xref: /freebsd/sys/contrib/openzfs/man/man8/zfs-send.8 (revision f374ba41f55c1a127303d92d830dd58eef2f5243)
1.\"
2.\" CDDL HEADER START
3.\"
4.\" The contents of this file are subject to the terms of the
5.\" Common Development and Distribution License (the "License").
6.\" You may not use this file except in compliance with the License.
7.\"
8.\" You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
9.\" or https://opensource.org/licenses/CDDL-1.0.
10.\" See the License for the specific language governing permissions
11.\" and limitations under the License.
12.\"
13.\" When distributing Covered Code, include this CDDL HEADER in each
14.\" file and include the License file at usr/src/OPENSOLARIS.LICENSE.
15.\" If applicable, add the following below this CDDL HEADER, with the
16.\" fields enclosed by brackets "[]" replaced with your own identifying
17.\" information: Portions Copyright [yyyy] [name of copyright owner]
18.\"
19.\" CDDL HEADER END
20.\"
21.\" Copyright (c) 2009 Sun Microsystems, Inc. All Rights Reserved.
22.\" Copyright 2011 Joshua M. Clulow <josh@sysmgr.org>
23.\" Copyright (c) 2011, 2019 by Delphix. All rights reserved.
24.\" Copyright (c) 2013 by Saso Kiselkov. All rights reserved.
25.\" Copyright (c) 2014, Joyent, Inc. All rights reserved.
26.\" Copyright (c) 2014 by Adam Stevko. All rights reserved.
27.\" Copyright (c) 2014 Integros [integros.com]
28.\" Copyright 2019 Richard Laager. All rights reserved.
29.\" Copyright 2018 Nexenta Systems, Inc.
30.\" Copyright 2019 Joyent, Inc.
31.\"
32.Dd January 12, 2023
33.Dt ZFS-SEND 8
34.Os
35.
36.Sh NAME
37.Nm zfs-send
38.Nd generate backup stream of ZFS dataset
39.Sh SYNOPSIS
40.Nm zfs
41.Cm send
42.Op Fl DLPVbcehnpsvw
43.Op Fl R Op Fl X Ar dataset Ns Oo , Ns Ar dataset Oc Ns …
44.Op Oo Fl I Ns | Ns Fl i Oc Ar snapshot
45.Ar snapshot
46.Nm zfs
47.Cm send
48.Op Fl DLPVcensvw
49.Op Fl i Ar snapshot Ns | Ns Ar bookmark
50.Ar filesystem Ns | Ns Ar volume Ns | Ns Ar snapshot
51.Nm zfs
52.Cm send
53.Fl -redact Ar redaction_bookmark
54.Op Fl DLPVcenpv
55.Op Fl i Ar snapshot Ns | Ns Ar bookmark
56.Ar snapshot
57.Nm zfs
58.Cm send
59.Op Fl PVenv
60.Fl t
61.Ar receive_resume_token
62.Nm zfs
63.Cm send
64.Op Fl PVnv
65.Fl S Ar filesystem
66.Nm zfs
67.Cm redact
68.Ar snapshot redaction_bookmark
69.Ar redaction_snapshot Ns …
70.
71.Sh DESCRIPTION
72.Bl -tag -width ""
73.It Xo
74.Nm zfs
75.Cm send
76.Op Fl DLPVbcehnpsvw
77.Op Fl R Op Fl X Ar dataset Ns Oo , Ns Ar dataset Oc Ns …
78.Op Oo Fl I Ns | Ns Fl i Oc Ar snapshot
79.Ar snapshot
80.Xc
81Creates a stream representation of the second
82.Ar snapshot ,
83which is written to standard output.
84The output can be redirected to a file or to a different system
85.Po for example, using
86.Xr ssh 1
87.Pc .
88By default, a full stream is generated.
89.Bl -tag -width "-D"
90.It Fl D , -dedup
91Deduplicated send is no longer supported.
92This flag is accepted for backwards compatibility, but a regular,
93non-deduplicated stream will be generated.
94.It Fl I Ar snapshot
95Generate a stream package that sends all intermediary snapshots from the first
96snapshot to the second snapshot.
97For example,
98.Fl I Em @a Em fs@d
99is similar to
100.Fl i Em @a Em fs@b Ns \&; Fl i Em @b Em fs@c Ns \&; Fl i Em @c Em fs@d .
101The incremental source may be specified as with the
102.Fl i
103option.
104.It Fl L , -large-block
105Generate a stream which may contain blocks larger than 128 KiB.
106This flag has no effect if the
107.Sy large_blocks
108pool feature is disabled, or if the
109.Sy recordsize
110property of this filesystem has never been set above 128 KiB.
111The receiving system must have the
112.Sy large_blocks
113pool feature enabled as well.
114See
115.Xr zpool-features 7
116for details on ZFS feature flags and the
117.Sy large_blocks
118feature.
119.It Fl P , -parsable
120Print machine-parsable verbose information about the stream package generated.
121.It Fl R , -replicate
122Generate a replication stream package, which will replicate the specified
123file system, and all descendent file systems, up to the named snapshot.
124When received, all properties, snapshots, descendent file systems, and clones
125are preserved.
126.Pp
127If the
128.Fl i
129or
130.Fl I
131flags are used in conjunction with the
132.Fl R
133flag, an incremental replication stream is generated.
134The current values of properties, and current snapshot and file system names are
135set when the stream is received.
136If the
137.Fl F
138flag is specified when this stream is received, snapshots and file systems that
139do not exist on the sending side are destroyed.
140If the
141.Fl R
142flag is used to send encrypted datasets, then
143.Fl w
144must also be specified.
145.It Fl V , -proctitle
146Set the process title to a per-second report of how much data has been sent.
147.It Fl X , -exclude Ar dataset Ns Oo , Ns Ar dataset Oc Ns …
148With
149.Fl R ,
150.Fl X
151specifies a set of datasets (and, hence, their descendants),
152to be excluded from the send stream.
153The root dataset may not be excluded.
154.Fl X Ar a Fl X Ar b
155is equivalent to
156.Fl X Ar a , Ns Ar b .
157.It Fl e , -embed
158Generate a more compact stream by using
159.Sy WRITE_EMBEDDED
160records for blocks which are stored more compactly on disk by the
161.Sy embedded_data
162pool feature.
163This flag has no effect if the
164.Sy embedded_data
165feature is disabled.
166The receiving system must have the
167.Sy embedded_data
168feature enabled.
169If the
170.Sy lz4_compress
171feature is active on the sending system, then the receiving system must have
172that feature enabled as well.
173Datasets that are sent with this flag may not be
174received as an encrypted dataset, since encrypted datasets cannot use the
175.Sy embedded_data
176feature.
177See
178.Xr zpool-features 7
179for details on ZFS feature flags and the
180.Sy embedded_data
181feature.
182.It Fl b , -backup
183Sends only received property values whether or not they are overridden by local
184settings, but only if the dataset has ever been received.
185Use this option when you want
186.Nm zfs Cm receive
187to restore received properties backed up on the sent dataset and to avoid
188sending local settings that may have nothing to do with the source dataset,
189but only with how the data is backed up.
190.It Fl c , -compressed
191Generate a more compact stream by using compressed WRITE records for blocks
192which are compressed on disk and in memory
193.Po see the
194.Sy compression
195property for details
196.Pc .
197If the
198.Sy lz4_compress
199feature is active on the sending system, then the receiving system must have
200that feature enabled as well.
201If the
202.Sy large_blocks
203feature is enabled on the sending system but the
204.Fl L
205option is not supplied in conjunction with
206.Fl c ,
207then the data will be decompressed before sending so it can be split into
208smaller block sizes.
209Streams sent with
210.Fl c
211will not have their data recompressed on the receiver side using
212.Fl o Sy compress Ns = Ar value .
213The data will stay compressed as it was from the sender.
214The new compression property will be set for future data.
215Note that uncompressed data from the sender will still attempt to
216compress on the receiver, unless you specify
217.Fl o Sy compress Ns = Em off .
218.It Fl w , -raw
219For encrypted datasets, send data exactly as it exists on disk.
220This allows backups to be taken even if encryption keys are not currently
221loaded.
222The backup may then be received on an untrusted machine since that machine will
223not have the encryption keys to read the protected data or alter it without
224being detected.
225Upon being received, the dataset will have the same encryption
226keys as it did on the send side, although the
227.Sy keylocation
228property will be defaulted to
229.Sy prompt
230if not otherwise provided.
231For unencrypted datasets, this flag will be equivalent to
232.Fl Lec .
233Note that if you do not use this flag for sending encrypted datasets, data will
234be sent unencrypted and may be re-encrypted with a different encryption key on
235the receiving system, which will disable the ability to do a raw send to that
236system for incrementals.
237.It Fl h , -holds
238Generate a stream package that includes any snapshot holds (created with the
239.Nm zfs Cm hold
240command), and indicating to
241.Nm zfs Cm receive
242that the holds be applied to the dataset on the receiving system.
243.It Fl i Ar snapshot
244Generate an incremental stream from the first
245.Ar snapshot
246.Pq the incremental source
247to the second
248.Ar snapshot
249.Pq the incremental target .
250The incremental source can be specified as the last component of the snapshot
251name
252.Po the
253.Sy @
254character and following
255.Pc
256and it is assumed to be from the same file system as the incremental target.
257.Pp
258If the destination is a clone, the source may be the origin snapshot, which must
259be fully specified
260.Po for example,
261.Em pool/fs@origin ,
262not just
263.Em @origin
264.Pc .
265.It Fl n , -dryrun
266Do a dry-run
267.Pq Qq No-op
268send.
269Do not generate any actual send data.
270This is useful in conjunction with the
271.Fl v
272or
273.Fl P
274flags to determine what data will be sent.
275In this case, the verbose output will be written to standard output
276.Po contrast with a non-dry-run, where the stream is written to standard output
277and the verbose output goes to standard error
278.Pc .
279.It Fl p , -props
280Include the dataset's properties in the stream.
281This flag is implicit when
282.Fl R
283is specified.
284The receiving system must also support this feature.
285Sends of encrypted datasets must use
286.Fl w
287when using this flag.
288.It Fl s , -skip-missing
289Allows sending a replication stream even when there are snapshots missing in the
290hierarchy.
291When a snapshot is missing, instead of throwing an error and aborting the send,
292a warning is printed to the standard error stream and the dataset to which it
293belongs
294and its descendents are skipped.
295This flag can only be used in conjunction with
296.Fl R .
297.It Fl v , -verbose
298Print verbose information about the stream package generated.
299This information includes a per-second report of how much data has been sent.
300.Pp
301The format of the stream is committed.
302You will be able to receive your streams on future versions of ZFS.
303.El
304.It Xo
305.Nm zfs
306.Cm send
307.Op Fl DLPVcenvw
308.Op Fl i Ar snapshot Ns | Ns Ar bookmark
309.Ar filesystem Ns | Ns Ar volume Ns | Ns Ar snapshot
310.Xc
311Generate a send stream, which may be of a filesystem, and may be incremental
312from a bookmark.
313If the destination is a filesystem or volume, the pool must be read-only, or the
314filesystem must not be mounted.
315When the stream generated from a filesystem or volume is received, the default
316snapshot name will be
317.Qq --head-- .
318.Bl -tag -width "-D"
319.It Fl D , -dedup
320Deduplicated send is no longer supported.
321This flag is accepted for backwards compatibility, but a regular,
322non-deduplicated stream will be generated.
323.It Fl L , -large-block
324Generate a stream which may contain blocks larger than 128 KiB.
325This flag has no effect if the
326.Sy large_blocks
327pool feature is disabled, or if the
328.Sy recordsize
329property of this filesystem has never been set above 128 KiB.
330The receiving system must have the
331.Sy large_blocks
332pool feature enabled as well.
333See
334.Xr zpool-features 7
335for details on ZFS feature flags and the
336.Sy large_blocks
337feature.
338.It Fl P , -parsable
339Print machine-parsable verbose information about the stream package generated.
340.It Fl c , -compressed
341Generate a more compact stream by using compressed WRITE records for blocks
342which are compressed on disk and in memory
343.Po see the
344.Sy compression
345property for details
346.Pc .
347If the
348.Sy lz4_compress
349feature is active on the sending system, then the receiving system must have
350that feature enabled as well.
351If the
352.Sy large_blocks
353feature is enabled on the sending system but the
354.Fl L
355option is not supplied in conjunction with
356.Fl c ,
357then the data will be decompressed before sending so it can be split into
358smaller block sizes.
359.It Fl w , -raw
360For encrypted datasets, send data exactly as it exists on disk.
361This allows backups to be taken even if encryption keys are not currently
362loaded.
363The backup may then be received on an untrusted machine since that machine will
364not have the encryption keys to read the protected data or alter it without
365being detected.
366Upon being received, the dataset will have the same encryption
367keys as it did on the send side, although the
368.Sy keylocation
369property will be defaulted to
370.Sy prompt
371if not otherwise provided.
372For unencrypted datasets, this flag will be equivalent to
373.Fl Lec .
374Note that if you do not use this flag for sending encrypted datasets, data will
375be sent unencrypted and may be re-encrypted with a different encryption key on
376the receiving system, which will disable the ability to do a raw send to that
377system for incrementals.
378.It Fl e , -embed
379Generate a more compact stream by using
380.Sy WRITE_EMBEDDED
381records for blocks which are stored more compactly on disk by the
382.Sy embedded_data
383pool feature.
384This flag has no effect if the
385.Sy embedded_data
386feature is disabled.
387The receiving system must have the
388.Sy embedded_data
389feature enabled.
390If the
391.Sy lz4_compress
392feature is active on the sending system, then the receiving system must have
393that feature enabled as well.
394Datasets that are sent with this flag may not be received as an encrypted
395dataset,
396since encrypted datasets cannot use the
397.Sy embedded_data
398feature.
399See
400.Xr zpool-features 7
401for details on ZFS feature flags and the
402.Sy embedded_data
403feature.
404.It Fl i Ar snapshot Ns | Ns Ar bookmark
405Generate an incremental send stream.
406The incremental source must be an earlier snapshot in the destination's history.
407It will commonly be an earlier snapshot in the destination's file system, in
408which case it can be specified as the last component of the name
409.Po the
410.Sy #
411or
412.Sy @
413character and following
414.Pc .
415.Pp
416If the incremental target is a clone, the incremental source can be the origin
417snapshot, or an earlier snapshot in the origin's filesystem, or the origin's
418origin, etc.
419.It Fl n , -dryrun
420Do a dry-run
421.Pq Qq No-op
422send.
423Do not generate any actual send data.
424This is useful in conjunction with the
425.Fl v
426or
427.Fl P
428flags to determine what data will be sent.
429In this case, the verbose output will be written to standard output
430.Po contrast with a non-dry-run, where the stream is written to standard output
431and the verbose output goes to standard error
432.Pc .
433.It Fl v , -verbose
434Print verbose information about the stream package generated.
435This information includes a per-second report of how much data has been sent.
436.El
437.It Xo
438.Nm zfs
439.Cm send
440.Fl -redact Ar redaction_bookmark
441.Op Fl DLPVcenpv
442.Op Fl i Ar snapshot Ns | Ns Ar bookmark
443.Ar snapshot
444.Xc
445Generate a redacted send stream.
446This send stream contains all blocks from the snapshot being sent that aren't
447included in the redaction list contained in the bookmark specified by the
448.Fl -redact
449(or
450.Fl d )
451flag.
452The resulting send stream is said to be redacted with respect to the snapshots
453the bookmark specified by the
454.Fl -redact No flag was created with .
455The bookmark must have been created by running
456.Nm zfs Cm redact
457on the snapshot being sent.
458.Pp
459This feature can be used to allow clones of a filesystem to be made available on
460a remote system, in the case where their parent need not (or needs to not) be
461usable.
462For example, if a filesystem contains sensitive data, and it has clones where
463that sensitive data has been secured or replaced with dummy data, redacted sends
464can be used to replicate the secured data without replicating the original
465sensitive data, while still sharing all possible blocks.
466A snapshot that has been redacted with respect to a set of snapshots will
467contain all blocks referenced by at least one snapshot in the set, but will
468contain none of the blocks referenced by none of the snapshots in the set.
469In other words, if all snapshots in the set have modified a given block in the
470parent, that block will not be sent; but if one or more snapshots have not
471modified a block in the parent, they will still reference the parent's block, so
472that block will be sent.
473Note that only user data will be redacted.
474.Pp
475When the redacted send stream is received, we will generate a redacted
476snapshot.
477Due to the nature of redaction, a redacted dataset can only be used in the
478following ways:
479.Bl -enum -width "a."
480.It
481To receive, as a clone, an incremental send from the original snapshot to one
482of the snapshots it was redacted with respect to.
483In this case, the stream will produce a valid dataset when received because all
484blocks that were redacted in the parent are guaranteed to be present in the
485child's send stream.
486This use case will produce a normal snapshot, which can be used just like other
487snapshots.
488.
489.It
490To receive an incremental send from the original snapshot to something
491redacted with respect to a subset of the set of snapshots the initial snapshot
492was redacted with respect to.
493In this case, each block that was redacted in the original is still redacted
494(redacting with respect to additional snapshots causes less data to be redacted
495(because the snapshots define what is permitted, and everything else is
496redacted)).
497This use case will produce a new redacted snapshot.
498.It
499To receive an incremental send from a redaction bookmark of the original
500snapshot that was created when redacting with respect to a subset of the set of
501snapshots the initial snapshot was created with respect to
502anything else.
503A send stream from such a redaction bookmark will contain all of the blocks
504necessary to fill in any redacted data, should it be needed, because the sending
505system is aware of what blocks were originally redacted.
506This will either produce a normal snapshot or a redacted one, depending on
507whether the new send stream is redacted.
508.It
509To receive an incremental send from a redacted version of the initial
510snapshot that is redacted with respect to a subject of the set of snapshots the
511initial snapshot was created with respect to.
512A send stream from a compatible redacted dataset will contain all of the blocks
513necessary to fill in any redacted data.
514This will either produce a normal snapshot or a redacted one, depending on
515whether the new send stream is redacted.
516.It
517To receive a full send as a clone of the redacted snapshot.
518Since the stream is a full send, it definitionally contains all the data needed
519to create a new dataset.
520This use case will either produce a normal snapshot or a redacted one, depending
521on whether the full send stream was redacted.
522.El
523.Pp
524These restrictions are detected and enforced by
525.Nm zfs Cm receive ;
526a redacted send stream will contain the list of snapshots that the stream is
527redacted with respect to.
528These are stored with the redacted snapshot, and are used to detect and
529correctly handle the cases above.
530Note that for technical reasons,
531raw sends and redacted sends cannot be combined at this time.
532.It Xo
533.Nm zfs
534.Cm send
535.Op Fl PVenv
536.Fl t
537.Ar receive_resume_token
538.Xc
539Creates a send stream which resumes an interrupted receive.
540The
541.Ar receive_resume_token
542is the value of this property on the filesystem or volume that was being
543received into.
544See the documentation for
545.Nm zfs Cm receive Fl s
546for more details.
547.It Xo
548.Nm zfs
549.Cm send
550.Op Fl PVnv
551.Op Fl i Ar snapshot Ns | Ns Ar bookmark
552.Fl S
553.Ar filesystem
554.Xc
555Generate a send stream from a dataset that has been partially received.
556.Bl -tag -width "-L"
557.It Fl S , -saved
558This flag requires that the specified filesystem previously received a resumable
559send that did not finish and was interrupted.
560In such scenarios this flag
561enables the user to send this partially received state.
562Using this flag will always use the last fully received snapshot
563as the incremental source if it exists.
564.El
565.It Xo
566.Nm zfs
567.Cm redact
568.Ar snapshot redaction_bookmark
569.Ar redaction_snapshot Ns …
570.Xc
571Generate a new redaction bookmark.
572In addition to the typical bookmark information, a redaction bookmark contains
573the list of redacted blocks and the list of redaction snapshots specified.
574The redacted blocks are blocks in the snapshot which are not referenced by any
575of the redaction snapshots.
576These blocks are found by iterating over the metadata in each redaction snapshot
577to determine what has been changed since the target snapshot.
578Redaction is designed to support redacted zfs sends; see the entry for
579.Nm zfs Cm send
580for more information on the purpose of this operation.
581If a redact operation fails partway through (due to an error or a system
582failure), the redaction can be resumed by rerunning the same command.
583.El
584.Ss Redaction
585ZFS has support for a limited version of data subsetting, in the form of
586redaction.
587Using the
588.Nm zfs Cm redact
589command, a
590.Sy redaction bookmark
591can be created that stores a list of blocks containing sensitive information.
592When provided to
593.Nm zfs Cm send ,
594this causes a
595.Sy redacted send
596to occur.
597Redacted sends omit the blocks containing sensitive information,
598replacing them with REDACT records.
599When these send streams are received, a
600.Sy redacted dataset
601is created.
602A redacted dataset cannot be mounted by default, since it is incomplete.
603It can be used to receive other send streams.
604In this way datasets can be used for data backup and replication,
605with all the benefits that zfs send and receive have to offer,
606while protecting sensitive information from being
607stored on less-trusted machines or services.
608.Pp
609For the purposes of redaction, there are two steps to the process.
610A redact step, and a send/receive step.
611First, a redaction bookmark is created.
612This is done by providing the
613.Nm zfs Cm redact
614command with a parent snapshot, a bookmark to be created, and a number of
615redaction snapshots.
616These redaction snapshots must be descendants of the parent snapshot,
617and they should modify data that is considered sensitive in some way.
618Any blocks of data modified by all of the redaction snapshots will
619be listed in the redaction bookmark, because it represents the truly sensitive
620information.
621When it comes to the send step, the send process will not send
622the blocks listed in the redaction bookmark, instead replacing them with
623REDACT records.
624When received on the target system, this will create a
625redacted dataset, missing the data that corresponds to the blocks in the
626redaction bookmark on the sending system.
627The incremental send streams from
628the original parent to the redaction snapshots can then also be received on
629the target system, and this will produce a complete snapshot that can be used
630normally.
631Incrementals from one snapshot on the parent filesystem and another
632can also be done by sending from the redaction bookmark, rather than the
633snapshots themselves.
634.Pp
635In order to make the purpose of the feature more clear, an example is provided.
636Consider a zfs filesystem containing four files.
637These files represent information for an online shopping service.
638One file contains a list of usernames and passwords, another contains purchase
639histories,
640a third contains click tracking data, and a fourth contains user preferences.
641The owner of this data wants to make it available for their development teams to
642test against, and their market research teams to do analysis on.
643The development teams need information about user preferences and the click
644tracking data, while the market research teams need information about purchase
645histories and user preferences.
646Neither needs access to the usernames and passwords.
647However, because all of this data is stored in one ZFS filesystem,
648it must all be sent and received together.
649In addition, the owner of the data
650wants to take advantage of features like compression, checksumming, and
651snapshots, so they do want to continue to use ZFS to store and transmit their
652data.
653Redaction can help them do so.
654First, they would make two clones of a snapshot of the data on the source.
655In one clone, they create the setup they want their market research team to see;
656they delete the usernames and passwords file,
657and overwrite the click tracking data with dummy information.
658In another, they create the setup they want the development teams
659to see, by replacing the passwords with fake information and replacing the
660purchase histories with randomly generated ones.
661They would then create a redaction bookmark on the parent snapshot,
662using snapshots on the two clones as redaction snapshots.
663The parent can then be sent, redacted, to the target
664server where the research and development teams have access.
665Finally, incremental sends from the parent snapshot to each of the clones can be
666sent
667to and received on the target server; these snapshots are identical to the
668ones on the source, and are ready to be used, while the parent snapshot on the
669target contains none of the username and password data present on the source,
670because it was removed by the redacted send operation.
671.
672.Sh EXAMPLES
673.\" These are, respectively, examples 12, 13 from zfs.8
674.\" Make sure to update them bidirectionally
675.Ss Example 1 : No Remotely Replicating ZFS Data
676The following commands send a full stream and then an incremental stream to a
677remote machine, restoring them into
678.Em poolB/received/fs@a
679and
680.Em poolB/received/fs@b ,
681respectively.
682.Em poolB
683must contain the file system
684.Em poolB/received ,
685and must not initially contain
686.Em poolB/received/fs .
687.Bd -literal -compact -offset Ds
688.No # Nm zfs Cm send Ar pool/fs@a |
689.No "   " Nm ssh Ar host Nm zfs Cm receive Ar poolB/received/fs Ns @ Ns Ar a
690.No # Nm zfs Cm send Fl i Ar a pool/fs@b |
691.No "   " Nm ssh Ar host Nm zfs Cm receive Ar poolB/received/fs
692.Ed
693.
694.Ss Example 2 : No Using the Nm zfs Cm receive Fl d No Option
695The following command sends a full stream of
696.Ar poolA/fsA/fsB@snap
697to a remote machine, receiving it into
698.Ar poolB/received/fsA/fsB@snap .
699The
700.Ar fsA/fsB@snap
701portion of the received snapshot's name is determined from the name of the sent
702snapshot.
703.Ar poolB
704must contain the file system
705.Ar poolB/received .
706If
707.Ar poolB/received/fsA
708does not exist, it is created as an empty file system.
709.Bd -literal -compact -offset Ds
710.No # Nm zfs Cm send Ar poolA/fsA/fsB@snap |
711.No "   " Nm ssh Ar host Nm zfs Cm receive Fl d Ar poolB/received
712.Ed
713.
714.Sh SEE ALSO
715.Xr zfs-bookmark 8 ,
716.Xr zfs-receive 8 ,
717.Xr zfs-redact 8 ,
718.Xr zfs-snapshot 8
719