
86 info [--object OBJECTDEF] [--image-opts] [-f FMT] [--output=OFMT]
88 [--backing-chain] [-U] FILENAME
89 map [--object OBJECTDEF] [--image-opts] [-f FMT] [--start-offset=OFF‐
91 SET] [--max-length=LEN] [--output=OFMT] [-U] FILENAME
92 measure [--output=OFMT] [-O OUTPUT_FMT] [-o OPTIONS] [--size N |
94 [--object OBJECTDEF] [--image-opts] [-f FMT] [-l SNAPSHOT_PARAM] FILE‐
95 NAME]
96 snapshot [--object OBJECTDEF] [--image-opts] [-U] [-q] [-l | -a SNAP‐
98 SHOT | -c SNAPSHOT | -d SNAPSHOT] FILENAME
99 rebase [--object OBJECTDEF] [--image-opts] [-U] [-q] [-f FMT] [-t
101 CACHE] [-T SRC_CACHE] [-p] [-u] -b BACKING_FILE [-F BACKING_FMT] FILE‐
102 NAME
103 resize [--object OBJECTDEF] [--image-opts] [-f FMT] [--prealloca‐
105 tion=PREALLOC] [-q] [--shrink] FILENAME [+ | -]SIZE
106 Command parameters:
108 FILENAME is a disk image filename.
110 FMT is the disk image format. It is guessed automatically in most
112 cases. See below for a description of the supported disk formats.
113 SIZE is the disk image size in bytes. Optional suffixes k or K (kilo‐
115 byte, 1024) M (megabyte, 1024k) and G (gigabyte, 1024M) and T (ter‐
116 abyte, 1024G) are supported. b is ignored.
117 OUTPUT_FILENAME is the destination disk image filename.
119 OUTPUT_FMT is the destination format.
121 OPTIONS is a comma separated list of format specific options in a
123 name=value format. Use -o ? for an overview of the options supported by
124 the used format or see the format descriptions below for details.
125 SNAPSHOT_PARAM is param used for internal snapshot, format is 'snap‐
127 shot.id=[ID],snapshot.name=[NAME]' or '[ID_OR_NAME]'.
128 --object OBJECTDEF
130 is a QEMU user creatable object definition. See the qemu(1) man‐
131 ual page for a description of the object properties. The most
132 common object type is a secret, which is used to supply pass‐
133 words and/or encryption keys.
134 --image-opts
136 Indicates that the source FILENAME parameter is to be inter‐
137 preted as a full option string, not a plain filename. This
138 parameter is mutually exclusive with the -f parameter.
139 --target-image-opts
141 Indicates that the OUTPUT_FILENAME parameter(s) are to be inter‐
142 preted as a full option string, not a plain filename. This
143 parameter is mutually exclusive with the -O parameters. It is
144 currently required to also use the -n parameter to skip image
145 creation. This restriction may be relaxed in a future release.
146 --force-share (-U)
148 If specified, qemu-img will open the image in shared mode,
149 allowing other QEMU processes to open it in write mode. For
150 example, this can be used to get the image information (with
151 'info' subcommand) when the image is used by a running guest.
152 Note that this could produce inconsistent results because of
153 concurrent metadata changes, etc. This option is only allowed
154 when opening images in read-only mode.
155 --backing-chain
157 Will enumerate information about backing files in a disk image
158 chain. Refer below for further description.
159 -c Indicates that target image must be compressed (qcow format
161 only).
162 -h With or without a command, shows help and lists the supported
164 formats.
165 -p Display progress bar (compare, convert and rebase commands
167 only). If the -p option is not used for a command that supports
168 it, the progress is reported when the process receives a SIGUSR1
169 or SIGINFO signal.
170 -q Quiet mode - do not print any output (except errors). There's no
172 progress bar in case both -q and -p options are used.
173 -S SIZE
175 Indicates the consecutive number of bytes that must contain only
176 zeros for qemu-img to create a sparse image during conversion.
177 This value is rounded down to the nearest 512 bytes. You may use
178 the common size suffixes like k for kilobytes.
179 -t CACHE
181 Specifies the cache mode that should be used with the (destina‐
182 tion) file. See the documentation of the emulator's -drive
183 cache=... option for allowed values.
184 -T SRC_CACHE
186 Specifies the cache mode that should be used with the source
187 file(s). See the documentation of the emulator's -drive
188 cache=... option for allowed values.
189 Parameters to compare subcommand:
191 -f First image format
193 -F Second image format
195 -s Strict mode - fail on different image size or sector allocation
197 Parameters to convert subcommand:
199 --bitmaps
201 Additionally copy all persistent bitmaps from the top layer of
202 the source
203 -n Skip the creation of the target volume
205 -m Number of parallel coroutines for the convert process
207 -W Allow out-of-order writes to the destination. This option
209 improves performance, but is only recommended for preallocated
210 devices like host devices or other raw block devices.
211 -C Try to use copy offloading to move data from source image to
213 target. This may improve performance if the data is remote, such
214 as with NFS or iSCSI backends, but will not automatically spar‐
215 sify zero sectors, and may result in a fully allocated target
216 image depending on the host support for getting allocation
217 information.
218 --salvage
220 Try to ignore I/O errors when reading. Unless in quiet mode
221 (-q), errors will still be printed. Areas that cannot be read
222 from the source will be treated as containing only zeroes.
223 --target-is-zero
225 Assume that reading the destination image will always return
226 zeros. This parameter is mutually exclusive with a destination
227 image that has a backing file. It is required to also use the -n
228 parameter to skip image creation.
229 Parameters to dd subcommand:
231 bs=BLOCK_SIZE
233 Defines the block size
234 count=BLOCKS
236 Sets the number of input blocks to copy
237 if=INPUT
239 Sets the input file
240 of=OUTPUT
242 Sets the output file
243 skip=BLOCKS
245 Sets the number of input blocks to skip
246 Parameters to snapshot subcommand:
248 snapshot
250 Is the name of the snapshot to create, apply or delete
251 -a Applies a snapshot (revert disk to saved state)
253 -c Creates a snapshot
255 -d Deletes a snapshot
257 -l Lists all snapshots in the given image
259 Command description:
261 amend [--object OBJECTDEF] [--image-opts] [-p] [-q] [-f FMT] [-t CACHE]
263 [--force] -o OPTIONS FILENAME
264 Amends the image format specific OPTIONS for the image file
265 FILENAME. Not all file formats support this operation.
266 The set of options that can be amended are dependent on the
268 image format, but note that amending the backing chain relation‐
269 ship should instead be performed with qemu-img rebase.
270 --force allows some unsafe operations. Currently for -f luks, it
272 allows to erase the last encryption key, and to overwrite an
273 active encryption key.
274 bench [-c COUNT] [-d DEPTH] [-f FMT] [--flush-interval=FLUSH_INTERVAL]
276 [-i AIO] [-n] [--no-drain] [-o OFFSET] [--pattern=PATTERN] [-q] [-s
277 BUFFER_SIZE] [-S STEP_SIZE] [-t CACHE] [-w] [-U] FILENAME
278 Run a simple sequential I/O benchmark on the specified image. If
279 -w is specified, a write test is performed, otherwise a read
280 test is performed.
281 A total number of COUNT I/O requests is performed, each BUF‐
283 FER_SIZE bytes in size, and with DEPTH requests in parallel. The
284 first request starts at the position given by OFFSET, each fol‐
285 lowing request increases the current position by STEP_SIZE. If
286 STEP_SIZE is not given, BUFFER_SIZE is used for its value.
287 If FLUSH_INTERVAL is specified for a write test, the request
289 queue is drained and a flush is issued before new writes are
290 made whenever the number of remaining requests is a multiple of
291 FLUSH_INTERVAL. If additionally --no-drain is specified, a flush
292 is issued without draining the request queue first.
293 if -i is specified, AIO option can be used to specify different
295 AIO backends: threads, native or io_uring.
296 If -n is specified, the native AIO backend is used if possible.
298 On Linux, this option only works if -t none or -t directsync is
299 specified as well.
300 For write tests, by default a buffer filled with zeros is writ‐
302 ten. This can be overridden with a pattern byte specified by
303 PATTERN.
304 bitmap (--merge SOURCE | --add | --remove | --clear | --enable | --dis‐
306 able)... [-b SOURCE_FILE [-F SOURCE_FMT]] [-g GRANULARITY] [--object
307 OBJECTDEF] [--image-opts | -f FMT] FILENAME BITMAP
308 Perform one or more modifications of the persistent bitmap BIT‐
309 MAP in the disk image FILENAME. The various modifications are:
310 --add to create BITMAP, enabled to record future edits.
312 --remove to remove BITMAP.
314 --clear to clear BITMAP.
316 --enable to change BITMAP to start recording future edits.
318 --disable to change BITMAP to stop recording future edits.
320 --merge to merge the contents of the SOURCE bitmap into BITMAP.
322 Additional options include -g which sets a non-default GRANULAR‐
324 ITY for --add, and -b and -F which select an alternative source
325 file for all SOURCE bitmaps used by --merge.
326 To see what bitmaps are present in an image, use qemu-img info.
328 check [--object OBJECTDEF] [--image-opts] [-q] [-f FMT] [--output=OFMT]
330 [-r [leaks | all]] [-T SRC_CACHE] [-U] FILENAME
331 Perform a consistency check on the disk image FILENAME. The com‐
332 mand can output in the format OFMT which is either human or
333 json. The JSON output is an object of QAPI type ImageCheck.
334 If -r is specified, qemu-img tries to repair any inconsistencies
336 found during the check. -r leaks repairs only cluster leaks,
337 whereas -r all fixes all kinds of errors, with a higher risk of
338 choosing the wrong fix or hiding corruption that has already
339 occurred.
340 Only the formats qcow2, qed and vdi support consistency checks.
342 In case the image does not have any inconsistencies, check exits
344 with 0. Other exit codes indicate the kind of inconsistency
345 found or if another error occurred. The following table summa‐
346 rizes all exit codes of the check subcommand:
347 0 Check completed, the image is (now) consistent
349 1 Check not completed because of internal errors
351 2 Check completed, image is corrupted
353 3 Check completed, image has leaked clusters, but is not
355 corrupted
356 63 Checks are not supported by the image format
358 If -r is specified, exit codes representing the image state
360 refer to the state after (the attempt at) repairing it. That is,
361 a successful -r all will yield the exit code 0, independently of
362 the image state before.
363 commit [--object OBJECTDEF] [--image-opts] [-q] [-f FMT] [-t CACHE] [-b
365 BASE] [-d] [-p] FILENAME
366 Commit the changes recorded in FILENAME in its base image or
367 backing file. If the backing file is smaller than the snapshot,
368 then the backing file will be resized to be the same size as the
369 snapshot. If the snapshot is smaller than the backing file, the
370 backing file will not be truncated. If you want the backing
371 file to match the size of the smaller snapshot, you can safely
372 truncate it yourself once the commit operation successfully com‐
373 pletes.
374 The image FILENAME is emptied after the operation has succeeded.
376 If you do not need FILENAME afterwards and intend to drop it,
377 you may skip emptying FILENAME by specifying the -d flag.
378 If the backing chain of the given image file FILENAME has more
380 than one layer, the backing file into which the changes will be
381 committed may be specified as BASE (which has to be part of
382 FILENAME's backing chain). If BASE is not specified, the immedi‐
383 ate backing file of the top image (which is FILENAME) will be
384 used. Note that after a commit operation all images between BASE
385 and the top image will be invalid and may return garbage data
386 when read. For this reason, -b implies -d (so that the top image
387 stays valid).
388 compare [--object OBJECTDEF] [--image-opts] [-f FMT] [-F FMT] [-T
390 SRC_CACHE] [-p] [-q] [-s] [-U] FILENAME1 FILENAME2
391 Check if two images have the same content. You can compare
392 images with different format or settings.
393 The format is probed unless you specify it by -f (used for FILE‐
395 NAME1) and/or -F (used for FILENAME2) option.
396 By default, images with different size are considered identical
398 if the larger image contains only unallocated and/or zeroed sec‐
399 tors in the area after the end of the other image. In addition,
400 if any sector is not allocated in one image and contains only
401 zero bytes in the second one, it is evaluated as equal. You can
402 use Strict mode by specifying the -s option. When compare runs
403 in Strict mode, it fails in case image size differs or a sector
404 is allocated in one image and is not allocated in the second
405 one.
406 By default, compare prints out a result message. This message
408 displays information that both images are same or the position
409 of the first different byte. In addition, result message can
410 report different image size in case Strict mode is used.
411 Compare exits with 0 in case the images are equal and with 1 in
413 case the images differ. Other exit codes mean an error occurred
414 during execution and standard error output should contain an
415 error message. The following table sumarizes all exit codes of
416 the compare subcommand:
417 0 Images are identical
419 1 Images differ
421 2 Error on opening an image
423 3 Error on checking a sector allocation
425 4 Error on reading data
427 convert [--object OBJECTDEF] [--image-opts] [--target-image-opts]
429 [--target-is-zero] [--bitmaps] [-U] [-C] [-c] [-p] [-q] [-n] [-f FMT]
430 [-t CACHE] [-T SRC_CACHE] [-O OUTPUT_FMT] [-B BACKING_FILE] [-o
431 OPTIONS] [-l SNAPSHOT_PARAM] [-S SPARSE_SIZE] [-m NUM_COROUTINES] [-W]
432 FILENAME [FILENAME2 [...]] OUTPUT_FILENAME
433 Convert the disk image FILENAME or a snapshot SNAPSHOT_PARAM to
434 disk image OUTPUT_FILENAME using format OUTPUT_FMT. It can be
435 optionally compressed (-c option) or use any format specific
436 options like encryption (-o option).
437 Only the formats qcow and qcow2 support compression. The com‐
439 pression is read-only. It means that if a compressed sector is
440 rewritten, then it is rewritten as uncompressed data.
441 Image conversion is also useful to get smaller image when using
443 a growable format such as qcow: the empty sectors are detected
444 and suppressed from the destination image.
445 SPARSE_SIZE indicates the consecutive number of bytes (defaults
447 to 4k) that must contain only zeros for qemu-img to create a
448 sparse image during conversion. If SPARSE_SIZE is 0, the source
449 will not be scanned for unallocated or zero sectors, and the
450 destination image will always be fully allocated.
451 You can use the BACKING_FILE option to force the output image to
453 be created as a copy on write image of the specified base image;
454 the BACKING_FILE should have the same content as the input's
455 base image, however the path, image format, etc may differ.
456 If a relative path name is given, the backing file is looked up
458 relative to the directory containing OUTPUT_FILENAME.
459 If the -n option is specified, the target volume creation will
461 be skipped. This is useful for formats such as rbd if the target
462 volume has already been created with site specific options that
463 cannot be supplied through qemu-img.
464 Out of order writes can be enabled with -W to improve perfor‐
466 mance. This is only recommended for preallocated devices like
467 host devices or other raw block devices. Out of order write does
468 not work in combination with creating compressed images.
469 NUM_COROUTINES specifies how many coroutines work in parallel
471 during the convert process (defaults to 8).
472 create [--object OBJECTDEF] [-q] [-f FMT] [-b BACKING_FILE] [-F BACK‐
474 ING_FMT] [-u] [-o OPTIONS] FILENAME [SIZE]
475 Create the new disk image FILENAME of size SIZE and format FMT.
476 Depending on the file format, you can add one or more OPTIONS
477 that enable additional features of this format.
478 If the option BACKING_FILE is specified, then the image will
480 record only the differences from BACKING_FILE. No size needs to
481 be specified in this case. BACKING_FILE will never be modified
482 unless you use the commit monitor command (or qemu-img commit).
483 If a relative path name is given, the backing file is looked up
485 relative to the directory containing FILENAME.
486 Note that a given backing file will be opened to check that it
488 is valid. Use the -u option to enable unsafe backing file mode,
489 which means that the image will be created even if the associ‐
490 ated backing file cannot be opened. A matching backing file must
491 be created or additional options be used to make the backing
492 file specification valid when you want to use an image created
493 this way.
494 The size can also be specified using the SIZE option with -o, it
496 doesn't need to be specified separately in this case.
497 dd [--image-opts] [-U] [-f FMT] [-O OUTPUT_FMT] [bs=BLOCK_SIZE]
499 [count=BLOCKS] [skip=BLOCKS] if=INPUT of=OUTPUT
500 dd copies from INPUT file to OUTPUT file converting it from FMT
501 format to OUTPUT_FMT format.
502 The data is by default read and written using blocks of 512
504 bytes but can be modified by specifying BLOCK_SIZE. If
505 count=BLOCKS is specified dd will stop reading input after read‐
506 ing BLOCKS input blocks.
507 The size syntax is similar to dd(1)'s size syntax.
509 info [--object OBJECTDEF] [--image-opts] [-f FMT] [--output=OFMT]
511 [--backing-chain] [-U] FILENAME
512 Give information about the disk image FILENAME. Use it in par‐
513 ticular to know the size reserved on disk which can be different
514 from the displayed size. If VM snapshots are stored in the disk
515 image, they are displayed too.
516 If a disk image has a backing file chain, information about each
518 disk image in the chain can be recursively enumerated by using
519 the option --backing-chain.
520 For instance, if you have an image chain like:
522 base.qcow2 <- snap1.qcow2 <- snap2.qcow2
524 To enumerate information about each disk image in the above
526 chain, starting from top to base, do:
527 qemu-img info --backing-chain snap2.qcow2
529 The command can output in the format OFMT which is either human
531 or json. The JSON output is an object of QAPI type ImageInfo;
532 with --backing-chain, it is an array of ImageInfo objects.
533 --output=human reports the following information (for every
535 image in the chain):
536 image The image file name
538 file format
540 The image format
541 virtual size
543 The size of the guest disk
544 disk size
546 How much space the image file occupies on the host file
547 system (may be shown as 0 if this information is unavail‐
548 able, e.g. because there is no file system)
549 cluster_size
551 Cluster size of the image format, if applicable
552 encrypted
554 Whether the image is encrypted (only present if so)
555 cleanly shut down
557 This is shown as no if the image is dirty and will have
558 to be auto-repaired the next time it is opened in qemu.
559 backing file
561 The backing file name, if present
562 backing file format
564 The format of the backing file, if the image enforces it
565 Snapshot list
567 A list of all internal snapshots
568 Format specific information
570 Further information whose structure depends on the image
571 format. This section is a textual representation of the
572 respective ImageInfoSpecific* QAPI object (e.g. ImageIn‐
573 foSpecificQCow2 for qcow2 images).
574 map [--object OBJECTDEF] [--image-opts] [-f FMT] [--start-offset=OFF‐
576 SET] [--max-length=LEN] [--output=OFMT] [-U] FILENAME
577 Dump the metadata of image FILENAME and its backing file chain.
578 In particular, this commands dumps the allocation state of every
579 sector of FILENAME, together with the topmost file that allo‐
580 cates it in the backing file chain.
581 Two option formats are possible. The default format (human)
583 only dumps known-nonzero areas of the file. Known-zero parts of
584 the file are omitted altogether, and likewise for parts that are
585 not allocated throughout the chain. qemu-img output will iden‐
586 tify a file from where the data can be read, and the offset in
587 the file. Each line will include four fields, the first three
588 of which are hexadecimal numbers. For example the first line
589 of:
590 Offset Length Mapped to File
592 0 0x20000 0x50000 /tmp/overlay.qcow2
593 0x100000 0x10000 0x95380000 /tmp/backing.qcow2
594 means that 0x20000 (131072) bytes starting at offset 0 in the
596 image are available in /tmp/overlay.qcow2 (opened in raw format)
597 starting at offset 0x50000 (327680). Data that is compressed,
598 encrypted, or otherwise not available in raw format will cause
599 an error if human format is in use. Note that file names can
600 include newlines, thus it is not safe to parse this output for‐
601 mat in scripts.
602 The alternative format json will return an array of dictionaries
604 in JSON format. It will include similar information in the
605 start, length, offset fields; it will also include other more
606 specific information:
607 · whether the sectors contain actual data or not (boolean field
609 data; if false, the sectors are either unallocated or stored
610 as optimized all-zero clusters);
611 · whether the data is known to read as zero (boolean field
613 zero);
614 · in order to make the output shorter, the target file is
616 expressed as a depth; for example, a depth of 2 refers to the
617 backing file of the backing file of FILENAME.
618 In JSON format, the offset field is optional; it is absent in
620 cases where human format would omit the entry or exit with an
621 error. If data is false and the offset field is present, the
622 corresponding sectors in the file are not yet in use, but they
623 are preallocated.
624 For more information, consult include/block/block.h in QEMU's
626 source code.
627 measure [--output=OFMT] [-O OUTPUT_FMT] [-o OPTIONS] [--size N |
629 [--object OBJECTDEF] [--image-opts] [-f FMT] [-l SNAPSHOT_PARAM] FILE‐
630 NAME]
631 Calculate the file size required for a new image. This informa‐
632 tion can be used to size logical volumes or SAN LUNs appropri‐
633 ately for the image that will be placed in them. The values
634 reported are guaranteed to be large enough to fit the image.
635 The command can output in the format OFMT which is either human
636 or json. The JSON output is an object of QAPI type BlockMea‐
637 sureInfo.
638 If the size N is given then act as if creating a new empty image
640 file using qemu-img create. If FILENAME is given then act as if
641 converting an existing image file using qemu-img convert. The
642 format of the new file is given by OUTPUT_FMT while the format
643 of an existing file is given by FMT.
644 A snapshot in an existing image can be specified using SNAP‐
646 SHOT_PARAM.
647 The following fields are reported:
649 required size: 524288
651 fully allocated size: 1074069504
652 bitmaps size: 0
653 The required size is the file size of the new image. It may be
655 smaller than the virtual disk size if the image format supports
656 compact representation.
657 The fully allocated size is the file size of the new image once
659 data has been written to all sectors. This is the maximum size
660 that the image file can occupy with the exception of internal
661 snapshots, dirty bitmaps, vmstate data, and other advanced image
662 format features.
663 The bitmaps size is the additional size required in order to
665 copy bitmaps from a source image in addition to the guest-visi‐
666 ble data; the line is omitted if either source or destination
667 lacks bitmap support, or 0 if bitmaps are supported but there is
668 nothing to copy.
669 snapshot [--object OBJECTDEF] [--image-opts] [-U] [-q] [-l | -a SNAP‐
671 SHOT | -c SNAPSHOT | -d SNAPSHOT] FILENAME
672 List, apply, create or delete snapshots in image FILENAME.
673 rebase [--object OBJECTDEF] [--image-opts] [-U] [-q] [-f FMT] [-t
675 CACHE] [-T SRC_CACHE] [-p] [-u] -b BACKING_FILE [-F BACKING_FMT] FILE‐
676 NAME
677 Changes the backing file of an image. Only the formats qcow2 and
678 qed support changing the backing file.
679 The backing file is changed to BACKING_FILE and (if the image
681 format of FILENAME supports this) the backing file format is
682 changed to BACKING_FMT. If BACKING_FILE is specified as "" (the
683 empty string), then the image is rebased onto no backing file
684 (i.e. it will exist independently of any backing file).
685 If a relative path name is given, the backing file is looked up
687 relative to the directory containing FILENAME.
688 CACHE specifies the cache mode to be used for FILENAME, whereas
690 SRC_CACHE specifies the cache mode for reading backing files.
691 There are two different modes in which rebase can operate:
693 Safe mode
695 This is the default mode and performs a real rebase oper‐
696 ation. The new backing file may differ from the old one
697 and qemu-img rebase will take care of keeping the
698 guest-visible content of FILENAME unchanged.
699 In order to achieve this, any clusters that differ
701 between BACKING_FILE and the old backing file of FILENAME
702 are merged into FILENAME before actually changing the
703 backing file.
704 Note that the safe mode is an expensive operation, compa‐
706 rable to converting an image. It only works if the old
707 backing file still exists.
708 Unsafe mode
710 qemu-img uses the unsafe mode if -u is specified. In this
711 mode, only the backing file name and format of FILENAME
712 is changed without any checks on the file contents. The
713 user must take care of specifying the correct new backing
714 file, or the guest-visible content of the image will be
715 corrupted.
716 This mode is useful for renaming or moving the backing
718 file to somewhere else. It can be used without an acces‐
719 sible old backing file, i.e. you can use it to fix an
720 image whose backing file has already been moved/renamed.
721 You can use rebase to perform a "diff" operation on two disk
723 images. This can be useful when you have copied or cloned a
724 guest, and you want to get back to a thin image on top of a tem‐
725 plate or base image.
726 Say that base.img has been cloned as modified.img by copying it,
728 and that the modified.img guest has run so there are now some
729 changes compared to base.img. To construct a thin image called
730 diff.qcow2 that contains just the differences, do:
731 qemu-img create -f qcow2 -b modified.img diff.qcow2
733 qemu-img rebase -b base.img diff.qcow2
734 At this point, modified.img can be discarded, since base.img +
736 diff.qcow2 contains the same information.
737 resize [--object OBJECTDEF] [--image-opts] [-f FMT] [--prealloca‐
739 tion=PREALLOC] [-q] [--shrink] FILENAME [+ | -]SIZE
740 Change the disk image as if it had been created with SIZE.
741 Before using this command to shrink a disk image, you MUST use
743 file system and partitioning tools inside the VM to reduce allo‐
744 cated file systems and partition sizes accordingly. Failure to
745 do so will result in data loss!
746 When shrinking images, the --shrink option must be given. This
748 informs qemu-img that the user acknowledges all loss of data
749 beyond the truncated image's end.
750 After using this command to grow a disk image, you must use file
752 system and partitioning tools inside the VM to actually begin
753 using the new space on the device.
754 When growing an image, the --preallocation option may be used to
756 specify how the additional image area should be allocated on the
757 host. See the format description in the Notes section which
758 values are allowed. Using this option may result in slightly
759 more data being allocated than necessary.
760
762 Supported image file formats:
763 raw
765 Raw disk image format (default). This format has the advantage of
766 being simple and easily exportable to all other emulators. If your
767 file system supports holes (for example in ext2 or ext3 on Linux or
768 NTFS on Windows), then only the written sectors will reserve space.
769 Use qemu-img info to know the real size used by the image or ls -ls
770 on Unix/Linux.
771 Supported options:
773 preallocation
775 Preallocation mode (allowed values: off, falloc, full). fal‐
776 loc mode preallocates space for image by calling posix_fallo‐
777 cate(). full mode preallocates space for image by writing
778 data to underlying storage. This data may or may not be
779 zero, depending on the storage location.
780 qcow2
782 QEMU image format, the most versatile format. Use it to have smaller
783 images (useful if your filesystem does not supports holes, for exam‐
784 ple on Windows), optional AES encryption, zlib based compression and
785 support of multiple VM snapshots.
786 Supported options:
788 compat Determines the qcow2 version to use. compat=0.10 uses the
790 traditional image format that can be read by any QEMU since
791 0.10. compat=1.1 enables image format extensions that only
792 QEMU 1.1 and newer understand (this is the default). Amongst
793 others, this includes zero clusters, which allow efficient
794 copy-on-read for sparse images.
795 backing_file
797 File name of a base image (see create subcommand)
798 backing_fmt
800 Image format of the base image
801 encryption
803 If this option is set to on, the image is encrypted with
804 128-bit AES-CBC.
805 The use of encryption in qcow and qcow2 images is considered
807 to be flawed by modern cryptography standards, suffering from
808 a number of design problems:
809 · The AES-CBC cipher is used with predictable initialization
811 vectors based on the sector number. This makes it vulnera‐
812 ble to chosen plaintext attacks which can reveal the exis‐
813 tence of encrypted data.
814 · The user passphrase is directly used as the encryption key.
816 A poorly chosen or short passphrase will compromise the
817 security of the encryption.
818 · In the event of the passphrase being compromised there is
820 no way to change the passphrase to protect data in any qcow
821 images. The files must be cloned, using a different encryp‐
822 tion passphrase in the new file. The original file must
823 then be securely erased using a program like shred, though
824 even this is ineffective with many modern storage technolo‐
825 gies.
826 · Initialization vectors used to encrypt sectors are based on
828 the guest virtual sector number, instead of the host physi‐
829 cal sector. When a disk image has multiple internal snap‐
830 shots this means that data in multiple physical sectors is
831 encrypted with the same initialization vector. With the CBC
832 mode, this opens the possibility of watermarking attacks if
833 the attack can collect multiple sectors encrypted with the
834 same IV and some predictable data. Having multiple qcow2
835 images with the same passphrase also exposes this weakness
836 since the passphrase is directly used as the key.
837 Use of qcow / qcow2 encryption is thus strongly discouraged.
839 Users are recommended to use an alternative encryption tech‐
840 nology such as the Linux dm-crypt / LUKS system.
841 cluster_size
843 Changes the qcow2 cluster size (must be between 512 and 2M).
844 Smaller cluster sizes can improve the image file size whereas
845 larger cluster sizes generally provide better performance.
846 preallocation
848 Preallocation mode (allowed values: off, metadata, falloc,
849 full). An image with preallocated metadata is initially
850 larger but can improve performance when the image needs to
851 grow. falloc and full preallocations are like the same
852 options of raw format, but sets up metadata also.
853 lazy_refcounts
855 If this option is set to on, reference count updates are
856 postponed with the goal of avoiding metadata I/O and improv‐
857 ing performance. This is particularly interesting with
858 cache=writethrough which doesn't batch metadata updates. The
859 tradeoff is that after a host crash, the reference count
860 tables must be rebuilt, i.e. on the next open an (automatic)
861 qemu-img check -r all is required, which may take some time.
862 This option can only be enabled if compat=1.1 is specified.
864 nocow If this option is set to on, it will turn off COW of the
866 file. It's only valid on btrfs, no effect on other file sys‐
867 tems.
868 Btrfs has low performance when hosting a VM image file, even
870 more when the guest on the VM also using btrfs as file sys‐
871 tem. Turning off COW is a way to mitigate this bad perfor‐
872 mance. Generally there are two ways to turn off COW on btrfs:
873 · Disable it by mounting with nodatacow, then all newly cre‐
875 ated files will be NOCOW
876 · For an empty file, add the NOCOW file attribute. That's
878 what this option does.
879 Note: this option is only valid to new or empty files. If
881 there is an existing file which is COW and has data blocks
882 already, it couldn't be changed to NOCOW by setting nocow=on.
883 One can issue lsattr filename to check if the NOCOW flag is
884 set or not (Capital 'C' is NOCOW flag).
885 Other
887 QEMU also supports various other image file formats for compatibil‐
888 ity with older QEMU versions or other hypervisors, including VMDK,
889 VDI, VHD (vpc), VHDX, qcow1 and QED. For a full list of supported
890 formats see qemu-img --help. For a more detailed description of
891 these formats, see the QEMU block drivers reference documentation.
892 The main purpose of the block drivers for these formats is image
894 conversion. For running VMs, it is recommended to convert the disk
895 images to either raw or qcow2 in order to achieve good performance.
896
898 Fabrice Bellard
899
901 2021, The QEMU Project Developers
9025.1.0 Jan 11, 2021 QEMU-IMG(1)