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
47 .Op Fl C Ar cachesize Ns Op Ns Cm \&: Ns Ar readahead
50 .\" .Op Fl s Ar linkpath
57 This manual page documents the
59 utility which provides miscellaneous functions related to managing a
62 For a general introduction to the
64 file system, its features, and
65 examples on how to set up and maintain one, see
68 The options are as follows:
69 .Bl -tag -width indent
73 Tell the mirror commands to use a 2-way protocol, which allows
74 automatic negotiation of transaction id ranges.
75 This option is automatically enabled by the
81 will not attempt to break-up large initial bulk transfers into smaller pieces.
82 This can save time but if the link is lost in the middle of the
83 initial bulk transfer you will have to start over from scratch.
85 Specify a bandwidth limit in bytes per second for mirroring streams.
86 This option is typically used to prevent batch mirroring operations from
87 loading down the machine.
88 The bandwidth may be suffixed with
92 to specify values in kilobytes, megabytes, and gigabytes per second.
93 If no suffix is specified, bytes per second is assumed.
94 .It Fl C Ar cachesize Ns Op Ns Cm \&: Ns Ar readahead
95 Set the memory cache size for any raw
102 for megabytes is allowed,
103 else the cache size is specified in bytes.
105 The read-behind/read-ahead defaults to 4
109 This option is typically only used with diagnostic commands
110 as kernel-supported commands will use the kernel's buffer cache.
111 .It Fl c Ar cyclefile
112 When pruning and reblocking you can instruction
114 to start at the object id stored in the specified file.
115 If the file does not exist
117 will start at the beginning.
121 specific period of time and is unable to complete the operation it will
122 write out the current object id so the next run can pick up where it left off.
125 runs to completion it will delete
128 Specify the volumes making up a
132 is a colon-separated list of devices, each specifying a
136 When maintaining a streaming mirroring this option specifies the
137 minimum delay after a batch ends before the next batch is allowed
139 The default is five seconds.
141 This passes the -p <port> option to ssh when using a remote
142 specification for the source and/or destination.
144 Decrease verboseness.
145 May be specified multiple times.
147 Specify recursion for those commands which support it.
149 When pruning and reblocking you can tell the utility to stop after a
150 certain period of time.
151 This option is used along with the
153 option to prune or reblock a portion of the file system incrementally.
155 Increase verboseness.
156 May be specified multiple times.
158 Force "yes" for any interactive question.
161 The commands are as follows:
162 .Bl -tag -width indent
163 .\" ==== synctid ====
164 .It Cm synctid Ar filesystem Op Cm quick
165 Generates a guaranteed, formal 64 bit transaction id representing the
166 current state of the specified
169 The file system will be synced to the media.
173 keyword is specified the file system will be soft-synced, meaning that a
174 crash might still undo the state of the file system as of the transaction
175 id returned but any new modifications will occur after the returned
176 transaction id as expected.
178 This operation does not create a snapshot.
179 It is meant to be used
180 to track temporary fine-grained changes to a subset of files and
181 will only remain valid for
183 snapshot access purposes for the
185 period configured for the PFS.
186 If you desire a real snapshot then the
188 directive may be what you are looking for.
190 .It Cm bstats Op Ar interval
193 B-tree statistics until interrupted.
196 seconds between each display.
197 The default interval is one second.
198 .\" ==== iostats ====
199 .It Cm iostats Op Ar interval
203 statistics until interrupted.
206 seconds between each display.
207 The default interval is one second.
208 .\" ==== history ====
209 .It Cm history Ar path ...
210 Show the modification history for
212 file's inode and data.
213 .\" ==== blockmap ====
215 Dump the blockmap for the file system.
218 blockmap is two-layer
219 blockmap representing the maximum possible file system size of 1 Exabyte.
220 Needless to say the second layer is only present for blocks which exist.
222 blockmap represents 8-Megabyte blocks, called big-blocks.
223 Each big-block has an append
224 point, a free byte count, and a typed zone id which allows content to be
225 reverse engineered to some degree.
229 allocations essentially appended to a selected big-block using
230 the append offset and deducted from the free byte count.
231 When space is freed the free byte count is adjusted but
233 does not track holes in big-blocks for reallocation.
234 A big-block must be completely freed, either
235 through normal file system operations or through reblocking, before
238 Data blocks can be shared by deducting the space used from the free byte
239 count for each shared references, though
241 does not yet make use of this feature.
242 This means the free byte count can legally go negative.
244 This command needs the
248 .It Cm show Op Ar lo Ns Cm \&: Ns Ar objid
250 By default this command will validate all B-Tree
251 linkages and CRCs, including data CRCs, and will report the most verbose
252 information it can dig up.
253 Any errors will show up with a
255 in column 1 along with various
258 If you specify a localization and object id field,
259 .Ar lo Ns Cm \&: Ns Ar objid ,
261 search for the key printing nodes as it recurses down, and then
262 will iterate forwards.
266 the command will report less information about the inode contents.
270 the command will not report the content of the inode or other typed
275 the command will not report volume header information, big-block fill
276 ratios, mirror transaction ids, or report or check data CRCs.
277 B-tree CRCs and linkages are still checked.
279 This command needs the
282 .\" ==== show-undo ====
288 This command needs the
292 .\" Dump the B-tree, record, large-data, and small-data blockmaps, showing
293 .\" physical block assignments and free space percentages.
294 .\" ==== namekey1 ====
295 .It Cm namekey1 Ar filename
298 64 bit directory hash for the specified file name, using
299 the original directory hash algorithm in version 1 of the file system.
300 The low 32 bits are used as an iterator for hash collisions and will be
302 .\" ==== namekey2 ====
303 .It Cm namekey2 Ar filename
306 64 bit directory hash for the specified file name, using
307 the new directory hash algorithm in version 2 of the file system.
308 The low 32 bits are still used as an iterator but will start out containing
309 part of the hash key.
310 .\" ==== namekey32 ====
311 .It Cm namekey32 Ar filename
312 Generate the top 32 bits of a
314 64 bit directory hash for the specified file name.
317 Shows extended information about all the mounted
320 At the moment volume identification, big-blocks information, space and
321 and pseudo-filesystem (PFS) details are shown.
322 .\" ==== cleanup ====
323 .It Cm cleanup Op Ar filesystem ...
324 This is a meta-command which executes snapshot, prune, rebalance and reblock
325 commands on the specified
330 is specified this command will clean-up all
332 file systems in use, including PFS's.
333 To do this it will scan all
337 mounts, extract PFS id's, and clean-up each PFS found.
339 This command will access a snapshots
340 directory and a configuration file for each
342 creating them if necessary.
343 .Bl -tag -width indent
344 .It Nm HAMMER No version 2-
345 The configuration file is
347 in the snapshots directory which defaults to
348 .Pa <pfs>/snapshots .
349 .It Nm HAMMER No version 3+
350 The configuration file is saved in file system meta-data, see
353 The snapshots directory defaults to
354 .Pa /var/hammer/<pfs>
355 .Pa ( /var/hammer/root
359 The format of the configuration file is:
360 .Bd -literal -offset indent
361 snapshots <period> <retention-time> [any]
362 prune <period> <max-runtime>
363 rebalance <period> <max-runtime>
364 reblock <period> <max-runtime>
365 recopy <period> <max-runtime>
369 .Bd -literal -offset indent
370 snapshots 1d 60d # 0d 0d for PFS /tmp, /var/tmp, /usr/obj
377 Time is given with a suffix of
383 meaning day, hour, minute and second.
387 directive has a period of 0 and a retention time of 0
388 then snapshot generation is disabled, removal of old snapshots are
389 disabled, and prunes will use
390 .Cm prune-everything .
393 directive has a period of 0 but a non-zero retention time
394 then this command will not create any new snapshots but will remove old
395 snapshots it finds based on the retention time.
397 By default only snapshots in the form
398 .Ql snap- Ns Ar yyyymmdd Ns Op - Ns Ar HHMM
402 directive is specified as a third argument on the
404 config line then any softlink of the form
405 .Ql *- Ns Ar yyyymmdd Ns Op - Ns Ar HHMM
407 .Ql *. Ns Ar yyyymmdd Ns Op - Ns Ar HHMM
410 A prune max-runtime of 0 means unlimited.
412 If period hasn't passed since the previous
415 For example a day has passed when midnight is passed (localtime).
423 The default configuration file will create a daily snapshot, do a daily
424 pruning, rebalancing and reblocking run and a monthly recopy run.
425 Reblocking is defragmentation with a level of 95%,
426 and recopy is full defragmentation.
428 By default prune and rebalance operations are time limited to 5 minutes,
429 reblock operations to a bit over 5 minutes,
430 and recopy operations to a bit over 10 minutes.
431 Reblocking and recopy runs are each broken down into four separate functions:
432 btree, inodes, dirs and data.
433 Each function is time limited to the time given in the configuration file,
434 but the btree, inodes and dirs functions usually does not take very long time,
435 full defragmentation is always used for these three functions.
436 Also note that this directive will by default disable snapshots on
443 The defaults may be adjusted by modifying the configuration file.
444 The pruning and reblocking commands automatically maintain a cyclefile
445 for incremental operation.
446 If you interrupt (^C) the program the cyclefile will be updated,
448 may continue to run in the background for a few seconds until the
450 ioctl detects the interrupt.
453 PFS option can be set to use another location for the snapshots directory.
455 Work on this command is still in progress.
457 An ability to remove snapshots dynamically as the
458 file system becomes full.
460 .It Cm config Op Ar filesystem Op Ar configfile
463 If zero or one arguments are specified this function dumps the current
464 configuration file to stdout.
465 Zero arguments specifies the PFS containing the current directory.
466 This configuration file is stored in file system meta-data.
467 If two arguments are specified this function installs a new config file.
471 versions less than 3 the configuration file is by default stored in
472 .Pa <pfs>/snapshots/config ,
473 but in all later versions the configuration file is stored in file system
475 .\" ==== viconfig ====
476 .It Cm viconfig Op Ar filesystem
479 Edit the configuration file and reinstall into file system meta-data when done.
480 Zero arguments specifies the PFS containing the current directory.
482 .It Cm expand Ar filesystem Ar device
483 This command will format
485 and add all of its space to
489 All existing data contained on
491 will be destroyed by this operation!
496 file system, formatting will be denied.
497 You can overcome this sanity check
500 to erase the beginning sectors of the device.
501 Also remember that you have to specify
503 together with any other device that make up the file system,
508 .\" ==== snapshot ====
509 .It Cm snapshot Oo Ar filesystem Oc Ar snapshot-dir
510 .It Cm snapshot Ar filesystem Ar snapshot-dir Op Ar note
511 Takes a snapshot of the file system either explicitly given by
513 or implicitly derived from the
515 argument and creates a symlink in the directory provided by
517 pointing to the snapshot.
520 is not a directory, it is assumed to be a format string passed to
522 with the current time as parameter.
525 refers to an existing directory, a default format string of
527 is assumed and used as name for the newly created symlink.
529 Snapshot is a per PFS operation, so a
531 file system and each PFS in it have to be snapshot separately.
533 Example, assuming that
541 are file systems on their own, the following invocations:
542 .Bd -literal -offset indent
543 hammer snapshot /mysnapshots
545 hammer snapshot /mysnapshots/%Y-%m-%d
547 hammer snapshot /obj /mysnapshots/obj-%Y-%m-%d
549 hammer snapshot /usr /my/snaps/usr "note"
552 Would create symlinks similar to:
553 .Bd -literal -offset indent
554 /mysnapshots/snap-20080627-1210 -> /@@0x10d2cd05b7270d16
556 /mysnapshots/2008-06-27 -> /@@0x10d2cd05b7270d16
558 /mysnapshots/obj-2008-06-27 -> /obj@@0x10d2cd05b7270d16
560 /my/snaps/usr/snap-20080627-1210 -> /usr@@0x10d2cd05b7270d16
565 version 3+ file system the snapshot is also recorded in file system meta-data
566 along with the optional
572 .It Cm snap Ar path Op Ar note
575 Create a snapshot for the PFS containing
577 and create a snapshot softlink.
578 If the path specified is a
579 directory a standard snapshot softlink will be created in the directory.
580 The snapshot softlink points to the base of the mounted PFS.
581 .It Cm snaplo Ar path Op Ar note
584 Create a snapshot for the PFS containing
586 and create a snapshot softlink.
587 If the path specified is a
588 directory a standard snapshot softlink will be created in the directory.
589 The snapshot softlink points into the directory it is contained in.
590 .It Cm snapq Ar dir Op Ar note
593 Create a snapshot for the PFS containing the specified directory but do
594 not create a softlink.
595 Instead output a path which can be used to access
596 the directory via the snapshot.
598 An absolute or relative path may be specified.
599 The path will be used as-is as a prefix in the path output to stdout.
601 snap and snapshot directives the snapshot transaction id will be registered
602 in the file system meta-data.
603 .It Cm snaprm Bro Ar path | transid Brc Ar ...
606 Remove a snapshot given its softlink or transaction id.
607 If specifying a transaction id
608 the snapshot is removed from file system meta-data but you are responsible
609 for removing any related softlinks.
610 .It Cm snapls Op Ar path ...
613 Dump the snapshot meta-data for PFSs containing each
615 listing all available snapshots and their notes.
616 If no arguments are specified snapshots for the PFS containing the
617 current directory are listed.
618 This is the definitive list of snapshots for the file system.
620 .It Cm prune Ar softlink-dir
621 Prune the file system based on previously created snapshot softlinks.
622 Pruning is the act of deleting file system history.
625 command will delete file system history such that
626 the file system state is retained for the given snapshots,
627 and all history after the latest snapshot.
628 By setting the per PFS parameter
630 history is guaranteed to be saved at least this time interval.
631 All other history is deleted.
633 The target directory is expected to contain softlinks pointing to
634 snapshots of the file systems you wish to retain.
635 The directory is scanned non-recursively and the mount points and
636 transaction ids stored in the softlinks are extracted and sorted.
637 The file system is then explicitly pruned according to what is found.
638 Cleaning out portions of the file system is as simple as removing a
639 snapshot softlink and then running the
643 As a safety measure pruning only occurs if one or more softlinks are found
646 snapshot id extension.
647 Currently the scanned softlink directory must contain softlinks pointing
651 The softlinks may specify absolute or relative paths.
652 Softlinks must use 20-character
654 transaction ids, as might be returned from
655 .Nm Cm synctid Ar filesystem .
657 Pruning is a per PFS operation, so a
659 file system and each PFS in it have to be pruned separately.
661 Note that pruning a file system may not immediately free-up space,
662 though typically some space will be freed if a large number of records are
664 The file system must be reblocked to completely recover all available space.
666 Example, lets say your that you didn't set
668 and snapshot directory contains the following links:
669 .Bd -literal -offset indent
670 lrwxr-xr-x 1 root wheel 29 May 31 17:57 snap1 ->
671 /usr/obj/@@0x10d2cd05b7270d16
673 lrwxr-xr-x 1 root wheel 29 May 31 17:58 snap2 ->
674 /usr/obj/@@0x10d2cd13f3fde98f
676 lrwxr-xr-x 1 root wheel 29 May 31 17:59 snap3 ->
677 /usr/obj/@@0x10d2cd222adee364
680 If you were to run the
682 command on this directory, then the
685 mount will be pruned to retain the above three snapshots.
686 In addition, history for modifications made to the file system older than
687 the oldest snapshot will be destroyed and history for potentially fine-grained
688 modifications made to the file system more recently than the most recent
689 snapshot will be retained.
691 If you then delete the
693 softlink and rerun the
696 history for modifications pertaining to that snapshot would be destroyed.
700 file system versions 3+ this command also scans the snapshots stored
701 in the file system meta-data and includes them in the prune.
702 .\" ==== prune-everything ====
703 .It Cm prune-everything Ar filesystem
704 This command will remove all historical records from the file system.
705 This directive is not normally used on a production system.
707 This command does not remove snapshot softlinks but will delete all
708 snapshots recorded in file system meta-data (for file system version 3+).
709 The user is responsible for deleting any softlinks.
710 .\" ==== rebalance ====
711 .It Cm rebalance Ar filesystem Op Ar saturation_percentage
712 This command will rebalance the B-tree, nodes with small number of
713 elements will be combined and element counts will be smoothed out
716 The saturation percentage is between 50% and 100%.
717 The default is 75% (the
719 suffix is not needed).
720 .\" ==== reblock* ====
721 .It Cm reblock Ar filesystem Op Ar fill_percentage
722 .It Cm reblock-btree Ar filesystem Op Ar fill_percentage
723 .It Cm reblock-inodes Ar filesystem Op Ar fill_percentage
724 .It Cm reblock-dirs Ar filesystem Op Ar fill_percentage
725 .It Cm reblock-data Ar filesystem Op Ar fill_percentage
726 Attempt to defragment and free space for reuse by reblocking a live
729 Big-blocks cannot be reused by
731 until they are completely free.
732 This command also has the effect of reordering all elements, effectively
733 defragmenting the file system.
735 The default fill percentage is 100% and will cause the file system to be
736 completely defragmented.
737 All specified element types will be reallocated and rewritten.
738 If you wish to quickly free up space instead try specifying
739 a smaller fill percentage, such as 90% or 80% (the
741 suffix is not needed).
743 Since this command may rewrite the entire contents of the disk it is
744 best to do it incrementally from a
750 options to limit the run time.
751 The file system would thus be defragmented over long period of time.
753 It is recommended that separate invocations be used for each data type.
754 B-tree nodes, inodes, and directories are typically the most important
755 elements needing defragmentation.
756 Data can be defragmented over a longer period of time.
758 Reblocking is a per PFS operation, so a
760 file system and each PFS in it have to be reblocked separately.
761 .\" ==== pfs-status ====
762 .It Cm pfs-status Ar dirpath ...
763 Retrieve the mirroring configuration parameters for the specified
765 file systems or pseudo-filesystems (PFS's).
766 .\" ==== pfs-master ====
767 .It Cm pfs-master Ar dirpath Op Ar options
768 Create a pseudo-filesystem (PFS) inside a
771 Up to 65535 such file systems can be created.
772 Each PFS uses an independent inode numbering space making it suitable
773 for use as a replication source or target.
777 directive creates a PFS that you can read, write, and use as a mirroring
780 It is recommended to use a
782 mount to access a PFS, for more information see
784 .\" ==== pfs-slave ====
785 .It Cm pfs-slave Ar dirpath Op Ar options
786 Create a pseudo-filesystem (PFS) inside a
789 Up to 65535 such file systems can be created.
790 Each PFS uses an independent inode numbering space making it suitable
791 for use as a replication source or target.
795 directive creates a PFS that you can use as a mirroring target.
796 You will not be able to access a slave PFS until you have completed the
797 first mirroring operation with it as the target (its root directory will
798 not exist until then).
800 Access to the pfs-slave via the special softlink, as described in the
805 dynamically modify the snapshot transaction id by returning a dynamic result
810 A PFS can only be truly destroyed with the
813 Removing the softlink will not destroy the underlying PFS.
815 It is recommended to use a
817 mount to access a PFS, for more information see
819 .\" ==== pfs-update ====
820 .It Cm pfs-update Ar dirpath Op Ar options
821 Update the configuration parameters for an existing
823 file system or pseudo-filesystem.
824 Options that may be specified:
825 .Bl -tag -width indent
826 .It Cm sync-beg-tid= Ns Ar 0x16llx
827 This is the automatic snapshot access starting transaction id for
829 This parameter is normally updated automatically by the
833 It is important to note that accessing a mirroring slave
834 with a transaction id greater than the last fully synchronized transaction
835 id can result in an unreliable snapshot since you will be accessing
836 data that is still undergoing synchronization.
838 Manually modifying this field is dangerous and can result in a broken mirror.
839 .It Cm sync-end-tid= Ns Ar 0x16llx
840 This is the current synchronization point for mirroring slaves.
841 This parameter is normally updated automatically by the
845 Manually modifying this field is dangerous and can result in a broken mirror.
846 .It Cm shared-uuid= Ns Ar uuid
847 Set the shared UUID for this file system.
848 All mirrors must have the same shared UUID.
849 For safety purposes the
851 directives will refuse to operate on a target with a different shared UUID.
853 Changing the shared UUID on an existing, non-empty mirroring target,
854 including an empty but not completely pruned target,
855 can lead to corruption of the mirroring target.
856 .It Cm unique-uuid= Ns Ar uuid
857 Set the unique UUID for this file system.
858 This UUID should not be used anywhere else,
859 even on exact copies of the file system.
860 .It Cm label= Ns Ar string
861 Set a descriptive label for this file system.
862 .It Cm snapshots= Ns Ar string
863 Specify the snapshots directory which
866 will use to manage this PFS.
867 .Bl -tag -width indent
868 .It Nm HAMMER No version 2-
869 The snapshots directory does not need to be configured for
870 PFS masters and will default to
871 .Pa <pfs>/snapshots .
873 PFS slaves are mirroring slaves so you cannot configure a snapshots
874 directory on the slave itself to be managed by the slave's machine.
875 In fact, the slave will likely have a
877 sub-directory mirrored
878 from the master, but that directory contains the configuration the master
879 is using for its copy of the file system, not the configuration that we
880 want to use for our slave.
882 It is recommended that
883 .Pa <fs>/var/slaves/<name>
884 be configured for a PFS slave, where
890 is an appropriate label.
891 .It Nm HAMMER No version 3+
892 The snapshots directory does not need to be configured for PFS masters or
894 The snapshots directory defaults to
895 .Pa /var/hammer/<pfs>
896 .Pa ( /var/hammer/root
900 You can control snapshot retention on your slave independent of the master.
901 .It Cm snapshots-clear
904 directory path for this PFS.
905 .It Cm prune-min= Ns Ar N Ns Cm d
906 .It Cm prune-min= Ns Oo Ar N Ns Cm d/ Oc Ns \
907 Ar hh Ns Op Cm \&: Ns Ar mm Ns Op Cm \&: Ns Ar ss
908 Set the minimum fine-grained data retention period.
910 always retains fine-grained history up to the most recent snapshot.
911 You can extend the retention period further by specifying a non-zero
913 Any snapshot softlinks within the retention period are ignored
914 for the purposes of pruning (the fine grained history is retained).
915 Number of days, hours, minutes and seconds are given as
920 Because the transaction id in the snapshot softlink cannot be used
921 to calculate a timestamp,
923 uses the earlier of the
927 field of the softlink to
928 determine which snapshots fall within the retention period.
929 Users must be sure to retain one of these two fields when manipulating
932 .\" ==== pfs-upgrade ====
933 .It Cm pfs-upgrade Ar dirpath
934 Upgrade a PFS from slave to master operation.
935 The PFS will be rolled back to the current end synchronization transaction id
936 (removing any partial synchronizations), and will then become writable.
940 currently supports only single masters and using
941 this command can easily result in file system corruption
942 if you don't know what you are doing.
944 This directive will refuse to run if any programs have open descriptors
945 in the PFS, including programs chdir'd into the PFS.
946 .\" ==== pfs-downgrade ====
947 .It Cm pfs-downgrade Ar dirpath
948 Downgrade a master PFS from master to slave operation
949 The PFS becomes read-only and access will be locked to its
952 This directive will refuse to run if any programs have open descriptors
953 in the PFS, including programs chdir'd into the PFS.
954 .\" ==== pfs-destroy ====
955 .It Cm pfs-destroy Ar dirpath
956 This permanently destroys a PFS.
958 This directive will refuse to run if any programs have open descriptors
959 in the PFS, including programs chdir'd into the PFS.
960 .\" ==== mirror-read ====
961 .It Cm mirror-read Ar filesystem Op Ar begin-tid
962 Generate a mirroring stream to stdout.
963 The stream ends when the transaction id space has been exhausted.
964 .\" ==== mirror-read-stream ====
965 .It Cm mirror-read-stream Ar filesystem Op Ar begin-tid
966 Generate a mirroring stream to stdout.
967 Upon completion the stream is paused until new data is synced to the
968 master, then resumed.
969 Operation continues until the pipe is broken.
970 .\" ==== mirror-write ====
971 .It Cm mirror-write Ar filesystem
972 Take a mirroring stream on stdin.
974 This command will fail if the
976 configuration field for the two file systems do not match.
978 If the target PFS does not exist this command will ask you whether
979 you want to create a compatible PFS slave for the target or not.
980 .\" ==== mirror-dump ====
986 to dump an ASCII representation of the mirroring stream.
987 .\" ==== mirror-copy ====
988 .\".It Cm mirror-copy Ar [[user@]host:]filesystem [[user@]host:]filesystem
990 Oo Oo Ar user Ns Cm @ Oc Ns Ar host Ns Cm \&: Oc Ns Ar filesystem \
991 Oo Oo Ar user Ns Cm @ Oc Ns Ar host Ns Cm \&: Oc Ns Ar filesystem
992 This is a shortcut which pipes a
997 If a remote host specification is made the program forks a
1003 on the appropriate host.
1004 The source may be a master or slave PFS, and the target must be a slave PFS.
1006 This command also established full duplex communication and turns on
1007 the two-way protocol feature which automatically negotiates transaction id
1008 ranges without having to use a cyclefile.
1009 If the operation completes successfully the target PFS's
1012 Note that you must re-chdir into the target PFS to see the updated information.
1013 If you do not you will still be in the previous snapshot.
1015 If the target PFS does not exist this command will ask you whether
1016 you want to create a compatible PFS slave for the target or not.
1017 .\" ==== mirror-stream ====
1018 .\".It Cm mirror-stream Ar [[user@]host:]filesystem [[user@]host:]filesystem
1019 .It Cm mirror-stream \
1020 Oo Oo Ar user Ns Cm @ Oc Ns Ar host Ns Cm \&: Oc Ns Ar filesystem \
1021 Oo Oo Ar user Ns Cm @ Oc Ns Ar host Ns Cm \&: Oc Ns Ar filesystem
1022 This command works similarly to
1024 but does not exit after the initial mirroring completes.
1025 The mirroring operation will resume as changes continue to be made to the
1027 The command is commonly used with
1031 options to keep the mirroring target in sync with the source on a continuing
1034 If the pipe is broken the command will automatically retry after sleeping
1036 The time slept will be 15 seconds plus the time given in the
1040 This command also detects the initial-mirroring case and spends some
1041 time scanning the B-Tree to find good break points, allowing the initial
1042 bulk mirroring operation to be broken down into about 20 separate pieces.
1043 This means that the user can kill and restart the operation and it will
1044 not have to start from scratch once it has gotten past the first chunk.
1047 option may be used to disable this feature and perform an initial bulk
1049 .\" ==== version ====
1050 .It Cm version Ar filesystem
1051 This command returns the
1053 file system version for the specified
1055 as well as the range of versions supported in the kernel.
1058 option may be used to remove the summary at the end.
1059 .\" ==== version-upgrade ====
1060 .It Cm version-upgrade Ar filesystem Ar version Op Cm force
1061 This command upgrades the
1066 Once upgraded a file system may not be downgraded.
1067 If you wish to upgrade a file system to a version greater or equal to the
1068 work-in-progress version number you must specify the
1071 Use of WIP versions should be relegated to testing and may require wiping
1072 the file system as development progresses, even though the WIP version might
1076 This command operates on the entire
1078 file system and is not a per PFS operation.
1079 All PFS's will be affected.
1080 .Bl -tag -width indent
1083 default version, first
1088 New directory entry layout.
1089 This version is using a new directory hash key.
1092 New snapshot management, using file system meta-data for saving
1093 configuration file and snapshots (transaction ids etc.).
1094 Also default snapshots directory has changed.
1098 New undo/flush, giving faster sync.
1102 .Sh PSEUDO-FILESYSTEM (PFS) NOTES
1103 The root of a PFS is not hooked into the primary
1105 file system as a directory.
1108 creates a special softlink called
1110 (exactly 10 characters long) in the primary
1114 then modifies the contents of the softlink as read by
1116 and thus what you see with an
1118 command or if you were to
1121 If the PFS is a master the link reflects the current state of the PFS.
1122 If the PFS is a slave the link reflects the last completed snapshot, and the
1123 contents of the link will change when the next snapshot is completed, and
1128 utility employs numerous safeties to reduce user foot-shooting.
1131 directive requires that the target be configured as a slave and that the
1133 field of the mirroring source and target match.
1134 .Sh UPGRADE INSTRUCTIONS HAMMER V1 TO V2
1135 This upgrade changes the way directory entries are stored.
1136 It is possible to upgrade a V1 file system to V2 in place, but
1137 directories created prior to the upgrade will continue to use
1140 Note that the slave mirroring code in the target kernel had bugs in
1141 V1 which can create an incompatible root directory on the slave.
1144 master created after the upgrade with a
1146 slave created prior to the upgrade.
1148 Any directories created after upgrading will use a new layout.
1149 .Sh UPGRADE INSTRUCTIONS HAMMER V2 TO V3
1150 This upgrade adds meta-data elements to the B-Tree.
1151 It is possible to upgrade a V2 file system to V3 in place.
1152 After issuing the upgrade be sure to run a
1155 to perform post-upgrade tasks.
1157 After making this upgrade running a
1162 directory for each PFS mount into
1163 .Pa /var/hammer/<pfs> .
1166 root mount will migrate
1169 .Pa /var/hammer/root .
1170 Migration occurs only once and only if you have not specified
1171 a snapshots directory in the PFS configuration.
1172 If you have specified a snapshots directory in the PFS configuration no
1173 automatic migration will occur.
1175 For slaves, if you desire, you can migrate your snapshots
1176 config to the new location manually and then clear the
1177 snapshot directory configuration in the slave PFS.
1178 The new snapshots hierarchy is designed to work with
1179 both master and slave PFSs equally well.
1181 In addition, the old config file will be moved to file system meta-data,
1182 editable via the new
1186 The old config file will be deleted.
1187 Migration occurs only once.
1189 The V3 file system has new
1191 directives for creating snapshots.
1192 All snapshot directives, including the original, will create
1193 meta-data entries for the snapshots and the pruning code will
1194 automatically incorporate these entries into its list and
1195 expire them the same way it expires softlinks.
1196 If you by accident blow away your snapshot softlinks you can use the
1198 directive to get a definitive list from the file system meta-data and
1199 regenerate them from that list.
1204 to backup file systems your scripts may be using the
1206 directive to generate transaction ids.
1207 This directive does not create a snapshot.
1208 You will have to modify your scripts to use the
1210 directive to generate the linkbuf for the softlink you create, or
1211 use one of the other
1216 directive will continue to work as expected and in V3 it will also
1217 record the snapshot transaction id in file system meta-data.
1218 You may also want to make use of the new
1220 tag for the meta-data.
1223 If you used to remove snapshot softlinks with
1225 you should probably start using the
1227 directive instead to also remove the related meta-data.
1228 The pruning code scans the meta-data so just removing the
1229 softlink is not sufficient.
1230 .Sh UPGRADE INSTRUCTIONS HAMMER V3 TO V4
1231 This upgrade changes undo/flush, giving faster sync.
1232 It is possible to upgrade a V3 file system to V4 in place.
1233 This upgrade reformats the UNDO FIFO (typically 1GB), so upgrade might take
1234 a minute or two depending.
1236 Version 4 allows the UNDO FIFO to be flushed without also having
1237 to flush the volume header, removing 2 of the 4 disk syncs typically
1238 required for an fsync() and removing 1 of the 2 disk syncs typically
1239 required for a flush sequence.
1243 .Bl -tag -width ".It Pa <fs>/var/slaves/<name>" -compact
1244 .It Pa <pfs>/snapshots
1245 default per PFS snapshots directory
1248 .It Pa /var/hammer/<pfs>
1249 default per PFS snapshots directory (not root)
1252 .It Pa /var/hammer/root
1253 default snapshots directory for root directory
1256 .It Pa <snapshots>/config
1263 .It Pa <fs>/var/slaves/<name>
1264 recommended slave PFS snapshots directory
1272 .Xr periodic.conf 5 ,
1273 .Xr mount_hammer 8 ,
1279 utility first appeared in
1282 .An Matthew Dillon Aq dillon@backplane.com
projects / dragonfly.git / blob