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 1.51 2008/09/30 23:13:08 dillon Exp $
34 .Dd September 28, 2008
39 .Nd HAMMER file system utility
45 .Op Fl f Ar blkdev[:blkdev]*
46 .\" .Op Fl s Ar linkpath
52 This manual page documents the
54 utility which provides miscellaneous functions related to managing a
57 For a general introduction to the
59 file system, its features, and
60 examples on how to set up and maintain one, see
63 The options are as follows:
64 .Bl -tag -width indent
68 Tell the mirror commands to use a 2-way protocol, which allows
69 automatic negotiation of transaction id ranges. This option is
70 automatically enabled by the
74 Specify recursion for those commands which support it.
76 Specify a bandwidth limit in bytes per second for mirroring streams.
77 This option is typically used to prevent batch mirroring operations from
78 loading down the machine.
79 The bandwidth may be suffixed with
85 values in kilobytes, megabytes, and gigabytes per second.
87 When pruning and reblocking you can instruction
90 object id stored in the specified file.
91 If the file does not exist
93 will start at the beginning.
97 specific period of time and is unable to complete the operation it will
98 write out the current object id so the next run can pick up where it left
102 runs to completion it will delete the cyclefile.
103 .It Fl f Ar blkdev[:blkdev]*
104 Specify the volumes making up a
108 When maintaining a streaming mirroring this option specifies the
109 minimum delay after a batch ends before the next batch is allowed
111 The default is five seconds.
113 Decrease verbosement. May be specified multiple times.
115 When pruning and reblocking you can tell the utility to stop after a
116 certain period of time. This option is used along with the cycle file
117 option to prune or reblock a portion of the file system incrementally.
119 Increase verboseness. May be specified multiple times.
122 The commands are as follows:
123 .Bl -tag -width indent
124 .\" ==== synctid ====
125 .It Ar synctid Ar filesystem Op quick
126 Generates a guaranteed, formal 64 bit transaction id representing the
127 current state of the specified
129 file system. The file system will
130 be synced to the media.
134 keyword is specified the file system will be soft-synced, meaning that a
135 crash might still undo the state of the file system as of the transaction
136 id returned but any new modifications will occur after the returned
137 transaction id as expected.
139 .It Ar bstats Op interval
142 B-tree statistics until interrupted.
145 seconds between each display.
146 The default interval is one second.
147 .\" ==== iostats ====
148 .It Ar iostats Op interval
151 I/O statistics until interrupted.
154 seconds between each display.
155 The default interval is one second.
156 .\" ==== history ====
157 .It Ar history Ar path ...
158 Show the modification history for
160 file's inode and data.
163 Dump the B-tree. This command needs the
167 .\" Dump the B-tree, record, large-data, and small-data blockmaps, showing
168 .\" physical block assignments and free space percentages.
169 .\" ==== namekey ====
170 .It Ar namekey Ar filename
173 64 bit directory hash for the specified file name.
174 The low 32 bits are used as an iterator for hash collisions and will be
176 .\" ==== namekey32 ====
177 .It Ar namekey32 Ar filename
178 Generate the top 32 bits of a
180 64 bit directory hash for the specified
182 .\" ==== cleanup ====
183 .It Ar cleanup Op Ar filesystem ...
184 This is a meta-command which executes snapshot, pruning, and reblocking
185 commands on the specified
190 is specified this command will clean-up all
192 file systems in use, including PFS's.
193 To do this it will scan all
197 mounts, extract PFS id's, and clean-up each PFS found.
199 This command will by default access a
205 creating them if necessary.
206 The format of the configuration file is:
207 .Bd -literal -offset indent
208 snapshots <period> <retensiontime>
209 prune <period> <max-runtime>
210 reblock <period> <1/3 max-runtime>
211 recopy <period> <1/3 max-runtime>
214 snapshots 1d 60d # 0 60d for PFS /tmp, /var/tmp, /usr/obj
220 Time is in seconds, it can be given with a suffix of
226 meaning day, hour, minute and second.
227 A snapshots period of 0 disables snapshot generation and prunes using
228 .Ar prune-everything ,
229 if no snapshots exists.
230 A prune max-runtime of 0 means unlimited.
232 It is recommended to run
234 once a day, depending on the configured period,
239 .Bd -literal -offset indent
240 15 2 * * * hammer cleanup >>/var/log/hammer.log 2>&1
243 The default configuration file will create a daily snapshot, do a daily
244 pruning and reblocking run and a monthly recopy run.
245 Reblocking is defragmentation with a level of 95%,
246 and recopy is full defragmentation.
248 By default prune and reblock operations are limited to 5 minutes per function,
249 and recopy operations are limited to 10 minutes per function.
250 Reblocking and recopy runs are each broken down into three separate functions
251 (btree, inodes and data)
252 and are thus by default limited to 15 and 30 minutes respectively.
253 Also note that this directive will by default disable snapshots on
260 The defaults may be adjusted by modifying the
263 The pruning and reblocking commands automatically maintain a cyclefile
264 for incremental operation.
265 If you interrupt (^C) the program the cyclefile will be updated, but a sub-command
266 may continue to run in the background for a few seconds until the
268 ioctl detects the interrupt.
271 PFS option can be set to use another location for the snapshots directory.
273 Work on this command is still in progress.
274 Expected additions: An ability to remove snapshots dynamically as the
275 file system becomes full.
277 .It Ar prune Ar softlink-dir
278 Prune the file system based on previously created snapshot softlinks.
279 Pruning is the act of deleting file system history.
283 will delete file system history such that
284 the file system state is retained for the given snapshots,
285 and all history after the latest snapshot,
286 but all other history is deleted.
288 The target directory is expected to contain softlinks pointing to
289 snapshots of the file systems you wish to retain. The directory is scanned
290 non-recursively and the mount points and transaction ids stored in the
291 softlinks are extracted and sorted.
292 The file system is then explicitly pruned according to what is found.
293 Cleaning out portions of the file system is as simple as removing a softlink
298 As a safety measure pruning only occurs if one or more softlinks are found
299 containing the @@ snapshot id extension.
300 Currently the scanned softlink directory must contain softlinks pointing
303 mount. The softlinks may specify absolute or relative
304 paths. Softlinks must use 20-character (@@0x%016llx) transaction ids,
305 as might be returned from
306 .Dq Nm Ar synctid filesystem .
308 Pruning is a per PFS operation, so a
310 file system and each PFS in it have to be pruned separately.
312 Note that pruning a file system may not immediately free-up space,
313 though typically some space will be freed if a large number of records are
314 pruned out. The file system must be reblocked to completely recover all
317 Example, lets say your snapshot directory contains the following links:
318 .Bd -literal -offset indent
319 lrwxr-xr-x 1 root wheel 29 May 31 17:57 snap1 ->
320 /usr/obj/@@0x10d2cd05b7270d16
322 lrwxr-xr-x 1 root wheel 29 May 31 17:58 snap2 ->
323 /usr/obj/@@0x10d2cd13f3fde98f
325 lrwxr-xr-x 1 root wheel 29 May 31 17:59 snap3 ->
326 /usr/obj/@@0x10d2cd222adee364
329 If you were to run the
331 command on this directory, then the
334 mount will be pruned to retain the above three snapshots.
335 In addition, history for modifications made to the file system older than the oldest
336 snapshot will be destroyed and history for potentially fine-grained modifications made
337 to the file system more recently than the most recent snapshot will be
340 If you then delete the snap2 softlink and rerun the
343 history for modifications pertaining to that snapshot would be destroyed.
344 .\" ==== prune-everything ====
345 .It Ar prune-everything Ar filesystem
346 This command will remove all historical records from the file system.
347 This directive is not normally used on a production system.
348 .\" ==== snapshot ====
349 .It Ar snapshot Ar snapshot-dir
350 .It Ar snapshot Ar filesystem snapshot-dir
351 Takes a snapshot of the file system either explicitly given by
353 or implicitly derived from the
355 argument and creates a symlink in the directory provided by
357 pointing to the snapshot.
360 is not a directory, it is assumed to be a format string
363 with the current time as parameter.
366 refers to an existing directory, a default format string of "snap-%Y%d%m-%H%M"
367 is assumed and used as name for the newly created symlink.
369 Snapshot is a per PFS operation, so a
371 file system and each PFS in it have to be snapshot separately.
373 Example, assuming that
379 is a file system on its own, the following invocations:
380 .Bd -literal -offset indent
381 hammer snapshot /mysnapshots
383 hammer snapshot /mysnapshots/%Y-%m-%d
385 hammer snapshot /obj /mysnapshots/obj-%Y-%m-%d
388 would create symlinks similar to:
389 .Bd -literal -offset indent
390 /mysnapshots/snap-20080627-1210 -> /@@0x10d2cd05b7270d16
392 /mysnapshots/2008-06-27 -> /@@0x10d2cd05b7270d16
394 /mysnapshots/obj-2008-06-27 -> /obj@@0x10d2cd05b7270d16
396 .\" ==== reblock ====
397 .It Ar reblock Ar filesystem Op Ar fill_percentage
398 .It Ar reblock-btree Ar filesystem Op Ar fill_percentage
399 .It Ar reblock-inodes Ar filesystem Op Ar fill_percentage
400 .It Ar reblock-dirs Ar filesystem Op Ar fill_percentage
401 .It Ar reblock-data Ar filesystem Op Ar fill_percentage
402 Attempt to defragment and free space for reuse by reblocking a live
405 Big blocks cannot be reused by
407 until they are completely free.
408 This command also has the effect of reordering all elements, effectively
409 defragmenting the file system.
411 The default fill percentage is 100% and will cause the file system to be
412 completely defragmented. All specified element types will be reallocated
413 and rewritten. If you wish to quickly free up space instead try specifying
414 a smaller fill percentage, such as 90% or 80% (the
416 suffix is not needed).
418 Since this command may rewrite the entire contents of the disk it is
419 best to do it incrementally from a
425 options to limit the run time.
426 The file system would thus be defragmented over long period of time.
428 It is recommended that separate invocations be used for each data type.
429 B-tree nodes, inodes, and directories are typically the most important
430 elements needing defragmentation. Data can be defragmented over a longer
433 Reblocking is a per PFS operation, so a
435 file system and each PFS in it have to be reblocked separately.
436 .\" ==== pfs-status ====
437 .It Ar pfs-status Ar dirpath ...
438 Retrieve the mirroring configuration parameters for the specified
440 file systems or pseudo-filesystems.
441 .\" ==== pfs-master ====
442 .It Ar pfs-master Ar dirpath Op options
443 Create a pseudo-filesystem (PFS) inside a
446 Up to 65535 such file systems can be created.
447 Each PFS uses an independent inode numbering space making it suitable
448 for use as a replication source or target.
452 directive creates a PFS that you can read, write, and use as a mirroring
454 .\" ==== pfs-slave ====
455 .It Ar pfs-slave Ar dirpath Op options
456 Create a pseudo-filesystem (PFS) inside a
459 Up to 65535 such file systems can be created.
460 Each PFS uses an independent inode numbering space making it suitable
461 for use as a replication source or target.
465 directive creates a PFS that you can use as a mirroring target.
466 You will not be able to access a slave PFS until you have completed the
467 first mirroring operation with it as the target (its root directory will
468 not exist until then).
470 Access to the pfs-slave via the special softlink,
471 as described in the PFS NOTES below, allows
474 dynamically modify the snapshot transaction id by returning a dynamic result
479 A PFS can only be truly destroyed with the
482 Removing the softlink will not destroy the underlying PFS.
483 .\" ==== pfs-update ====
484 .It Ar pfs-update Ar dirpath Op options
485 Update the configuration parameters for an existing
488 or pseudo-filesystem. Options that may be specified:
489 .Bl -tag -width indent
490 .It sync-beg-tid=0x16llx
491 This is the automatic snapshot access starting transaction id for mirroring slaves.
492 This parameter is normally updated automatically by the
496 It is important to note that accessing a mirroring slave
497 with a transaction id greater than the last fully synchronized transaction
498 id can result in an unreliable snapshot since you will be accessing
499 data that is still undergoing synchronization.
501 Manually modifying this field is dangerous and can result in a broken
503 .It sync-end-tid=0x16llx
504 This is the current synchronization point for mirroring slaves.
505 This parameter is normally updated automatically by the
509 Manually modifying this field is dangerous and can result in a broken
511 .It shared-uuid=<uuid>
512 Set the shared UUID for this file system. All mirrors must have the same
513 shared UUID. For safety purposes the
515 directives will refuse
516 to operate on a target with a different shared UUID.
518 Changing the shared UUID on an existing, non-empty mirroring target,
519 including an empty but not completely pruned target, can lead
520 to corruption of the mirroring target.
521 .It unique-uuid=<uuid>
522 Set the unique UUID for this file system. This UUID should not be used
523 anywhere else, even on exact copies of the file system.
525 Set a descriptive label for this file system.
526 .It snapshots=<string>
527 Specify the snapshots directory which
531 this PFS. The snapshots directory does not need to be configured for
532 PFS masters and will default to
533 .Pa <pfs>/snapshots .
535 PFS slaves are mirroring slaves so you cannot configure a snapshots
536 directory on the slave itself to be managed by the slave's machine.
537 In fact, the slave will likely have a
539 sub-directory mirrored
540 from the master, but that directory contains the configuration the master
541 is using for its copy of the file system, not the configuration that we
542 want to use for our slave.
544 It is recommended that
545 .Pa <fs>/var/slaves/<name>
546 be configured for a PFS slave, where
552 is an appropriate label.
553 You can control snapshot
554 retention on your slave independent of the master.
556 Zero out the snapshots directory path for this PFS.
558 .\" ==== pfs-upgrade ====
559 .It Ar pfs-upgrade Ar dirpath
560 Upgrade a PFS from slave to master operation. The PFS will be rolled back
561 to the current end synchronization tid (removing any partial synchronizations),
562 and will then becomes writable.
566 currently supports only single masters and using
567 this command can easily result in file system corruption if you don't
568 know what you are doing.
570 This directive will refuse to run if any programs have open descriptors
571 in the PFS, including programs chdir'd into the PFS.
572 .\" ==== pfs-downgrade ====
573 .It Ar pfs-downgrade Ar dirpath
574 Downgrade a master PFS from master to slave operation. The PFS becomes
575 read-only and access will be locked to its
578 This directive will refuse to run if any programs have open descriptors
579 in the PFS, including programs chdir'd into the PFS.
580 .\" ==== pfs-destroy ====
581 .It Ar pfs-destroy Ar dirpath
582 This permanently destroys a PFS.
584 This directive will refuse to run if any programs have open descriptors
585 in the PFS, including programs chdir'd into the PFS.
586 .\" ==== mirror-read ====
587 .It Ar mirror-read Ar filesystem Op Ar <begin-tid>
588 Generate a mirroring stream to stdout.
589 The stream ends when the transaction id space has been exhausted.
590 .\" ==== mirror-read-stream ====
591 .It Ar mirror-read-stream Ar filesystem Op Ar <begin-tid>
592 Generate a mirroring stream to stdout.
593 Upon completion the stream is paused until new data is synced to the
594 master, then resumed.
595 Operation continues until the pipe is broken.
596 .\" ==== mirror-write ====
597 .It Ar mirror-write Ar filesystem
598 Take a mirroring stream on stdin.
600 This command will fail if the
602 configuration field for the two file systems do not match.
603 .\" ==== mirror-dump ====
610 representation of the mirroring stream.
611 .\" ==== mirror-copy ====
612 .It Ar mirror-copy Ar [[user@]host:]filesystem Ar [[user@]host:]filesystem
613 This is a shortcut which pipes a
617 command. If a remote host specification is made the program forks a
623 on the appropriate host.
624 The source may be a master or slave PFS, and the target must be a slave PFS.
626 This command also established full duplex communication and turns on
627 the two-way protocol feature which automatically negotiates transaction id ranges
628 without having to use a cycle file.
629 If the operation completes successfully the target PFS's
632 be updated. Note that you must re-chdir into the target PFS to see the
633 updated information. If you do not you will still be in the previous snapshot.
634 .\" ==== mirror-stream ====
635 .It Ar mirror-stream Ar [[user@]host:]filesystem Ar [[user@]host:]filesystem
636 This command works similarly to
638 but does not exit unless the pipe is broken.
639 This command will resume the mirroring operation whenever the master is
640 synced. The command is commonly used with
644 options to keep the mirroring target in sync with the source on a continuing
648 .Sh PSEUDO FILESYSTEM (PFS) NOTES
649 The root of a PFS is not hooked into the primary
655 creates a special softlink called "@@PFS%05d" (exactly 10
656 characters long) in the primary
660 then modifies the contents of the softlink as read by
662 and thus what you see with an
664 command or if you were to
667 If the PFS is a master the link reflects the current state of the PFS.
668 If the PFS is a slave the link reflects the last completed snapshot, and the
669 contents of the link will change when the next snapshot is completed, and
672 PFS support is currently very new and experimental. The
675 employs numerous safeties to reduce user foot-shooting.
678 directive requires that the target be configured as a slave and that the
680 field of the mirroring source and target match.
684 .Bl -tag -width ".It Pa <fs>/var/slaves/<name>" -compact
686 default per PFS snapshots directory
687 .It Pa <snapshots>/config
691 .It Pa <fs>/var/slaves/<name>
692 recommended slave PFS snapshots directory
702 utility first appeared in
705 .An Matthew Dillon Aq dillon@backplane.com