1 <?xml version="1.0" encoding="ISO-8859-1"?>
2 <!DOCTYPE refentry PUBLIC "-//OASIS//DTD DocBook XML V4.1.2//EN"
3 "http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd"
5 <!-- entities files to use -->
6 <!ENTITY % global_entities SYSTEM 'global.entities'>
10 <refentry id='amanda.conf.5'>
13 <refentrytitle>amanda.conf</refentrytitle>
14 <manvolnum>5</manvolnum>
20 <refname>amanda.conf</refname>
21 <refpurpose>Main configuration file for Amanda, the Advanced Maryland Automatic Network Disk Archiver</refpurpose>
27 <!-- body begins here -->
29 <refsect1><title>DESCRIPTION</title>
30 <para>&amconf; is the main configuration file for Amanda. This manpage lists the
31 relevant sections and parameters of this file for quick reference.</para>
32 <para> The file <emphasis remap='B'><CONFIG_DIR>/<config>/amanda.conf</emphasis> is loaded.</para>
35 <refsect1><title>SYNTAX</title>
37 <para>There are a number of configuration parameters that control the
38 behavior of the Amanda programs.
39 All have default values,
40 so you need not specify the parameter in
41 <emphasis remap='B'>amanda.conf</emphasis>
42 if the default is suitable.</para>
44 <refsect2><title>COMMENTS</title>
46 <para>Lines starting with # are ignored, as are blank lines.
47 Comments may be placed on a line with a directive by starting
49 The remainder of the line is ignored.</para>
52 <refsect2><title>KEYWORDS AND IDENTIFIERS</title>
54 <para>Keywords are case insensitive, i.e.
55 <amkeyword>mailto</amkeyword>
57 <amkeyword>MailTo</amkeyword>
58 are treated the same. Also, the characters
59 '<amkeyword>-</amkeyword>' and '<amkeyword>_</amkeyword>'
60 are interchangeable in all predefined Amanda keywords:
61 <amkeyword>device_property</amkeyword>
63 <amkeyword>device-property</amkeyword>
64 have the same meaning. This manpage uses the dashed versions, but the
65 underscored versions will be accepted for backward compatibility</para>
67 <para>Identifiers are names which are defined in the configuration itself, such
68 as dumptypes or interfaces. Identifiers are are case-insensitive, but
70 '<amkeyword>-</amkeyword>' vs. '<amkeyword>_</amkeyword>'.
71 Identifiers should be quoted in the configuration file, although For historical
72 reasons, the quotes are optional.</para>
74 <para>Strings are always quoted with double quotes ("), and any double quotes
75 or backslashes within the string are escaped with a backslash:
77 tapelist "/path/to/tapelist"
78 property "escaped-string" "escaping: \\ (backslash) and \" (double-quote)"
82 <para>To summarize, then:
85 logdir "logs" # required sensitive sensitive
86 send-amreport-on strange # prohibited insensitive insensitive
87 tapetype "EXABYTE" # optional insensitive sensitive
89 define dumptype "dt" { # optional insensitive sensitive
90 "dumptype-common" # optional insensitive sensitive
91 strategy noinc # prohibited insensitive insensitive
98 <refsect2><title>VALUE SUFFIXES</title>
100 <para>Integer arguments may have one of the following (case insensitive) suffixes,
101 some of which have a multiplier effect:</para>
103 <variablelist remap='TP'>
105 <term><amkeyword>b byte bytes</amkeyword></term>
107 <para>Some number of bytes.</para>
111 <term><amkeyword>bps</amkeyword></term>
113 <para>Some number of bytes per second.</para>
117 <term><amkeyword>k kb kbyte kbytes kilobyte kilobytes</amkeyword></term>
119 <para>Some number of kilobytes (bytes*1024).</para>
123 <term><amkeyword>kps kbps</amkeyword></term>
125 <para>Some number of kilobytes per second (bytes*1024).</para>
126 <para>It is the default multiplier for all size options.</para>
130 <term><amkeyword>m mb meg mbyte mbytes megabyte megabytes</amkeyword></term>
132 <para>Some number of megabytes (bytes*1024*1024).</para>
136 <term><amkeyword>mps mbps</amkeyword></term>
138 <para>Some number of megabytes per second (bytes*1024*1024).</para>
142 <term><amkeyword>g gb gbyte gbytes gigabyte gigabytes</amkeyword></term>
144 <para>Some number of gigabytes (bytes*1024*1024*1024).</para>
148 <term><amkeyword>t tb tbyte tbytes terabyte terabytes</amkeyword></term>
150 <para>Some number of terabytes (bytes*1024*1024*1024*1024).</para>
154 <term><amkeyword>tape tapes</amkeyword></term>
156 <para>Some number of tapes.</para>
160 <term><amkeyword>day days</amkeyword></term>
162 <para>Some number of days.</para>
166 <term><amkeyword>week weeks</amkeyword></term>
168 <para>Some number of weeks (days*7).</para>
170 <note><para>The value
171 <amkeyword>inf</amkeyword>
172 may be used in most places where an integer is expected
173 to mean an infinite amount.</para>
175 <para>Boolean arguments may have any of the values
176 <amkeyword>1</amkeyword>,
177 <amkeyword>y</amkeyword>,
178 <amkeyword>yes</amkeyword>,
179 <amkeyword>t</amkeyword>,
180 <amkeyword>true</amkeyword>
182 <amkeyword>on</amkeyword>
183 to indicate a true state, or
184 <amkeyword>0</amkeyword>,
185 <amkeyword>n</amkeyword>,
186 <amkeyword>no</amkeyword>,
187 <amkeyword>f</amkeyword>,
188 <amkeyword>false</amkeyword>
190 <amkeyword>off</amkeyword>
191 to indicate a false state.
192 If no argument is given,
193 <amkeyword>true</amkeyword>
202 <refsect2><title>PARAMETER ORDER</title>
204 <para>In general, the order in which parameters occur in the configuration file
205 does not matter, with the exception of subsection inheritance. For example, if
206 dumptype "normal-encrypt" which inherits from dumptype "normal", then "normal"
207 must appear first in the configuration file.</para>
211 <refsect2><title>STRINGS</title>
212 <para>Quoted strings in Amanda follow a common, C-like syntax. Printable
213 characters and whitespace are kept as-is, except that the backslash character
214 (\) is used as an escape character, and a double-quote ends the string. The allowed
217 ESCAPE SEQUENCE BECOMES
226 \001 - \377 (character specified in octal)
228 Illegally quoted strings are handled on a "best-effort" basis, which may lead to
233 finserver "/data/finance/XYZ Corp's \"real\" finances" finance-high eth0 -1
234 property "syspath" "C:\\WINDOWS\\SYSTEM"
239 <refsect2><title>SUBSECTIONS AND INHERITANCE</title>
241 <para>Amanda configuration files may include various
242 <emphasis>subsections</emphasis>, each defining a set of configuration
243 directives. Each type of subsection is described below. Note that all types
244 of subsections can <emphasis>inherit</emphasis> from other subsections of the
245 same type by naming the "parent" section in the "child" subsection. For
249 define dumptype global {
254 define dumptype nocomp {
255 global # inherit the parameters in dumptype 'global'
258 </programlisting></para>
260 <para>Note that multiple inheritance is also supported by simply naming
261 multiple parent sections in a child. Parents are implicitly expanded in place
262 in a child, and the last occurrence of each parameter takes precedence. For
265 define tapetype par1 {
271 define tapetype par2 {
276 define tapetype child {
282 In this example, 'child' will have a filemark of 32k, a speed of 400bps, and a
283 length of 200M.</para>
289 <title>GLOBAL PARAMETERS</title>
291 <variablelist remap='TP'>
293 <term><amkeyword>org</amkeyword> <amtype>string</amtype></term>
296 <amdefault>"daily"</amdefault>.
297 A descriptive name for the configuration.
298 This string appears in the Subject line of mail reports.
299 Each Amanda configuration should have a different string to keep
300 mail reports distinct.</para>
304 <term><amkeyword>mailer</amkeyword> <amtype>string</amtype></term>
306 <para>Default found by configure. A mail program that can send mail with '<emphasis remap='I'>MAILER -s "subject" user < message_file</emphasis>'.</para>
310 <term><amkeyword>mailto</amkeyword> <amtype>string</amtype></term>
313 A space separated list of recipients for mail reports. If not specified, amdump will not send any mail.</para>
317 <term><amkeyword>send-amreport-on</amkeyword> [ <amkeyword>all</amkeyword> | <amkeyword>strange</amkeyword> | <amkeyword>error</amkeyword> | <amkeyword>never</amkeyword> ]</term>
320 <amkeyword>all</amkeyword>.
321 Specify which types of messages will trigger an email from amreport. amreport is used by amdump and amflush.</para>
323 <variablelist remap='TP'>
325 <term><amkeyword>all</amkeyword></term>
327 <para>Send an email on any message.</para>
331 <term><amkeyword>strange</amkeyword></term>
333 <para>Send an email on strange or error message. A strange message occurs when the dump succeeded, but returned one or more errors unknown to Amanda.</para>
337 <term><amkeyword>error</amkeyword></term>
339 <para>Send an email only on error messages.</para>
343 <term><amkeyword>never</amkeyword></term>
345 <para>Never send an email.</para>
353 <term><amkeyword>report-use-media</amkeyword> <amtype>boolean</amtype></term>
356 <amdefault>True</amdefault> if <amkeyword>max-dle-by-volume</amkeyword>
358 <amdefault>False</amdefault> if it is set.
359 If the reporter must print the list of media used in the run.</para>
364 <term><amkeyword>report-next-media</amkeyword> <amtype>boolean</amtype></term>
367 <amdefault>True</amdefault> if <amkeyword>max-dle-by-volume</amkeyword>
369 <amdefault>False</amdefault> if it is set.
370 If the reporter must print the list of media expected for the next run.</para>
375 <term><amkeyword>max-dle-by-volume</amkeyword> <amtype>int</amtype></term>
378 <amdefault>1000000000</amdefault>.
379 The maximum number of dle written to a single volume.</para>
384 <term><amkeyword>dumpcycle</amkeyword> <amtype>int</amtype></term>
387 <amdefault>10 days</amdefault>.
388 The number of days in the backup cycle.
389 Each disk will get a full backup at least this often.
390 Setting this to zero tries to do a full backup each run.</para>
391 <note><para>This parameter may also be set in a specific
392 <amkeyword>dumptype</amkeyword>
394 This value sets the default for all
395 <amkeyword>dumptype</amkeyword>s
397 <emphasis remap='B'>amanda.conf</emphasis>
399 <amkeyword>dumptype</amkeyword>s
406 <term><amkeyword>runspercycle</amkeyword> <amtype>int</amtype></term>
409 <amdefault>same as dumpcycle</amdefault>.
410 The number of amdump runs in
411 <amkeyword>dumpcycle</amkeyword>
413 A value of 0 means the same value as
414 <amkeyword>dumpcycle</amkeyword>.
415 A value of -1 means guess the number of runs from the &tapelist; file,
416 which is the number of tapes used in the last
417 <amkeyword>dumpcycle</amkeyword>
419 <amkeyword>runtapes</amkeyword>.</para>
424 <term><amkeyword>tapecycle</amkeyword> <amtype>int</amtype></term>
427 <amdefault>15 tapes</amdefault>.
429 Specifies the number of "active" volumes - volumes that Amanda will not
430 overwrite. While Amanda is always willing to write to a new volume, it refuses
431 to overwrite a volume unless at least '<amkeyword>tapecycle</amkeyword> -1'
432 volumes have been written since.</para>
434 <para> It is considered good administrative practice to set the
435 <amkeyword>tapecycle</amkeyword> parameter slightly lower than the actual
436 number of tapes in use. This allows the administrator to more easily cope
437 with damaged or misplaced tapes or schedule adjustments that call for
438 slight adjustments in the rotation order.</para>
440 <para>Note: Amanda is commonly misconfigured with <amkeyword>tapecycle</amkeyword>
441 equal to the number of tapes per <amkeyword>dumpcycle</amkeyword>. In this
442 misconfiguration, amanda may erase a full dump before a new one is
443 completed. Recovery is then impossible. The
444 <amkeyword>tapecycle</amkeyword> must be at least one tape larger than the
445 number of tapes per dumpcycle.</para>
448 The number of tapes per dumpcycle is calculated by multiplying the number of
449 &amdump; runs per dump cycle <amkeyword>runspercycle</amkeyword> (the
450 number of &amdump; runs per dump cycle) and <emphasis
451 remap='B'>runtapes</emphasis> (the number of tapes used per run). Typically
452 <amkeyword>tapecycle</amkeyword> is set to two or four times the tapes
453 per dumpcycle.</para>
459 <term><amkeyword>usetimestamps</amkeyword>
460 <amtype>bool</amtype></term>
462 <para>Default: <amkeyword>Yes</amkeyword>.
463 This option allows Amanda to track multiple runs per calendar
464 day. The only reason one might disable it is that Amanda versions before 2.5.1
465 can't read logfiles written when this option was enabled.
471 <term><amkeyword>label-new-tapes</amkeyword>
472 <amtype>string</amtype></term>
474 <para>Deprecated, use <amkeyword>autolabel</amkeyword> option with options <amkeyword>volume-error empty</amkeyword> to get equivalent behavior.</para>
475 <para>Default: not set.
476 When set, this directive will cause Amanda to automatically write an Amanda
477 tape label to any blank tape she encounters.</para>
482 <term><amkeyword>autolabel</amkeyword>
483 <amtype>string</amtype>
484 [<amkeyword>any</amkeyword>]
485 [<amkeyword>other-config</amkeyword>]
486 [<amkeyword>non-amanda</amkeyword>]
487 [<amkeyword>volume-error</amkeyword>]
488 [<amkeyword>empty</amkeyword>]</term>
490 <para>Default: not set.
491 When set, this directive will cause Amanda to automatically write an Amanda
492 tape label to most volume she encounters. This option is DANGEROUS
493 because when set, Amanda may erase near-failing tapes or tapes accidentally
494 loaded in the wrong slot.</para>
495 <para>When using this directive, specify the template for new tape
496 labels. The template can contains many variables that are substituted by their
500 $o : org configuration
501 $b : barcode of the volume
502 $s : slot number, can specify a minimun number of digit:
505 </programlisting></para>
506 <para>The template can contain some number of contiguous '%'
507 characters, which will be replaced with a generated number. Be sure to
508 specify enough '%' characters that you do not run out of tape labels.
510 <markup>"DailySet1-%%%"</markup>,
511 <markup>"$c-%%%"</markup>,
512 <markup>"$m-%%%"</markup>,
513 <markup>"$m-$b"</markup>
515 <para>The generared label can be used only if it match the
516 <amkeyword>labelstr</amkeyword> setting. The volume will not be used if the
517 generated label doesn't match the <amkeyword>labelstr</amkeyword> setting.
519 <para>Note that many devices cannot distinguish an empty tape from an error
520 condition, so it may is often necessary to include
521 <amkeyword>volume-error</amkeyword> as an autolabel condition.</para>
522 <variablelist remap='TP'>
524 <term><amkeyword>any</amkeyword></term>
525 <listitem>equivalent to '<amkeyword>other-config non-amanda volume-error empty</amkeyword>'
529 <term><amkeyword>other-config</amkeyword></term>
530 <listitem>Label volumes with a valid Amanda label that do not match our
531 <amkeyword>labelstr</amkeyword>. Danger: this may erase volumes
532 from other Amanda configurations without warning!
536 <term><amkeyword>non-amanda</amkeyword></term>
537 <listitem>Label volumes which do not start with data that resembles an
538 Amanda header. Danger: this may erase volumes from other backup applications
543 <term><amkeyword>volume-error</amkeyword></term>
544 <listitem>Label volumes where an error occurs while trying to read the label.
545 Danger: this may erase arbitrary volumes due to transient errors.
549 <term><amkeyword>empty</amkeyword></term>
550 <listitem>Label volumes where a read returns 0 bytes.
558 <term><amkeyword>meta-autolabel</amkeyword>
559 <amtype>string</amtype></term>
561 <para>Default: not set. When set and if the changer support meta-label, this directive will cause Amanda to automatically add a meta-label to a meta-volume.</para>
562 <para>A meta-volume is a containers that contains many volumes, eg. a removable
563 hard-disk for use with chg-disk, each hard disk have many slots (volume).
564 The meta-label is the label to put on the meta-volume.</para>
565 <para>When using this directive, specify the template for new meta
566 labels. The template can contains many variables that are substituted by their
570 $o : org configuration
571 </programlisting></para>
572 <para>The template should contain some number of contiguous '%'
573 characters, which will be replaced with a generated number. Be sure to
574 specify enough '%' characters that you do not run out of meta labels.
576 <markup>"DailySet1-%%%"</markup>,
577 <markup>"$o-%%%"</markup>,
583 <term><amkeyword>dumpuser</amkeyword> <amtype>string</amtype></term>
586 <amdefault>"amanda"</amdefault>.
587 The login name Amanda uses to run the backups.
588 The backup client hosts must allow access
589 from the tape server host as this user via
590 <markup>.rhosts</markup>
592 <markup>.amandahosts</markup>,
593 depending on how the Amanda software was built.</para>
597 <term><amkeyword>printer</amkeyword> <amtype>string</amtype></term>
599 <para>Printer to use when doing tape labels.
601 <amkeyword>lbl-templ</amkeyword>
602 <amkeyword>tapetype</amkeyword>
608 <term><amkeyword>tapedev</amkeyword> <amtype>string</amtype></term>
611 <amdefault>"null:"</amdefault>.
612 This parameter can either specify a device (explicitly or by referencing a device definition - see <manref name="amanda-devices" vol="7" />)
613 or a tape changer (explicitly or by referencing a device definition - see <manref name="amanda-changers" vol="7" />).</para>
619 <term><amkeyword>device-property</amkeyword> <amtype>string</amtype> <amtype>string</amtype></term>
621 <para>These options can set various device properties. See
622 <manref name="amanda-devices" vol="7"/>
623 for more information on device properties and their syntax.
624 Both strings are always quoted; the first string contains the name of
625 the property to set, and the second contains its value. For example, to set
626 a fixed block size of 128k, write:</para>
628 device-property "BLOCK_SIZE" "128k"
635 <term><amkeyword>property</amkeyword> [<amkeyword>append</amkeyword>] <amtype>string</amtype> <amtype>string</amtype>+</term>
637 <para>These options can set various properties, they can be used by third
638 party software to store information in the configuration file.
639 Both strings are quoted; the first string contains the name of
640 the property to set, and the others contains its values.
641 <amkeyword>append</amkeyword> keyword append the values to the list of values for that property.
647 <term><amkeyword>tpchanger</amkeyword> <amtype>string</amtype></term>
649 <para>Default: not set. The tape changer to use. In most cases, only one of
650 <amkeyword>tpchanger</amkeyword> or <amkeyword>tapedev</amkeyword> is
651 specified, although for backward compatibility both may be specified if
652 <amkeyword>tpchanger</amkeyword> gives the name of an old changer script.
653 See <manref name="amanda-changers" vol="7" /> for more information on
654 configuring changers.</para>
659 <term><amkeyword>interactivity</amkeyword> <amtype>string</amtype></term>
661 <para>Default: not set. The interactivity module Amanda should use to interact with
662 the user. See <manref name="amanda-interactivity" vol="7" /> for a list of
668 <term><amkeyword>taperscan</amkeyword> <amtype>string</amtype></term>
670 <para>Default: traditional. The taperscan module amanda should use to find a tape
671 to write to. See <manref name="amanda-taperscan" vol="7" /> for a list of
677 <term><amkeyword>changerdev</amkeyword> <amtype>string</amtype></term>
680 <amdefault>"dev/null"</amdefault>.
681 A tape changer configuration parameter.
682 Usage depends on the particular changer defined with the
683 <amkeyword>tpchanger</amkeyword>
688 <term><amkeyword>changerfile</amkeyword> <amtype>string</amtype></term>
691 <amdefault>"/usr/adm/amanda/log/changer-status"</amdefault>.
692 A tape changer configuration parameter.
693 Usage depends on the particular changer defined with the
694 <amkeyword>tpchanger</amkeyword>
699 <term><amkeyword>runtapes</amkeyword> <amtype>int</amtype></term>
702 <amdefault>1</amdefault>.
703 The maximum number of tapes used in a single run.
704 If a tape changer is not configured, this option is not used
705 and should be commented out of the configuration file.</para>
706 <para>If a tape changer is configured, this may be set larger than one to
707 let Amanda write to more than one tape.</para>
708 <para>Note that this is an upper bound on the number of tapes,
709 and Amanda may use less.</para>
714 <term><amkeyword>maxdumpsize</amkeyword> <amtype>int</amtype></term>
717 <amdefault><amkeyword>runtapes</amkeyword>*<amkeyword>tape-length</amkeyword></amdefault>.
718 Maximum number of bytes the planner will schedule for a run.</para>
719 <para>The default unit is Kbytes if it is not specified.</para>
723 <term><amkeyword>taperalgo</amkeyword> [ <amkeyword>first</amkeyword> | <amkeyword>firstfit</amkeyword> | <amkeyword>largest</amkeyword> | <amkeyword>largestfit</amkeyword> | <amkeyword>smallest</amkeyword> | <amkeyword>last</amkeyword> ]</term>
726 <amkeyword>first</amkeyword>.
727 The algorithm used to choose which dump image to send to the taper.</para>
729 <variablelist remap='TP'>
731 <term><amkeyword>first</amkeyword></term>
733 <para>First in, first out.</para>
737 <term><amkeyword>firstfit</amkeyword></term>
739 <para>The first dump image that will fit on the current tape.</para>
743 <term><amkeyword>largest</amkeyword></term>
745 <para>The largest dump image.</para>
749 <term><amkeyword>largestfit</amkeyword></term>
751 <para>The largest dump image that will fit on the current tape.</para>
755 <term><amkeyword>smallest</amkeyword></term>
757 <para>The smallest dump image.</para>
761 <term><amkeyword>last</amkeyword></term>
763 <para>Last in, first out.</para>
771 <term><amkeyword>taper-parallel-write</amkeyword> <amtype>int</amtype></term>
773 <para>Default: <amdefault>1</amdefault>.
774 Amanda can write simultaneously up to that number of volume at any given
775 time. The changer must have as many drives.</para>
780 <term><amkeyword>eject-volume</amkeyword> <amtype>int</amtype></term>
782 <para>Default: <amdefault>no</amdefault>.
783 Set to <amkeyword>yes</amkeyword> if you want the volume to be ejected
784 after Amanda wrote data to it. It works only with some changer and device.</para>
789 <term><amkeyword>labelstr</amkeyword> <amtype>string</amtype></term>
792 <amdefault>".*"</amdefault>.
793 The tape label constraint regular expression.
794 All tape labels generated (see
795 <manref name="amlabel" vol="8"/>)
796 and used by this configuration must match the regular expression.
797 If multiple configurations are run from the same tape server host,
798 it is helpful to set their labels to different strings
799 (for example, "DAILY[0-9][0-9]*" vs.
800 "ARCHIVE[0-9][0-9]*")
801 to avoid overwriting each other's tapes.</para>
805 <term><amkeyword>tapetype</amkeyword> <amtype>string</amtype></term>
808 <amdefault>no default</amdefault>.
809 The type of tape drive associated with
810 <amkeyword>tapedev</amkeyword>
812 <amkeyword>tpchanger</amkeyword>.
813 This refers to one of the defined <amkeyword>tapetype</amkeyword>s
814 in the config file (see below), which specify various tape parameters,
816 <amkeyword>length</amkeyword>,
817 <amkeyword>filemark</amkeyword>
819 <amkeyword>speed</amkeyword>
820 of the tape media and device.</para>
824 <term><amkeyword>ctimeout</amkeyword> <amtype>int</amtype></term>
827 <amdefault>30 seconds</amdefault>.
828 Maximum amount of time that
829 <emphasis remap='B'>amcheck</emphasis>
830 will wait for each client host.</para>
834 <term><amkeyword>dtimeout</amkeyword> <amtype>int</amtype></term>
837 <amdefault>1800 seconds</amdefault>.
838 Amount of idle time per disk on a given client that a
839 <emphasis remap='B'>dumper</emphasis>
841 <command>amdump</command>
842 will wait before it fails with a data timeout error.</para>
847 <term><amkeyword>etimeout</amkeyword> <amtype>int</amtype></term>
850 <amdefault>300 seconds</amdefault>.
851 Amount of time per estimate on a given client that the
852 <emphasis remap='B'>planner</emphasis> step of
853 <command>amdump</command> will wait to get the dump size estimates
854 (note: Amanda runs up to 3 estimates for each DLE). For instance,
855 with the default of 300 seconds and four DLE's, each estimating level
856 0 and level 1 on client A, <emphasis remap='B'>planner</emphasis>
857 will wait up to 40 minutes for that machine. A negative value will
858 be interpreted as a total amount of time to wait
859 per client instead of per disk.</para>
864 <term><amkeyword>connect-tries</amkeyword> <amtype>int</amtype></term>
867 <amdefault>3</amdefault>.
868 How many times the server will try a connection.</para>
873 <term><amkeyword>req-tries</amkeyword> <amtype>int</amtype></term>
876 <amdefault>3</amdefault>.
877 How many times the server will resend a REQ packet if it doesn't get the ACK packet.</para>
882 <term><amkeyword>netusage</amkeyword> <amtype>int</amtype></term>
885 <amdefault>80000 Kbps</amdefault>.
886 The maximum network bandwidth allocated to Amanda, in Kbytes per second.
888 <emphasis remap='B'>interface</emphasis>
893 <term><amkeyword>inparallel</amkeyword> <amtype>int</amtype></term>
896 <amdefault>10</amdefault>.
897 The maximum number of backups that Amanda will attempt to run in parallel.
898 Amanda will stay within the constraints of network bandwidth and
899 holding disk space available, so it doesn't hurt to set
900 this number a bit high. Some contention can occur with larger numbers
901 of backups, but this effect is relatively small on most systems.</para>
906 <term><amkeyword>displayunit</amkeyword> "k|m|g|t"</term>
909 <amdefault>"k"</amdefault>.
910 The unit used to print many numbers, k=kilo, m=mega, g=giga, t=tera.
916 <term><amkeyword>dumporder</amkeyword> <amtype>string</amtype></term>
919 <amdefault>"tttTTTTTTT"</amdefault>.
920 The priority order of each dumper:
927 b: smallest bandwidth
929 </programlisting></para>
935 <term><amkeyword>maxdumps</amkeyword> <amtype>int</amtype></term>
938 <amdefault>1</amdefault>.
939 The maximum number of backups from a single host that Amanda will
940 attempt to run in parallel. See also the
941 <emphasis remap='B'>inparallel</emphasis>
944 <para>Note that this parameter may also be set in a specific
945 <amkeyword>dumptype</amkeyword>
947 This value sets the default for all
948 <amkeyword>dumptype</amkeyword>s
950 <emphasis remap='B'>amanda.conf</emphasis>
952 <amkeyword>dumptype</amkeyword>s
958 <term><amkeyword>bumpsize</amkeyword> <amtype>int</amtype></term>
961 <amdefault>10 Mbytes</amdefault>.
962 The minimum savings required to trigger an automatic
963 bump from one incremental level to the next, expressed as size.
964 If Amanda determines that the next higher backup level
965 will be this much smaller than the current level,
966 it will do the next level.
967 The value of this parameter is used only if the parameter
968 <emphasis>bumppercent</emphasis> is set to 0.
970 <para>The default unit is Kbytes if it is not specified.</para>
972 The global setting of this parameter can be overwritten inside of a
977 <amkeyword>bumppercent</amkeyword>,
978 <amkeyword>bumpmult</amkeyword> and
979 <amkeyword>bumpdays</amkeyword>.</para>
984 <term><amkeyword>bumppercent</amkeyword> <amtype>int</amtype></term>
987 <amdefault>0</amdefault>.
988 The minimum savings required to trigger an automatic
989 bump from one incremental level to the next, expressed as percentage of the
990 current size of the DLE (size of current level 0).
991 If Amanda determines that the next higher backup level
992 will be this much smaller than the current level,
993 it will do the next level.
996 If this parameter is set to 0, the value of the
997 parameter <emphasis>bumpsize</emphasis> is used to trigger bumping.
1000 The global setting of this parameter can be overwritten inside of a
1001 dumptype-definition.
1004 See also the options
1005 <amkeyword>bumpsize</amkeyword>,
1006 <amkeyword>bumpmult</amkeyword> and
1007 <amkeyword>bumpdays</amkeyword>.</para>
1012 <term><amkeyword>bumpmult</amkeyword> <amtype>float</amtype></term>
1015 <amdefault>1.5</amdefault>.
1016 The bump size multiplier.
1018 <amkeyword>bumpsize</amkeyword>
1019 by this factor for each level.
1020 This prevents active filesystems from
1021 bumping too much by making it harder to bump to the next level.
1022 For example, with the default
1023 <amkeyword>bumpsize</amkeyword>
1025 <amkeyword>bumpmult</amkeyword>
1026 set to 2.0, the bump threshold will be 10 Mbytes for level one, 20
1027 Mbytes for level two, 40 Mbytes for level three, and so on.</para>
1029 The global setting of this parameter can be overwritten inside of a
1030 dumptype-definition.
1035 <term><amkeyword>bumpdays</amkeyword> <amtype>int</amtype></term>
1038 <amdefault>2 days</amdefault>.
1039 To insure redundancy in the dumps, Amanda keeps filesystems at the
1040 same incremental level for at least
1041 <amkeyword>bumpdays</amkeyword>
1042 days, even if the other bump threshold criteria are met.</para>
1044 The global setting of this parameter can be overwritten inside of a
1045 dumptype-definition.
1051 <term><amkeyword>diskfile</amkeyword> <amtype>string</amtype></term>
1054 <amdefault>"disklist"</amdefault>.
1055 The file name for the
1056 <emphasis remap='I'>disklist</emphasis>
1057 file holding client hosts, disks and other client dumping information.</para>
1061 <term><amkeyword>infofile</amkeyword> <amtype>string</amtype></term>
1064 <amdefault>"/usr/adm/amanda/curinfo"</amdefault>.
1065 The file or directory name for the historical information database.
1066 If Amanda was configured to use DBM databases, this is the base file
1068 If it was configured to use text formated databases (the default),
1069 this is the base directory and within here will be a directory per
1070 client, then a directory per disk, then a text file of data.</para>
1074 <term><amkeyword>logdir</amkeyword> <amtype>string</amtype></term>
1077 <amdefault>"/usr/adm/amanda"</amdefault>.
1078 The directory for the
1079 <command>amdump</command>
1081 <emphasis remap='B'>log</emphasis>
1086 <term><amkeyword>indexdir</amkeyword> <amtype>string</amtype></term>
1089 <amdefault>"/usr/adm/amanda/index"</amdefault>.
1090 The directory where index files (backup image catalogues) are stored.
1092 only generated for filesystems whose
1093 <amkeyword>dumptype</amkeyword>
1095 <amkeyword>index</amkeyword>
1096 option enabled.</para>
1100 <term><amkeyword>tapelist</amkeyword> <amtype>string</amtype></term>
1103 <amdefault>"tapelist"</amdefault>.
1104 The file name for the active &tapelist;.
1105 Amanda maintains this file with information about the active set of tapes.</para>
1109 <term><amkeyword>device-output-buffer-size</amkeyword> <amtype>int</amtype></term>
1112 <amdefault>1280k</amdefault>.
1113 Controls the amount of memory used by Amanda
1114 to hold data as it is read from the network or disk before it is written to
1115 the output device. Higher values may be
1116 useful on fast tape drives and optical media.</para>
1117 <para>The default unit is bytes if it is not specified.</para>
1121 <term><amkeyword>tapebufs</amkeyword> <amtype>int</amtype></term>
1124 <amdefault>20</amdefault>.
1125 This option is deprecated; use
1126 the <amkeyword>device-output-buffer-size</amkeyword> directive
1127 instead. <amkeyword>tapebufs</amkeyword> works the same way,
1128 but the number specified is multiplied by the device blocksize prior
1133 <term><amkeyword>reserve</amkeyword> <amtype>int</amtype></term>
1136 <amdefault>100</amdefault>.
1137 The part of holding-disk space that should be reserved for incremental
1138 backups if no tape is available, expressed as a percentage of the
1139 available holding-disk space (0-100).
1140 By default, when there is no tape to write to, degraded mode (incremental) backups
1141 will be performed to the holding disk. If full backups should also be allowed in this case,
1142 the amount of holding disk space reserved for incrementals should be lowered.</para>
1146 <term><amkeyword>autoflush</amkeyword> <amkeyword>no</amkeyword>|<amkeyword>yes</amkeyword>|<amkeyword>all</amkeyword></term>
1149 <amkeyword>no</amkeyword>.
1150 Whether an amdump run will flush the dumps from holding disk to tape.
1151 With <amkeyword>yes</amkeyword>, only dump matching the command line argument are flushed.
1152 With <amkeyword>all</amkeyword>, all dump are flushed.</para>
1156 <term><amkeyword>amrecover-do-fsf</amkeyword> <amtype>bool</amtype></term>
1158 <para>Deprecated; amrecover always uses fsf, and does not invoke amrestore.</para>
1160 <amkeyword>on</amkeyword>.
1161 Amrecover will call amrestore with the -f flag for faster positioning of the tape.</para>
1165 <term><amkeyword>amrecover-check-label</amkeyword> <amtype>bool</amtype></term>
1167 <para>Deprecated; amrecover always checks the label, and does not invoke amrestore.</para>
1169 <amkeyword>on</amkeyword>.
1170 Amrecover will call amrestore with the -l flag to check the label.</para>
1174 <term><amkeyword>amrecover-changer</amkeyword> <amtype>string</amtype></term>
1176 <para>Default: not set.
1177 Amrecover will use the changer if you use 'settape <string>' and that string
1178 is the same as the <amkeyword>amrecover-changer</amkeyword> setting.</para>
1182 <term><amkeyword>columnspec</amkeyword> <amtype>string</amtype></term>
1184 <para> default: "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"</para>
1185 <para>Defines the width of columns <emphasis remap='B'>amreport</emphasis>
1187 <emphasis remap='I'>String</emphasis>
1188 is a comma (',') separated list of triples. Each triple consists
1189 of three parts which are separated by a equal sign ('=') and a colon (':') (see the example).
1190 These four parts specify:
1194 <para>the name of the column, which may be:
1197 Compress (compression ratio)
1198 Disk (client disk name)
1199 DumpRate (dump rate in KBytes/sec)
1200 DumpTime (total dump time in hours:minutes)
1201 HostName (client host name)
1203 OrigKB (original image size in KBytes)
1204 OutKB (output image size in KBytes)
1205 TapeRate (tape writing rate in KBytes/sec)
1206 TapeTime (total tape time in hours:minutes)
1207 </programlisting></para>
1212 <para>the amount of space to display before the column (used to get whitespace between columns).</para>
1215 <para>the width of the column itself.
1216 If set to a negative value, the width will be calculated on demand to fit the largest entry in
1220 <para>the precision of the column, number of digit after the decimal point for number.</para>
1222 </orderedlist></para>
1224 <para>Here is an example:
1226 columnspec "Disk=1:18,HostName=0:10,OrigKB=::2,OutKB=1:7"
1227 </programlisting></para>
1229 The above will display the disk information in 18 characters
1230 and put one space before it. The hostname column will be 10 characters wide with
1231 no space to the left. The Original KBytes print 2 decimal digit. The output KBytes column is seven characters wide
1232 with one space before it.</para>
1237 <term><amkeyword>includefile</amkeyword> <amtype>string</amtype></term>
1240 <amdefault>no default</amdefault>.
1241 The name of an Amanda configuration file to include within the current file.
1242 Useful for sharing dumptypes, tapetypes and interface definitions among several configurations.
1243 Relative pathnames are relative to the configuration directory.
1249 <term><amkeyword>debug-days</amkeyword> <amtype>int</amtype></term>
1252 <amdefault>3</amdefault>.
1253 The number of days the debug files are kept.</para>
1258 <term><amkeyword>debug-auth</amkeyword> <amtype>int</amtype></term>
1261 <amdefault>0</amdefault>.
1262 Debug level of the auth module</para>
1267 <term><amkeyword>debug-event</amkeyword> <amtype>int</amtype></term>
1270 <amdefault>0</amdefault>.
1271 Debug level of the event module</para>
1276 <term><amkeyword>debug-holding</amkeyword> <amtype>int</amtype></term>
1279 <amdefault>0</amdefault>.
1280 Debug level of the holdingdisk module</para>
1285 <term><amkeyword>debug-protocol</amkeyword> <amtype>int</amtype></term>
1288 <amdefault>0</amdefault>.
1289 Debug level of the protocol module</para>
1294 <term><amkeyword>debug-planner</amkeyword> <amtype>int</amtype></term>
1297 <amdefault>0</amdefault>.
1298 Debug level of the planner process</para>
1303 <term><amkeyword>debug-driver</amkeyword> <amtype>int</amtype></term>
1306 <amdefault>0</amdefault>.
1307 Debug level of the driver process</para>
1312 <term><amkeyword>debug-dumper</amkeyword> <amtype>int</amtype></term>
1315 <amdefault>0</amdefault>.
1316 Debug level of the dumper process</para>
1321 <term><amkeyword>debug-chunker</amkeyword> <amtype>int</amtype></term>
1324 <amdefault>0</amdefault>.
1325 Debug level of the chunker process</para>
1330 <term><amkeyword>debug-taper</amkeyword> <amtype>int</amtype></term>
1333 <amdefault>0</amdefault>.
1334 Debug level of the taper process</para>
1339 <term><amkeyword>debug-recovery</amkeyword> <amtype>int</amtype></term>
1342 <amdefault>1</amdefault>.
1343 Debug level of all recovery process</para>
1348 <term><amkeyword>flush-threshold-dumped</amkeyword> <amtype>int</amtype></term>
1350 <para>Default: <amdefault>0</amdefault>.
1351 Amanda will not begin writing data to a new volume until the amount of
1352 data on the holding disk is at least this percentage of the volume
1353 size. In other words, Amanda will not begin until the
1354 amount of data on the holding disk is greater than the tape
1355 length times this parameter.
1356 This parameter may be larger than 100%, for example to keep
1357 more recent dumps on the holding disk for faster recovery.</para>
1358 <para> Needless to say, your holding disk must be big enough
1359 that this criterion could be satisfied. If the holding disk cannot
1360 be used for a particular dump (because, for example, there is no
1361 remaining holding space) then Amanda will disregard the constraint
1362 specified by this setting and start a new volume anyway. Once
1363 writing to a volume has begun, this constraint is not applied unless
1364 and until a new volume is needed.</para>
1365 <para> The value of this parameter may not exceed than that of
1366 the <amkeyword>flush-threshold-scheduled</amkeyword>
1372 <term><amkeyword>flush-threshold-scheduled</amkeyword> <amtype>int</amtype></term>
1374 <para>Default: <amdefault>0</amdefault>.
1375 Amanda will not begin writing data to a new volume until the sum of the
1376 amount of data on the holding disk and the estimated amount of data
1377 remaining to be dumped during this run is at least this percentage
1378 of the volume size. In other words, Amanda will not begin until the
1379 inequality <inlineequation><mathphrase>h + s > t ×
1380 d</mathphrase></inlineequation> is satisfied,
1381 where <mathphrase>h</mathphrase> is the amount of data on the
1382 holding disk, <mathphrase>s</mathphrase> is the total amount of
1383 data scheduled for this run but not dumped
1384 yet, <mathphrase>t</mathphrase> is the capacity of a volume,
1385 and <mathphrase>d</mathphrase> is this parameter, expressed as a
1386 percentage. This parameter may be larger than 100%.</para>
1387 <para> Needless to say, your holding disk must be big enough
1388 that this criterion could be satisfied. If the holding disk cannot
1389 be used for a particular dump (because, for example, there is no
1390 remaining holding space) then Amanda will disregard the constraint
1391 specified by this setting and start a new volume anyway. Once
1392 writing to a volume has begun, this constraint is not applied unless
1393 and until a new volume is needed.</para>
1394 <para> The value of this parameter may not be less than that of
1395 the <amkeyword>flush-threshold-dumped</amkeyword>
1396 or <amkeyword>taperflush</amkeyword> parameters.</para>
1400 <!-- the entity '#215' below is U+00D7 MULTIPLICATION SIGN; the 'times' entity
1401 is not as portable as one might hope. -->
1403 <term><amkeyword>taperflush</amkeyword> <amtype>int</amtype></term>
1405 <para>Default: <amdefault>0</amdefault>.
1406 At the end of a run, Amanda will start a new tape to flush remaining
1407 data if there is more data on the holding disk at the end of a run
1408 than this setting allows; the amount is specified as a percentage of
1409 the capacity of a single
1410 volume. In other words, at the end of a run, Amanda will begin
1412 inequality <inlineequation><mathphrase>h > t ×
1413 f</mathphrase></inlineequation> is satisfied,
1414 where <mathphrase>h</mathphrase> is the amount of data remaining on the
1415 holding disk from this or previous runs, <mathphrase>t</mathphrase>
1416 is the capacity of a volume,
1417 and <mathphrase>f</mathphrase> is this parameter, expressed as a
1418 percentage. This parameter may be greater than 100%.</para>
1419 <para> The value of this parameter may not exceed that of
1420 the <amkeyword>flush-threshold-scheduled</amkeyword>
1421 parameter.; <amkeyword>autoflush</amkeyword> must be set to 'yes' if
1422 <amkeyword>taperflush</amkeyword> is greater than 0.</para>
1427 <term><amkeyword>reserved-udp-port</amkeyword> <amtype>int</amtype>,<amtype>int</amtype></term>
1429 <para>Default: --with-udpportrange or
1430 <amdefault>512,1023</amdefault>.
1431 Reserved udp port that will be used (bsd, bsdudp).
1432 Range is inclusive.</para>
1437 <term><amkeyword>reserved-tcp-port</amkeyword> <amtype>int</amtype>,<amtype>int</amtype></term>
1439 <para>Default: --with-low-tcpportrange or
1440 <amdefault>512,1023</amdefault>.
1441 Reserved tcp port that will be used (bsdtcp).
1442 Range is inclusive.</para>
1447 <term><amkeyword>unreserved-tcp-port</amkeyword> <amtype>int</amtype>,<amtype>int</amtype></term>
1449 <para>Default: --with-tcpportrange or
1450 <amdefault>1024,65535</amdefault>.
1451 Unreserved tcp port that will be used (bsd, bsdudp).
1452 Range is inclusive.</para>
1457 <term><amkeyword>recovery-limit</amkeyword> [ <amtype>string</amtype> | <amkeyword>same-host</amkeyword> | <amkeyword>server</amkeyword>]</term>
1459 <para>Default: none (no limitations). This parameter limits the hosts
1460 that may do recoveries. Hosts are identified by their authenticated
1461 peer name, as described in <manref name="amanda-auth" vol="7"/>; if this is
1462 not available and the recovery-limit parameter is present, recovery will be
1463 denied. The arguments to the parameter are strings giving host match
1464 expressions (see <manref name="amanda-match" vol="7"/>) or the special
1465 keywords <amkeyword>same-host</amkeyword> or <amkeyword>server</amkeyword>.
1466 The <amkeyword>same-host</amkeyword> keyword requires an exact
1467 match to the hostname of the DLE being recovered.
1468 The <amkeyword>server</amkeyword> keyword require the connection come from
1469 the fqdn of the server.
1470 Specifying no arguments at all will disable all
1471 recoveries from any host.</para>
1473 <para>Note that match expressions can be constructed to be
1474 forgiving of e.g., fully-qualified vs. unqualified hostnames, but
1475 <amkeyword>same-host</amkeyword> requires an exact match.</para>
1477 <para>The error messages that appear in amrecover are intentionally vague to
1478 avoid information leakage. Consult the amindexd debug log for more details
1479 on the reasons a recovery was rejected.</para>
1481 <para>Recovery limits can be refined on a per-DLE basis using the dumptype
1482 parameter of the same name. Note that the default value will apply to any
1483 dumpfiles for disks which no longer appear in the disklist; thus leaving the
1484 global parameter at its default value but setting it for all DLEs is not
1485 sufficient to maintain secure backups.</para>
1491 <term><amkeyword>tmpdir</amkeyword> <amtype>string</amtype></term>
1493 <para>Default: none (system default). Set it to a directory with lots of free
1494 space if sort in amindexd fail with 'No space left on device'.</para>
1500 <refsect1><title>HOLDINGDISK SECTION</title>
1501 <para>The <emphasis remap='B'>amanda.conf</emphasis>
1502 file may define one or more holding disks used as buffers to hold
1503 backup images before they are written to tape.
1504 The syntax is:</para>
1507 define holdingdisk <emphasis remap='I'>name</emphasis> {
1508 <emphasis remap='I'>holdingdisk-option</emphasis> <emphasis remap='I'>holdingdisk-value</emphasis>
1509 <literal>...</literal>
1513 <para>The { must appear at the end of a line, and the } on its own line.</para>
1514 <para><emphasis remap='I'>Name</emphasis>
1515 is a logical name for this holding disk.</para>
1517 <para>The options and values are:</para>
1519 <variablelist remap='TP'>
1521 <term><amkeyword>comment</amkeyword> <amtype>string</amtype></term>
1523 <para>Default: not set.
1524 A comment string describing this holding disk.</para>
1529 <term><amkeyword>directory</amkeyword> <amtype>string</amtype></term>
1532 <amdefault>"/dumps/amanda"</amdefault>.
1533 The path to this holding area.</para>
1538 <term><amkeyword>use</amkeyword> <amtype>int</amtype></term>
1541 <amdefault>0 Gb</amdefault>.
1542 Amount of space that can be used in this holding disk area.
1543 If the value is zero, all available space on the file system is used.
1544 If the value is negative, Amanda will use all available space minus that value.</para>
1549 <term><amkeyword>chunksize</amkeyword> <amtype>int</amtype></term>
1552 <amdefault>1 Gb</amdefault>.
1553 Holding disk chunk size. Dumps larger than the specified size will be stored in multiple
1554 holding disk files. The size of each chunk will not exceed the specified value.
1555 However, even though dump images are split in the holding disk, they are concatenated as
1556 they are written to tape, so each dump image still corresponds to a single continuous
1557 tape section.</para>
1558 <para>The default unit is Kbytes if it is not specified.</para>
1560 <para>If 0 is specified, Amanda will create holding disk chunks as large as
1561 ((INT_MAX/1024)-64) Kbytes.</para>
1563 <para>Each holding disk chunk includes a 32 Kbyte header, so the minimum
1564 chunk size is 64 Kbytes (but that would be really silly).</para>
1566 <para>Operating systems that are limited to a maximum file size of 2 Gbytes
1567 actually cannot handle files that large.
1568 They must be at least one byte less than 2 Gbytes.
1569 Since Amanda works with 32 Kbyte blocks, and
1570 to handle the final read at the end of the chunk, the chunk size
1571 should be at least 64 Kbytes (2 * 32 Kbytes) smaller than the maximum
1572 file size, e.g. 2047 Mbytes.</para>
1578 <refsect1><title>DUMPTYPE SECTION</title>
1579 <para>The &amconf; file may define multiple sets of backup options
1580 and refer to them by name from the &disklist; file.
1581 For instance, one set of options might be defined for file systems
1582 that can benefit from high compression, another set that does not compress well,
1583 another set for file systems that should always get a full backup and so on.</para>
1585 <para>A set of backup options are entered in a
1586 <amkeyword>dumptype</amkeyword>
1587 section, which looks like this:</para>
1590 define dumptype "<emphasis remap='I'>name</emphasis>" {
1591 <emphasis remap='I'>dumptype-option</emphasis> <emphasis remap='I'>dumptype-value</emphasis>
1592 <literal>...</literal>
1596 <para>The { must appear at the end of a line, and the } on its own line.</para>
1597 <para><emphasis remap='I'>Name</emphasis>
1598 is the name of this set of backup options.
1599 It is referenced from the &disklist; file.</para>
1601 <para>Some of the options in a
1602 <amkeyword>dumptype</amkeyword>
1603 section are the same as those in the main part of &amconf;.
1604 The main option value is used to set the default for all
1605 <amkeyword>dumptype</amkeyword>
1606 sections. For instance, setting
1607 <amkeyword>dumpcycle</amkeyword>
1608 to 50 in the main part of the config file causes all following
1609 <amkeyword>dumptype</amkeyword>
1610 sections to start with that value,
1611 but the value may be changed on a section by section basis.
1612 Changes to variables in the main part of the config file must be
1613 done before (earlier in the file) any
1614 <amkeyword>dumptype</amkeyword>s
1617 <para>The dumptype options and values are:</para>
1619 <variablelist remap='TP'>
1621 <term><amkeyword>auth</amkeyword> <amtype>string</amtype></term>
1624 <amdefault>"bsdtcp"</amdefault>.
1625 Type of authorization to perform between tape server and backup client hosts. See <manref name="amanda-auth" vol="7"/> for more detail.</para>
1630 <term><amkeyword>amandad-path</amkeyword> <amtype>string</amtype></term>
1633 <amdefault>"$libexec/amandad"</amdefault>.
1634 Specify the amandad path of the client, only use with rsh/ssh authentification.
1640 <term><amkeyword>client-username</amkeyword> <amtype>string</amtype></term>
1643 <amdefault>CLIENT_LOGIN</amdefault>.
1644 Specify the username to connect on the client, only use with rsh/ssh authentification.
1650 <term><amkeyword>client-port</amkeyword> [ <amtype>int</amtype> | <amtype>string</amtype> ]</term>
1653 <amdefault>"amanda"</amdefault>.
1654 Specifies the port to connect to on the client. It can be a service name or a numeric port number.
1660 <!-- bumping parameters yanked from the global section above -->
1663 <term><amkeyword>bumpsize</amkeyword> <amtype>int</amtype></term>
1666 <amdefault>10 Mbytes</amdefault>.
1667 The minimum savings required to trigger an automatic
1668 bump from one incremental level to the next, expressed as size.
1669 If Amanda determines that the next higher backup level
1670 will be this much smaller than the current level,
1671 it will do the next level.
1672 The value of this parameter is used only if the parameter
1673 <emphasis>bumppercent</emphasis> is set to 0.
1675 <para>The default unit is Kbytes if it is not specified.</para>
1677 See also the options
1678 <amkeyword>bumppercent</amkeyword>,
1679 <amkeyword>bumpmult</amkeyword> and
1680 <amkeyword>bumpdays</amkeyword>.</para>
1685 <term><amkeyword>bumppercent</amkeyword> <amtype>int</amtype></term>
1688 <amdefault>0</amdefault>.
1689 The minimum savings required to trigger an automatic
1690 bump from one incremental level to the next, expressed as percentage of the
1691 current size of the DLE (size of current level 0).
1692 If Amanda determines that the next higher backup level
1693 will be this much smaller than the current level,
1694 it will do the next level.
1697 If this parameter is set to 0, the value of the
1698 parameter <emphasis>bumpsize</emphasis> is used to trigger bumping.
1701 See also the options
1702 <amkeyword>bumpsize</amkeyword>,
1703 <amkeyword>bumpmult</amkeyword> and
1704 <amkeyword>bumpdays</amkeyword>.</para>
1709 <term><amkeyword>bumpmult</amkeyword> <amtype>float</amtype></term>
1712 <amdefault>1.5</amdefault>.
1713 The bump size multiplier.
1715 <amkeyword>bumpsize</amkeyword>
1716 by this factor for each level.
1717 This prevents active filesystems from
1718 bumping too much by making it harder to bump to the next level.
1719 For example, with the default
1720 <amkeyword>bumpsize</amkeyword>
1722 <amkeyword>bumpmult</amkeyword>
1723 set to 2.0, the bump threshold will be 10 Mbytes for level one, 20
1724 Mbytes for level two, 40 Mbytes for level three, and so on.</para>
1728 <term><amkeyword>bumpdays</amkeyword> <amtype>int</amtype></term>
1731 <amdefault>2 days</amdefault>.
1732 To insure redundancy in the dumps, Amanda keeps filesystems at the
1733 same incremental level for at least
1734 <amkeyword>bumpdays</amkeyword>
1735 days, even if the other bump threshold criteria are met.</para>
1740 <term><amkeyword>comment</amkeyword> <amtype>string</amtype></term>
1742 <para>Default: not set.
1743 A comment string describing this set of backup options.</para>
1747 <term><amkeyword>comprate</amkeyword> <amtype>float</amtype> [, <amtype>float</amtype> ]</term>
1750 <amdefault>0.50, 0.50</amdefault>.
1751 The expected full and incremental compression factor for dumps.
1752 It is only used if Amanda does not have any history information on
1753 compression rates for a filesystem, so should not usually need to be set.
1754 However, it may be useful for the first time a very large filesystem that
1755 compresses very little is backed up.</para>
1759 <term><amkeyword>compress</amkeyword> [ <amkeyword>none</amkeyword> | <amkeyword>client</amkeyword> | <amkeyword>server</amkeyword> ] [ <amkeyword>best</amkeyword> | <amkeyword>fast</amkeyword> | <amkeyword>custom</amkeyword> ]</term>
1762 <amkeyword>client fast</amkeyword>.
1763 If Amanda does compression of the backup images, it can do so either on the backup client
1764 host before it crosses the network or on the tape server host as it goes from the
1765 network into the holding disk or to tape. Which place to do compression (if at all) depends on how well the
1766 dump image usually compresses, the speed and load on the client or server, network capacity,
1767 holding disk capacity, availability of tape hardware compression, etc.</para>
1769 <para>For either type of compression, Amanda also allows the selection
1770 of three styles of compression.
1771 <amkeyword>best</amkeyword>
1772 is the best compression available, often at the expense of CPU overhead.
1773 <amkeyword>fast</amkeyword>
1774 is often not as good a compression as
1775 <amkeyword>best</amkeyword>,
1776 but usually less CPU overhead. Or to specify <amkeyword>custom</amkeyword>
1777 to use your own compression method. (See dumptype custom-compress in example/amanda.conf for reference)</para>
1779 <para>So the <amkeyword>compress</amkeyword> options line may be one of:</para>
1781 <variablelist remap='TP'>
1783 <term>compress none</term><listitem><para/></listitem>
1786 <term>compress client fast</term><listitem><para/></listitem>
1789 <term>compress client best</term><listitem><para/></listitem>
1792 <term>compress client custom</term>
1794 <para>Specify <amkeyword>client-custom-compress</amkeyword> "PROG"</para>
1795 <para>PROG must not contain white space and it must accept -d for uncompress.</para>
1799 <term>compress server fast</term><listitem><para/></listitem>
1802 <term>compress server best</term><listitem><para/></listitem>
1805 <term>compress server custom</term>
1807 <para>Specify <amkeyword>server-custom-compress</amkeyword> "PROG"</para>
1808 <para>PROG must not contain white space and it must accept -d for uncompress.</para>
1812 <para>Note that some tape devices do compression and this option has nothing
1813 to do with whether that is used. If hardware compression is used (usually via a particular tape device name
1814 or <emphasis remap='B'>mt</emphasis> option), Amanda (software) compression should be disabled.</para>
1819 <term><amkeyword>client-custom-compress</amkeyword> <amtype>string</amtype></term>
1821 <para>Default: none.
1822 The program to use to perform compression/decompression on the client; used with
1823 "compress client custom". Must not contain whitespace. Must accept -d to uncompress.</para>
1828 <term><amkeyword>server-custom-compress</amkeyword> <amtype>string</amtype></term>
1830 <para>Default: none.
1831 The program to use to perform compression/decompression on the server; used with
1832 "compress server custom". Must not contain whitespace. Must accept -d to uncompress.</para>
1837 <term><amkeyword>dumpcycle</amkeyword> <amtype>int</amtype></term>
1839 <para>Default: <amdefault>10 days</amdefault>.
1840 The number of days in the backup cycle. Each disk using this set of options will get a full
1841 backup at least this of
1842 ten. Setting this to zero tries to do a full backup each run.</para>
1847 <term><amkeyword>encrypt</amkeyword> [ <amkeyword>none</amkeyword> | <amkeyword>client</amkeyword> | <amkeyword>server</amkeyword> ]</term>
1849 <para>Default: not set.
1850 To encrypt backup images, it can do so either on the backup client host before it crosses the network or on the tape
1851 server host as it goes from the network into the holding disk or to tape.</para>
1853 <para>So the <amkeyword>encrypt</amkeyword> options line may be one
1856 <variablelist remap='TP'>
1858 <term>encrypt none</term><listitem><para/></listitem>
1861 <term>encrypt client</term>
1863 <para>Specify client-encrypt "PROG"</para>
1864 <para>PROG must not contain white space.</para>
1865 <para>Specify client-decrypt-option "decryption-parameter" Default: "-d"</para>
1866 <para>decryption-parameter must not contain white space.</para>
1867 <para>(See dumptype client-encrypt-nocomp in example/amanda.conf for reference)</para>
1871 <term>encrypt server</term>
1873 <para>Specify server-encrypt "PROG"</para>
1874 <para>PROG must not contain white space.</para>
1875 <para>Specify server-decrypt-option "decryption-parameter" Default: "-d"</para>
1876 <para>decryption-parameter must not contain white space.</para>
1877 <para>(See dumptype server-encrypt-fast in example/amanda.conf for reference)</para>
1881 <para>Note that current logic assumes compression then encryption during
1882 backup(thus decrypt then uncompress during restore). So specifying
1883 client-encryption AND server-compression is not supported.
1884 <emphasis remap='I'>amcrypt</emphasis> which is a wrapper of
1885 <emphasis remap='I'>aespipe</emphasis> is provided as a reference
1886 symmetric encryption program.</para>
1891 <term><amkeyword>client-encrypt</amkeyword> <amtype>string</amtype></term>
1893 <para>Default: none.
1894 The program to use to perform encryption/decryption on the client; used with
1895 "encrypt client". Must not contain whitespace.</para>
1900 <term><amkeyword>client-decrypt-option</amkeyword> <amtype>string</amtype></term>
1903 The option that can be passed to client-encrypt to make it decrypt instead.
1904 Must not contain whitespace.</para>
1909 <term><amkeyword>server-encrypt</amkeyword> <amtype>string</amtype></term>
1911 <para>Default: none.
1912 The program to use to perform encryption/decryption on the server; used with
1913 "encrypt server". Must not contain whitespace.</para>
1918 <term><amkeyword>server-decrypt-option</amkeyword> <amtype>string</amtype></term>
1921 The option that can be passed to server-encrypt to make it decrypt instead.
1922 Must not contain whitespace.</para>
1927 <term><amkeyword>estimate</amkeyword> [ <amkeyword>client</amkeyword> | <amkeyword>calcsize</amkeyword> | <amkeyword>server</amkeyword> ]+</term>
1929 <para>Default: <amkeyword>client</amkeyword>.
1930 Determine the way Amanda estimates the size of each DLE before beginning a backup. This is a list of acceptable estimate methods, and Amanda applies the first method supported by the application. The methods are:</para>
1931 <variablelist remap='TP'>
1935 <para>Use the same program as the dumping program. This is the most
1936 accurate method to do estimates, but it can take a long time.</para>
1940 <term>calcsize</term>
1942 <para>Use a faster program to do estimates, but the result is less
1949 <para>Use only statistics from the previous few runs to give an estimate.
1950 This very quick, but the result is not accurate if your disk
1951 usage changes from day to day. If this method is specified, but the
1952 server does not have enough data to make an estimate, then the option
1953 is internally moved to the end of the list, thereby preferring 'client'
1954 or 'calcsize' in this case.</para>
1963 <term><amkeyword>exclude</amkeyword> [ <amkeyword>list</amkeyword> | <amkeyword>file</amkeyword> ][[<amkeyword>optional</amkeyword>][<amkeyword>append</amkeyword>][ <amtype>string</amtype> ]+]</term>
1966 <amkeyword>file</amkeyword>. Exclude is the opposite of
1967 <amkeyword>include</amkeyword> and specifies files that will be excluded from
1968 the backup. The format of the exclude expressions depends on the application,
1969 and some applications do not support excluding files at all.</para>
1971 <para>There are two exclude parameters,
1972 <amkeyword>exclude</amkeyword> <amkeyword>file</amkeyword>
1974 <amkeyword>exclude</amkeyword> <amkeyword>list.</amkeyword>
1976 <amkeyword>exclude</amkeyword> <amkeyword>file</amkeyword>, the
1977 <amtype>string</amtype>
1978 is an exclude expression. With
1979 <amkeyword>exclude</amkeyword> <amkeyword>list</amkeyword>
1981 <amtype>string</amtype>
1982 is a file name on the client containing &gnutar; exclude expressions.
1983 The path to the specified exclude list file, if present (see description of
1984 'optional' below), must be readable by the Amanda user.</para>
1986 <para>All exclude expressions are concatenated in one file and passed to the
1987 application as an <option>--exclude-from</option> argument.</para>
1989 <para>For &gnutar;, exclude expressions must always be specified as relative to the
1990 top-level directory of the DLE, and must start with "./". See the manpages for individual
1991 applications for more information on supported exclude expressions.</para>
1994 <amkeyword>append</amkeyword> keyword, the
1995 <amtype>string</amtype> is appended to the current list, without it, the
1996 <amtype>string</amtype> overwrites the list.</para>
1998 <para>If <amkeyword>optional</amkeyword>
1999 is specified for <amkeyword>exclude</amkeyword> <amkeyword>list</amkeyword>,
2000 then amcheck will not complain if the file doesn't exist or is not readable.</para>
2002 <para>For <amkeyword>exclude</amkeyword> <amkeyword>list</amkeyword>,
2003 if the file name is relative, the disk name being backed up is prepended.
2004 So if this is entered:</para>
2007 exclude list ".amanda.excludes"
2010 <para>the actual file used would be
2011 <filename>/var/.amanda.excludes</filename>
2012 for a backup of <filename>/var</filename>,
2013 <filename>/usr/local/.amanda.excludes</filename> for a backup of
2014 <filename>/usr/local</filename>, and so on.</para>
2019 <term><amkeyword>holdingdisk</amkeyword> [ <amkeyword>never</amkeyword> | <amkeyword>auto</amkeyword> | <amkeyword>required</amkeyword> ]</term>
2022 <amkeyword>auto</amkeyword>.
2023 Whether a holding disk should be used for these backups or whether they should go directly to tape.
2024 If the holding disk is a portion of another file system that Amanda
2025 is backing up, that file system should refer to a dumptype with
2026 <amkeyword>holdingdisk</amkeyword>
2028 <emphasis remap='I'>never</emphasis>
2029 to avoid backing up the holding disk into itself.</para>
2031 <variablelist remap='TP'>
2033 <term><amkeyword>never</amkeyword>|no|false|off</term>
2035 <para>Never use a holdingdisk, the dump will always go directly to tape. There will be no dump if you have a tape error.</para>
2039 <term><amkeyword>auto</amkeyword>|yes|true|on</term>
2041 <para>Use the holding disk, unless there is a problem with the holding disk, the dump won't fit there or the medium doesn't require spooling (e.g., VFS device)</para>
2045 <term><amkeyword>required</amkeyword></term>
2047 <para>Always dump to holdingdisk, never directly to tape. There will be no dump if it doesn't fit on holdingdisk</para>
2056 <term><amkeyword>ignore</amkeyword> <amtype>boolean</amtype></term>
2059 <amkeyword>no</amkeyword>.
2060 Whether disks associated with this backup type should be backed up or not.
2061 This option is useful when the
2062 <emphasis remap='I'>disklist</emphasis>
2063 file is shared among several configurations,
2064 some of which should not back up all the listed file systems.</para>
2068 <term><amkeyword>include</amkeyword> [ <amkeyword>list</amkeyword> | <amkeyword>file</amkeyword> ][[<amkeyword>optional</amkeyword>][<amkeyword>append</amkeyword>][ <amtype>string</amtype> ]+]</term>
2071 <amkeyword>file</amkeyword>
2073 There are two include lists,
2074 <amkeyword>include file</amkeyword>
2076 <amkeyword>include list.</amkeyword>
2078 <amkeyword>include file</amkeyword>
2080 <amtype>string</amtype>
2081 is a glob expression. With
2082 <amkeyword>include list</amkeyword>
2084 <amtype>string</amtype>
2085 is a file name on the client containing glob expressions.</para>
2087 <para>All include expressions are expanded by Amanda, concatenated in one file and passed to &gnutar; as a
2088 <option>--files-from</option> argument. They must start with "./" and contain no other "/".</para>
2090 <para>Include expressions must always be specified as relative to the
2091 head directory of the DLE.</para>
2093 <note><para>For globbing to work at all, even the limited single level,
2094 the top level directory of the DLE must be readable by the Amanda user.</para>
2097 <para>With the <amkeyword>append</amkeyword> keyword, the
2098 <amtype>string</amtype> is appended to the current list, without it, the
2099 <amtype>string</amtype> overwrites the list.</para>
2102 <amkeyword>optional</amkeyword> is specified for
2103 <amkeyword>include list</amkeyword>, then amcheck will not complain if the file
2104 doesn't exist or is not readable.</para>
2106 <para>For <amkeyword>include list</amkeyword>,
2107 If the file name is relative, the disk name being backed up is prepended.</para>
2112 <term><amkeyword>index</amkeyword> <amtype>boolean</amtype></term>
2114 <para>Default: <amkeyword>no</amkeyword>.
2115 Whether an index (catalogue) of the backup should be generated and saved in
2116 <amkeyword>indexdir</amkeyword>.
2117 These catalogues are used by the <emphasis remap='B'>amrecover</emphasis> utility.</para>
2122 <term><amkeyword>kencrypt</amkeyword> <amtype>boolean</amtype></term>
2125 <amkeyword>no</amkeyword>.
2126 Whether the backup image should be encrypted by Kerberos as it is sent
2127 across the network from the backup client host to the tape server host.</para>
2131 <term><amkeyword>maxdumps</amkeyword> <amtype>int</amtype></term>
2134 <amdefault>1</amdefault>.
2135 The maximum number of backups from a single host that Amanda will attempt to run in parallel.
2136 See also the main section parameter <amkeyword>inparallel</amkeyword>.</para>
2140 <term><amkeyword>maxpromoteday</amkeyword> <amtype>int</amtype></term>
2143 <amdefault>10000</amdefault>.
2144 The maximum number of day for a promotion, set it 0 if you don't want
2145 promotion, set it to 1 or 2 if your disks get overpromoted.</para>
2149 <term><amkeyword>max-warnings</amkeyword> <amtype>int</amtype></term>
2152 <amdefault>20</amdefault>.
2153 The maximum number of error lines in the report for a dle. A value of '0'
2154 means unlimited. This is useful to reduce the size of the log file and the
2155 size of the report. All errors are put in separate files if a dle have more
2160 <term><amkeyword>priority</amkeyword> [ <amkeyword>low</amkeyword> | <amkeyword>medium</amkeyword> | <amkeyword>high</amkeyword> ]</term>
2162 <para>Default: <amkeyword>medium</amkeyword>.
2163 When there is no tape to write to, Amanda will do incremental backups
2164 in priority order to the holding disk. The priority may be
2165 high (2), medium (1), low (0) or a number of your choice.</para>
2169 <term><amkeyword>program</amkeyword> [ <amkeyword>"DUMP"</amkeyword> | <amkeyword>"GNUTAR"</amkeyword> | <amkeyword>"APPLICATION"</amkeyword> ]</term>
2172 <amdefault>"DUMP"</amdefault>.
2173 The type of backup to perform. Valid values are:</para>
2175 <variablelist remap='TP'>
2177 <term><amkeyword>"DUMP"</amkeyword></term>
2179 <para>The native operating system backup program.</para>
2183 <term><amkeyword>"GNUTAR"</amkeyword></term>
2185 <para>To use GNU-tar or to do PC backups using Samba.</para>
2189 <term><amkeyword>"APPLICATION"</amkeyword></term>
2191 <para>To use an application, see the <emphasis>application</emphasis> option.</para>
2198 <term><amkeyword>application</amkeyword> <amtype>string</amtype></term>
2200 <para>No default. Must be the name of an application if <emphasis>program</emphasis> is set to <emphasis remap='I'>APPLICATION</emphasis>. See APPLICATION SECTION below.</para>
2204 <term><amkeyword>script</amkeyword> <amtype>string</amtype></term>
2206 <para>No default. Must be the name of a script. You can have many script. See SCRIPT SECTION below.</para>
2210 <term><amkeyword>property</amkeyword> [<amkeyword>append</amkeyword>] <amtype>string</amtype> <amtype>string</amtype>+</term>
2212 <para>These options can set various properties, they can be used by third
2213 party software to store information in the configuration file.
2214 Both strings are quoted; the first string contains the name of
2215 the property to set, and the others contains its values.
2216 <amkeyword>append</amkeyword> keyword append the values to the list of values for that property.
2222 <term><amkeyword>record</amkeyword> <amtype>boolean</amtype></term>
2225 <emphasis remap='I'>yes</emphasis>.
2226 Whether to ask the backup program to update its database (e.g. <filename>/var/lib/dumpdates</filename>
2227 for DUMP or <filename>/usr/local/var/amanda/gnutar-lists</filename> for GNUTAR) of time stamps.
2228 This is normally enabled for daily backups and turned off for periodic archival runs.</para>
2232 <term><amkeyword>skip-full</amkeyword> <amtype>boolean</amtype></term>
2235 <amkeyword>no</amkeyword>. If <amkeyword>true</amkeyword> and
2236 <emphasis remap='B'>planner</emphasis> has scheduled a full backup, these disks will be skipped, and
2237 full backups should be run off-line on these days. It was reported that Amanda only schedules
2238 level 1 incrementals in this configuration; this is probably a bug.</para>
2242 <term><amkeyword>skip-incr</amkeyword> <amtype>boolean</amtype></term>
2245 <amkeyword>no</amkeyword>. If <amkeyword>true</amkeyword> and
2246 <emphasis remap='B'>planner</emphasis> has scheduled an incremental backup, these disks will be skipped.</para>
2250 <term><amkeyword>ssh-keys</amkeyword> <amtype>string</amtype></term>
2252 <para>Default: not set.
2253 The key file the ssh auth will use, it must be the private key. If this parameter is not specified, then the default ssh key will be used.</para>
2257 <term><amkeyword>starttime</amkeyword> <amtype>int</amtype></term>
2259 <para>Default: not set.
2260 Backup of these disks will not start until after this time of day.
2261 The value should be hh*100+mm, e.g. 6:30PM (18:30) would be entered as
2262 <literal>1830</literal>.</para>
2266 <term><amkeyword>strategy</amkeyword> [ <amkeyword>standard</amkeyword> | <amkeyword>nofull</amkeyword> | <amkeyword>noinc</amkeyword> | <amkeyword>skip</amkeyword> | <amkeyword>incronly</amkeyword> ]</term>
2268 <para>Default: <amkeyword>standard</amkeyword>.
2269 Strategy to use when planning what level of backup to run next. Values are:</para>
2271 <variablelist remap='TP'>
2273 <term><amkeyword>standard</amkeyword></term>
2275 <para>The standard Amanda schedule.</para>
2279 <term><amkeyword>nofull</amkeyword></term>
2281 <para>Never do full backups, only level 1 incrementals.</para>
2285 <term><amkeyword>noinc</amkeyword></term>
2287 <para>Never do incremental backups, only full dumps.</para>
2291 <term><amkeyword>skip</amkeyword></term>
2293 <para>Treat this DLE as if it doesn't exist (useful to disable DLEs
2294 when sharing the <emphasis remap='I'>disklist</emphasis> file between
2295 multiple configurations). Skipped DLEs will not be checked or
2296 dumped, and will not be matched by disklist expressions.</para>
2300 <term><amkeyword>incronly</amkeyword></term>
2302 <para>Only do incremental dumps. <command>amadmin force</command> should be used to tell
2303 Amanda that a full dump has been performed off-line, so that it resets to level 1.</para>
2311 <term><amkeyword>allow-split</amkeyword> <amtype>bool</amtype></term>
2313 <para>Default: true.
2314 If true, then dumps with this dumptype can be split on the storage media. If false, then
2315 the dump will be written in a single file on the media. See "Dump Splitting Configuration"
2321 <term><amkeyword>tape-splitsize</amkeyword> <amtype>int</amtype></term>
2323 <para>Deprecated. See "Dump Splitting Configuration" below.</para>
2324 <para>Default: not set.
2325 Split dump file on tape into pieces of a specified size.
2326 The default unit is Kbytes if it is not specified.</para>
2331 <term><amkeyword>split-diskbuffer</amkeyword> <amtype>string</amtype></term>
2332 <para>Deprecated. See "Dump Splitting Configuration" below.</para>
2334 <para>Default: not set.
2335 When dumping a split dump in PORT-WRITE mode (usually meaning "no holding disk"), buffer the split chunks to a file in the directory specified by this option.
2341 <term><amkeyword>fallback-splitsize</amkeyword> <amtype>int</amtype></term>
2343 <para>Deprecated. See "Dump Splitting Configuration" below.</para>
2345 <amdefault>10M</amdefault>.
2346 This specifies the part size used when no <amkeyword>split-diskbuffer</amkeyword> is specified, or when it is too small or does not exist,
2347 and thus the maximum amount of memory consumed for in-memory splitting.
2348 The default unit is Kbytes if it is not specified.</para>
2353 <term><amkeyword>recovery-limit</amkeyword>
2354 [ <amkeyword>server</amkeyword> | <amkeyword>same-host</amkeyword> | <amtype>string</amtype> ]*</term>
2357 <para>Default: global value. This parameter overrides the global
2358 <amkeyword>recovery-limit</amkeyword> parameter for DLEs of this
2365 <term><amkeyword>dump-limit</amkeyword>
2366 [ <amkeyword>server</amkeyword> | <amkeyword>same-host</amkeyword> ]*</term>
2369 <para>Default: <amkeyword>server</amkeyword>. Specify which host can initiate
2370 a backup of the dle. With <amkeyword>server</amkeyword>, the server
2371 can initiate a backup with the <command>amdump</command> command.
2372 With <amkeyword>same-host</amkeyword>, the client can initiate a backup with
2373 the <command>amdump_client</command> command.</para>
2379 <para>The following <amkeyword>dumptype</amkeyword> entries are predefined by Amanda:</para>
2381 <programlisting remap='.nf'>
2382 define dumptype "no-compress" {
2385 define dumptype "compress-fast" {
2386 compress client fast
2388 define dumptype "compress-best" {
2389 compress client best
2391 define dumptype "srvcompress" {
2392 compress server fast
2394 define dumptype "bsd-auth" {
2397 define dumptype "bsdtcp-auth" {
2400 define dumptype "no-record" {
2403 define dumptype "no-hold" {
2406 define dumptype "no-full" {
2411 <para>In addition to options in a
2412 <amkeyword>dumptype</amkeyword>
2413 section, one or more other
2414 <amkeyword>dumptype</amkeyword>
2415 names may be supplied as identifiers, which make this
2416 <amkeyword>dumptype</amkeyword>
2417 inherit options from other previously defined
2418 <amkeyword>dumptype</amkeyword>s.
2419 For instance, two sections might be the same except for the
2420 <amkeyword>record</amkeyword> option:</para>
2422 <programlisting remap='.nf'>
2423 define dumptype "normal" {
2424 comment "Normal backup, no compression, do indexing"
2429 define dumptype "testing" {
2430 comment "Test backup, no compression, do indexing, no recording"
2436 <para>Amanda provides a
2437 <amkeyword>dumptype</amkeyword>
2439 <emphasis remap='I'>global</emphasis>
2441 <emphasis remap='B'>amanda.conf</emphasis>
2443 <amkeyword>dumptype</amkeyword>s
2445 This provides an easy place to make changes that will affect every
2446 <amkeyword>dumptype</amkeyword>, although you must be careful
2447 that every dumptype explicitly inherits from the
2448 <emphasis remap='I'>global</emphasis>
2449 dumptype - Amanda does not do so automatically.</para>
2452 <refsect1><title>TAPETYPE SECTION</title>
2453 <para>The <emphasis remap='B'>amanda.conf</emphasis>
2454 file may define multiple types of tape media and devices.
2455 The information is entered in a
2456 <amkeyword>tapetype</amkeyword>
2457 section, which looks like this in the config file:</para>
2460 define tapetype "<amtype>name</amtype>" {
2461 <emphasis remap='I'>tapetype-option</emphasis> <emphasis remap='I'>tapetype-value</emphasis>
2462 <literal>...</literal>
2466 <para>The { must appear at the end of a line, and the } on its own line.</para>
2467 <para><emphasis remap='I'>Name</emphasis>
2468 is the name of this type of tape medium/device.
2469 It is referenced from the
2470 <amkeyword>tapetype</amkeyword>
2471 option in the main part of the config file.</para>
2473 <para>The tapetype options and values are:</para>
2474 <variablelist remap='TP'>
2476 <term><amkeyword>comment</amkeyword> <amtype>string</amtype></term>
2478 <para>Default: not set.
2479 A comment string describing this set of tape information.</para>
2483 <term><amkeyword>filemark</amkeyword> <amtype>int</amtype></term>
2486 <amdefault>1 kbytes</amdefault>.
2487 How large a file mark (tape mark) is, measured in kbytes.
2488 If the size is only known in some linear measurement (e.g. inches),
2489 convert it to kbytes using the device density.</para>
2493 <term><amkeyword>length</amkeyword> <amtype>int</amtype></term>
2496 <amdefault>2000 kbytes</amdefault>. How much data will fit on a tape, expressed in kbytes.</para>
2498 <para>Note that this value is only used by Amanda to schedule which backups will be run.
2499 Once the backups start, Amanda will continue to write to a tape until it gets an error,
2500 regardless of what value is entered for <amkeyword>length</amkeyword>
2501 (but see <manref name="amanda-devices" vol="7" /> for exceptions).
2507 <term><amkeyword>blocksize</amkeyword> <amtype>int</amtype></term>
2510 <amdefault>32 kbytes</amdefault>.
2511 How much data will be written in each tape record, expressed in kbytes. This is similar to the <emphasis remap="I">BLOCK_SIZE</emphasis> device property, but if the blocksize is not a multiple of 1024 bytes, then this parameter cannot be used to specify it, and the property must be used instead.</para>
2515 <term><amkeyword>readblocksize</amkeyword> <amtype>int</amtype></term>
2517 <para>Default: <amdefault>32 kytes</amdefault>
2518 How much data will be read in each tape record. This can be used to override a
2519 device's block size for reads only. This may be useful, for example, in
2520 reading a tape written with a 256k block size when Amanda is configured to use
2521 128k blocks. This unusual feature is not supported by all operating systems and
2524 <para>The default unit is Kbytes if it is not specified.</para>
2529 <term><amkeyword>part-size</amkeyword> <amtype>int</amtype></term>
2531 <para>If this is set to zero (default), then no splitting will take place,
2532 and the entire dump will fail, if end-of-medium is encountered before
2533 the dump is complete, unless the device property LEOM is true, and the
2534 device can detect EOM. See "Dump Splitting Configuration" below.</para>
2539 <term><amkeyword>part-cache-type</amkeyword> [ <amkeyword>none</amkeyword> | <amkeyword>disk</amkeyword> | <amkeyword>memory</amkeyword> ] </term>
2541 <para>Default: none. When part caching is required, this parameter specifies
2542 the type of caching that will be used. The options include no caching
2543 (<amkeyword>none</amkeyword>), in which case a failed part will cause the
2544 entire dump to fail; on-disk caching (<amkeyword>disk</amkeyword>), for
2545 which <amkeyword>part-cache-dir</amkeyword> must be set properly; and
2546 in-memory caching (<amkeyword>memory</amkeyword>), which on most systems
2547 severely restrains the size of the part that can be written. See "Dump
2548 Splitting Configuration" below.</para>
2553 <term><amkeyword>part-cache-dir</amkeyword> <amtype>string</amtype></term>
2555 <para>Default: none.
2556 The directory in which part-cache files can be written when caching on disk.
2557 See "Dump Splitting Configuration" below.</para>
2562 <term><amkeyword>part-cache-max-size</amkeyword> <amtype>int</amtype></term>
2564 <para>Default: none.
2565 The maximum part size to use when caching is in effect. This is used to limit
2566 the part size when disk or memory space for caching is constrained. This value
2567 must be greater than zero.</para>
2572 <term><amkeyword>speed</amkeyword> <amtype>int</amtype></term>
2575 <amdefault>200 bps</amdefault>.
2576 How fast the drive will accept data, in bytes per second.
2577 This parameter is NOT currently used by Amanda.</para>
2581 <term><amkeyword>lbl-templ</amkeyword> <amtype>string</amtype></term>
2583 <para>Default: not set.
2584 A PostScript template file used by
2585 <emphasis remap='B'>amreport</emphasis>
2586 to generate labels. Several sample files are provided with the Amanda sources in the
2587 <emphasis remap='I'>example</emphasis> directory.
2589 <manref name="amreport" vol="8"/>
2590 man page for more information.</para>
2595 <para>In addition to options, another
2596 <amkeyword>tapetype</amkeyword>
2597 name may be supplied as an identifier, which makes this
2598 <amkeyword>tapetype</amkeyword>
2599 inherit options from another
2600 <amkeyword>tapetype</amkeyword>.
2601 For instance, the only difference between a DLT4000 tape drive using
2602 Compact-III tapes and one using Compact-IV tapes is the length of the tape.
2603 So they could be entered as:</para>
2605 <programlisting remap='.nf'>
2606 define tapetype "DLT4000-III" {
2607 comment "DLT4000 tape drives with Compact-III tapes"
2608 length 12500 mbytes # 10 Gig tapes with some compression
2609 filemark 2000 kbytes
2612 define tapetype "DLT4000-IV" {
2614 comment "DLT4000 tape drives with Compact-IV tapes"
2615 length 25000 mbytes # 20 Gig tapes with some compression
2619 <!-- If anybody has read this, where do all those stupid
2620 pseudo-duplicate FOM-tapetype-entries come from??
2626 <refsect1><title>INTERFACE SECTION</title>
2628 <emphasis remap='B'>amanda.conf</emphasis>
2629 file may define multiple types of network interfaces.
2630 The information is entered in an <amkeyword>interface</amkeyword>
2631 section, which looks like this:</para>
2634 define interface "<emphasis remap='I'>name</emphasis>" {
2635 <emphasis remap='I'>interface-option</emphasis> <emphasis remap='I'>interface-value</emphasis>
2636 <literal>...</literal>
2640 <para>The { must appear at the end of a line, and the } on its own line.</para>
2641 <para><emphasis remap='I'>name</emphasis>
2642 is the name of this type of network interface. It is referenced from the
2643 <emphasis remap='I'>disklist</emphasis> file.</para>
2645 <para>Note that these sections define network interface characteristics,
2646 not the actual interface that will be used. Nor do they impose limits on the bandwidth that will
2647 actually be taken up by Amanda.
2648 Amanda computes the estimated bandwidth each file system backup will take
2649 based on the estimated size and time, then compares that plus any other running
2650 backups with the limit as another of the criteria when deciding whether
2651 to start the backup. Once a backup starts, Amanda will use as much of the network as it can
2652 leaving throttling up to the operating system and network hardware.</para>
2654 <para>The interface options and values are:</para>
2655 <variablelist remap='TP'>
2657 <term><amkeyword>comment</amkeyword> <amtype>string</amtype></term>
2659 <para>Default: not set.
2660 A comment string describing this set of network information.</para>
2664 <term><amkeyword>use</amkeyword> <amtype>int</amtype></term>
2667 <amdefault>80000 Kbps</amdefault>.
2668 The speed of the interface in Kbytes per second.</para>
2673 <para>In addition to options, another
2674 <amkeyword>interface</amkeyword>
2675 name may be supplied as an identifier, which makes this
2676 <amkeyword>interface</amkeyword>
2677 inherit options from another
2678 <amkeyword>interface</amkeyword>.
2679 At the moment, this is of little use.</para>
2682 <refsect1><title>APPLICATION SECTION</title>
2684 <emphasis remap='B'>amanda.conf</emphasis>
2685 file may define multiple types of application.
2686 The information is entered in a <amkeyword>application</amkeyword>
2687 section, which looks like this:</para>
2690 define application "<emphasis remap='I'>name</emphasis>" {
2691 <emphasis remap='I'>application-option</emphasis> <emphasis remap='I'>application-value</emphasis>
2692 <literal>...</literal>
2695 <para>The { must appear at the end of a line, and the } on its own line.</para>
2697 <para><emphasis remap='I'>name</emphasis>
2698 is the name of this type of application. It is referenced from the
2699 <emphasis remap='I'>dumptype</emphasis></para>
2701 <para>The application options and values are:</para>
2702 <variablelist remap='TP'>
2704 <term><amkeyword>client-name</amkeyword> <amtype>string</amtype></term>
2707 specifies an application name that is in the amanda-client.conf on the client.
2708 The setting from that application will be merged with the current application.
2709 If <amkeyword>client-name</amkeyword> is set then it is an error
2710 if that application is not defined on the client.</para>
2711 <para>If <amkeyword>client-name</amkeyword> is not set then the
2712 merge is done with the application that have the name equal to the plugin.
2713 eg. if the plugin is 'amgtar', then the setting from the application 'amgtar'
2714 is used if it is defined.
2719 <term><amkeyword>comment</amkeyword> <amtype>string</amtype></term>
2721 <para>Default: not set.
2722 A comment string describing this application.</para>
2726 <term><amkeyword>plugin</amkeyword> <amtype>string</amtype></term>
2728 <para>No default. Must be set to the name of the program. This program must be in the <emphasis remap='I'>$libexecdir/amanda/application</emphasis> directory on the client.</para>
2732 <term><amkeyword>property</amkeyword> [<amkeyword>append</amkeyword>] [<amkeyword>priority</amkeyword>] <amtype>string</amtype> <amtype>string</amtype>+</term>
2734 <para>No default. You can set property for the application, each application have a different set of property. Both strings are quoted; the first string contains the name of
2735 the property to set, and the others contains its values.
2736 <amkeyword>append</amkeyword> keyword append the values to the list of values for that property.
2737 <amkeyword>priority</amkeyword> keyword disallow the setting of that property on the client.
2744 <refsect1><title>SCRIPT SECTION</title>
2746 <emphasis remap='B'>amanda.conf</emphasis>
2747 file may define multiple types of script.
2748 The information is entered in a <amkeyword>script</amkeyword>
2749 section, which looks like this:</para>
2752 define script "<emphasis remap='I'>name</emphasis>" {
2753 <emphasis remap='I'>script-option</emphasis> <emphasis remap='I'>script-value</emphasis>
2754 <literal>...</literal>
2757 <para>The { must appear at the end of a line, and the } on its own line.</para>
2759 <para><emphasis remap='I'>name</emphasis>
2760 is the name of this type of script. It is referenced from the
2761 <emphasis remap='I'>dumptype</emphasis></para>
2763 <para>The script options and values are:</para>
2764 <variablelist remap='TP'>
2766 <term><amkeyword>client-name</amkeyword> <amtype>string</amtype></term>
2769 specifies a script name that is in the amanda-client.conf on the client.
2770 The setting from that script will be merged with the currect script.
2771 If <amkeyword>client-name</amkeyword> is set then it is an error if that script is not defined on the client.</para>
2772 <para>If <amkeyword>client-name</amkeyword> is not set then the
2773 merge is done with the script that have the name equal to the plugin.
2774 eg. if the plugin is 'amlog-script', then the setting from the script
2775 'amlog-script' is used.
2780 <term><amkeyword>comment</amkeyword> <amtype>string</amtype></term>
2782 <para>Default: not set.
2783 A comment string describing this script.</para>
2787 <term><amkeyword>plugin</amkeyword> <amtype>string</amtype></term>
2789 <para>No default. Must be set to the name of the program. This program must be in the <emphasis remap='I'>$libexecdir/amanda/application</emphasis> directory on the client and/or server.</para>
2793 <term><amkeyword>order</amkeyword> <amtype>int</amtype></term>
2795 <para>Default: <amdefault>5000</amdefault>. Scripts are executed in that order, it is useful if you have many scripts and they must be executed in a spefific order.</para>
2799 <term><amkeyword>single-execution</amkeyword> <amtype>boolean</amtype></term>
2801 <para>Default: <amdefault>no</amdefault>. The script is executed for each dle. If <amdefault>yes</amdefault>, the script is executed one time only.</para>
2805 <term><amkeyword>execute-where</amkeyword> [ <amkeyword>client</amkeyword> | <amkeyword>server</amkeyword> ]</term>
2807 <para>Default: <amkeyword>client</amkeyword>. Where the script must be executed, on the client or server.</para>
2811 <term><amkeyword>execute-on</amkeyword> <amtype>execute_on</amtype> [,<amtype>execute_on</amtype>]*</term>
2813 <para>No default. When the script must be executed, you can specify many of them:</para>
2815 <variablelist remap='TP'>
2817 <term><amkeyword>pre-amcheck</amkeyword></term>
2819 <para>Execute before the amcheck command for all dle. Can only be run on server.</para>
2823 <term><amkeyword>pre-dle-amcheck</amkeyword></term>
2825 <para>Execute before the amcheck command for the dle.</para>
2829 <term><amkeyword>pre-host-amcheck</amkeyword></term>
2831 <para>Execute before the amcheck command for all dle for the client.</para>
2835 <term><amkeyword>post-amcheck</amkeyword></term>
2837 <para>Execute after the amcheck command for all dle. Can only be run on server.</para>
2841 <term><amkeyword>post-dle-amcheck</amkeyword></term>
2843 <para>Execute after the amcheck command for the dle.</para>
2847 <term><amkeyword>post-host-amcheck</amkeyword></term>
2849 <para>Execute after the amcheck command for all dle for the client.</para>
2853 <term><amkeyword>pre-estimate</amkeyword></term>
2855 <para>Execute before the estimate command for all dle. Can only be run on server.</para>
2859 <term><amkeyword>pre-dle-estimate</amkeyword></term>
2861 <para>Execute before the estimate command for the dle.</para>
2865 <term><amkeyword>pre-host-estimate</amkeyword></term>
2867 <para>Execute before the estimate command for all dle for the client.</para>
2871 <term><amkeyword>post-estimate</amkeyword></term>
2873 <para>Execute after the estimate command for all dle. Can only be run on server.</para>
2877 <term><amkeyword>post-dle-estimate</amkeyword></term>
2879 <para>Execute after the estimate command for the dle.</para>
2883 <term><amkeyword>post-host-estimate</amkeyword></term>
2885 <para>Execute after the estimate command for all dle for the client.</para>
2889 <term><amkeyword>pre-backup</amkeyword></term>
2891 <para>Execute before the backup command for all dle. Can only be run on server.</para>
2895 <term><amkeyword>pre-dle-backup</amkeyword></term>
2897 <para>Execute before the backup command for the dle.</para>
2901 <term><amkeyword>pre-host-backup</amkeyword></term>
2903 <para>Execute before the backup command for all dle for the client. It can't be run on client, it must be run on server</para>
2907 <term><amkeyword>post-backup</amkeyword></term>
2909 <para>Execute after the backup command for all dle. Can only be run on server.</para>
2913 <term><amkeyword>post-dle-backup</amkeyword></term>
2915 <para>Execute after the backup command for the dle.</para>
2919 <term><amkeyword>post-host-backup</amkeyword></term>
2921 <para>Execute after the backup command for all dle for the client. It can't be run on client, it must be run on server</para>
2925 <term><amkeyword>pre-recover</amkeyword></term>
2927 <para>Execute before any level is recovered.</para>
2931 <term><amkeyword>post-recover</amkeyword></term>
2933 <para>Execute after all levels are recovered.</para>
2937 <term><amkeyword>pre-level-recover</amkeyword></term>
2939 <para>Execute before each level recovery.</para>
2943 <term><amkeyword>post-level-recover</amkeyword></term>
2945 <para>Execute after each level recovery.</para>
2949 <term><amkeyword>inter-level-recover</amkeyword></term>
2951 <para>Execute between two levels of recovery.</para>
2955 <para>If you recover level 0 and 2 of the disk /usr with amrecover, it will execute:</para>
2957 script --pre-recover
2958 script --pre-level-recover --level 0
2960 script --post-level-recover --level 0
2961 script --inter-level-recover --level 0 --level 2
2962 script --pre-level-recover --level 2
2964 script --post-level-recover --level 2
2965 script --post-recover
2970 <term><amkeyword>property</amkeyword> [<amkeyword>append</amkeyword>] [<amkeyword>priority</amkeyword>] <amtype>string</amtype> <amtype>string</amtype>+</term>
2972 <para>No default. You can set property for the script, each script have a different set of property. Both strings are quoted; the first string contains the name of
2973 the property to set, and the others contains its values.
2974 <amkeyword>append</amkeyword> keyword append the values to the list of values for that property.
2975 <amkeyword>priority</amkeyword> keyword disallow the setting of that property on the client.
2982 <refsect1><title>DEVICE SECTION</title>
2983 <para>Backend storage devices are specified in
2984 <emphasis remap='B'>amanda.conf</emphasis>
2985 in the form of "device" sections, which look like this:</para>
2988 define device <emphasis remap='I'>name</emphasis> {
2989 commend "<emphasis remap='I'>comment (optional)</emphasis>"
2990 tapedev "<emphasis remap='I'>device-specifier</emphasis>"
2991 device-property "<emphasis remap='I'>prop-name</emphasis>" "<emphasis remap='I'>prop-value</emphasis>"
2992 <literal>...</literal>
2996 <para>The { must appear at the end of a line, and the } on its own line.</para>
2997 <para><emphasis remap='I'>name</emphasis> is the user-specified name of
2998 this device. It is referenced from the global <emphasis
2999 remap='I'>tapedev</emphasis> parameter. The <emphasis
3000 remap='I'>device-specifier</emphasis> specifies the device name to use;
3001 see <manref name="amanda-devices" vol="7"/>.
3002 As with most sections, the <emphasis remap='I'>comment</emphasis>
3003 parmeter is optional and only for the user's convenience.</para>
3005 <para>An arbitrary number of <emphasis
3006 remap='I'>device-property</emphasis> parameters can be specified.
3008 <manref name="amanda-devices" vol="7"/>
3009 for information on device properties.</para>
3013 <refsect1><title>CHANGER SECTION</title>
3014 <para>Changers are described in
3015 <emphasis remap='B'>amanda.conf</emphasis>
3016 in the form of "changer" sections, which look like this:</para>
3019 define changer <emphasis remap='I'>name</emphasis> {
3020 comment "<emphasis remap='I'>comment (optional)</emphasis>"
3021 tpchanger "<emphasis remap='I'>changer-spec</emphasis>"
3022 changerdev "<emphasis remap='I'>device-name</emphasis>"
3023 changerfile "<emphasis remap='I'>state-file</emphasis>"
3024 <literal>...</literal>
3028 <para>The { must appear at the end of a line, and the } on its own line.</para>
3029 <para><emphasis remap='I'>name</emphasis> is the user-specified name of this
3030 device. The remaining parameters are specific to the changer type selected.
3033 <para>See <manref name="amanda-changers" vol="7" /> for more information on configuring changers.</para>
3037 <refsect1><title>INTERACTIVITY SECTION</title>
3039 <emphasis remap='B'>amanda.conf</emphasis>
3040 file may define multiple interactivyt methods, although only one will be used - that specified
3041 by the <amkeyword>interactivity</amkeyword> parameter.
3042 The information is entered in a <amkeyword>interactivity</amkeyword>
3043 section, which looks like this:</para>
3046 define interactivity <emphasis remap='I'>name</emphasis> {
3047 <emphasis remap='I'>interactivity-option</emphasis> <emphasis remap='I'>interactivity-value</emphasis>
3048 <literal>...</literal>
3052 <para>The { must appear at the end of a line, and the } on its own line.</para>
3053 <para><emphasis remap='I'>name</emphasis> is the user-specified name of this
3054 interactivity. The remaining parameters are specific to the interactivity type
3058 <para>The interactivity options and values are:</para>
3059 <variablelist remap='TP'>
3061 <term><amkeyword>comment</amkeyword> <amtype>string</amtype></term>
3063 <para>Default: not set.
3064 A comment string describing this interactivity.</para>
3068 <term><amkeyword>plugin</amkeyword> <amtype>string</amtype></term>
3070 <para>No default. Must be set to the name of the interactivity module, as described in <manref name="amanda-interactivity" vol="7" />.</para>
3074 <term><amkeyword>property</amkeyword> [<amkeyword>append</amkeyword>] <amtype>string</amtype> <amtype>string</amtype>+</term>
3076 <para>No default. You can set arbitrary properties for the interactivity. Each interactivity module has a different set of properties. The first string contains the name of
3077 the property to set, and the others contains its values. All strings should be quoted.
3078 The <amkeyword>append</amkeyword> keyword appends the given values to an existing list of values for that property.
3084 <para>See <manref name="amanda-interactivity" vol="7" /> for more information on configuring interactivity methods.</para>
3088 <refsect1><title>TAPERSCAN SECTION</title>
3090 <emphasis remap='B'>amanda.conf</emphasis>
3091 file may define multiple taperscan methods, although only one will be used - that specified
3092 by the <amkeyword>taperscan</amkeyword> parameter.
3093 The information is entered in a <amkeyword>taperscan</amkeyword>
3094 section, which looks like this:</para>
3097 define taperscan <emphasis remap='I'>name</emphasis> {
3098 <emphasis remap='I'>taperscan-option</emphasis> <emphasis remap='I'>taperscan-value</emphasis>
3099 <literal>...</literal>
3103 <para>The { must appear at the end of a line, and the } on its own line.</para>
3104 <para><emphasis remap='I'>name</emphasis> is the user-specified name of this
3105 taperscan. The remaining parameters are specific to the taperscan type
3109 <para>The taperscan options and values are:</para>
3110 <variablelist remap='TP'>
3112 <term><amkeyword>comment</amkeyword> <amtype>string</amtype></term>
3114 <para>Default: not set.
3115 A comment string describing this taperscan.</para>
3119 <term><amkeyword>plugin</amkeyword> <amtype>string</amtype></term>
3121 <para>No default. Must be set to the name of the taperscan module. See <manref name="amanda-taperscan" vol="7" /> for a list of defined taperscan modules.</para>
3125 <term><amkeyword>property</amkeyword> [<amkeyword>append</amkeyword>] <amtype>string</amtype> <amtype>string</amtype>+</term>
3127 <para>No default. Operates just like properties for interactivity methods, above.</para>
3132 <para>See <manref name="amanda-taperscan" vol="7" /> for more information on configuring taperscan.</para>
3136 <refsect1><title>Dump Splitting Configuration</title>
3138 <para>Amanda can "split" dumps into parts while writing them to storage
3139 media. This allows Amanda to recover gracefully from a failure while
3140 writing a part to a volume, by simply selecting a new volume and
3141 re-writing the dump from the beginning of the failed part. Parts also
3142 allow Amanda to seek directly to the required data, although this
3143 functionality is not yet used.</para>
3145 <para>In order to support re-writing from the beginning of a failed part,
3146 Amanda must have access to the contents of the part after it has been
3147 partially written. If the dump is being read from holding disk, then
3148 the part contents are availble there. Otherwise, the part must be
3149 cached, and this can be done memory or on disk. In either of the
3150 latter cases, the cache must have enough space to hold an entire
3153 <para>Because it is common for a single Amanda configuration to use both
3154 holding-disk (FILE-WRITE) and direct (known as PORT-WRITE) dumps, Amanda
3155 allows the configuration of different split sizes for the two cases. This
3156 allows, for example, for a part size appropriate to large tapes when
3157 performing FILE-WRITE dumps, with a part size limited by available disk
3158 or memory when performing PORT-WRITE dumps.</para>
3160 <para>Selecting a proper split size is a delicate matter. If the parts are
3161 too large, substantial storage space may be wasted in failed parts. If
3162 too small, large dumps will be split into innumerable tiny dumpfiles,
3163 adding to restoration complexity; furthermore, an excess of filemarks
3164 will cause slower tape drive operation and reduce the usable space on
3165 tape. A good rule of thumb is 1/10 of the size of a volume of storage
3168 <para>In versions of Amanda through 3.1.*, splitting was controlled by the
3169 dumptype parameters <amkeyword>tape-splitsize</amkeyword>,
3170 <amkeyword>split-diskbuffer</amkeyword>, and
3171 <amkeyword>fallback-splitsize</amkeyword>. These keywords had
3172 confusing and non-intuitive interactions, and have since been
3175 <para>If the deprecated keywords are not present, subsequent versions
3176 of Amanda use the dumptype parameter
3177 <amkeyword>allow-split</amkeyword> to control whether a DLE can be
3178 split, and the <emphasis>tapetype</emphasis> parameters
3179 <amkeyword>part-size</amkeyword>,
3180 <amkeyword>part-cache-type</amkeyword>,
3181 <amkeyword>part-cache-dir</amkeyword>, and
3182 <amkeyword>part-cache-max-size</amkeyword>. The
3183 <amkeyword>part-size</amkeyword> specifies the "normal" part size,
3184 while the <amkeyword>part-cache-*</amkeyword> parameters describe
3185 how to behave when caching is required (on PORT-WRITE). Full
3186 details on these parameters are given above.</para>
3191 <manref name="amanda-client.conf" vol="5"/>,
3192 <manref name="amanda-applications" vol="7"/>,
3193 <manref name="amanda-auth" vol="7"/>,
3194 <manref name="amanda-changers" vol="7"/>,
3195 <manref name="amanda-devices" vol="7"/>,
3196 <manref name="amanda-interactivity" vol="7"/>,
3197 <manref name="amanda-scripts" vol="7"/>
3198 <manref name="amanda-taperscan" vol="7"/>