1 .\" Copyright (c) 2007 The DragonFly Project. All rights reserved.
3 .\" This code is derived from software contributed to The DragonFly Project
4 .\" by Matthew Dillon <dillon@backplane.com>
6 .\" Redistribution and use in source and binary forms, with or without
7 .\" modification, are permitted provided that the following conditions
10 .\" 1. Redistributions of source code must retain the above copyright
11 .\" notice, this list of conditions and the following disclaimer.
12 .\" 2. Redistributions in binary form must reproduce the above copyright
13 .\" notice, this list of conditions and the following disclaimer in
14 .\" the documentation and/or other materials provided with the
16 .\" 3. Neither the name of The DragonFly Project nor the names of its
17 .\" contributors may be used to endorse or promote products derived
18 .\" from this software without specific, prior written permission.
20 .\" THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
21 .\" ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
22 .\" LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
23 .\" FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
24 .\" COPYRIGHT HOLDERS OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
25 .\" INCIDENTAL, SPECIAL, EXEMPLARY OR CONSEQUENTIAL DAMAGES (INCLUDING,
26 .\" BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
27 .\" LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED
28 .\" AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
29 .\" OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
30 .\" OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
33 .\" $DragonFly: src/sbin/hammer/hammer.8,v 1.58 2008/11/13 02:04:27 dillon Exp $
40 .Nd HAMMER file system utility
49 .\" .Op Fl s Ar linkpath
53 .Op Fl C Ar cachesize Ns Op Ns Cm \&: Ns Ar readahead
58 This manual page documents the
60 utility which provides miscellaneous functions related to managing a
63 For a general introduction to the
65 file system, its features, and
66 examples on how to set up and maintain one, see
69 The options are as follows:
70 .Bl -tag -width indent
74 Tell the mirror commands to use a 2-way protocol, which allows
75 automatic negotiation of transaction id ranges.
76 This option is automatically enabled by the
80 Specify a bandwidth limit in bytes per second for mirroring streams.
81 This option is typically used to prevent batch mirroring operations from
82 loading down the machine.
83 The bandwidth may be suffixed with
87 to specify values in kilobytes, megabytes, and gigabytes per second.
88 If no suffix is specified, bytes per second is assumed.
90 Unfortunately this is only applicable to the pre-compression bandwidth
91 when compression is used, so a better solution would probably be to
98 When pruning, rebalancing or reblocking you can tell the utility
99 to start at the object id stored in the specified file.
100 If the file does not exist
102 will start at the beginning.
106 specific period of time and is unable to complete the operation it will
107 write out the current object id so the next run can pick up where it left off.
110 runs to completion it will delete
113 Specify the volumes making up a
117 is a colon-separated list of devices, each specifying a
121 When maintaining a streaming mirroring this option specifies the
122 minimum delay after a batch ends before the next batch is allowed
124 The default is five seconds.
131 specification for the source and/or destination.
133 Decrease verboseness.
134 May be specified multiple times.
136 Specify recursion for those commands which support it.
138 When pruning, rebalancing or reblocking you can tell the utility to stop
139 after a certain period of time.
140 This option is used along with the
142 option to prune, rebalance or reblock incrementally.
144 Increase verboseness.
145 May be specified multiple times.
147 Force "yes" for any interactive question.
151 will not attempt to break-up large initial bulk transfers into smaller
153 This can save time but if the link is lost in the middle of the
154 initial bulk transfer you will have to start over from scratch.
155 This option is not recommended.
156 For more information see the
159 .It Fl C Ar cachesize Ns Op Ns Cm \&: Ns Ar readahead
160 Set the memory cache size for any raw
167 for megabytes is allowed,
168 else the cache size is specified in bytes.
170 The read-behind/read-ahead defaults to 4
174 This option is typically only used with diagnostic commands
175 as kernel-supported commands will use the kernel's buffer cache.
176 .It Fl S Ar splitsize
177 Specify the bulk splitup size in bytes for mirroring streams.
182 will do an initial run-through of the data to calculate good
183 transaction ids to cut up the bulk transfers, creating
184 restart points in case the stream is interrupted.
185 If we don't do this and the stream is interrupted it might
186 have to start all over again.
187 The default is a splitsize of 4G.
189 At the moment the run-through is disk-bandwidth-heavy but some
190 future version will limit the run-through to just the B-Tree
191 records and not the record data.
193 The splitsize may be suffixed with
197 to specify values in kilobytes, megabytes, or gigabytes.
198 If no suffix is specified, bytes is assumed.
200 When mirroring very large filesystems the minimum recommended
202 A small split size may wind up generating a great deal of overhead
203 but very little actual incremental data and is not recommended.
205 Enable compression for any remote ssh specifications.
208 option has already been reserved for other purposes so we had to use
210 This option is typically used with the mirroring directives.
212 Force "yes" for any interactive question.
215 The commands are as follows:
216 .Bl -tag -width indent
217 .\" ==== synctid ====
218 .It Cm synctid Ar filesystem Op Cm quick
219 Generates a guaranteed, formal 64 bit transaction id representing the
220 current state of the specified
223 The file system will be synced to the media.
227 keyword is specified the file system will be soft-synced, meaning that a
228 crash might still undo the state of the file system as of the transaction
229 id returned but any new modifications will occur after the returned
230 transaction id as expected.
232 This operation does not create a snapshot.
233 It is meant to be used
234 to track temporary fine-grained changes to a subset of files and
235 will only remain valid for
237 snapshot access purposes for the
239 period configured for the PFS.
240 If you desire a real snapshot then the
242 directive may be what you are looking for.
244 .It Cm bstats Op Ar interval
247 B-Tree statistics until interrupted.
250 seconds between each display.
251 The default interval is one second.
252 .\" ==== iostats ====
253 .It Cm iostats Op Ar interval
257 statistics until interrupted.
260 seconds between each display.
261 The default interval is one second.
262 .\" ==== history ====
263 .It Cm history Ar path ...
264 Show the modification history for
266 file's inode and data.
267 .\" ==== blockmap ====
269 Dump the blockmap for the file system.
272 blockmap is two-layer
273 blockmap representing the maximum possible file system size of 1 Exabyte.
274 Needless to say the second layer is only present for blocks which exist.
276 blockmap represents 8-Megabyte blocks, called big-blocks.
277 Each big-block has an append
278 point, a free byte count, and a typed zone id which allows content to be
279 reverse engineered to some degree.
283 allocations are essentially appended to a selected big-block using
284 the append offset and deducted from the free byte count.
285 When space is freed the free byte count is adjusted but
287 does not track holes in big-blocks for reallocation.
288 A big-block must be completely freed, either
289 through normal file system operations or through reblocking, before
292 Data blocks can be shared by deducting the space used from the free byte
293 count for each shared references, though
295 does not yet make use of this feature.
296 This means the free byte count can legally go negative.
298 This command needs the
301 .\" ==== checkmap ====
303 Check the blockmap allocation count.
305 will scan the B-Tree, collect allocation information, and
306 construct a blockmap in-memory. It will then check that blockmap
307 against the on-disk blockmap.
309 .It Cm show Op Ar lo Ns Cm \&: Ns Ar objid
311 By default this command will validate all B-Tree
312 linkages and CRCs, including data CRCs, and will report the most verbose
313 information it can dig up.
314 Any errors will show up with a
316 in column 1 along with various
319 If you specify a localization field or a localization:obj_id field,
320 .Ar lo Ns Cm \&: Ns Ar objid ,
322 search for the key printing nodes as it recurses down, and then
323 will iterate forwards.
324 These fields are specified in HEX.
325 Note that the pfsid is the top 16 bits of the 32 bit localization
326 field so PFS #1 would be 00010000.
330 the command will report less information about the inode contents.
334 the command will not report the content of the inode or other typed
339 the command will not report volume header information, big-block fill
340 ratios, mirror transaction ids, or report or check data CRCs.
341 B-Tree CRCs and linkages are still checked.
343 This command needs the
346 .\" ==== show-undo ====
352 This command needs the
356 .\" Dump the B-Tree, record, large-data, and small-data blockmaps, showing
357 .\" physical block assignments and free space percentages.
358 .\" ==== namekey1 ====
359 .It Cm namekey1 Ar filename
362 64 bit directory hash for the specified file name, using
363 the original directory hash algorithm in version 1 of the file system.
364 The low 32 bits are used as an iterator for hash collisions and will be
366 .\" ==== namekey2 ====
367 .It Cm namekey2 Ar filename
370 64 bit directory hash for the specified file name, using
371 the new directory hash algorithm in version 2 of the file system.
372 The low 32 bits are still used as an iterator but will start out containing
373 part of the hash key.
374 .\" ==== namekey32 ====
375 .It Cm namekey32 Ar filename
376 Generate the top 32 bits of a
378 64 bit directory hash for the specified file name.
381 Shows extended information about all the mounted
384 The information is divided into sections:
385 .Bl -tag -width indent
386 .It Volume identification
387 General information, like the label of the
389 filesystem, the number of volumes it contains, the FSID, and the
392 .It Big block information
393 Big block statistics, such as total, used, reserved and free big blocks.
394 .It Space information
395 Information about space used on the filesystem.
396 Currently total size, used, reserved and free space are displayed.
398 Basic information about the PFSs currently present on a
403 is the ID of the PFS, with 0 being the root PFS.
405 is the current snapshot count on the PFS.
407 displays the mount point of the PFS is currently mounted on (if any).
409 .\" ==== cleanup ====
410 .It Cm cleanup Op Ar filesystem ...
411 This is a meta-command which executes snapshot, prune, rebalance and reblock
412 commands on the specified
417 is specified this command will clean-up all
419 file systems in use, including PFS's.
420 To do this it will scan all
424 mounts, extract PFS id's, and clean-up each PFS found.
426 This command will access a snapshots
427 directory and a configuration file for each
429 creating them if necessary.
430 .Bl -tag -width indent
431 .It Nm HAMMER No version 2-
432 The configuration file is
434 in the snapshots directory which defaults to
435 .Pa <pfs>/snapshots .
436 .It Nm HAMMER No version 3+
437 The configuration file is saved in file system meta-data, see
440 The snapshots directory defaults to
441 .Pa /var/hammer/<pfs>
442 .Pa ( /var/hammer/root
446 The format of the configuration file is:
447 .Bd -literal -offset indent
448 snapshots <period> <retention-time> [any]
449 prune <period> <max-runtime>
450 rebalance <period> <max-runtime>
451 reblock <period> <max-runtime>
452 recopy <period> <max-runtime>
456 .Bd -literal -offset indent
457 snapshots 1d 60d # 0d 0d for PFS /tmp, /var/tmp, /usr/obj
464 Time is given with a suffix of
470 meaning day, hour, minute and second.
474 directive has a period of 0 and a retention time of 0
475 then snapshot generation is disabled, removal of old snapshots are
476 disabled, and prunes will use
477 .Cm prune-everything .
480 directive has a period of 0 but a non-zero retention time
481 then this command will not create any new snapshots but will remove old
482 snapshots it finds based on the retention time.
484 By default only snapshots in the form
485 .Ql snap- Ns Ar yyyymmdd Ns Op - Ns Ar HHMM
489 directive is specified as a third argument on the
491 config line then any softlink of the form
492 .Ql *- Ns Ar yyyymmdd Ns Op - Ns Ar HHMM
494 .Ql *. Ns Ar yyyymmdd Ns Op - Ns Ar HHMM
497 A prune max-runtime of 0 means unlimited.
499 If period hasn't passed since the previous
502 For example a day has passed when midnight is passed (localtime).
510 The default configuration file will create a daily snapshot, do a daily
511 pruning, rebalancing and reblocking run and a monthly recopy run.
512 Reblocking is defragmentation with a level of 95%,
513 and recopy is full defragmentation.
515 By default prune and rebalance operations are time limited to 5 minutes,
516 reblock operations to a bit over 5 minutes,
517 and recopy operations to a bit over 10 minutes.
518 Reblocking and recopy runs are each broken down into four separate functions:
519 btree, inodes, dirs and data.
520 Each function is time limited to the time given in the configuration file,
521 but the btree, inodes and dirs functions usually does not take very long time,
522 full defragmentation is always used for these three functions.
523 Also note that this directive will by default disable snapshots on
530 The defaults may be adjusted by modifying the configuration file.
531 The pruning and reblocking commands automatically maintain a cyclefile
532 for incremental operation.
533 If you interrupt (^C) the program the cyclefile will be updated,
535 may continue to run in the background for a few seconds until the
537 ioctl detects the interrupt.
540 PFS option can be set to use another location for the snapshots directory.
542 Work on this command is still in progress.
544 An ability to remove snapshots dynamically as the
545 file system becomes full.
547 .It Cm config Op Ar filesystem Op Ar configfile
550 Show or change configuration for
552 If zero or one arguments are specified this function dumps the current
553 configuration file to stdout.
554 Zero arguments specifies the PFS containing the current directory.
555 This configuration file is stored in file system meta-data.
556 If two arguments are specified this function installs a new config file.
560 versions less than 3 the configuration file is by default stored in
561 .Pa <pfs>/snapshots/config ,
562 but in all later versions the configuration file is stored in file system
564 .\" ==== viconfig ====
565 .It Cm viconfig Op Ar filesystem
568 Edit the configuration file and reinstall into file system meta-data when done.
569 Zero arguments specifies the PFS containing the current directory.
570 .\" ==== volume-add ====
571 .It Cm volume-add Ar device Ar filesystem
572 This command will format
574 and add all of its space to
578 All existing data contained on
580 will be destroyed by this operation!
585 file system, formatting will be denied.
586 You can overcome this sanity check
589 to erase the beginning sectors of the device.
590 Also remember that you have to specify
592 together with any other device that make up the file system,
597 .\" ==== volume-del ====
598 .It Cm volume-del Ar device Ar filesystem
599 This command will remove volume
604 Remember that you have to remove
606 from the colon-separated list in
610 .\" ==== snapshot ====
611 .It Cm snapshot Oo Ar filesystem Oc Ar snapshot-dir
612 .It Cm snapshot Ar filesystem Ar snapshot-dir Op Ar note
613 Takes a snapshot of the file system either explicitly given by
615 or implicitly derived from the
617 argument and creates a symlink in the directory provided by
619 pointing to the snapshot.
622 is not a directory, it is assumed to be a format string passed to
624 with the current time as parameter.
627 refers to an existing directory, a default format string of
629 is assumed and used as name for the newly created symlink.
631 Snapshot is a per PFS operation, so a
633 file system and each PFS in it have to be snapshot separately.
635 Example, assuming that
643 are file systems on their own, the following invocations:
644 .Bd -literal -offset indent
645 hammer snapshot /mysnapshots
647 hammer snapshot /mysnapshots/%Y-%m-%d
649 hammer snapshot /obj /mysnapshots/obj-%Y-%m-%d
651 hammer snapshot /usr /my/snaps/usr "note"
654 Would create symlinks similar to:
655 .Bd -literal -offset indent
656 /mysnapshots/snap-20080627-1210 -> /@@0x10d2cd05b7270d16
658 /mysnapshots/2008-06-27 -> /@@0x10d2cd05b7270d16
660 /mysnapshots/obj-2008-06-27 -> /obj@@0x10d2cd05b7270d16
662 /my/snaps/usr/snap-20080627-1210 -> /usr@@0x10d2cd05b7270d16
667 version 3+ file system the snapshot is also recorded in file system meta-data
668 along with the optional
674 .It Cm snap Ar path Op Ar note
677 Create a snapshot for the PFS containing
679 and create a snapshot softlink.
680 If the path specified is a
681 directory a standard snapshot softlink will be created in the directory.
682 The snapshot softlink points to the base of the mounted PFS.
683 .It Cm snaplo Ar path Op Ar note
686 Create a snapshot for the PFS containing
688 and create a snapshot softlink.
689 If the path specified is a
690 directory a standard snapshot softlink will be created in the directory.
691 The snapshot softlink points into the directory it is contained in.
692 .It Cm snapq Ar dir Op Ar note
695 Create a snapshot for the PFS containing the specified directory but do
696 not create a softlink.
697 Instead output a path which can be used to access
698 the directory via the snapshot.
700 An absolute or relative path may be specified.
701 The path will be used as-is as a prefix in the path output to stdout.
703 snap and snapshot directives the snapshot transaction id will be registered
704 in the file system meta-data.
705 .It Cm snaprm Ar path Ar ...
706 .It Cm snaprm Ar transid Ar ...
707 .It Cm snaprm Ar filesystem Ar transid Ar ...
710 Remove a snapshot given its softlink or transaction id.
711 If specifying a transaction id
712 the snapshot is removed from file system meta-data but you are responsible
713 for removing any related softlinks.
715 If a softlink path is specified the filesystem and transaction id
716 is derived from the contents of the softlink.
717 If just a transaction id is specified it is assumed to be a snapshot
718 in the HAMMER filesystem you are currently chdir'd into.
719 You can also specify the filesystem and transaction id explicitly.
720 .It Cm snapls Op Ar path ...
723 Dump the snapshot meta-data for PFSs containing each
725 listing all available snapshots and their notes.
726 If no arguments are specified snapshots for the PFS containing the
727 current directory are listed.
728 This is the definitive list of snapshots for the file system.
730 .It Cm prune Ar softlink-dir
731 Prune the file system based on previously created snapshot softlinks.
732 Pruning is the act of deleting file system history.
735 command will delete file system history such that
736 the file system state is retained for the given snapshots,
737 and all history after the latest snapshot.
738 By setting the per PFS parameter
740 history is guaranteed to be saved at least this time interval.
741 All other history is deleted.
743 The target directory is expected to contain softlinks pointing to
744 snapshots of the file systems you wish to retain.
745 The directory is scanned non-recursively and the mount points and
746 transaction ids stored in the softlinks are extracted and sorted.
747 The file system is then explicitly pruned according to what is found.
748 Cleaning out portions of the file system is as simple as removing a
749 snapshot softlink and then running the
753 As a safety measure pruning only occurs if one or more softlinks are found
756 snapshot id extension.
757 Currently the scanned softlink directory must contain softlinks pointing
761 The softlinks may specify absolute or relative paths.
762 Softlinks must use 20-character
764 transaction ids, as might be returned from
765 .Nm Cm synctid Ar filesystem .
767 Pruning is a per PFS operation, so a
769 file system and each PFS in it have to be pruned separately.
771 Note that pruning a file system may not immediately free-up space,
772 though typically some space will be freed if a large number of records are
774 The file system must be reblocked to completely recover all available space.
776 Example, lets say your that you didn't set
778 and snapshot directory contains the following links:
779 .Bd -literal -offset indent
780 lrwxr-xr-x 1 root wheel 29 May 31 17:57 snap1 ->
781 /usr/obj/@@0x10d2cd05b7270d16
783 lrwxr-xr-x 1 root wheel 29 May 31 17:58 snap2 ->
784 /usr/obj/@@0x10d2cd13f3fde98f
786 lrwxr-xr-x 1 root wheel 29 May 31 17:59 snap3 ->
787 /usr/obj/@@0x10d2cd222adee364
790 If you were to run the
792 command on this directory, then the
795 mount will be pruned to retain the above three snapshots.
796 In addition, history for modifications made to the file system older than
797 the oldest snapshot will be destroyed and history for potentially fine-grained
798 modifications made to the file system more recently than the most recent
799 snapshot will be retained.
801 If you then delete the
803 softlink and rerun the
806 history for modifications pertaining to that snapshot would be destroyed.
810 file system versions 3+ this command also scans the snapshots stored
811 in the file system meta-data and includes them in the prune.
812 .\" ==== prune-everything ====
813 .It Cm prune-everything Ar filesystem
814 This command will remove all historical records from the file system.
815 This directive is not normally used on a production system.
817 This command does not remove snapshot softlinks but will delete all
818 snapshots recorded in file system meta-data (for file system version 3+).
819 The user is responsible for deleting any softlinks.
821 Pruning is a per PFS operation, so a
823 file system and each PFS in it have to be pruned separately.
824 .\" ==== rebalance ====
825 .It Cm rebalance Ar filesystem Op Ar saturation_percentage
826 This command will rebalance the B-Tree, nodes with small number of
827 elements will be combined and element counts will be smoothed out
830 The saturation percentage is between 50% and 100%.
831 The default is 75% (the
833 suffix is not needed).
835 Rebalancing is a per PFS operation, so a
837 file system and each PFS in it have to be rebalanced separately.
838 .\" ==== reblock* ====
839 .It Cm reblock Ar filesystem Op Ar fill_percentage
840 .It Cm reblock-btree Ar filesystem Op Ar fill_percentage
841 .It Cm reblock-inodes Ar filesystem Op Ar fill_percentage
842 .It Cm reblock-dirs Ar filesystem Op Ar fill_percentage
843 .It Cm reblock-data Ar filesystem Op Ar fill_percentage
844 Attempt to defragment and free space for reuse by reblocking a live
847 Big-blocks cannot be reused by
849 until they are completely free.
850 This command also has the effect of reordering all elements, effectively
851 defragmenting the file system.
853 The default fill percentage is 100% and will cause the file system to be
854 completely defragmented.
855 All specified element types will be reallocated and rewritten.
856 If you wish to quickly free up space instead try specifying
857 a smaller fill percentage, such as 90% or 80% (the
859 suffix is not needed).
861 Since this command may rewrite the entire contents of the disk it is
862 best to do it incrementally from a
868 options to limit the run time.
869 The file system would thus be defragmented over long period of time.
871 It is recommended that separate invocations be used for each data type.
872 B-Tree nodes, inodes, and directories are typically the most important
873 elements needing defragmentation.
874 Data can be defragmented over a longer period of time.
876 Reblocking is a per PFS operation, so a
878 file system and each PFS in it have to be reblocked separately.
879 .\" ==== pfs-status ====
880 .It Cm pfs-status Ar dirpath ...
881 Retrieve the mirroring configuration parameters for the specified
883 file systems or pseudo-filesystems (PFS's).
884 .\" ==== pfs-master ====
885 .It Cm pfs-master Ar dirpath Op Ar options
886 Create a pseudo-filesystem (PFS) inside a
889 Up to 65535 such file systems can be created.
890 Each PFS uses an independent inode numbering space making it suitable
891 for use as a replication source or target.
895 directive creates a PFS that you can read, write, and use as a mirroring
898 It is recommended to use a
900 mount to access a PFS, for more information see
902 .\" ==== pfs-slave ====
903 .It Cm pfs-slave Ar dirpath Op Ar options
904 Create a pseudo-filesystem (PFS) inside a
907 Up to 65535 such file systems can be created.
908 Each PFS uses an independent inode numbering space making it suitable
909 for use as a replication source or target.
913 directive creates a PFS that you can use as a mirroring target.
914 You will not be able to access a slave PFS until you have completed the
915 first mirroring operation with it as the target (its root directory will
916 not exist until then).
918 Access to the pfs-slave via the special softlink, as described in the
923 dynamically modify the snapshot transaction id by returning a dynamic result
928 A PFS can only be truly destroyed with the
931 Removing the softlink will not destroy the underlying PFS.
933 It is recommended to use a
935 mount to access a PFS, for more information see
937 .\" ==== pfs-update ====
938 .It Cm pfs-update Ar dirpath Op Ar options
939 Update the configuration parameters for an existing
941 file system or pseudo-filesystem.
942 Options that may be specified:
943 .Bl -tag -width indent
944 .It Cm sync-beg-tid= Ns Ar 0x16llx
945 This is the automatic snapshot access starting transaction id for
947 This parameter is normally updated automatically by the
951 It is important to note that accessing a mirroring slave
952 with a transaction id greater than the last fully synchronized transaction
953 id can result in an unreliable snapshot since you will be accessing
954 data that is still undergoing synchronization.
956 Manually modifying this field is dangerous and can result in a broken mirror.
957 .It Cm sync-end-tid= Ns Ar 0x16llx
958 This is the current synchronization point for mirroring slaves.
959 This parameter is normally updated automatically by the
963 Manually modifying this field is dangerous and can result in a broken mirror.
964 .It Cm shared-uuid= Ns Ar uuid
965 Set the shared UUID for this file system.
966 All mirrors must have the same shared UUID.
967 For safety purposes the
969 directives will refuse to operate on a target with a different shared UUID.
971 Changing the shared UUID on an existing, non-empty mirroring target,
972 including an empty but not completely pruned target,
973 can lead to corruption of the mirroring target.
974 .It Cm unique-uuid= Ns Ar uuid
975 Set the unique UUID for this file system.
976 This UUID should not be used anywhere else,
977 even on exact copies of the file system.
978 .It Cm label= Ns Ar string
979 Set a descriptive label for this file system.
980 .It Cm snapshots= Ns Ar string
981 Specify the snapshots directory which
984 will use to manage this PFS.
985 .Bl -tag -width indent
986 .It Nm HAMMER No version 2-
987 The snapshots directory does not need to be configured for
988 PFS masters and will default to
989 .Pa <pfs>/snapshots .
991 PFS slaves are mirroring slaves so you cannot configure a snapshots
992 directory on the slave itself to be managed by the slave's machine.
993 In fact, the slave will likely have a
995 sub-directory mirrored
996 from the master, but that directory contains the configuration the master
997 is using for its copy of the file system, not the configuration that we
998 want to use for our slave.
1000 It is recommended that
1001 .Pa <fs>/var/slaves/<name>
1002 be configured for a PFS slave, where
1008 is an appropriate label.
1009 .It Nm HAMMER No version 3+
1010 The snapshots directory does not need to be configured for PFS masters or
1012 The snapshots directory defaults to
1013 .Pa /var/hammer/<pfs>
1014 .Pa ( /var/hammer/root
1018 You can control snapshot retention on your slave independent of the master.
1019 .It Cm snapshots-clear
1022 directory path for this PFS.
1023 .It Cm prune-min= Ns Ar N Ns Cm d
1024 .It Cm prune-min= Ns Oo Ar N Ns Cm d/ Oc Ns \
1025 Ar hh Ns Op Cm \&: Ns Ar mm Ns Op Cm \&: Ns Ar ss
1026 Set the minimum fine-grained data retention period.
1028 always retains fine-grained history up to the most recent snapshot.
1029 You can extend the retention period further by specifying a non-zero
1031 Any snapshot softlinks within the retention period are ignored
1032 for the purposes of pruning (the fine grained history is retained).
1033 Number of days, hours, minutes and seconds are given as
1038 Because the transaction id in the snapshot softlink cannot be used
1039 to calculate a timestamp,
1041 uses the earlier of the
1045 field of the softlink to
1046 determine which snapshots fall within the retention period.
1047 Users must be sure to retain one of these two fields when manipulating
1050 .\" ==== pfs-upgrade ====
1051 .It Cm pfs-upgrade Ar dirpath
1052 Upgrade a PFS from slave to master operation.
1053 The PFS will be rolled back to the current end synchronization transaction id
1054 (removing any partial synchronizations), and will then become writable.
1058 currently supports only single masters and using
1059 this command can easily result in file system corruption
1060 if you don't know what you are doing.
1062 This directive will refuse to run if any programs have open descriptors
1063 in the PFS, including programs chdir'd into the PFS.
1064 .\" ==== pfs-downgrade ====
1065 .It Cm pfs-downgrade Ar dirpath
1066 Downgrade a master PFS from master to slave operation.
1067 The PFS becomes read-only and access will be locked to its
1070 This directive will refuse to run if any programs have open descriptors
1071 in the PFS, including programs chdir'd into the PFS.
1072 .\" ==== pfs-destroy ====
1073 .It Cm pfs-destroy Ar dirpath
1074 This permanently destroys a PFS.
1076 This directive will refuse to run if any programs have open descriptors
1077 in the PFS, including programs chdir'd into the PFS.
1078 .\" ==== mirror-read ====
1079 .It Cm mirror-read Ar filesystem Op Ar begin-tid
1080 Generate a mirroring stream to stdout.
1081 The stream ends when the transaction id space has been exhausted.
1082 .\" ==== mirror-read-stream ====
1083 .It Cm mirror-read-stream Ar filesystem Op Ar begin-tid
1084 Generate a mirroring stream to stdout.
1085 Upon completion the stream is paused until new data is synced to the
1088 Operation continues until the pipe is broken.
1091 command for more details.
1092 .\" ==== mirror-write ====
1093 .It Cm mirror-write Ar filesystem
1094 Take a mirroring stream on stdin.
1096 This command will fail if the
1098 configuration field for the two file systems do not match.
1101 command for more details.
1103 If the target PFS does not exist this command will ask you whether
1104 you want to create a compatible PFS slave for the target or not.
1105 .\" ==== mirror-dump ====
1111 to dump an ASCII representation of the mirroring stream.
1112 .\" ==== mirror-copy ====
1113 .\".It Cm mirror-copy Ar [[user@]host:]filesystem [[user@]host:]filesystem
1114 .It Cm mirror-copy \
1115 Oo Oo Ar user Ns Cm @ Oc Ns Ar host Ns Cm \&: Oc Ns Ar filesystem \
1116 Oo Oo Ar user Ns Cm @ Oc Ns Ar host Ns Cm \&: Oc Ns Ar filesystem
1117 This is a shortcut which pipes a
1122 If a remote host specification is made the program forks a
1128 on the appropriate host.
1129 The source may be a master or slave PFS, and the target must be a slave PFS.
1131 This command also establishes full duplex communication and turns on
1132 the 2-way protocol feature
1134 which automatically negotiates transaction id
1135 ranges without having to use a cyclefile.
1136 If the operation completes successfully the target PFS's
1139 Note that you must re-chdir into the target PFS to see the updated information.
1140 If you do not you will still be in the previous snapshot.
1142 If the target PFS does not exist this command will ask you whether
1143 you want to create a compatible PFS slave for the target or not.
1144 .\" ==== mirror-stream ====
1145 .\".It Cm mirror-stream Ar [[user@]host:]filesystem [[user@]host:]filesystem
1146 .It Cm mirror-stream \
1147 Oo Oo Ar user Ns Cm @ Oc Ns Ar host Ns Cm \&: Oc Ns Ar filesystem \
1148 Oo Oo Ar user Ns Cm @ Oc Ns Ar host Ns Cm \&: Oc Ns Ar filesystem
1149 This is a shortcut which pipes a
1150 .Cm mirror-read-stream
1154 This command works similarly to
1156 but does not exit after the initial mirroring completes.
1157 The mirroring operation will resume as changes continue to be made to the
1159 The command is commonly used with
1163 options to keep the mirroring target in sync with the source on a continuing
1166 If the pipe is broken the command will automatically retry after sleeping
1168 The time slept will be 15 seconds plus the time given in the
1172 This command also detects the initial-mirroring case and spends some
1173 time scanning the B-Tree to find good break points, allowing the initial
1174 bulk mirroring operation to be broken down into 100MB pieces.
1175 This means that the user can kill and restart the operation and it will
1176 not have to start from scratch once it has gotten past the first chunk.
1179 option may be used to change the size of pieces and the
1181 option may be used to disable this feature and perform an initial bulk
1183 .\" ==== version ====
1184 .It Cm version Ar filesystem
1185 This command returns the
1187 file system version for the specified
1189 as well as the range of versions supported in the kernel.
1192 option may be used to remove the summary at the end.
1193 .\" ==== version-upgrade ====
1194 .It Cm version-upgrade Ar filesystem Ar version Op Cm force
1195 This command upgrades the
1200 Once upgraded a file system may not be downgraded.
1201 If you wish to upgrade a file system to a version greater or equal to the
1202 work-in-progress version number you must specify the
1205 Use of WIP versions should be relegated to testing and may require wiping
1206 the file system as development progresses, even though the WIP version might
1210 This command operates on the entire
1212 file system and is not a per PFS operation.
1213 All PFS's will be affected.
1214 .Bl -tag -width indent
1217 default version, first
1222 New directory entry layout.
1223 This version is using a new directory hash key.
1226 New snapshot management, using file system meta-data for saving
1227 configuration file and snapshots (transaction ids etc.).
1228 Also default snapshots directory has changed.
1232 New undo/redo/flush, giving HAMMER a much faster sync and fsync.
1235 .Sh PSEUDO-FILESYSTEM (PFS) NOTES
1236 The root of a PFS is not hooked into the primary
1238 file system as a directory.
1241 creates a special softlink called
1243 (exactly 10 characters long) in the primary
1247 then modifies the contents of the softlink as read by
1249 and thus what you see with an
1251 command or if you were to
1254 If the PFS is a master the link reflects the current state of the PFS.
1255 If the PFS is a slave the link reflects the last completed snapshot, and the
1256 contents of the link will change when the next snapshot is completed, and
1261 utility employs numerous safeties to reduce user foot-shooting.
1264 directive requires that the target be configured as a slave and that the
1266 field of the mirroring source and target match.
1267 .Sh UPGRADE INSTRUCTIONS HAMMER V1 TO V2
1268 This upgrade changes the way directory entries are stored.
1269 It is possible to upgrade a V1 file system to V2 in place, but
1270 directories created prior to the upgrade will continue to use
1273 Note that the slave mirroring code in the target kernel had bugs in
1274 V1 which can create an incompatible root directory on the slave.
1277 master created after the upgrade with a
1279 slave created prior to the upgrade.
1281 Any directories created after upgrading will use a new layout.
1282 .Sh UPGRADE INSTRUCTIONS HAMMER V2 TO V3
1283 This upgrade adds meta-data elements to the B-Tree.
1284 It is possible to upgrade a V2 file system to V3 in place.
1285 After issuing the upgrade be sure to run a
1288 to perform post-upgrade tasks.
1290 After making this upgrade running a
1295 directory for each PFS mount into
1296 .Pa /var/hammer/<pfs> .
1299 root mount will migrate
1302 .Pa /var/hammer/root .
1303 Migration occurs only once and only if you have not specified
1304 a snapshots directory in the PFS configuration.
1305 If you have specified a snapshots directory in the PFS configuration no
1306 automatic migration will occur.
1308 For slaves, if you desire, you can migrate your snapshots
1309 config to the new location manually and then clear the
1310 snapshot directory configuration in the slave PFS.
1311 The new snapshots hierarchy is designed to work with
1312 both master and slave PFSs equally well.
1314 In addition, the old config file will be moved to file system meta-data,
1315 editable via the new
1319 The old config file will be deleted.
1320 Migration occurs only once.
1322 The V3 file system has new
1324 directives for creating snapshots.
1325 All snapshot directives, including the original, will create
1326 meta-data entries for the snapshots and the pruning code will
1327 automatically incorporate these entries into its list and
1328 expire them the same way it expires softlinks.
1329 If you by accident blow away your snapshot softlinks you can use the
1331 directive to get a definitive list from the file system meta-data and
1332 regenerate them from that list.
1337 to backup file systems your scripts may be using the
1339 directive to generate transaction ids.
1340 This directive does not create a snapshot.
1341 You will have to modify your scripts to use the
1343 directive to generate the linkbuf for the softlink you create, or
1344 use one of the other
1349 directive will continue to work as expected and in V3 it will also
1350 record the snapshot transaction id in file system meta-data.
1351 You may also want to make use of the new
1353 tag for the meta-data.
1356 If you used to remove snapshot softlinks with
1358 you should probably start using the
1360 directive instead to also remove the related meta-data.
1361 The pruning code scans the meta-data so just removing the
1362 softlink is not sufficient.
1363 .Sh UPGRADE INSTRUCTIONS HAMMER V3 TO V4
1364 This upgrade changes undo/flush, giving faster sync.
1365 It is possible to upgrade a V3 file system to V4 in place.
1366 This upgrade reformats the UNDO FIFO (typically 1GB), so upgrade might take
1367 a minute or two depending.
1369 Version 4 allows the UNDO FIFO to be flushed without also having
1370 to flush the volume header, removing 2 of the 4 disk syncs typically
1373 and removing 1 of the 2 disk syncs typically
1374 required for a flush sequence.
1375 Version 4 also implements the REDO log (see below) which is capable
1376 of fsync()ing with either one disk flush or zero disk flushes.
1377 .Sh FSYNC FLUSH MODES
1379 implements five different fsync flush modes via the
1380 .Va vfs.hammer.fsync_mode
1383 version 4+ file systems.
1387 fsync mode 3 is set by default.
1388 REDO operation and recovery is enabled by default.
1389 .Bl -tag -width indent
1391 Full synchronous fsync semantics without REDO.
1394 will not generate REDOs.
1397 will completely sync
1398 the data and meta-data and double-flush the FIFO, including
1399 issuing two disk synchronization commands.
1400 The data is guaranteed
1401 to be on the media as of when
1404 Needless to say, this is slow.
1406 Relaxed asynchronous fsync semantics without REDO.
1408 This mode works the same as mode 0 except the last disk synchronization
1409 command is not issued.
1410 It is faster than mode 0 but not even remotely
1411 close to the speed you get with mode 2 or mode 3.
1413 Note that there is no chance of meta-data corruption when using this
1414 mode, it simply means that the data you wrote and then
1416 might not have made it to the media if the storage system crashes at a bad
1420 Full synchronous fsync semantics using REDO.
1421 NOTE: If not running
1422 a HAMMER version 4 filesystem or later mode 0 is silently used.
1425 will generate REDOs in the UNDO/REDO FIFO based on a heuristic.
1426 If this is sufficient to satisfy the
1428 operation the blocks
1429 will be written out and
1431 will wait for the I/Os to complete,
1432 and then followup with a disk sync command to guarantee the data
1433 is on the media before returning.
1434 This is slower than mode 3 and can result in significant disk or
1435 SSDs overheads, though not as bad as mode 0 or mode 1.
1438 Relaxed asynchronous fsync semantics using REDO.
1439 NOTE: If not running
1440 a HAMMER version 4 filesystem or later mode 1 is silently used.
1443 will generate REDOs in the UNDO/REDO FIFO based on a heuristic.
1444 If this is sufficient to satisfy the
1446 operation the blocks
1447 will be written out and
1449 will wait for the I/Os to complete,
1452 issue a disk synchronization command.
1454 Note that there is no chance of meta-data corruption when using this
1455 mode, it simply means that the data you wrote and then
1458 not have made it to the media if the storage system crashes at a bad
1461 This mode is the fastest production fsyncing mode available.
1462 This mode is equivalent to how the UFS fsync in the
1472 This mode is primarily designed
1473 for testing and should not be used on a production system.
1475 .Sh RESTORING FROM A SNAPSHOT BACKUP
1476 You restore a snapshot by copying it over to live, but there is a caveat.
1477 The mtime and atime fields for files accessed via a snapshot is locked
1478 to the ctime in order to keep the snapshot consistent, because neither
1479 mtime nor atime changes roll any history.
1481 In order to avoid unnecessary copying it is recommended that you use
1485 when doing the copyback. Also make sure you traverse the snapshot softlink
1486 by appending a ".", as in "<snapshotpath>/.", and you match up the directory
1491 If the following environment variables exist, they will be used by:
1492 .Bl -tag -width ".Ev EDITOR"
1494 The editor program specified in the variable
1496 will be invoked instead of the default editor, which is
1504 .Bl -tag -width ".It Pa <fs>/var/slaves/<name>" -compact
1505 .It Pa <pfs>/snapshots
1506 default per PFS snapshots directory
1509 .It Pa /var/hammer/<pfs>
1510 default per PFS snapshots directory (not root)
1513 .It Pa /var/hammer/root
1514 default snapshots directory for root directory
1517 .It Pa <snapshots>/config
1524 .It Pa <fs>/var/slaves/<name>
1525 recommended slave PFS snapshots directory
1534 .Xr periodic.conf 5 ,
1535 .Xr mount_hammer 8 ,
1541 utility first appeared in
1544 .An Matthew Dillon Aq dillon@backplane.com