1AMANDA.CONF(5) File formats and conventions AMANDA.CONF(5)
2
6 amanda.conf - Main configuration file for Amanda, the Advanced Maryland
7 Automatic Network Disk Archiver
8
10 amanda.conf(5) is the main configuration file for Amanda. This manpage
11 lists the relevant sections and parameters of this file for quick
12 reference.
13 The file <CONFIG_DIR>/<config>/amanda.conf is loaded.
15
17 There are a number of configuration parameters that control the
18 behavior of the Amanda programs. All have default values, so you need
19 not specify the parameter in amanda.conf if the default is suitable.
20 COMMENTS
22 Lines starting with # are ignored, as are blank lines. Comments may be
23 placed on a line with a directive by starting the comment with a #. The
24 remainder of the line is ignored.
25 KEYWORDS AND IDENTIFIERS
27 Keywords are case insensitive, i.e. mailto and MailTo are treated the
28 same. Also, the characters '-' and '_' are interchangeable in all
29 predefined Amanda keywords: device_property and device-property have
30 the same meaning. This manpage uses the dashed versions, but the
31 underscored versions will be accepted for backward compatibility
32 Identifiers are names which are defined in the configuration itself,
34 such as dumptypes or interfaces. Identifiers are are case-insensitive,
35 but sensitive to '-' vs. '_'. Identifiers should be quoted in the
36 configuration file, although For historical reasons, the quotes are
37 optional.
38 Strings are always quoted with double quotes ("), and any double quotes
40 or backslashes within the string are escaped with a backslash:
41 tapelist "/path/to/tapelist"
43 property "escaped-string" "escaping: \\ (backslash) and \" (double-quote)"
44 To summarize, then:
46 # QUOTES CASE -/_
48 logdir "logs" # required sensitive sensitive
49 send-amreport-on strange # prohibited insensitive insensitive
50 tapetype "EXABYTE" # optional insensitive sensitive
51 define dumptype "dt" { # optional insensitive sensitive
53 "dumptype-common" # optional insensitive sensitive
54 strategy noinc # prohibited insensitive insensitive
55 }
56 VALUE SUFFIXES
58 Integer arguments may have one of the following (case insensitive)
59 suffixes, some of which have a multiplier effect:
60 b byte bytes
62 Some number of bytes.
63 bps
65 Some number of bytes per second.
66 k kb kbyte kbytes kilobyte kilobytes
68 Some number of kilobytes (bytes*1024).
69 kps kbps
71 Some number of kilobytes per second (bytes*1024).
72 It is the default multiplier for all size options.
74 m mb meg mbyte mbytes megabyte megabytes
76 Some number of megabytes (bytes*1024*1024).
77 mps mbps
79 Some number of megabytes per second (bytes*1024*1024).
80 g gb gbyte gbytes gigabyte gigabytes
82 Some number of gigabytes (bytes*1024*1024*1024).
83 t tb tbyte tbytes terabyte terabytes
85 Some number of terabytes (bytes*1024*1024*1024*1024).
86 tape tapes
88 Some number of tapes.
89 day days
91 Some number of days.
92 week weeks
94 Some number of weeks (days*7).
95 Note
97 The value inf may be used in most places where an integer is
98 expected to mean an infinite amount.
99 Boolean arguments may have any of the values 1, y, yes, t, true
101 or on to indicate a true state, or 0, n, no, f, false or off to
102 indicate a false state. If no argument is given, true is
103 assumed.
104 PARAMETER ORDER
106 In general, the order in which parameters occur in the configuration
107 file does not matter, with the exception of subsection inheritance. For
108 example, if dumptype "normal-encrypt" which inherits from dumptype
109 "normal", then "normal" must appear first in the configuration file.
110 STRINGS
112 Quoted strings in Amanda follow a common, C-like syntax. Printable
113 characters and whitespace are kept as-is, except that the backslash
114 character (\) is used as an escape character, and a double-quote ends
115 the string. The allowed escape sequences are
116 ESCAPE SEQUENCE BECOMES
118 \\ \
119 \" "
120 \n (newline)
121 \t (tab)
122 \r (carriage return)
123 \f (form-feed)
124 \1 - \7
125 \01 - \77
126 \001 - \377 (character specified in octal)
127 Illegally quoted strings are handled on a "best-effort" basis, which
128 may lead to unexpected results.
129 Examples:
131 finserver "/data/finance/XYZ Corp's \"real\" finances" finance-high eth0 -1
133 property "syspath" "C:\\WINDOWS\\SYSTEM"
134 SUBSECTIONS AND INHERITANCE
136 Amanda configuration files may include various subsections, each
137 defining a set of configuration directives. Each type of subsection is
138 described below. Note that all types of subsections can inherit from
139 other subsections of the same type by naming the "parent" section in
140 the "child" subsection. For example:
141 define dumptype global {
143 record yes
144 index yes
145 }
146 define dumptype nocomp {
148 global # inherit the parameters in dumptype 'global'
149 compress none
150 }
151 Note that multiple inheritance is also supported by simply naming
153 multiple parent sections in a child. Parents are implicitly expanded in
154 place in a child, and the last occurrence of each parameter takes
155 precedence. For example,
156 define tapetype par1 {
158 comment "Parent 1"
159 filemark 8k
160 speed 300bps
161 length 200M
162 }
163 define tapetype par2 {
164 comment "Parent 2"
165 filemark 16k
166 speed 400bps
167 }
168 define tapetype child {
169 par1
170 par2
171 filemark 32k
172 }
173 In this example, 'child' will have a filemark of 32k, a speed of
174 400bps, and a length of 200M.
175
177 org string
178 Default: "daily". A descriptive name for the configuration. This
179 string appears in the Subject line of mail reports. Each Amanda
180 configuration should have a different string to keep mail reports
181 distinct.
182 mailer string
184 Default found by configure. A mail program that can send mail with
185 'MAILER -s "subject" user < message_file'.
186 mailto string
188 Default: none. A space separated list of recipients for mail
189 reports. If not specified, amdump will not send any mail.
190 send-amreport-on [ all | strange | error | never ]
192 Default: all. Specify which types of messages will trigger an email
193 from amreport. amreport is used by amdump and amflush.
194 all
196 Send an email on any message.
197 strange
199 Send an email on strange or error message. A strange message
200 occurs when the dump succeeded, but returned one or more errors
201 unknown to Amanda.
202 error
204 Send an email only on error messages.
205 never
207 Never send an email.
208 report-use-media boolean
210 Default: True if max-dle-by-volume is not set, False if it is set.
211 If the reporter must print the list of media used in the run.
212 report-next-media boolean
214 Default: True if max-dle-by-volume is not set, False if it is set.
215 If the reporter must print the list of media expected for the next
216 run.
217 max-dle-by-volume int
219 Default: 1000000000. The maximum number of dle written to a single
220 volume.
221 dumpcycle int
223 Default: 10 days. The number of days in the backup cycle. Each disk
224 will get a full backup at least this often. Setting this to zero
225 tries to do a full backup each run.
226 Note
228 This parameter may also be set in a specific dumptype (see
229 below). This value sets the default for all dumptypes so must
230 appear in amanda.conf before any dumptypes are defined.
231 runspercycle int
233 Default: same as dumpcycle. The number of amdump runs in dumpcycle
234 days. A value of 0 means the same value as dumpcycle. A value of -1
235 means guess the number of runs from the tapelist(5) file, which is
236 the number of tapes used in the last dumpcycle days / runtapes.
237 tapecycle int
239 Default: 15 tapes. Specifies the number of "active" volumes -
240 volumes that Amanda will not overwrite. While Amanda is always
241 willing to write to a new volume, it refuses to overwrite a volume
242 unless at least 'tapecycle -1' volumes have been written since.
243 It is considered good administrative practice to set the tapecycle
245 parameter slightly lower than the actual number of tapes in use.
246 This allows the administrator to more easily cope with damaged or
247 misplaced tapes or schedule adjustments that call for slight
248 adjustments in the rotation order.
249 Note: Amanda is commonly misconfigured with tapecycle equal to the
251 number of tapes per dumpcycle. In this misconfiguration, amanda may
252 erase a full dump before a new one is completed. Recovery is then
253 impossible. The tapecycle must be at least one tape larger than the
254 number of tapes per dumpcycle.
255 The number of tapes per dumpcycle is calculated by multiplying the
257 number of amdump runs per dump cycle runspercycle (the number of
258 amdump runs per dump cycle) and runtapes (the number of tapes used
259 per run). Typically tapecycle is set to two or four times the tapes
260 per dumpcycle.
261 usetimestamps bool
263 Default: Yes. This option allows Amanda to track multiple runs per
264 calendar day. The only reason one might disable it is that Amanda
265 versions before 2.5.1 can't read logfiles written when this option
266 was enabled.
267 label-new-tapes string
269 Deprecated, use autolabel option with options volume-error empty to
270 get equivalent behavior.
271 Default: not set. When set, this directive will cause Amanda to
273 automatically write an Amanda tape label to any blank tape she
274 encounters.
275 autolabel string [any] [other-config] [non-amanda] [volume-error]
277 [empty]
278 Default: not set. When set, this directive will cause Amanda to
279 automatically write an Amanda tape label to most volume she
280 encounters. This option is DANGEROUS because when set, Amanda may
281 erase near-failing tapes or tapes accidentally loaded in the wrong
282 slot.
283 When using this directive, specify the template for new tape
285 labels. The template can contains many variables that are
286 substituted by their values:
287 $c : config name
289 $o : org configuration
290 $b : barcode of the volume
291 $s : slot number, can specify a minimun number of digit:
292 $3s to get '001'
293 $m : meta label
294 The template can contain some number of contiguous '%' characters,
296 which will be replaced with a generated number. Be sure to specify
297 enough '%' characters that you do not run out of tape labels.
298 Example: "DailySet1-%%%", "$c-%%%", "$m-%%%", "$m-$b"
299 The generared label can be used only if it match the labelstr
301 setting. The volume will not be used if the generated label doesn't
302 match the labelstr setting.
303 Note that many devices cannot distinguish an empty tape from an
305 error condition, so it may is often necessary to include
306 volume-error as an autolabel condition.
307 any
309 equivalent to 'other-config non-amanda volume-error empty'
310 other-config
312 Label volumes with a valid Amanda label that do not match our
313 labelstr. Danger: this may erase volumes from other Amanda
314 configurations without warning!
315 non-amanda
317 Label volumes which do not start with data that resembles an
318 Amanda header. Danger: this may erase volumes from other backup
319 applications without warning!
320 volume-error
322 Label volumes where an error occurs while trying to read the
323 label. Danger: this may erase arbitrary volumes due to
324 transient errors.
325 empty
327 Label volumes where a read returns 0 bytes.
328 meta-autolabel string
330 Default: not set. When set and if the changer support meta-label,
331 this directive will cause Amanda to automatically add a meta-label
332 to a meta-volume.
333 A meta-volume is a containers that contains many volumes, eg. a
335 removable hard-disk for use with chg-disk, each hard disk have many
336 slots (volume). The meta-label is the label to put on the
337 meta-volume.
338 When using this directive, specify the template for new meta
340 labels. The template can contains many variables that are
341 substituted by their values:
342 $c : config name
344 $o : org configuration
345 The template should contain some number of contiguous '%'
347 characters, which will be replaced with a generated number. Be sure
348 to specify enough '%' characters that you do not run out of meta
349 labels. Example: "DailySet1-%%%", "$o-%%%",
350 dumpuser string
352 Default: "amanda". The login name Amanda uses to run the backups.
353 The backup client hosts must allow access from the tape server host
354 as this user via .rhosts or .amandahosts, depending on how the
355 Amanda software was built.
356 printer string
358 Printer to use when doing tape labels. See the lbl-templ tapetype
359 option.
360 tapedev string
362 Default: "null:". This parameter can either specify a device
363 (explicitly or by referencing a device definition - see amanda-
364 devices(7)) or a tape changer (explicitly or by referencing a
365 device definition - see amanda-changers(7)).
366 device-property string string
368 These options can set various device properties. See amanda-
369 devices(7) for more information on device properties and their
370 syntax. Both strings are always quoted; the first string contains
371 the name of the property to set, and the second contains its value.
372 For example, to set a fixed block size of 128k, write:
373 device-property "BLOCK_SIZE" "128k"
374 property [append] string string+
376 These options can set various properties, they can be used by third
377 party software to store information in the configuration file. Both
378 strings are quoted; the first string contains the name of the
379 property to set, and the others contains its values. append
380 keyword append the values to the list of values for that property.
381 tpchanger string
383 Default: not set. The tape changer to use. In most cases, only one
384 of tpchanger or tapedev is specified, although for backward
385 compatibility both may be specified if tpchanger gives the name of
386 an old changer script. See amanda-changers(7) for more information
387 on configuring changers.
388 interactivity string
390 Default: not set. The interactivity module Amanda should use to
391 interact with the user. See amanda-interactivity(7) for a list of
392 modules.
393 taperscan string
395 Default: traditional. The taperscan module amanda should use to
396 find a tape to write to. See amanda-taperscan(7) for a list of
397 modules.
398 changerdev string
400 Default: "dev/null". A tape changer configuration parameter. Usage
401 depends on the particular changer defined with the tpchanger
402 option.
403 changerfile string
405 Default: "/usr/adm/amanda/log/changer-status". A tape changer
406 configuration parameter. Usage depends on the particular changer
407 defined with the tpchanger option.
408 runtapes int
410 Default: 1. The maximum number of tapes used in a single run. If a
411 tape changer is not configured, this option is not used and should
412 be commented out of the configuration file.
413 If a tape changer is configured, this may be set larger than one to
415 let Amanda write to more than one tape.
416 Note that this is an upper bound on the number of tapes, and Amanda
418 may use less.
419 maxdumpsize int
421 Default: runtapes*tape-length. Maximum number of bytes the planner
422 will schedule for a run.
423 The default unit is Kbytes if it is not specified.
425 taperalgo [ first | firstfit | largest | largestfit | smallest | last ]
427 Default: first. The algorithm used to choose which dump image to
428 send to the taper.
429 first
431 First in, first out.
432 firstfit
434 The first dump image that will fit on the current tape.
435 largest
437 The largest dump image.
438 largestfit
440 The largest dump image that will fit on the current tape.
441 smallest
443 The smallest dump image.
444 last
446 Last in, first out.
447 taper-parallel-write int
449 Default: 1. Amanda can write simultaneously up to that number of
450 volume at any given time. The changer must have as many drives.
451 eject-volume int
453 Default: no. Set to yes if you want the volume to be ejected after
454 Amanda wrote data to it. It works only with some changer and
455 device.
456 labelstr string
458 Default: ".*". The tape label constraint regular expression. All
459 tape labels generated (see amlabel(8)) and used by this
460 configuration must match the regular expression. If multiple
461 configurations are run from the same tape server host, it is
462 helpful to set their labels to different strings (for example,
463 "DAILY[0-9][0-9]*" vs. "ARCHIVE[0-9][0-9]*") to avoid overwriting
464 each other's tapes.
465 tapetype string
467 Default: no default. The type of tape drive associated with tapedev
468 or tpchanger. This refers to one of the defined tapetypes in the
469 config file (see below), which specify various tape parameters,
470 like the length, filemark size, and speed of the tape media and
471 device.
472 ctimeout int
474 Default: 30 seconds. Maximum amount of time that amcheck will wait
475 for each client host.
476 dtimeout int
478 Default: 1800 seconds. Amount of idle time per disk on a given
479 client that a dumper running from within amdump will wait before it
480 fails with a data timeout error.
481 etimeout int
483 Default: 300 seconds. Amount of time per estimate on a given client
484 that the planner step of amdump will wait to get the dump size
485 estimates (note: Amanda runs up to 3 estimates for each DLE). For
486 instance, with the default of 300 seconds and four DLE's, each
487 estimating level 0 and level 1 on client A, planner will wait up to
488 40 minutes for that machine. A negative value will be interpreted
489 as a total amount of time to wait per client instead of per disk.
490 connect-tries int
492 Default: 3. How many times the server will try a connection.
493 req-tries int
495 Default: 3. How many times the server will resend a REQ packet if
496 it doesn't get the ACK packet.
497 netusage int
499 Default: 80000 Kbps. The maximum network bandwidth allocated to
500 Amanda, in Kbytes per second. See also the interface section.
501 inparallel int
503 Default: 10. The maximum number of backups that Amanda will attempt
504 to run in parallel. Amanda will stay within the constraints of
505 network bandwidth and holding disk space available, so it doesn't
506 hurt to set this number a bit high. Some contention can occur with
507 larger numbers of backups, but this effect is relatively small on
508 most systems.
509 displayunit "k|m|g|t"
511 Default: "k". The unit used to print many numbers, k=kilo, m=mega,
512 g=giga, t=tera.
513 dumporder string
515 Default: "tttTTTTTTT". The priority order of each dumper:
516 s: smallest size
518 S: largest size
519 t: smallest time
520 T: largest time
521 b: smallest bandwidth
522 B: largest bandwidth
523 maxdumps int
525 Default: 1. The maximum number of backups from a single host that
526 Amanda will attempt to run in parallel. See also the inparallel
527 option.
528 Note that this parameter may also be set in a specific dumptype
530 (see below). This value sets the default for all dumptypes so must
531 appear in amanda.conf before any dumptypes are defined.
532 bumpsize int
534 Default: 10 Mbytes. The minimum savings required to trigger an
535 automatic bump from one incremental level to the next, expressed as
536 size. If Amanda determines that the next higher backup level will
537 be this much smaller than the current level, it will do the next
538 level. The value of this parameter is used only if the parameter
539 bumppercent is set to 0.
540 The default unit is Kbytes if it is not specified.
542 The global setting of this parameter can be overwritten inside of a
544 dumptype-definition.
545 See also the options bumppercent, bumpmult and bumpdays.
547 bumppercent int
549 Default: 0. The minimum savings required to trigger an automatic
550 bump from one incremental level to the next, expressed as
551 percentage of the current size of the DLE (size of current level
552 0). If Amanda determines that the next higher backup level will be
553 this much smaller than the current level, it will do the next
554 level.
555 If this parameter is set to 0, the value of the parameter bumpsize
557 is used to trigger bumping.
558 The global setting of this parameter can be overwritten inside of a
560 dumptype-definition.
561 See also the options bumpsize, bumpmult and bumpdays.
563 bumpmult float
565 Default: 1.5. The bump size multiplier. Amanda multiplies bumpsize
566 by this factor for each level. This prevents active filesystems
567 from bumping too much by making it harder to bump to the next
568 level. For example, with the default bumpsize and bumpmult set to
569 2.0, the bump threshold will be 10 Mbytes for level one, 20 Mbytes
570 for level two, 40 Mbytes for level three, and so on.
571 The global setting of this parameter can be overwritten inside of a
573 dumptype-definition.
574 bumpdays int
576 Default: 2 days. To insure redundancy in the dumps, Amanda keeps
577 filesystems at the same incremental level for at least bumpdays
578 days, even if the other bump threshold criteria are met.
579 The global setting of this parameter can be overwritten inside of a
581 dumptype-definition.
582 diskfile string
584 Default: "disklist". The file name for the disklist file holding
585 client hosts, disks and other client dumping information.
586 infofile string
588 Default: "/usr/adm/amanda/curinfo". The file or directory name for
589 the historical information database. If Amanda was configured to
590 use DBM databases, this is the base file name for them. If it was
591 configured to use text formated databases (the default), this is
592 the base directory and within here will be a directory per client,
593 then a directory per disk, then a text file of data.
594 logdir string
596 Default: "/usr/adm/amanda". The directory for the amdump and log
597 files.
598 indexdir string
600 Default "/usr/adm/amanda/index". The directory where index files
601 (backup image catalogues) are stored. Index files are only
602 generated for filesystems whose dumptype has the index option
603 enabled.
604 tapelist string
606 Default: "tapelist". The file name for the active tapelist(5).
607 Amanda maintains this file with information about the active set of
608 tapes.
609 device-output-buffer-size int
611 Default: 1280k. Controls the amount of memory used by Amanda to
612 hold data as it is read from the network or disk before it is
613 written to the output device. Higher values may be useful on fast
614 tape drives and optical media.
615 The default unit is bytes if it is not specified.
617 tapebufs int
619 Default: 20. This option is deprecated; use the
620 device-output-buffer-size directive instead. tapebufs works the
621 same way, but the number specified is multiplied by the device
622 blocksize prior to use.
623 reserve int
625 Default: 100. The part of holding-disk space that should be
626 reserved for incremental backups if no tape is available, expressed
627 as a percentage of the available holding-disk space (0-100). By
628 default, when there is no tape to write to, degraded mode
629 (incremental) backups will be performed to the holding disk. If
630 full backups should also be allowed in this case, the amount of
631 holding disk space reserved for incrementals should be lowered.
632 autoflush no|yes|all
634 Default: no. Whether an amdump run will flush the dumps from
635 holding disk to tape. With yes, only dump matching the command line
636 argument are flushed. With all, all dump are flushed.
637 amrecover-do-fsf bool
639 Deprecated; amrecover always uses fsf, and does not invoke
640 amrestore.
641 Default: on. Amrecover will call amrestore with the -f flag for
643 faster positioning of the tape.
644 amrecover-check-label bool
646 Deprecated; amrecover always checks the label, and does not invoke
647 amrestore.
648 Default: on. Amrecover will call amrestore with the -l flag to
650 check the label.
651 amrecover-changer string
653 Default: not set. Amrecover will use the changer if you use
654 'settape <string>' and that string is the same as the
655 amrecover-changer setting.
656 columnspec string
658 default:
659 "HostName=0:-12:12,Disk=1:-11:11,Level=1:-1:1,OrigKB=1:-7:0,OutKB=1:-7:0,Compress=1:-6:1,DumpTime=1:-7:7,Dumprate=1:-6:1,TapeTime=1:-6:6,TapeRate=1:-6:1"
660 Defines the width of columns amreport should use. String is a
662 comma (',') separated list of triples. Each triple consists of
663 three parts which are separated by a equal sign ('=') and a colon
664 (':') (see the example). These four parts specify:
665 1. the name of the column, which may be:
667 Compress (compression ratio)
669 Disk (client disk name)
670 DumpRate (dump rate in KBytes/sec)
671 DumpTime (total dump time in hours:minutes)
672 HostName (client host name)
673 Level (dump level)
674 OrigKB (original image size in KBytes)
675 OutKB (output image size in KBytes)
676 TapeRate (tape writing rate in KBytes/sec)
677 TapeTime (total tape time in hours:minutes)
678 2. the amount of space to display before the column (used to get
680 whitespace between columns).
681 3. the width of the column itself. If set to a negative value, the
683 width will be calculated on demand to fit the largest entry in
684 this column.
685 4. the precision of the column, number of digit after the decimal
687 point for number.
688 Here is an example:
690 columnspec "Disk=1:18,HostName=0:10,OrigKB=::2,OutKB=1:7"
692 The above will display the disk information in 18 characters and
694 put one space before it. The hostname column will be 10 characters
695 wide with no space to the left. The Original KBytes print 2 decimal
696 digit. The output KBytes column is seven characters wide with one
697 space before it.
698 includefile string
700 Default: no default. The name of an Amanda configuration file to
701 include within the current file. Useful for sharing dumptypes,
702 tapetypes and interface definitions among several configurations.
703 Relative pathnames are relative to the configuration directory.
704 debug-days int
706 Default: 3. The number of days the debug files are kept.
707 debug-auth int
709 Default: 0. Debug level of the auth module
710 debug-event int
712 Default: 0. Debug level of the event module
713 debug-holding int
715 Default: 0. Debug level of the holdingdisk module
716 debug-protocol int
718 Default: 0. Debug level of the protocol module
719 debug-planner int
721 Default: 0. Debug level of the planner process
722 debug-driver int
724 Default: 0. Debug level of the driver process
725 debug-dumper int
727 Default: 0. Debug level of the dumper process
728 debug-chunker int
730 Default: 0. Debug level of the chunker process
731 debug-taper int
733 Default: 0. Debug level of the taper process
734 debug-recovery int
736 Default: 1. Debug level of all recovery process
737 flush-threshold-dumped int
739 Default: 0. Amanda will not begin writing data to a new volume
740 until the amount of data on the holding disk is at least this
741 percentage of the volume size. In other words, Amanda will not
742 begin until the amount of data on the holding disk is greater than
743 the tape length times this parameter. This parameter may be larger
744 than 100%, for example to keep more recent dumps on the holding
745 disk for faster recovery.
746 Needless to say, your holding disk must be big enough that this
748 criterion could be satisfied. If the holding disk cannot be used
749 for a particular dump (because, for example, there is no remaining
750 holding space) then Amanda will disregard the constraint specified
751 by this setting and start a new volume anyway. Once writing to a
752 volume has begun, this constraint is not applied unless and until a
753 new volume is needed.
754 The value of this parameter may not exceed than that of the
756 flush-threshold-scheduled parameter.
757 flush-threshold-scheduled int
759 Default: 0. Amanda will not begin writing data to a new volume
760 until the sum of the amount of data on the holding disk and the
761 estimated amount of data remaining to be dumped during this run is
762 at least this percentage of the volume size. In other words, Amanda
763 will not begin until the inequality h + s > t × d is satisfied,
764 where h is the amount of data on the holding disk, s is the total
765 amount of data scheduled for this run but not dumped yet, t is the
766 capacity of a volume, and d is this parameter, expressed as a
767 percentage. This parameter may be larger than 100%.
768 Needless to say, your holding disk must be big enough that this
770 criterion could be satisfied. If the holding disk cannot be used
771 for a particular dump (because, for example, there is no remaining
772 holding space) then Amanda will disregard the constraint specified
773 by this setting and start a new volume anyway. Once writing to a
774 volume has begun, this constraint is not applied unless and until a
775 new volume is needed.
776 The value of this parameter may not be less than that of the
778 flush-threshold-dumped or taperflush parameters.
779 taperflush int
781 Default: 0. At the end of a run, Amanda will start a new tape to
782 flush remaining data if there is more data on the holding disk at
783 the end of a run than this setting allows; the amount is specified
784 as a percentage of the capacity of a single volume. In other words,
785 at the end of a run, Amanda will begin a new tape if the inequality
786 h > t × f is satisfied, where h is the amount of data remaining on
787 the holding disk from this or previous runs, t is the capacity of a
788 volume, and f is this parameter, expressed as a percentage. This
789 parameter may be greater than 100%.
790 The value of this parameter may not exceed that of the
792 flush-threshold-scheduled parameter.; autoflush must be set to
793 'yes' if taperflush is greater than 0.
794 reserved-udp-port int,int
796 Default: --with-udpportrange or 512,1023. Reserved udp port that
797 will be used (bsd, bsdudp). Range is inclusive.
798 reserved-tcp-port int,int
800 Default: --with-low-tcpportrange or 512,1023. Reserved tcp port
801 that will be used (bsdtcp). Range is inclusive.
802 unreserved-tcp-port int,int
804 Default: --with-tcpportrange or 1024,65535. Unreserved tcp port
805 that will be used (bsd, bsdudp). Range is inclusive.
806 recovery-limit [ string | same-host | server]
808 Default: none (no limitations). This parameter limits the hosts
809 that may do recoveries. Hosts are identified by their authenticated
810 peer name, as described in amanda-auth(7); if this is not available
811 and the recovery-limit parameter is present, recovery will be
812 denied. The arguments to the parameter are strings giving host
813 match expressions (see amanda-match(7)) or the special keywords
814 same-host or server. The same-host keyword requires an exact match
815 to the hostname of the DLE being recovered. The server keyword
816 require the connection come from the fqdn of the server. Specifying
817 no arguments at all will disable all recoveries from any host.
818 Note that match expressions can be constructed to be forgiving of
820 e.g., fully-qualified vs. unqualified hostnames, but same-host
821 requires an exact match.
822 The error messages that appear in amrecover are intentionally vague
824 to avoid information leakage. Consult the amindexd debug log for
825 more details on the reasons a recovery was rejected.
826 Recovery limits can be refined on a per-DLE basis using the
828 dumptype parameter of the same name. Note that the default value
829 will apply to any dumpfiles for disks which no longer appear in the
830 disklist; thus leaving the global parameter at its default value
831 but setting it for all DLEs is not sufficient to maintain secure
832 backups.
833 tmpdir string
835 Default: none (system default). Set it to a directory with lots of
836 free space if sort in amindexd fail with 'No space left on device'.
837
839 The amanda.conf file may define one or more holding disks used as
840 buffers to hold backup images before they are written to tape. The
841 syntax is:
842 define holdingdisk name {
843 holdingdisk-option holdingdisk-value
844 ...
845 }
846 The { must appear at the end of a line, and the } on its own line.
848 Name is a logical name for this holding disk.
850 The options and values are:
852 comment string
854 Default: not set. A comment string describing this holding disk.
855 directory string
857 Default: "/dumps/amanda". The path to this holding area.
858 use int
860 Default: 0 Gb. Amount of space that can be used in this holding
861 disk area. If the value is zero, all available space on the file
862 system is used. If the value is negative, Amanda will use all
863 available space minus that value.
864 chunksize int
866 Default: 1 Gb. Holding disk chunk size. Dumps larger than the
867 specified size will be stored in multiple holding disk files. The
868 size of each chunk will not exceed the specified value. However,
869 even though dump images are split in the holding disk, they are
870 concatenated as they are written to tape, so each dump image still
871 corresponds to a single continuous tape section.
872 The default unit is Kbytes if it is not specified.
874 If 0 is specified, Amanda will create holding disk chunks as large
876 as ((INT_MAX/1024)-64) Kbytes.
877 Each holding disk chunk includes a 32 Kbyte header, so the minimum
879 chunk size is 64 Kbytes (but that would be really silly).
880 Operating systems that are limited to a maximum file size of 2
882 Gbytes actually cannot handle files that large. They must be at
883 least one byte less than 2 Gbytes. Since Amanda works with 32 Kbyte
884 blocks, and to handle the final read at the end of the chunk, the
885 chunk size should be at least 64 Kbytes (2 * 32 Kbytes) smaller
886 than the maximum file size, e.g. 2047 Mbytes.
887
889 The amanda.conf(5) file may define multiple sets of backup options and
890 refer to them by name from the disklist(5) file. For instance, one set
891 of options might be defined for file systems that can benefit from high
892 compression, another set that does not compress well, another set for
893 file systems that should always get a full backup and so on.
894 A set of backup options are entered in a dumptype section, which looks
896 like this:
897 define dumptype "name" {
898 dumptype-option dumptype-value
899 ...
900 }
901 The { must appear at the end of a line, and the } on its own line.
903 Name is the name of this set of backup options. It is referenced from
905 the disklist(5) file.
906 Some of the options in a dumptype section are the same as those in the
908 main part of amanda.conf(5). The main option value is used to set the
909 default for all dumptype sections. For instance, setting dumpcycle to
910 50 in the main part of the config file causes all following dumptype
911 sections to start with that value, but the value may be changed on a
912 section by section basis. Changes to variables in the main part of the
913 config file must be done before (earlier in the file) any dumptypes are
914 defined.
915 The dumptype options and values are:
917 auth string
919 Default: "bsdtcp". Type of authorization to perform between tape
920 server and backup client hosts. See amanda-auth(7) for more detail.
921 amandad-path string
923 Default: "$libexec/amandad". Specify the amandad path of the
924 client, only use with rsh/ssh authentification.
925 client-username string
927 Default: CLIENT_LOGIN. Specify the username to connect on the
928 client, only use with rsh/ssh authentification.
929 client-port [ int | string ]
931 Default: "amanda". Specifies the port to connect to on the client.
932 It can be a service name or a numeric port number.
933 bumpsize int
935 Default: 10 Mbytes. The minimum savings required to trigger an
936 automatic bump from one incremental level to the next, expressed as
937 size. If Amanda determines that the next higher backup level will
938 be this much smaller than the current level, it will do the next
939 level. The value of this parameter is used only if the parameter
940 bumppercent is set to 0.
941 The default unit is Kbytes if it is not specified.
943 See also the options bumppercent, bumpmult and bumpdays.
945 bumppercent int
947 Default: 0. The minimum savings required to trigger an automatic
948 bump from one incremental level to the next, expressed as
949 percentage of the current size of the DLE (size of current level
950 0). If Amanda determines that the next higher backup level will be
951 this much smaller than the current level, it will do the next
952 level.
953 If this parameter is set to 0, the value of the parameter bumpsize
955 is used to trigger bumping.
956 See also the options bumpsize, bumpmult and bumpdays.
958 bumpmult float
960 Default: 1.5. The bump size multiplier. Amanda multiplies bumpsize
961 by this factor for each level. This prevents active filesystems
962 from bumping too much by making it harder to bump to the next
963 level. For example, with the default bumpsize and bumpmult set to
964 2.0, the bump threshold will be 10 Mbytes for level one, 20 Mbytes
965 for level two, 40 Mbytes for level three, and so on.
966 bumpdays int
968 Default: 2 days. To insure redundancy in the dumps, Amanda keeps
969 filesystems at the same incremental level for at least bumpdays
970 days, even if the other bump threshold criteria are met.
971 comment string
973 Default: not set. A comment string describing this set of backup
974 options.
975 comprate float [, float ]
977 Default: 0.50, 0.50. The expected full and incremental compression
978 factor for dumps. It is only used if Amanda does not have any
979 history information on compression rates for a filesystem, so
980 should not usually need to be set. However, it may be useful for
981 the first time a very large filesystem that compresses very little
982 is backed up.
983 compress [ none | client | server ] [ best | fast | custom ]
985 Default: client fast. If Amanda does compression of the backup
986 images, it can do so either on the backup client host before it
987 crosses the network or on the tape server host as it goes from the
988 network into the holding disk or to tape. Which place to do
989 compression (if at all) depends on how well the dump image usually
990 compresses, the speed and load on the client or server, network
991 capacity, holding disk capacity, availability of tape hardware
992 compression, etc.
993 For either type of compression, Amanda also allows the selection of
995 three styles of compression. best is the best compression
996 available, often at the expense of CPU overhead. fast is often not
997 as good a compression as best, but usually less CPU overhead. Or to
998 specify custom to use your own compression method. (See dumptype
999 custom-compress in example/amanda.conf for reference)
1000 So the compress options line may be one of:
1002 compress none
1004 compress client fast
1006 compress client best
1008 compress client custom
1010 Specify client-custom-compress "PROG"
1011 PROG must not contain white space and it must accept -d for
1013 uncompress.
1014 compress server fast
1016 compress server best
1018 compress server custom
1020 Specify server-custom-compress "PROG"
1021 PROG must not contain white space and it must accept -d for
1023 uncompress.
1024 Note that some tape devices do compression and this option has
1026 nothing to do with whether that is used. If hardware compression is
1027 used (usually via a particular tape device name or mt option),
1028 Amanda (software) compression should be disabled.
1029 client-custom-compress string
1031 Default: none. The program to use to perform
1032 compression/decompression on the client; used with "compress client
1033 custom". Must not contain whitespace. Must accept -d to uncompress.
1034 server-custom-compress string
1036 Default: none. The program to use to perform
1037 compression/decompression on the server; used with "compress server
1038 custom". Must not contain whitespace. Must accept -d to uncompress.
1039 dumpcycle int
1041 Default: 10 days. The number of days in the backup cycle. Each disk
1042 using this set of options will get a full backup at least this of
1043 ten. Setting this to zero tries to do a full backup each run.
1044 encrypt [ none | client | server ]
1046 Default: not set. To encrypt backup images, it can do so either on
1047 the backup client host before it crosses the network or on the tape
1048 server host as it goes from the network into the holding disk or to
1049 tape.
1050 So the encrypt options line may be one of:
1052 encrypt none
1054 encrypt client
1056 Specify client-encrypt "PROG"
1057 PROG must not contain white space.
1059 Specify client-decrypt-option "decryption-parameter" Default:
1061 "-d"
1062 decryption-parameter must not contain white space.
1064 (See dumptype client-encrypt-nocomp in example/amanda.conf for
1066 reference)
1067 encrypt server
1069 Specify server-encrypt "PROG"
1070 PROG must not contain white space.
1072 Specify server-decrypt-option "decryption-parameter" Default:
1074 "-d"
1075 decryption-parameter must not contain white space.
1077 (See dumptype server-encrypt-fast in example/amanda.conf for
1079 reference)
1080 Note that current logic assumes compression then encryption during
1082 backup(thus decrypt then uncompress during restore). So specifying
1083 client-encryption AND server-compression is not supported. amcrypt
1084 which is a wrapper of aespipe is provided as a reference symmetric
1085 encryption program.
1086 client-encrypt string
1088 Default: none. The program to use to perform encryption/decryption
1089 on the client; used with "encrypt client". Must not contain
1090 whitespace.
1091 client-decrypt-option string
1093 Default: -d. The option that can be passed to client-encrypt to
1094 make it decrypt instead. Must not contain whitespace.
1095 server-encrypt string
1097 Default: none. The program to use to perform encryption/decryption
1098 on the server; used with "encrypt server". Must not contain
1099 whitespace.
1100 server-decrypt-option string
1102 Default: -d. The option that can be passed to server-encrypt to
1103 make it decrypt instead. Must not contain whitespace.
1104 estimate [ client | calcsize | server ]+
1106 Default: client. Determine the way Amanda estimates the size of
1107 each DLE before beginning a backup. This is a list of acceptable
1108 estimate methods, and Amanda applies the first method supported by
1109 the application. The methods are:
1110 client
1112 Use the same program as the dumping program. This is the most
1113 accurate method to do estimates, but it can take a long time.
1114 calcsize
1116 Use a faster program to do estimates, but the result is less
1117 accurate.
1118 server
1120 Use only statistics from the previous few runs to give an
1121 estimate. This very quick, but the result is not accurate if
1122 your disk usage changes from day to day. If this method is
1123 specified, but the server does not have enough data to make an
1124 estimate, then the option is internally moved to the end of the
1125 list, thereby preferring 'client' or 'calcsize' in this case.
1126 exclude [ list | file ][[optional][append][ string ]+]
1128 Default: file. Exclude is the opposite of include and specifies
1129 files that will be excluded from the backup. The format of the
1130 exclude expressions depends on the application, and some
1131 applications do not support excluding files at all.
1132 There are two exclude parameters, exclude file and exclude list.
1134 With exclude file, the string is an exclude expression. With
1135 exclude list , the string is a file name on the client containing
1136 GNU-tar exclude expressions. The path to the specified exclude list
1137 file, if present (see description of 'optional' below), must be
1138 readable by the Amanda user.
1139 All exclude expressions are concatenated in one file and passed to
1141 the application as an --exclude-from argument.
1142 For GNU-tar, exclude expressions must always be specified as
1144 relative to the top-level directory of the DLE, and must start with
1145 "./". See the manpages for individual applications for more
1146 information on supported exclude expressions.
1147 With the append keyword, the string is appended to the current
1149 list, without it, the string overwrites the list.
1150 If optional is specified for exclude list, then amcheck will not
1152 complain if the file doesn't exist or is not readable.
1153 For exclude list, if the file name is relative, the disk name being
1155 backed up is prepended. So if this is entered:
1156 exclude list ".amanda.excludes"
1157 the actual file used would be /var/.amanda.excludes for a backup of
1158 /var, /usr/local/.amanda.excludes for a backup of /usr/local, and
1159 so on.
1160 holdingdisk [ never | auto | required ]
1162 Default: auto. Whether a holding disk should be used for these
1163 backups or whether they should go directly to tape. If the holding
1164 disk is a portion of another file system that Amanda is backing up,
1165 that file system should refer to a dumptype with holdingdisk set to
1166 never to avoid backing up the holding disk into itself.
1167 never|no|false|off
1169 Never use a holdingdisk, the dump will always go directly to
1170 tape. There will be no dump if you have a tape error.
1171 auto|yes|true|on
1173 Use the holding disk, unless there is a problem with the
1174 holding disk, the dump won't fit there or the medium doesn't
1175 require spooling (e.g., VFS device)
1176 required
1178 Always dump to holdingdisk, never directly to tape. There will
1179 be no dump if it doesn't fit on holdingdisk
1180 ignore boolean
1182 Default: no. Whether disks associated with this backup type should
1183 be backed up or not. This option is useful when the disklist file
1184 is shared among several configurations, some of which should not
1185 back up all the listed file systems.
1186 include [ list | file ][[optional][append][ string ]+]
1188 Default: file ".". There are two include lists, include file and
1189 include list. With include file , the string is a glob expression.
1190 With include list , the string is a file name on the client
1191 containing glob expressions.
1192 All include expressions are expanded by Amanda, concatenated in one
1194 file and passed to GNU-tar as a --files-from argument. They must
1195 start with "./" and contain no other "/".
1196 Include expressions must always be specified as relative to the
1198 head directory of the DLE.
1199 Note
1201 For globbing to work at all, even the limited single level, the
1202 top level directory of the DLE must be readable by the Amanda
1203 user.
1204 With the append keyword, the string is appended to the current
1205 list, without it, the string overwrites the list.
1206 If optional is specified for include list, then amcheck will not
1208 complain if the file doesn't exist or is not readable.
1209 For include list, If the file name is relative, the disk name being
1211 backed up is prepended.
1212 index boolean
1214 Default: no. Whether an index (catalogue) of the backup should be
1215 generated and saved in indexdir. These catalogues are used by the
1216 amrecover utility.
1217 kencrypt boolean
1219 Default: no. Whether the backup image should be encrypted by
1220 Kerberos as it is sent across the network from the backup client
1221 host to the tape server host.
1222 maxdumps int
1224 Default: 1. The maximum number of backups from a single host that
1225 Amanda will attempt to run in parallel. See also the main section
1226 parameter inparallel.
1227 maxpromoteday int
1229 Default: 10000. The maximum number of day for a promotion, set it 0
1230 if you don't want promotion, set it to 1 or 2 if your disks get
1231 overpromoted.
1232 max-warnings int
1234 Default: 20. The maximum number of error lines in the report for a
1235 dle. A value of '0' means unlimited. This is useful to reduce the
1236 size of the log file and the size of the report. All errors are put
1237 in separate files if a dle have more errors.
1238 priority [ low | medium | high ]
1240 Default: medium. When there is no tape to write to, Amanda will do
1241 incremental backups in priority order to the holding disk. The
1242 priority may be high (2), medium (1), low (0) or a number of your
1243 choice.
1244 program [ "DUMP" | "GNUTAR" | "APPLICATION" ]
1246 Default: "DUMP". The type of backup to perform. Valid values are:
1247 "DUMP"
1249 The native operating system backup program.
1250 "GNUTAR"
1252 To use GNU-tar or to do PC backups using Samba.
1253 "APPLICATION"
1255 To use an application, see the application option.
1256 application string
1258 No default. Must be the name of an application if program is set to
1259 APPLICATION. See APPLICATION SECTION below.
1260 script string
1262 No default. Must be the name of a script. You can have many script.
1263 See SCRIPT SECTION below.
1264 property [append] string string+
1266 These options can set various properties, they can be used by third
1267 party software to store information in the configuration file. Both
1268 strings are quoted; the first string contains the name of the
1269 property to set, and the others contains its values. append
1270 keyword append the values to the list of values for that property.
1271 record boolean
1273 Default: yes. Whether to ask the backup program to update its
1274 database (e.g. /etc/dumpdates for DUMP or
1275 /usr/local/var/amanda/gnutar-lists for GNUTAR) of time stamps. This
1276 is normally enabled for daily backups and turned off for periodic
1277 archival runs.
1278 skip-full boolean
1280 Default: no. If true and planner has scheduled a full backup, these
1281 disks will be skipped, and full backups should be run off-line on
1282 these days. It was reported that Amanda only schedules level 1
1283 incrementals in this configuration; this is probably a bug.
1284 skip-incr boolean
1286 Default: no. If true and planner has scheduled an incremental
1287 backup, these disks will be skipped.
1288 ssh-keys string
1290 Default: not set. The key file the ssh auth will use, it must be
1291 the private key. If this parameter is not specified, then the
1292 default ssh key will be used.
1293 starttime int
1295 Default: not set. Backup of these disks will not start until after
1296 this time of day. The value should be hh*100+mm, e.g. 6:30PM
1297 (18:30) would be entered as 1830.
1298 strategy [ standard | nofull | noinc | skip | incronly ]
1300 Default: standard. Strategy to use when planning what level of
1301 backup to run next. Values are:
1302 standard
1304 The standard Amanda schedule.
1305 nofull
1307 Never do full backups, only level 1 incrementals.
1308 noinc
1310 Never do incremental backups, only full dumps.
1311 skip
1313 Treat this DLE as if it doesn't exist (useful to disable DLEs
1314 when sharing the disklist file between multiple
1315 configurations). Skipped DLEs will not be checked or dumped,
1316 and will not be matched by disklist expressions.
1317 incronly
1319 Only do incremental dumps. amadmin force should be used to
1320 tell Amanda that a full dump has been performed off-line, so
1321 that it resets to level 1.
1322 allow-split bool
1324 Default: true. If true, then dumps with this dumptype can be split
1325 on the storage media. If false, then the dump will be written in a
1326 single file on the media. See "Dump Splitting Configuration" below.
1327 tape-splitsize int
1329 Deprecated. See "Dump Splitting Configuration" below.
1330 Default: not set. Split dump file on tape into pieces of a
1332 specified size. The default unit is Kbytes if it is not specified.
1333 split-diskbuffer string
1335 Deprecated. See "Dump Splitting Configuration" below. Default: not
1337 set. When dumping a split dump in PORT-WRITE mode (usually meaning
1338 "no holding disk"), buffer the split chunks to a file in the
1339 directory specified by this option.
1340 fallback-splitsize int
1342 Deprecated. See "Dump Splitting Configuration" below.
1343 Default: 10M. This specifies the part size used when no
1345 split-diskbuffer is specified, or when it is too small or does not
1346 exist, and thus the maximum amount of memory consumed for in-memory
1347 splitting. The default unit is Kbytes if it is not specified.
1348 recovery-limit [ server | same-host | string ]*
1350 Default: global value. This parameter overrides the global
1351 recovery-limit parameter for DLEs of this dumptype.
1352 dump-limit [ server | same-host ]*
1354 Default: server. Specify which host can initiate a backup of the
1355 dle. With server, the server can initiate a backup with the amdump
1356 command. With same-host, the client can initiate a backup with the
1357 amdump_client command.
1358 The following dumptype entries are predefined by Amanda:
1360 define dumptype "no-compress" {
1361 compress none
1362 }
1363 define dumptype "compress-fast" {
1364 compress client fast
1365 }
1366 define dumptype "compress-best" {
1367 compress client best
1368 }
1369 define dumptype "srvcompress" {
1370 compress server fast
1371 }
1372 define dumptype "bsd-auth" {
1373 auth "bsd"
1374 }
1375 define dumptype "bsdtcp-auth" {
1376 auth "bsdtcp"
1377 }
1378 define dumptype "no-record" {
1379 record no
1380 }
1381 define dumptype "no-hold" {
1382 holdingdisk no
1383 }
1384 define dumptype "no-full" {
1385 skip-full yes
1386 }
1387 In addition to options in a dumptype section, one or more other
1389 dumptype names may be supplied as identifiers, which make this dumptype
1390 inherit options from other previously defined dumptypes. For instance,
1391 two sections might be the same except for the record option:
1392 define dumptype "normal" {
1393 comment "Normal backup, no compression, do indexing"
1394 no-compress
1395 index yes
1396 maxdumps 2
1397 }
1398 define dumptype "testing" {
1399 comment "Test backup, no compression, do indexing, no recording"
1400 "normal"
1401 record no
1402 }
1403 Amanda provides a dumptype named global in the sample amanda.conf file
1405 that all dumptypes should reference. This provides an easy place to
1406 make changes that will affect every dumptype, although you must be
1407 careful that every dumptype explicitly inherits from the global
1408 dumptype - Amanda does not do so automatically.
1409
1411 The amanda.conf file may define multiple types of tape media and
1412 devices. The information is entered in a tapetype section, which looks
1413 like this in the config file:
1414 define tapetype "name" {
1415 tapetype-option tapetype-value
1416 ...
1417 }
1418 The { must appear at the end of a line, and the } on its own line.
1420 Name is the name of this type of tape medium/device. It is referenced
1422 from the tapetype option in the main part of the config file.
1423 The tapetype options and values are:
1425 comment string
1427 Default: not set. A comment string describing this set of tape
1428 information.
1429 filemark int
1431 Default: 1 kbytes. How large a file mark (tape mark) is, measured
1432 in kbytes. If the size is only known in some linear measurement
1433 (e.g. inches), convert it to kbytes using the device density.
1434 length int
1436 Default: 2000 kbytes. How much data will fit on a tape, expressed
1437 in kbytes.
1438 Note that this value is only used by Amanda to schedule which
1440 backups will be run. Once the backups start, Amanda will continue
1441 to write to a tape until it gets an error, regardless of what value
1442 is entered for length (but see amanda-devices(7) for exceptions).
1443 blocksize int
1445 Default: 32 kbytes. How much data will be written in each tape
1446 record, expressed in kbytes. This is similar to the BLOCK_SIZE
1447 device property, but if the blocksize is not a multiple of 1024
1448 bytes, then this parameter cannot be used to specify it, and the
1449 property must be used instead.
1450 readblocksize int
1452 Default: 32 kytes How much data will be read in each tape record.
1453 This can be used to override a device's block size for reads only.
1454 This may be useful, for example, in reading a tape written with a
1455 256k block size when Amanda is configured to use 128k blocks. This
1456 unusual feature is not supported by all operating systems and tape
1457 devices.
1458 The default unit is Kbytes if it is not specified.
1460 part-size int
1462 If this is set to zero (default), then no splitting will take
1463 place, and the entire dump will fail, if end-of-medium is
1464 encountered before the dump is complete, unless the device property
1465 LEOM is true, and the device can detect EOM. See "Dump Splitting
1466 Configuration" below.
1467 part-cache-type [ none | disk | memory ]
1469 Default: none. When part caching is required, this parameter
1470 specifies the type of caching that will be used. The options
1471 include no caching (none), in which case a failed part will cause
1472 the entire dump to fail; on-disk caching (disk), for which
1473 part-cache-dir must be set properly; and in-memory caching
1474 (memory), which on most systems severely restrains the size of the
1475 part that can be written. See "Dump Splitting Configuration" below.
1476 part-cache-dir string
1478 Default: none. The directory in which part-cache files can be
1479 written when caching on disk. See "Dump Splitting Configuration"
1480 below.
1481 part-cache-max-size int
1483 Default: none. The maximum part size to use when caching is in
1484 effect. This is used to limit the part size when disk or memory
1485 space for caching is constrained. This value must be greater than
1486 zero.
1487 speed int
1489 Default: 200 bps. How fast the drive will accept data, in bytes per
1490 second. This parameter is NOT currently used by Amanda.
1491 lbl-templ string
1493 Default: not set. A PostScript template file used by amreport to
1494 generate labels. Several sample files are provided with the Amanda
1495 sources in the example directory. See the amreport(8) man page for
1496 more information.
1497 In addition to options, another tapetype name may be supplied as an
1499 identifier, which makes this tapetype inherit options from another
1500 tapetype. For instance, the only difference between a DLT4000 tape
1501 drive using Compact-III tapes and one using Compact-IV tapes is the
1502 length of the tape. So they could be entered as:
1503 define tapetype "DLT4000-III" {
1504 comment "DLT4000 tape drives with Compact-III tapes"
1505 length 12500 mbytes # 10 Gig tapes with some compression
1506 filemark 2000 kbytes
1507 speed 1536 kps
1508 }
1509 define tapetype "DLT4000-IV" {
1510 "DLT4000-III"
1511 comment "DLT4000 tape drives with Compact-IV tapes"
1512 length 25000 mbytes # 20 Gig tapes with some compression
1513 }
1514
1516 The amanda.conf file may define multiple types of network interfaces.
1517 The information is entered in an interface section, which looks like
1518 this:
1519 define interface "name" {
1520 interface-option interface-value
1521 ...
1522 }
1523 The { must appear at the end of a line, and the } on its own line.
1525 name is the name of this type of network interface. It is referenced
1527 from the disklist file.
1528 Note that these sections define network interface characteristics, not
1530 the actual interface that will be used. Nor do they impose limits on
1531 the bandwidth that will actually be taken up by Amanda. Amanda computes
1532 the estimated bandwidth each file system backup will take based on the
1533 estimated size and time, then compares that plus any other running
1534 backups with the limit as another of the criteria when deciding whether
1535 to start the backup. Once a backup starts, Amanda will use as much of
1536 the network as it can leaving throttling up to the operating system and
1537 network hardware.
1538 The interface options and values are:
1540 comment string
1542 Default: not set. A comment string describing this set of network
1543 information.
1544 use int
1546 Default: 80000 Kbps. The speed of the interface in Kbytes per
1547 second.
1548 In addition to options, another interface name may be supplied as an
1550 identifier, which makes this interface inherit options from another
1551 interface. At the moment, this is of little use.
1552
1554 The amanda.conf file may define multiple types of application. The
1555 information is entered in a application section, which looks like this:
1556 define application "name" {
1557 application-option application-value
1558 ...
1559 }
1560 The { must appear at the end of a line, and the } on its own line.
1562 name is the name of this type of application. It is referenced from the
1564 dumptype
1565 The application options and values are:
1567 client-name string
1569 No default, specifies an application name that is in the
1570 amanda-client.conf on the client. The setting from that application
1571 will be merged with the current application. If client-name is set
1572 then it is an error if that application is not defined on the
1573 client.
1574 If client-name is not set then the merge is done with the
1576 application that have the name equal to the plugin. eg. if the
1577 plugin is 'amgtar', then the setting from the application 'amgtar'
1578 is used if it is defined.
1579 comment string
1581 Default: not set. A comment string describing this application.
1582 plugin string
1584 No default. Must be set to the name of the program. This program
1585 must be in the $libexecdir/amanda/application directory on the
1586 client.
1587 property [append] [priority] string string+
1589 No default. You can set property for the application, each
1590 application have a different set of property. Both strings are
1591 quoted; the first string contains the name of the property to set,
1592 and the others contains its values. append keyword append the
1593 values to the list of values for that property. priority keyword
1594 disallow the setting of that property on the client.
1595
1597 The amanda.conf file may define multiple types of script. The
1598 information is entered in a script section, which looks like this:
1599 define script "name" {
1600 script-option script-value
1601 ...
1602 }
1603 The { must appear at the end of a line, and the } on its own line.
1605 name is the name of this type of script. It is referenced from the
1607 dumptype
1608 The script options and values are:
1610 client-name string
1612 No default, specifies a script name that is in the
1613 amanda-client.conf on the client. The setting from that script will
1614 be merged with the currect script. If client-name is set then it is
1615 an error if that script is not defined on the client.
1616 If client-name is not set then the merge is done with the script
1618 that have the name equal to the plugin. eg. if the plugin is
1619 'amlog-script', then the setting from the script 'amlog-script' is
1620 used.
1621 comment string
1623 Default: not set. A comment string describing this script.
1624 plugin string
1626 No default. Must be set to the name of the program. This program
1627 must be in the $libexecdir/amanda/application directory on the
1628 client and/or server.
1629 order int
1631 Default: 5000. Scripts are executed in that order, it is useful if
1632 you have many scripts and they must be executed in a spefific
1633 order.
1634 single-execution boolean
1636 Default: no. The script is executed for each dle. If yes, the
1637 script is executed one time only.
1638 execute-where [ client | server ]
1640 Default: client. Where the script must be executed, on the client
1641 or server.
1642 execute-on execute_on [,execute_on]*
1644 No default. When the script must be executed, you can specify many
1645 of them:
1646 pre-amcheck
1648 Execute before the amcheck command for all dle. Can only be run
1649 on server.
1650 pre-dle-amcheck
1652 Execute before the amcheck command for the dle.
1653 pre-host-amcheck
1655 Execute before the amcheck command for all dle for the client.
1656 post-amcheck
1658 Execute after the amcheck command for all dle. Can only be run
1659 on server.
1660 post-dle-amcheck
1662 Execute after the amcheck command for the dle.
1663 post-host-amcheck
1665 Execute after the amcheck command for all dle for the client.
1666 pre-estimate
1668 Execute before the estimate command for all dle. Can only be
1669 run on server.
1670 pre-dle-estimate
1672 Execute before the estimate command for the dle.
1673 pre-host-estimate
1675 Execute before the estimate command for all dle for the client.
1676 post-estimate
1678 Execute after the estimate command for all dle. Can only be run
1679 on server.
1680 post-dle-estimate
1682 Execute after the estimate command for the dle.
1683 post-host-estimate
1685 Execute after the estimate command for all dle for the client.
1686 pre-backup
1688 Execute before the backup command for all dle. Can only be run
1689 on server.
1690 pre-dle-backup
1692 Execute before the backup command for the dle.
1693 pre-host-backup
1695 Execute before the backup command for all dle for the client.
1696 It can't be run on client, it must be run on server
1697 post-backup
1699 Execute after the backup command for all dle. Can only be run
1700 on server.
1701 post-dle-backup
1703 Execute after the backup command for the dle.
1704 post-host-backup
1706 Execute after the backup command for all dle for the client. It
1707 can't be run on client, it must be run on server
1708 pre-recover
1710 Execute before any level is recovered.
1711 post-recover
1713 Execute after all levels are recovered.
1714 pre-level-recover
1716 Execute before each level recovery.
1717 post-level-recover
1719 Execute after each level recovery.
1720 inter-level-recover
1722 Execute between two levels of recovery.
1723 If you recover level 0 and 2 of the disk /usr with amrecover, it
1725 will execute:
1726 script --pre-recover
1727 script --pre-level-recover --level 0
1728 #recovering level 0
1729 script --post-level-recover --level 0
1730 script --inter-level-recover --level 0 --level 2
1731 script --pre-level-recover --level 2
1732 #recovering level 2
1733 script --post-level-recover --level 2
1734 script --post-recover
1735 property [append] [priority] string string+
1737 No default. You can set property for the script, each script have a
1738 different set of property. Both strings are quoted; the first
1739 string contains the name of the property to set, and the others
1740 contains its values. append keyword append the values to the list
1741 of values for that property. priority keyword disallow the setting
1742 of that property on the client.
1743
1745 Backend storage devices are specified in amanda.conf in the form of
1746 "device" sections, which look like this:
1747 define device name {
1748 commend "comment (optional)"
1749 tapedev "device-specifier"
1750 device-property "prop-name" "prop-value"
1751 ...
1752 }
1753 The { must appear at the end of a line, and the } on its own line.
1755 name is the user-specified name of this device. It is referenced from
1757 the global tapedev parameter. The device-specifier specifies the device
1758 name to use; see amanda-devices(7). As with most sections, the comment
1759 parmeter is optional and only for the user's convenience.
1760 An arbitrary number of device-property parameters can be specified.
1762 Again, see amanda-devices(7) for information on device properties.
1763
1765 Changers are described in amanda.conf in the form of "changer"
1766 sections, which look like this:
1767 define changer name {
1768 comment "comment (optional)"
1769 tpchanger "changer-spec"
1770 changerdev "device-name"
1771 changerfile "state-file"
1772 ...
1773 }
1774 The { must appear at the end of a line, and the } on its own line.
1776 name is the user-specified name of this device. The remaining
1778 parameters are specific to the changer type selected.
1779 See amanda-changers(7) for more information on configuring changers.
1781
1783 The amanda.conf file may define multiple interactivyt methods, although
1784 only one will be used - that specified by the interactivity parameter.
1785 The information is entered in a interactivity section, which looks like
1786 this:
1787 define interactivity name {
1788 interactivity-option interactivity-value
1789 ...
1790 }
1791 The { must appear at the end of a line, and the } on its own line.
1793 name is the user-specified name of this interactivity. The remaining
1795 parameters are specific to the interactivity type selected.
1796 The interactivity options and values are:
1798 comment string
1800 Default: not set. A comment string describing this interactivity.
1801 plugin string
1803 No default. Must be set to the name of the interactivity module, as
1804 described in amanda-interactivity(7).
1805 property [append] string string+
1807 No default. You can set arbitrary properties for the interactivity.
1808 Each interactivity module has a different set of properties. The
1809 first string contains the name of the property to set, and the
1810 others contains its values. All strings should be quoted. The
1811 append keyword appends the given values to an existing list of
1812 values for that property.
1813 See amanda-interactivity(7) for more information on configuring
1815 interactivity methods.
1816
1818 The amanda.conf file may define multiple taperscan methods, although
1819 only one will be used - that specified by the taperscan parameter. The
1820 information is entered in a taperscan section, which looks like this:
1821 define taperscan name {
1822 taperscan-option taperscan-value
1823 ...
1824 }
1825 The { must appear at the end of a line, and the } on its own line.
1827 name is the user-specified name of this taperscan. The remaining
1829 parameters are specific to the taperscan type selected.
1830 The taperscan options and values are:
1832 comment string
1834 Default: not set. A comment string describing this taperscan.
1835 plugin string
1837 No default. Must be set to the name of the taperscan module. See
1838 amanda-taperscan(7) for a list of defined taperscan modules.
1839 property [append] string string+
1841 No default. Operates just like properties for interactivity
1842 methods, above.
1843 See amanda-taperscan(7) for more information on configuring taperscan.
1845
1847 Amanda can "split" dumps into parts while writing them to storage
1848 media. This allows Amanda to recover gracefully from a failure while
1849 writing a part to a volume, by simply selecting a new volume and
1850 re-writing the dump from the beginning of the failed part. Parts also
1851 allow Amanda to seek directly to the required data, although this
1852 functionality is not yet used.
1853 In order to support re-writing from the beginning of a failed part,
1855 Amanda must have access to the contents of the part after it has been
1856 partially written. If the dump is being read from holding disk, then
1857 the part contents are availble there. Otherwise, the part must be
1858 cached, and this can be done memory or on disk. In either of the latter
1859 cases, the cache must have enough space to hold an entire part.
1860 Because it is common for a single Amanda configuration to use both
1862 holding-disk (FILE-WRITE) and direct (known as PORT-WRITE) dumps,
1863 Amanda allows the configuration of different split sizes for the two
1864 cases. This allows, for example, for a part size appropriate to large
1865 tapes when performing FILE-WRITE dumps, with a part size limited by
1866 available disk or memory when performing PORT-WRITE dumps.
1867 Selecting a proper split size is a delicate matter. If the parts are
1869 too large, substantial storage space may be wasted in failed parts. If
1870 too small, large dumps will be split into innumerable tiny dumpfiles,
1871 adding to restoration complexity; furthermore, an excess of filemarks
1872 will cause slower tape drive operation and reduce the usable space on
1873 tape. A good rule of thumb is 1/10 of the size of a volume of storage
1874 media.
1875 In versions of Amanda through 3.1.*, splitting was controlled by the
1877 dumptype parameters tape-splitsize, split-diskbuffer, and
1878 fallback-splitsize. These keywords had confusing and non-intuitive
1879 interactions, and have since been deprecated.
1880 If the deprecated keywords are not present, subsequent versions of
1882 Amanda use the dumptype parameter allow-split to control whether a DLE
1883 can be split, and the tapetype parameters part-size, part-cache-type,
1884 part-cache-dir, and part-cache-max-size. The part-size specifies the
1885 "normal" part size, while the part-cache-* parameters describe how to
1886 behave when caching is required (on PORT-WRITE). Full details on these
1887 parameters are given above.
1888
1890 amanda(8), amanda-client.conf(5), amanda-applications(7), amanda-
1891 auth(7), amanda-changers(7), amanda-devices(7), amanda-
1892 interactivity(7), amanda-scripts(7), amanda-taperscan(7)
1893 The Amanda Wiki: : http://wiki.zmanda.com/
1895
1897 James da Silva <jds@amanda.org>
1898 Stefan G. Weichinger <sgw@amanda.org>
1900Amanda 3.3.3 01/10/2013 AMANDA.CONF(5)