4 \chapter{Configuring the Director}
5 \label{DirectorChapter}
6 \index[general]{Director!Configuring the}
7 \index[general]{Configuring the Director}
9 Of all the configuration files needed to run {\bf Bacula}, the Director's is
10 the most complicated, and the one that you will need to modify the most often
11 as you add clients or modify the FileSets.
13 For a general discussion of configuration files and resources including the
14 data types recognized by {\bf Bacula}. Please see the
15 \ilink{Configuration}{ConfigureChapter} chapter of this manual.
17 \section{Director Resource Types}
18 \index[general]{Types!Director Resource}
19 \index[general]{Director Resource Types}
21 Director resource type may be one of the following:
23 Job, JobDefs, Client, Storage, Catalog, Schedule, FileSet, Pool, Director, or
24 Messages. We present them here in the most logical order for defining them:
26 Note, everything revolves around a job and is tied to a job in one
31 \ilink{Director}{DirectorResource4} -- to define the Director's
32 name and its access password used for authenticating the Console program.
33 Only a single Director resource definition may appear in the Director's
34 configuration file. If you have either {\bf /dev/random} or {\bf bc} on your
35 machine, Bacula will generate a random password during the configuration
36 process, otherwise it will be left blank.
38 \ilink{Job}{JobResource} -- to define the backup/restore Jobs
39 and to tie together the Client, FileSet and Schedule resources to be used
40 for each Job. Normally, you will Jobs of different names corresponding
41 to each client (i.e. one Job per client, but a different one with a different name
44 \ilink{JobDefs}{JobDefsResource} -- optional resource for
45 providing defaults for Job resources.
47 \ilink{Schedule}{ScheduleResource} -- to define when a Job is to
48 be automatically run by {\bf Bacula's} internal scheduler. You
49 may have any number of Schedules, but each job will reference only
52 \ilink{FileSet}{FileSetResource} -- to define the set of files
53 to be backed up for each Client. You may have any number of
54 FileSets but each Job will reference only one.
56 \ilink{Client}{ClientResource2} -- to define what Client is to be
57 backed up. You will generally have multiple Client definitions. Each
58 Job will reference only a single client.
60 \ilink{Storage}{StorageResource2} -- to define on what physical
61 device the Volumes should be mounted. You may have one or
62 more Storage definitions.
64 \ilink{Pool}{PoolResource} -- to define the pool of Volumes
65 that can be used for a particular Job. Most people use a
66 single default Pool. However, if you have a large number
67 of clients or volumes, you may want to have multiple Pools.
68 Pools allow you to restrict a Job (or a Client) to use
69 only a particular set of Volumes.
71 \ilink{Catalog}{CatalogResource} -- to define in what database to
72 keep the list of files and the Volume names where they are backed up.
73 Most people only use a single catalog. However, if you want to
74 scale the Director to many clients, multiple catalogs can be helpful.
75 Multiple catalogs require a bit more management because in general
76 you must know what catalog contains what data. Currently, all
77 Pools are defined in each catalog. This restriction will be removed
80 \ilink{Messages}{MessagesChapter} -- to define where error and
81 information messages are to be sent or logged. You may define
82 multiple different message resources and hence direct particular
83 classes of messages to different users or locations (files, ...).
86 \section{The Director Resource}
87 \label{DirectorResource4}
88 \index[general]{Director Resource}
89 \index[general]{Resource!Director}
91 The Director resource defines the attributes of the Directors running on the
92 network. In the current implementation, there is only a single Director
93 resource, but the final design will contain multiple Directors to maintain
94 index and media database redundancy.
100 Start of the Director resource. One and only one director resource must be
103 \item [Name = \lt{}name\gt{}]
105 \index[dir]{Directive!Name}
106 The director name used by the system administrator. This directive is
109 \item [Description = \lt{}text\gt{}]
110 \index[dir]{Description}
111 \index[dir]{Directive!Description}
112 The text field contains a description of the Director that will be displayed
113 in the graphical user interface. This directive is optional.
115 \item [Password = \lt{}UA-password\gt{}]
116 \index[dir]{Password}
117 \index[dir]{Directive!Password}
118 Specifies the password that must be supplied for the default Bacula
119 Console to be authorized. The same password must appear in the {\bf
120 Director} resource of the Console configuration file. For added
121 security, the password is never passed across the network but instead a
122 challenge response hash code created with the password. This directive
123 is required. If you have either {\bf /dev/random} or {\bf bc} on your
124 machine, Bacula will generate a random password during the configuration
125 process, otherwise it will be left blank and you must manually supply
128 The password is plain text. It is not generated through any special
129 process but as noted above, it is better to use random text for
132 \item [Messages = \lt{}Messages-resource-name\gt{}]
133 \index[dir]{Messages}
134 \index[dir]{Directive!Messages}
135 The messages resource specifies where to deliver Director messages that are
136 not associated with a specific Job. Most messages are specific to a job and
137 will be directed to the Messages resource specified by the job. However,
138 there are a few messages that can occur when no job is running. This
139 directive is required.
141 \item [Working Directory = \lt{}Directory\gt{}]
142 \index[dir]{Working Directory}
143 \index[dir]{Directive!Working Directory}
144 This directive is mandatory and specifies a directory in which the Director
145 may put its status files. This directory should be used only by Bacula but
146 may be shared by other Bacula daemons. However, please note, if this
147 directory is shared with other Bacula daemons (the File daemon and Storage
148 daemon), you must ensure that the {\bf Name} given to each daemon is
149 unique so that the temporary filenames used do not collide. By default
150 the Bacula configure process creates unique daemon names by postfixing them
151 with -dir, -fd, and -sd. Standard shell expansion of the {\bf
152 Directory} is done when the configuration file is read so that values such
153 as {\bf \$HOME} will be properly expanded. This directive is required.
154 The working directory specified must already exist and be
155 readable and writable by the Bacula daemon referencing it.
157 If you have specified a Director user and/or a Director group on your
158 ./configure line with {\bf {-}{-}with-dir-user} and/or
159 {\bf {-}{-}with-dir-group} the Working Directory owner and group will
160 be set to those values.
162 \item [Pid Directory = \lt{}Directory\gt{}]
163 \index[dir]{Pid Directory}
164 \index[dir]{Directive!Pid Directory}
165 This directive is mandatory and specifies a directory in which the Director
166 may put its process Id file. The process Id file is used to shutdown
167 Bacula and to prevent multiple copies of Bacula from running simultaneously.
168 Standard shell expansion of the {\bf Directory} is done when the
169 configuration file is read so that values such as {\bf \$HOME} will be
172 The PID directory specified must already exist and be
173 readable and writable by the Bacula daemon referencing it
175 Typically on Linux systems, you will set this to: {\bf /var/run}. If you are
176 not installing Bacula in the system directories, you can use the {\bf Working
177 Directory} as defined above. This directive is required.
179 \item [Scripts Directory = \lt{}Directory\gt{}]
180 \index[dir]{Scripts Directory}
181 \index[dir]{Directive!Scripts Directory}
182 This directive is optional and, if defined, specifies a directory in
183 which the Director will look for the Python startup script {\bf
184 DirStartup.py}. This directory may be shared by other Bacula daemons.
185 Standard shell expansion of the directory is done when the configuration
186 file is read so that values such as {\bf \$HOME} will be properly
189 \item [QueryFile = \lt{}Path\gt{}]
190 \index[dir]{QueryFile}
191 \index[dir]{Directive!QueryFile}
192 This directive is mandatory and specifies a directory and file in which
193 the Director can find the canned SQL statements for the {\bf Query}
194 command of the Console. Standard shell expansion of the {\bf Path} is
195 done when the configuration file is read so that values such as {\bf
196 \$HOME} will be properly expanded. This directive is required.
198 \item [Heartbeat Interval = \lt{}time-interval\gt{}]
199 \index[dir]{Heartbeat Interval}
200 \index[dir]{Directive!Heartbeat}
201 This directive is optional and if specified will cause the Director to
202 set a keepalive interval (heartbeat) in seconds on each of the sockets
203 it opens for the Client resource. This value will override any
204 specified at the Director level. It is implemented only on systems
205 (Linux, ...) that provide the {\bf setsockopt} TCP\_KEEPIDLE function.
206 The default value is zero, which means no change is made to the socket.
209 \label{DirMaxConJobs}
210 \item [Maximum Concurrent Jobs = \lt{}number\gt{}]
211 \index[dir]{Maximum Concurrent Jobs}
212 \index[dir]{Directive!Maximum Concurrent Jobs}
213 \index[general]{Simultaneous Jobs}
214 \index[general]{Concurrent Jobs}
215 where \lt{}number\gt{} is the maximum number of total Director Jobs that
216 should run concurrently. The default is set to 1, but you may set it to a
219 The Volume format becomes more complicated with
220 multiple simultaneous jobs, consequently, restores may take longer if
221 Bacula must sort through interleaved volume blocks from multiple simultaneous
222 jobs. This can be avoided by having each simultaneous job write to
223 a different volume or by using data spooling, which will first spool the data
224 to disk simultaneously, then write one spool file at a time to the volume
225 thus avoiding excessive interleaving of the different job blocks.
227 \item [FD Connect Timeout = \lt{}time\gt{}]
228 \index[dir]{FD Connect Timeout}
229 \index[dir]{Directive!FD Connect Timeout}
230 where {\bf time} is the time that the Director should continue
231 attempting to contact the File daemon to start a job, and after which
232 the Director will cancel the job. The default is 30 minutes.
234 \item [SD Connect Timeout = \lt{}time\gt{}]
235 \index[dir]{SD Connect Timeout}
236 \index[dir]{Directive!SD Connect Timeout}
237 where {\bf time} is the time that the Director should continue
238 attempting to contact the Storage daemon to start a job, and after which
239 the Director will cancel the job. The default is 30 minutes.
241 \item [DirAddresses = \lt{}IP-address-specification\gt{}]
242 \index[dir]{DirAddresses}
244 \index[general]{Address}
245 \index[dir]{Directive!DirAddresses}
246 Specify the ports and addresses on which the Director daemon will listen
247 for Bacula Console connections. Probably the simplest way to explain
248 this is to show an example:
253 ip = { addr = 1.2.3.4; port = 1205;}
255 addr = 1.2.3.4; port = http;}
264 ip = { addr = 1.2.3.4 }
265 ip = { addr = 201:220:222::2 }
267 addr = bluedot.thun.net
273 where ip, ip4, ip6, addr, and port are all keywords. Note, that the address
274 can be specified as either a dotted quadruple, or IPv6 colon notation, or as
275 a symbolic name (only in the ip specification). Also, port can be specified
276 as a number or as the mnemonic value from the /etc/services file. If a port
277 is not specified, the default will be used. If an ip section is specified,
278 the resolution can be made either by IPv4 or IPv6. If ip4 is specified, then
279 only IPv4 resolutions will be permitted, and likewise with ip6.
281 Please note that if you use the DirAddresses directive, you must
282 not use either a DirPort or a DirAddress directive in the same
285 \item [DirPort = \lt{}port-number\gt{}]
287 \index[dir]{Directive!DirPort}
288 Specify the port (a positive integer) on which the Director daemon will
289 listen for Bacula Console connections. This same port number must be
290 specified in the Director resource of the Console configuration file. The
291 default is 9101, so normally this directive need not be specified. This
292 directive should not be used if you specify DirAddresses (N.B plural)
295 \item [DirAddress = \lt{}IP-Address\gt{}]
296 \index[dir]{DirAddress}
297 \index[dir]{Directive!DirAddress}
298 This directive is optional, but if it is specified, it will cause the
299 Director server (for the Console program) to bind to the specified {\bf
300 IP-Address}, which is either a domain name or an IP address specified as a
301 dotted quadruple in string or quoted string format. If this directive is
302 not specified, the Director will bind to any available address (the
303 default). Note, unlike the DirAddresses specification noted above, this
304 directive only permits a single address to be specified. This directive
305 should not be used if you specify a DirAddresses (N.B. plural) directive.
307 \item [DirSourceAddress = \lt{}IP-Address\gt{}]
308 \index[fd]{DirSourceAddress}
309 \index[fd]{Directive!DirSourceAddress}
310 This record is optional, and if it is specified, it will cause the Director
311 server (when initiating connections to a storage or file daemon) to source
312 its connections from the specified address. Only a single IP address may be
313 specified. If this record is not specified, the Director server will source
314 its outgoing connections according to the system routing table (the default).
316 \item[Statistics Retention = \lt{}time\gt{}]
317 \index[dir]{StatisticsRetention}
318 \index[dir]{Directive!StatisticsRetention}
319 \label{PruneStatistics}
321 The \texttt{Statistics Retention} directive defines the length of time that
322 Bacula will keep statistics job records in the Catalog database after the
323 Job End time. (In \texttt{JobHistory} table) When this time period expires,
324 and if user runs \texttt{prune stats} command, Bacula will prune (remove)
325 Job records that are older than the specified period.
327 Theses statistics records aren't use for restore purpose, but mainly for
328 capacity planning, billings, etc. See \ilink{Statistics chapter} for
329 additional information.
331 See the \ilink{ Configuration chapter}{Time} of this manual for additional
332 details of time specification.
334 The default is 5 years.
336 \item[VerId = \lt{}string\gt{}]
337 \index[dir]{Directive!VerId}
338 where \lt{}string\gt{} is an identifier which can be used for support purpose.
339 This string is displayed using the \texttt{version} command.
341 \item[MaximumConsoleConnections = \lt{}number\gt{}]
342 \index[dir]{MaximumConsoleConnections}
343 \index[dir]{Directive!MaximumConsoleConnections}
345 where \lt{}number\gt{} is the maximum number of Console Connections that
346 could run concurrently. The default is set to 20, but you may set it to a
349 \item[SharedStorage = \lt{}yes\vb{}no\gt{}]
350 \index[dir]{SharedStorage}
351 \index[dir]{Directive!SharedStorage}
353 The \texttt{Shared Storage} directive is a Bacula Enterprise feature that
354 allows you to share volumes between different Storage resources. This
355 directive should be used \textbf{only} if all \texttt{Media Type} are
356 correctly set across all Devices.
358 The \texttt{Shared Storage} directive should be used when using the SAN
359 Shared Storage plugin or when accessing from the Director Storage resources
360 directly to Devices of an Autochanger.
362 When sharing volumes between different Storage resources, you will
363 need also to use the \texttt{reset-storageid} script before using the
364 \texttt{update slots} command. This script can be scheduled once a day in
368 $ /opt/bacula/scripts/reset-storageid MediaType StorageName
370 * update slots storage=StorageName drive=0
373 Please contact Bacula Systems support to get help on this advanced
378 The following is an example of a valid Director resource definition:
384 WorkingDirectory = "$HOME/bacula/bin/working"
385 Password = UA_password
386 PidDirectory = "$HOME/bacula/bin/working"
387 QueryFile = "$HOME/bacula/bin/query.sql"
393 \section{The Job Resource}
395 \index[general]{Resource!Job}
396 \index[general]{Job Resource}
398 The Job resource defines a Job (Backup, Restore, ...) that Bacula must
399 perform. Each Job resource definition contains the name of a Client and
400 a FileSet to backup, the Schedule for the Job, where the data
401 are to be stored, and what media Pool can be used. In effect, each Job
402 resource must specify What, Where, How, and When or FileSet, Storage,
403 Backup/Restore/Level, and Schedule respectively. Note, the FileSet must
404 be specified for a restore job for historical reasons, but it is no longer used.
406 Only a single type ({\bf Backup}, {\bf Restore}, ...) can be specified for any
407 job. If you want to backup multiple FileSets on the same Client or multiple
408 Clients, you must define a Job for each one.
410 Note, you define only a single Job to do the Full, Differential, and
411 Incremental backups since the different backup levels are tied together by
412 a unique Job name. Normally, you will have only one Job per Client, but
413 if a client has a really huge number of files (more than several million),
414 you might want to split it into to Jobs each with a different FileSet
415 covering only part of the total files.
417 Multiple Storage daemons are not currently supported for Jobs, so if
418 you do want to use multiple storage daemons, you will need to create
419 a different Job and ensure that for each Job that the combination of
420 Client and FileSet are unique. The Client and FileSet are what Bacula
421 uses to restore a client, so if there are multiple Jobs with the same
422 Client and FileSet or multiple Storage daemons that are used, the
423 restore will not work. This problem can be resolved by defining multiple
424 FileSet definitions (the names must be different, but the contents of
425 the FileSets may be the same).
432 \index[dir]{Directive!Job}
433 Start of the Job resource. At least one Job resource is required.
435 \item [Name = \lt{}name\gt{}]
437 \index[dir]{Directive!Name}
438 The Job name. This name can be specified on the {\bf Run} command in the
439 console program to start a job. If the name contains spaces, it must be
440 specified between quotes. It is generally a good idea to give your job the
441 same name as the Client that it will backup. This permits easy
442 identification of jobs.
444 When the job actually runs, the unique Job Name will consist of the name you
445 specify here followed by the date and time the job was scheduled for
446 execution. This directive is required.
448 \item [Enabled = \lt{}yes\vb{}no\gt{}]
450 \index[dir]{Directive!Enable}
451 This directive allows you to enable or disable automatic execution
452 via the scheduler of a Job.
454 \item [Type = \lt{}job-type\gt{}]
456 \index[dir]{Directive!Type}
457 The {\bf Type} directive specifies the Job type, which may be one of the
458 following: {\bf Backup}, {\bf Restore}, {\bf Verify}, or {\bf Admin}. This
459 directive is required. Within a particular Job Type, there are also Levels
460 as discussed in the next item.
466 Run a backup Job. Normally you will have at least one Backup job for each
467 client you want to save. Normally, unless you turn off cataloging, most all
468 the important statistics and data concerning files backed up will be placed
473 Run a restore Job. Normally, you will specify only one Restore job
474 which acts as a sort of prototype that you will modify using the console
475 program in order to perform restores. Although certain basic
476 information from a Restore job is saved in the catalog, it is very
477 minimal compared to the information stored for a Backup job -- for
478 example, no File database entries are generated since no Files are
481 {\bf Restore} jobs cannot be
482 automatically started by the scheduler as is the case for Backup, Verify
483 and Admin jobs. To restore files, you must use the {\bf restore} command
489 Run a verify Job. In general, {\bf verify} jobs permit you to compare the
490 contents of the catalog to the file system, or to what was backed up. In
491 addition, to verifying that a tape that was written can be read, you can
492 also use {\bf verify} as a sort of tripwire intrusion detection.
496 Run an admin Job. An {\bf Admin} job can be used to periodically run catalog
497 pruning, if you do not want to do it at the end of each {\bf Backup} Job.
498 Although an Admin job is recorded in the catalog, very little data is saved.
503 \item [Level = \lt{}job-level\gt{}]
505 \index[dir]{Directive!Level}
506 The Level directive specifies the default Job level to be run. Each
507 different Job Type (Backup, Restore, ...) has a different set of Levels
508 that can be specified. The Level is normally overridden by a different
509 value that is specified in the {\bf Schedule} resource. This directive
510 is not required, but must be specified either by a {\bf Level} directive
511 or as an override specified in the {\bf Schedule} resource.
513 For a {\bf Backup} Job, the Level may be one of the following:
519 When the Level is set to Full all files in the FileSet whether or not
520 they have changed will be backed up.
523 \index[dir]{Incremental}
524 When the Level is set to Incremental all files specified in the FileSet
525 that have changed since the last successful backup of the the same Job
526 using the same FileSet and Client, will be backed up. If the Director
527 cannot find a previous valid Full backup then the job will be upgraded
528 into a Full backup. When the Director looks for a valid backup record
529 in the catalog database, it looks for a previous Job with:
532 \item The same Job name.
533 \item The same Client name.
534 \item The same FileSet (any change to the definition of the FileSet such as
535 adding or deleting a file in the Include or Exclude sections constitutes a
537 \item The Job was a Full, Differential, or Incremental backup.
538 \item The Job terminated normally (i.e. did not fail or was not canceled).
539 \item The Job started no longer ago than {\bf Max Full Interval}.
542 If all the above conditions do not hold, the Director will upgrade the
543 Incremental to a Full save. Otherwise, the Incremental backup will be
544 performed as requested.
546 The File daemon (Client) decides which files to backup for an
547 Incremental backup by comparing start time of the prior Job (Full,
548 Differential, or Incremental) against the time each file was last
549 "modified" (st\_mtime) and the time its attributes were last
550 "changed"(st\_ctime). If the file was modified or its attributes
551 changed on or after this start time, it will then be backed up.
553 Some virus scanning software may change st\_ctime while
554 doing the scan. For example, if the virus scanning program attempts to
555 reset the access time (st\_atime), which Bacula does not use, it will
556 cause st\_ctime to change and hence Bacula will backup the file during
557 an Incremental or Differential backup. In the case of Sophos virus
558 scanning, you can prevent it from resetting the access time (st\_atime)
559 and hence changing st\_ctime by using the {\bf \verb:--:no-reset-atime}
560 option. For other software, please see their manual.
562 When Bacula does an Incremental backup, all modified files that are
563 still on the system are backed up. However, any file that has been
564 deleted since the last Full backup remains in the Bacula catalog,
565 which means that if between a Full save and the time you do a
566 restore, some files are deleted, those deleted files will also be
567 restored. The deleted files will no longer appear in the catalog
568 after doing another Full save.
570 In addition, if you move a directory rather than copy it, the files in
571 it do not have their modification time (st\_mtime) or their attribute
572 change time (st\_ctime) changed. As a consequence, those files will
573 probably not be backed up by an Incremental or Differential backup which
574 depend solely on these time stamps. If you move a directory, and wish
575 it to be properly backed up, it is generally preferable to copy it, then
578 However, to manage deleted files or directories changes in the
579 catalog during an Incremental backup you can use \texttt{accurate}
580 mode. This is quite memory consuming process. See \ilink{Accurate
581 mode}{accuratemode} for more details.
584 \index[dir]{Differential}
585 When the Level is set to Differential
586 all files specified in the FileSet that have changed since the last
587 successful Full backup of the same Job will be backed up.
588 If the Director cannot find a
589 valid previous Full backup for the same Job, FileSet, and Client,
590 backup, then the Differential job will be upgraded into a Full backup.
591 When the Director looks for a valid Full backup record in the catalog
592 database, it looks for a previous Job with:
595 \item The same Job name.
596 \item The same Client name.
597 \item The same FileSet (any change to the definition of the FileSet such as
598 adding or deleting a file in the Include or Exclude sections constitutes a
600 \item The Job was a FULL backup.
601 \item The Job terminated normally (i.e. did not fail or was not canceled).
602 \item The Job started no longer ago than {\bf Max Full Interval}.
605 If all the above conditions do not hold, the Director will upgrade the
606 Differential to a Full save. Otherwise, the Differential backup will be
607 performed as requested.
609 The File daemon (Client) decides which files to backup for a
610 differential backup by comparing the start time of the prior Full backup
611 Job against the time each file was last "modified" (st\_mtime) and the
612 time its attributes were last "changed" (st\_ctime). If the file was
613 modified or its attributes were changed on or after this start time, it
614 will then be backed up. The start time used is displayed after the {\bf
615 Since} on the Job report. In rare cases, using the start time of the
616 prior backup may cause some files to be backed up twice, but it ensures
617 that no change is missed. As with the Incremental option, you should
618 ensure that the clocks on your server and client are synchronized or as
619 close as possible to avoid the possibility of a file being skipped.
620 Note, on versions 1.33 or greater Bacula automatically makes the
621 necessary adjustments to the time between the server and the client so
622 that the times Bacula uses are synchronized.
624 When Bacula does a Differential backup, all modified files that are
625 still on the system are backed up. However, any file that has been
626 deleted since the last Full backup remains in the Bacula catalog, which
627 means that if between a Full save and the time you do a restore, some
628 files are deleted, those deleted files will also be restored. The
629 deleted files will no longer appear in the catalog after doing another
630 Full save. However, to remove deleted files from the catalog during a
631 Differential backup is quite a time consuming process and not currently
632 implemented in Bacula. It is, however, a planned future feature.
634 As noted above, if you move a directory rather than copy it, the
635 files in it do not have their modification time (st\_mtime) or
636 their attribute change time (st\_ctime) changed. As a
637 consequence, those files will probably not be backed up by an
638 Incremental or Differential backup which depend solely on these
639 time stamps. If you move a directory, and wish it to be
640 properly backed up, it is generally preferable to copy it, then
641 delete the original. Alternatively, you can move the directory, then
642 use the {\bf touch} program to update the timestamps.
644 %% TODO: merge this with incremental
645 However, to manage deleted files or directories changes in the
646 catalog during an Differential backup you can use \texttt{accurate}
647 mode. This is quite memory consuming process. See \ilink{Accurate
648 mode}{accuratemode} for more details.
650 Every once and a while, someone asks why we need Differential
651 backups as long as Incremental backups pickup all changed files.
652 There are possibly many answers to this question, but the one
653 that is the most important for me is that a Differential backup
655 all the Incremental and Differential backups since the last Full backup
656 into a single Differential backup. This has two effects: 1. It gives
657 some redundancy since the old backups could be used if the merged backup
658 cannot be read. 2. More importantly, it reduces the number of Volumes
659 that are needed to do a restore effectively eliminating the need to read
660 all the volumes on which the preceding Incremental and Differential
661 backups since the last Full are done.
665 For a {\bf Restore} Job, no level needs to be specified.
667 For a {\bf Verify} Job, the Level may be one of the following:
672 \index[dir]{InitCatalog}
673 does a scan of the specified {\bf FileSet} and stores the file
674 attributes in the Catalog database. Since no file data is saved, you
675 might ask why you would want to do this. It turns out to be a very
676 simple and easy way to have a {\bf Tripwire} like feature using {\bf
677 Bacula}. In other words, it allows you to save the state of a set of
678 files defined by the {\bf FileSet} and later check to see if those files
679 have been modified or deleted and if any new files have been added.
680 This can be used to detect system intrusion. Typically you would
681 specify a {\bf FileSet} that contains the set of system files that
682 should not change (e.g. /sbin, /boot, /lib, /bin, ...). Normally, you
683 run the {\bf InitCatalog} level verify one time when your system is
684 first setup, and then once again after each modification (upgrade) to
685 your system. Thereafter, when your want to check the state of your
686 system files, you use a {\bf Verify} {\bf level = Catalog}. This
687 compares the results of your {\bf InitCatalog} with the current state of
692 Compares the current state of the files against the state previously
693 saved during an {\bf InitCatalog}. Any discrepancies are reported. The
694 items reported are determined by the {\bf verify} options specified on
695 the {\bf Include} directive in the specified {\bf FileSet} (see the {\bf
696 FileSet} resource below for more details). Typically this command will
697 be run once a day (or night) to check for any changes to your system
700 Please note! If you run two Verify Catalog jobs on the same client at
701 the same time, the results will certainly be incorrect. This is because
702 Verify Catalog modifies the Catalog database while running in order to
705 \item [VolumeToCatalog]
706 \index[dir]{VolumeToCatalog}
707 This level causes Bacula to read the file attribute data written to the
708 Volume from the last backup Job for the job specified on the {\bf VerifyJob}
709 directive. The file attribute data are compared to the
710 values saved in the Catalog database and any differences are reported.
711 This is similar to the {\bf DiskToCatalog} level except that instead of
712 comparing the disk file attributes to the catalog database, the
713 attribute data written to the Volume is read and compared to the catalog
714 database. Although the attribute data including the signatures (MD5 or
715 SHA1) are compared, the actual file data is not compared (it is not in
718 Please note! If you run two Verify VolumeToCatalog jobs on the same
719 client at the same time, the results will certainly be incorrect. This
720 is because the Verify VolumeToCatalog modifies the Catalog database
723 \item [DiskToCatalog]
724 \index[dir]{DiskToCatalog}
725 This level causes Bacula to read the files as they currently are on
726 disk, and to compare the current file attributes with the attributes
727 saved in the catalog from the last backup for the job specified on the
728 {\bf VerifyJob} directive. This level differs from the {\bf VolumeToCatalog}
729 level described above by the fact that it doesn't compare against a
730 previous Verify job but against a previous backup. When you run this
731 level, you must supply the verify options on your Include statements.
732 Those options determine what attribute fields are compared.
734 This command can be very useful if you have disk problems because it
735 will compare the current state of your disk against the last successful
736 backup, which may be several jobs.
738 Note, the current implementation (1.32c) does not identify files that
742 \item [Accurate = \lt{}yes\vb{}no\gt{}]
743 \index[dir]{Accurate}
744 In accurate mode, the File daemon knowns exactly which files were present
745 after the last backup. So it is able to handle deleted or renamed files.
747 When restoring a FileSet for a specified date (including "most
748 recent"), Bacula is able to restore exactly the files and
749 directories that existed at the time of the last backup prior to
750 that date including ensuring that deleted files are actually deleted,
751 and renamed directories are restored properly.
753 In this mode, the File daemon must keep data concerning all files in
754 memory. So If you do not have sufficient memory, the backup may
755 either be terribly slow or fail.
757 %% $$ memory = \sum_{i=1}^{n}(strlen(path_i + file_i) + sizeof(CurFile))$$
759 For 500.000 files (a typical desktop linux system), it will require
760 approximately 64 Megabytes of RAM on your File daemon to hold the
761 required information.
763 \item [Verify Job = \lt{}Job-Resource-Name\gt{}]
764 \index[dir]{Verify Job}
765 \index[dir]{Directive!Verify Job}
766 If you run a verify job without this directive, the last job run will be
767 compared with the catalog, which means that you must immediately follow
768 a backup by a verify command. If you specify a {\bf Verify Job} Bacula
769 will find the last job with that name that ran. This permits you to run
770 all your backups, then run Verify jobs on those that you wish to be
771 verified (most often a {\bf VolumeToCatalog}) so that the tape just
774 \item [JobDefs = \lt{}JobDefs-Resource-Name\gt{}]
776 \index[dir]{Directive!JobDefs}
777 If a JobDefs-Resource-Name is specified, all the values contained in the
778 named JobDefs resource will be used as the defaults for the current Job.
779 Any value that you explicitly define in the current Job resource, will
780 override any defaults specified in the JobDefs resource. The use of
781 this directive permits writing much more compact Job resources where the
782 bulk of the directives are defined in one or more JobDefs. This is
783 particularly useful if you have many similar Jobs but with minor
784 variations such as different Clients. A simple example of the use of
785 JobDefs is provided in the default bacula-dir.conf file.
787 \item [Bootstrap = \lt{}bootstrap-file\gt{}]
788 \index[dir]{Bootstrap}
789 \index[dir]{Directive!Bootstrap}
790 The Bootstrap directive specifies a bootstrap file that, if provided,
791 will be used during {\bf Restore} Jobs and is ignored in other Job
792 types. The {\bf bootstrap} file contains the list of tapes to be used
793 in a restore Job as well as which files are to be restored.
794 Specification of this directive is optional, and if specified, it is
795 used only for a restore job. In addition, when running a Restore job
796 from the console, this value can be changed.
798 If you use the {\bf Restore} command in the Console program, to start a
799 restore job, the {\bf bootstrap} file will be created automatically from
800 the files you select to be restored.
802 For additional details of the {\bf bootstrap} file, please see
803 \ilink{Restoring Files with the Bootstrap File}{BootstrapChapter} chapter
806 \label{writebootstrap}
807 \item [Write Bootstrap = \lt{}bootstrap-file-specification\gt{}]
808 \index[dir]{Write Bootstrap}
809 \index[dir]{Directive!Write Bootstrap}
810 The {\bf writebootstrap} directive specifies a file name where Bacula
811 will write a {\bf bootstrap} file for each Backup job run. This
812 directive applies only to Backup Jobs. If the Backup job is a Full
813 save, Bacula will erase any current contents of the specified file
814 before writing the bootstrap records. If the Job is an Incremental
816 save, Bacula will append the current bootstrap record to the end of the
819 Using this feature, permits you to constantly have a bootstrap file that
820 can recover the current state of your system. Normally, the file
821 specified should be a mounted drive on another machine, so that if your
822 hard disk is lost, you will immediately have a bootstrap record
823 available. Alternatively, you should copy the bootstrap file to another
824 machine after it is updated. Note, it is a good idea to write a separate
825 bootstrap file for each Job backed up including the job that backs up
826 your catalog database.
828 If the {\bf bootstrap-file-specification} begins with a vertical bar
829 (\verb+|+), Bacula will use the specification as the name of a program to which
830 it will pipe the bootstrap record. It could for example be a shell
831 script that emails you the bootstrap record.
833 On versions 1.39.22 or greater, before opening the file or executing the
834 specified command, Bacula performs
835 \ilink{character substitution}{character substitution} like in RunScript
836 directive. To automatically manage your bootstrap files, you can use
837 this in your {\bf JobDefs} resources:
840 Write Bootstrap = "%c_%n.bsr"
845 For more details on using this file, please see the chapter entitled
846 \ilink{The Bootstrap File}{BootstrapChapter} of this manual.
848 \item [Client = \lt{}client-resource-name\gt{}]
850 \index[dir]{Directive!Client}
851 The Client directive specifies the Client (File daemon) that will be used in
852 the current Job. Only a single Client may be specified in any one Job. The
853 Client runs on the machine to be backed up, and sends the requested files to
854 the Storage daemon for backup, or receives them when restoring. For
855 additional details, see the
856 \ilink{Client Resource section}{ClientResource2} of this chapter.
857 This directive is required.
859 \item [FileSet = \lt{}FileSet-resource-name\gt{}]
861 \index[dir]{Directive!FileSet}
862 The FileSet directive specifies the FileSet that will be used in the
863 current Job. The FileSet specifies which directories (or files) are to
864 be backed up, and what options to use (e.g. compression, ...). Only a
865 single FileSet resource may be specified in any one Job. For additional
866 details, see the \ilink{FileSet Resource section}{FileSetResource} of
867 this chapter. This directive is required.
869 \item [Base = \lt{}job-resource-name, ...\gt{}]
871 \index[dir]{Directive!Base}
872 The Base directive permits to specify the list of jobs that will be used during
873 Full backup as base. This directive is optional. See the \ilink{Base Job
874 chapter}{basejobs} for more information.
876 \item [Messages = \lt{}messages-resource-name\gt{}]
877 \index[dir]{Messages}
878 \index[dir]{Directive!Messages}
879 The Messages directive defines what Messages resource should be used for
880 this job, and thus how and where the various messages are to be
881 delivered. For example, you can direct some messages to a log file, and
882 others can be sent by email. For additional details, see the
883 \ilink{Messages Resource}{MessagesChapter} Chapter of this manual. This
884 directive is required.
886 \item [Pool = \lt{}pool-resource-name\gt{}]
888 \index[dir]{Directive!Pool}
889 The Pool directive defines the pool of Volumes where your data can be
890 backed up. Many Bacula installations will use only the {\bf Default}
891 pool. However, if you want to specify a different set of Volumes for
892 different Clients or different Jobs, you will probably want to use
893 Pools. For additional details, see the \ilink{Pool Resource
894 section}{PoolResource} of this chapter. This directive is required.
896 \item [Full Backup Pool = \lt{}pool-resource-name\gt{}]
897 \index[dir]{Full Backup Pool}
898 \index[dir]{Directive!Full Backup Pool}
899 The {\it Full Backup Pool} specifies a Pool to be used for Full backups.
900 It will override any Pool specification during a Full backup. This
901 directive is optional.
903 \item [Differential Backup Pool = \lt{}pool-resource-name\gt{}]
904 \index[dir]{Differential Backup Pool}
905 \index[dir]{Directive!Differential Backup Pool}
906 The {\it Differential Backup Pool} specifies a Pool to be used for
907 Differential backups. It will override any Pool specification during a
908 Differential backup. This directive is optional.
910 \item [Incremental Backup Pool = \lt{}pool-resource-name\gt{}]
911 \index[dir]{Incremental Backup Pool}
912 \index[dir]{Directive!Incremental Backup Pool}
913 The {\it Incremental Backup Pool} specifies a Pool to be used for
914 Incremental backups. It will override any Pool specification during an
915 Incremental backup. This directive is optional.
917 \item [Schedule = \lt{}schedule-name\gt{}]
918 \index[dir]{Schedule}
919 \index[dir]{Directive!Schedule}
920 The Schedule directive defines what schedule is to be used for the Job.
921 The schedule in turn determines when the Job will be automatically
922 started and what Job level (i.e. Full, Incremental, ...) is to be run.
923 This directive is optional, and if left out, the Job can only be started
924 manually using the Console program. Although you may specify only a
925 single Schedule resource for any one job, the Schedule resource may
926 contain multiple {\bf Run} directives, which allow you to run the Job at
927 many different times, and each {\bf run} directive permits overriding
928 the default Job Level Pool, Storage, and Messages resources. This gives
929 considerable flexibility in what can be done with a single Job. For
930 additional details, see the \ilink{Schedule Resource
931 Chapter}{ScheduleResource} of this manual.
934 \item [Storage = \lt{}storage-resource-name\gt{}]
936 \index[dir]{Directive!Storage}
937 The Storage directive defines the name of the storage services where you
938 want to backup the FileSet data. For additional details, see the
939 \ilink{Storage Resource Chapter}{StorageResource2} of this manual.
940 The Storage resource may also be specified in the Job's Pool resource,
941 in which case the value in the Pool resource overrides any value
942 in the Job. This Storage resource definition is not required by either
943 the Job resource or in the Pool, but it must be specified in
944 one or the other, if not an error will result.
946 \item [Max Start Delay = \lt{}time\gt{}]
947 \index[dir]{Max Start Delay}
948 \index[dir]{Directive!Max Start Delay}
949 The time specifies the maximum delay between the scheduled time and the
950 actual start time for the Job. For example, a job can be scheduled to
951 run at 1:00am, but because other jobs are running, it may wait to run.
952 If the delay is set to 3600 (one hour) and the job has not begun to run
953 by 2:00am, the job will be canceled. This can be useful, for example,
954 to prevent jobs from running during day time hours. The default is 0
955 which indicates no limit.
957 \item [Max Run Time = \lt{}time\gt{}]
958 \index[dir]{Max Run Time}
959 \index[dir]{Directive!Max Run Time}
960 The time specifies the maximum allowed time that a job may run, counted
961 from when the job starts, ({\bf not} necessarily the same as when the
964 By default, the the watchdog thread will kill any Job that has run more
965 than 6 days. The maximum watchdog timeout is independent of MaxRunTime
966 and cannot be changed.
969 \item [Incremental|Differential Max Wait Time = \lt{}time\gt{}]
970 \index[dir]{Incremental Wait Run Time}
971 \index[dir]{Differential Wait Run Time}
972 \index[dir]{Directive!Differential Max Wait Time}
973 Theses directives have been deprecated in favor of
974 \texttt{Incremental|Differential Max Run Time} since bacula 2.3.18.
976 \item [Incremental Max Run Time = \lt{}time\gt{}]
977 \index[dir]{Incremental Max Run Time}
978 \index[dir]{Directive!Incremental Max Run Time}
979 The time specifies the maximum allowed time that an Incremental backup job may
980 run, counted from when the job starts, ({\bf not} necessarily the same as when
981 the job was scheduled).
983 \item [Differential Max Wait Time = \lt{}time\gt{}]
984 \index[dir]{Differential Max Run Time}
985 \index[dir]{Directive!Differential Max Run Time}
986 The time specifies the maximum allowed time that a Differential backup job may
987 run, counted from when the job starts, ({\bf not} necessarily the same as when
988 the job was scheduled).
990 \item [Max Run Sched Time = \lt{}time\gt{}]
991 \index[dir]{Max Run Sched Time}
992 \index[dir]{Directive!Max Run Sched Time}
994 The time specifies the maximum allowed time that a job may run, counted from
995 when the job was scheduled. This can be useful to prevent jobs from running
996 during working hours. We can see it like \texttt{Max Start Delay + Max Run
999 \item [Max Wait Time = \lt{}time\gt{}]
1000 \index[dir]{Max Wait Time}
1001 \index[dir]{Directive!Max Wait Time}
1002 The time specifies the maximum allowed time that a job may block waiting
1003 for a resource (such as waiting for a tape to be mounted, or waiting for
1004 the storage or file daemons to perform their duties), counted from the
1005 when the job starts, ({\bf not} necessarily the same as when the job was
1006 scheduled). This directive works as expected since bacula 2.3.18.
1008 \begin{figure}[htbp]
1010 \includegraphics[width=13cm]{\idir different_time.eps}
1011 \caption{Job time control directives}
1012 \label{fig:differenttime}
1015 \item [Maximum Bandwidth = \lt{}speed\gt{}]
1016 \index[dir]{Maximum Bandwidth}
1017 \index[dir]{Directive!Maximum Bandwidth}
1019 The speed parameter specifies the maximum allowed bandwidth that a job may
1020 use. The speed parameter should be specified in k/s, kb/s, m/s or mb/s.
1022 \item [Max Full Interval = \lt{}time\gt{}]
1023 \index[dir]{Max Full Interval}
1024 \index[dir]{Directive!Max Full Interval}
1025 The time specifies the maximum allowed age (counting from start time) of
1026 the most recent successful Full backup that is required in order to run
1027 Incremental or Differential backup jobs. If the most recent Full backup
1028 is older than this interval, Incremental and Differential backups will be
1029 upgraded to Full backups automatically. If this directive is not present,
1030 or specified as 0, then the age of the previous Full backup is not
1033 \label{PreferMountedVolumes}
1034 \item [Prefer Mounted Volumes = \lt{}yes\vb{}no\gt{}]
1035 \index[dir]{Prefer Mounted Volumes}
1036 \index[dir]{Directive!Prefer Mounted Volumes}
1037 If the Prefer Mounted Volumes directive is set to {\bf yes} (default
1038 yes), the Storage daemon is requested to select either an Autochanger or
1039 a drive with a valid Volume already mounted in preference to a drive
1040 that is not ready. This means that all jobs will attempt to append
1041 to the same Volume (providing the Volume is appropriate -- right Pool,
1042 ... for that job), unless you are using multiple pools.
1043 If no drive with a suitable Volume is available, it
1044 will select the first available drive. Note, any Volume that has
1045 been requested to be mounted, will be considered valid as a mounted
1046 volume by another job. This if multiple jobs start at the same time
1047 and they all prefer mounted volumes, the first job will request the
1048 mount, and the other jobs will use the same volume.
1050 If the directive is set to {\bf no}, the Storage daemon will prefer
1051 finding an unused drive, otherwise, each job started will append to the
1052 same Volume (assuming the Pool is the same for all jobs). Setting
1053 Prefer Mounted Volumes to no can be useful for those sites
1054 with multiple drive autochangers that prefer to maximize backup
1055 throughput at the expense of using additional drives and Volumes.
1056 This means that the job will prefer to use an unused drive rather
1057 than use a drive that is already in use.
1059 Despite the above, we recommend against setting this directive to
1061 it tends to add a lot of swapping of Volumes between the different
1062 drives and can easily lead to deadlock situations in the Storage
1063 daemon. We will accept bug reports against it, but we cannot guarantee
1064 that we will be able to fix the problem in a reasonable time.
1066 A better alternative for using multiple drives is to use multiple
1067 pools so that Bacula will be forced to mount Volumes from those Pools
1068 on different drives.
1070 \item [Prune Jobs = \lt{}yes\vb{}no\gt{}]
1071 \index[dir]{Prune Jobs}
1072 \index[dir]{Directive!Prune Jobs}
1073 Normally, pruning of Jobs from the Catalog is specified on a Client by
1074 Client basis in the Client resource with the {\bf AutoPrune} directive.
1075 If this directive is specified (not normally) and the value is {\bf
1076 yes}, it will override the value specified in the Client resource. The
1077 default is {\bf no}.
1080 \item [Prune Files = \lt{}yes\vb{}no\gt{}]
1081 \index[dir]{Prune Files}
1082 \index[dir]{Directive!Prune Files}
1083 Normally, pruning of Files from the Catalog is specified on a Client by
1084 Client basis in the Client resource with the {\bf AutoPrune} directive.
1085 If this directive is specified (not normally) and the value is {\bf
1086 yes}, it will override the value specified in the Client resource. The
1087 default is {\bf no}.
1089 \item [Prune Volumes = \lt{}yes\vb{}no\gt{}]
1090 \index[dir]{Prune Volumes}
1091 \index[dir]{Directive!Prune Volumes}
1092 Normally, pruning of Volumes from the Catalog is specified on a Pool by
1093 Pool basis in the Pool resource with the {\bf AutoPrune} directive.
1094 Note, this is different from File and Job pruning which is done on a
1095 Client by Client basis. If this directive is specified (not normally)
1096 and the value is {\bf yes}, it will override the value specified in the
1097 Pool resource. The default is {\bf no}.
1099 \item [RunScript \{\lt{}body-of-runscript\gt{}\}]
1100 \index[dir]{RunScript}
1101 \index[dir]{Directive!Run Script}
1103 The RunScript directive behaves like a resource in that it
1104 requires opening and closing braces around a number of directives
1105 that make up the body of the runscript.
1107 The specified {\bf Command} (see below for details) is run as an external
1108 program prior or after the current Job. This is optional. By default, the
1109 program is executed on the Client side like in \texttt{ClientRunXXXJob}.
1111 \textbf{Console} options are special commands that are sent to the director instead
1112 of the OS. At this time, console command ouputs are redirected to log with
1115 You can use following console command : \texttt{delete}, \texttt{disable},
1116 \texttt{enable}, \texttt{estimate}, \texttt{list}, \texttt{llist},
1117 \texttt{memory}, \texttt{prune}, \texttt{purge}, \texttt{reload},
1118 \texttt{status}, \texttt{setdebug}, \texttt{show}, \texttt{time},
1119 \texttt{trace}, \texttt{update}, \texttt{version}, \texttt{.client},
1120 \texttt{.jobs}, \texttt{.pool}, \texttt{.storage}. See console chapter for
1121 more information. You need to specify needed information on command line, nothing
1122 will be prompted. Example :
1125 Console = "prune files client=%c"
1126 Console = "update stats age=3"
1129 You can specify more than one Command/Console option per RunScript.
1131 You can use following options may be specified in the body
1134 \begin{tabular}{|c|c|c|l}
1135 Options & Value & Default & Information \\
1138 Runs On Success & Yes/No & {\it Yes} & Run command if JobStatus is successful\\
1140 Runs On Failure & Yes/No & {\it No} & Run command if JobStatus isn't successful\\
1142 Runs On Client & Yes/No & {\it Yes} & Run command on client\\
1144 Runs When & Before|After|Always|\textsl{AfterVSS} & {\it Never} & When run commands\\
1146 Fail Job On Error & Yes/No & {\it Yes} & Fail job if script returns
1147 something different from 0 \\
1149 Command & & & Path to your script\\
1151 Console & & & Console command\\
1156 Any output sent by the command to standard output will be included in the
1157 Bacula job report. The command string must be a valid program name or name
1160 In addition, the command string is parsed then fed to the OS,
1161 which means that the path will be searched to execute your specified
1162 command, but there is no shell interpretation, as a consequence, if you
1163 invoke complicated commands or want any shell features such as redirection
1164 or piping, you must call a shell script and do it inside that script.
1166 Before submitting the specified command to the operating system, Bacula
1167 performs character substitution of the following characters:
1169 \label{character substitution}
1175 %d = Daemon's name (Such as host-dir or host-fd)
1176 %D = Director's name (Also valid on file daemon)
1177 %e = Job Exit Status
1178 %f = Job FileSet (Only on director side)
1185 %p = Pool name (Only on director side)
1187 %t = Job type (Backup, ...)
1188 %v = Volume name (Only on director side)
1189 %w = Storage name (Only on director side)
1190 %x = Spooling enabled? ("yes" or "no")
1195 Some character substitutions are not available in all situations. The Job Exit
1196 Status code \%e edits the following values:
1198 \index[dir]{Exit Status}
1205 \item Unknown term code
1208 Thus if you edit it on a command line, you will need to enclose
1209 it within some sort of quotes.
1212 You can use these following shortcuts:\\
1214 \begin{tabular}{|c|c|c|c|c|c}
1215 Keyword & RunsOnSuccess & RunsOnFailure & FailJobOnError & Runs On Client & RunsWhen \\
1217 Run Before Job & & & Yes & No & Before \\
1219 Run After Job & Yes & No & & No & After \\
1221 Run After Failed Job & No & Yes & & No & After \\
1223 Client Run Before Job & & & Yes & Yes & Before \\
1225 Client Run After Job & Yes & No & & Yes & After \\
1233 Command = "/etc/init.d/apache stop"
1239 Command = "/etc/init.d/apache start"
1243 {\bf Notes about ClientRunBeforeJob}
1245 For compatibility reasons, with this shortcut, the command is executed
1246 directly when the client recieve it. And if the command is in error, other
1247 remote runscripts will be discarded. To be sure that all commands will be
1248 sent and executed, you have to use RunScript syntax.
1250 {\bf Special Windows Considerations}
1252 You can run scripts just after snapshots initializations with
1253 \textsl{AfterVSS} keyword.
1255 In addition, for a Windows client, please take
1256 note that you must ensure a correct path to your script. The script or
1257 program can be a .com, .exe or a .bat file. If you just put the program
1258 name in then Bacula will search using the same rules that cmd.exe uses
1259 (current directory, Bacula bin directory, and PATH). It will even try the
1260 different extensions in the same order as cmd.exe.
1261 The command can be anything that cmd.exe or command.com will recognize
1262 as an executable file.
1264 However, if you have slashes in the program name then Bacula figures you
1265 are fully specifying the name, so you must also explicitly add the three
1266 character extension.
1268 The command is run in a Win32 environment, so Unix like commands will not
1269 work unless you have installed and properly configured Cygwin in addition
1270 to and separately from Bacula.
1272 The System \%Path\% will be searched for the command. (under the
1273 environment variable dialog you have have both System Environment and
1274 User Environment, we believe that only the System environment will be
1275 available to bacula-fd, if it is running as a service.)
1277 System environment variables can be referenced with \%var\% and
1278 used as either part of the command name or arguments.
1280 So if you have a script in the Bacula\\bin directory then the following lines
1285 Client Run Before Job = systemstate
1287 Client Run Before Job = systemstate.bat
1289 Client Run Before Job = "systemstate"
1291 Client Run Before Job = "systemstate.bat"
1293 ClientRunBeforeJob = "\"C:/Program Files/Bacula/systemstate.bat\""
1297 The outer set of quotes is removed when the configuration file is parsed.
1298 You need to escape the inner quotes so that they are there when the code
1299 that parses the command line for execution runs so it can tell what the
1305 ClientRunBeforeJob = "\"C:/Program Files/Software
1306 Vendor/Executable\" /arg1 /arg2 \"foo bar\""
1310 The special characters
1314 will need to be quoted,
1315 if they are part of a filename or argument.
1317 If someone is logged in, a blank "command" window running the commands
1318 will be present during the execution of the command.
1320 Some Suggestions from Phil Stracchino for running on Win32 machines with
1321 the native Win32 File daemon:
1324 \item You might want the ClientRunBeforeJob directive to specify a .bat
1325 file which runs the actual client-side commands, rather than trying
1326 to run (for example) regedit /e directly.
1327 \item The batch file should explicitly 'exit 0' on successful completion.
1328 \item The path to the batch file should be specified in Unix form:
1330 ClientRunBeforeJob = "c:/bacula/bin/systemstate.bat"
1332 rather than DOS/Windows form:
1334 ClientRunBeforeJob =
1336 "c:\textbackslash{}bacula\textbackslash{}bin\textbackslash{}systemstate.bat"
1340 For Win32, please note that there are certain limitations:
1342 ClientRunBeforeJob = "C:/Program Files/Bacula/bin/pre-exec.bat"
1344 Lines like the above do not work because there are limitations of
1345 cmd.exe that is used to execute the command.
1346 Bacula prefixes the string you supply with {\bf cmd.exe /c }. To test that
1347 your command works you should type {\bf cmd /c "C:/Program Files/test.exe"} at a
1348 cmd prompt and see what happens. Once the command is correct insert a
1349 backslash (\textbackslash{}) before each double quote ("), and
1350 then put quotes around the whole thing when putting it in
1351 the director's .conf file. You either need to have only one set of quotes
1352 or else use the short name and don't put quotes around the command path.
1354 Below is the output from cmd's help as it relates to the command line
1355 passed to the /c option.
1358 If /C or /K is specified, then the remainder of the command line after
1359 the switch is processed as a command line, where the following logic is
1360 used to process quote (") characters:
1364 If all of the following conditions are met, then quote characters
1365 on the command line are preserved:
1368 \item exactly two quote characters.
1369 \item no special characters between the two quote characters,
1370 where special is one of:
1374 \item there are one or more whitespace characters between the
1375 the two quote characters.
1376 \item the string between the two quote characters is the name
1377 of an executable file.
1380 \item Otherwise, old behavior is to see if the first character is
1381 a quote character and if so, strip the leading character and
1382 remove the last quote character on the command line, preserving
1383 any text after the last quote character.
1388 The following example of the use of the Client Run Before Job directive was
1389 submitted by a user:\\
1390 You could write a shell script to back up a DB2 database to a FIFO. The shell
1400 db2 BACKUP DATABASE mercuryd TO $DIR/dbpipe WITHOUT PROMPTING &
1405 The following line in the Job resource in the bacula-dir.conf file:
1408 Client Run Before Job = "su - mercuryd -c \"/u01/mercuryd/backupdb.sh '%t'
1413 When the job is run, you will get messages from the output of the script
1414 stating that the backup has started. Even though the command being run is
1415 backgrounded with \&, the job will block until the "db2 BACKUP DATABASE"
1416 command, thus the backup stalls.
1418 To remedy this situation, the "db2 BACKUP DATABASE" line should be changed to
1423 db2 BACKUP DATABASE mercuryd TO $DIR/dbpipe WITHOUT PROMPTING > $DIR/backup.log
1428 It is important to redirect the input and outputs of a backgrounded command to
1429 /dev/null to prevent the script from blocking.
1431 \item [Run Before Job = \lt{}command\gt{}]
1432 \index[dir]{Run Before Job}
1433 \index[dir]{Directive!Run Before Job}
1434 \index[dir]{Directive!Run Before Job}
1435 The specified {\bf command} is run as an external program prior to running the
1436 current Job. This directive is not required, but if it is defined, and if the
1437 exit code of the program run is non-zero, the current Bacula job will be
1441 Run Before Job = "echo test"
1443 it's equivalent to :
1446 Command = "echo test"
1452 Lutz Kittler has pointed out that using the RunBeforeJob directive can be a
1453 simple way to modify your schedules during a holiday. For example, suppose
1454 that you normally do Full backups on Fridays, but Thursday and Friday are
1455 holidays. To avoid having to change tapes between Thursday and Friday when
1456 no one is in the office, you can create a RunBeforeJob that returns a
1457 non-zero status on Thursday and zero on all other days. That way, the
1458 Thursday job will not run, and on Friday the tape you inserted on Wednesday
1459 before leaving will be used.
1461 \item [Run After Job = \lt{}command\gt{}]
1462 \index[dir]{Run After Job}
1463 \index[dir]{Directive!Run After Job}
1464 The specified {\bf command} is run as an external program if the current
1465 job terminates normally (without error or without being canceled). This
1466 directive is not required. If the exit code of the program run is
1467 non-zero, Bacula will print a warning message. Before submitting the
1468 specified command to the operating system, Bacula performs character
1469 substitution as described above for the {\bf RunScript} directive.
1471 An example of the use of this directive is given in the
1472 \ilink{Tips Chapter}{JobNotification} of this manual.
1474 See the {\bf Run After Failed Job} if you
1475 want to run a script after the job has terminated with any
1478 \item [Run After Failed Job = \lt{}command\gt{}]
1479 \index[dir]{Run After Job}
1480 \index[dir]{Directive!Run After Job}
1481 The specified {\bf command} is run as an external program after the current
1482 job terminates with any error status. This directive is not required. The
1483 command string must be a valid program name or name of a shell script. If
1484 the exit code of the program run is non-zero, Bacula will print a
1485 warning message. Before submitting the specified command to the
1486 operating system, Bacula performs character substitution as described above
1487 for the {\bf RunScript} directive. Note, if you wish that your script
1488 will run regardless of the exit status of the Job, you can use this :
1491 Command = "echo test"
1495 RunsOnSuccess = yes # default, you can drop this line
1499 An example of the use of this directive is given in the
1500 \ilink{Tips Chapter}{JobNotification} of this manual.
1503 \item [Client Run Before Job = \lt{}command\gt{}]
1504 \index[dir]{Client Run Before Job}
1505 \index[dir]{Directive!Client Run Before Job}
1506 This directive is the same as {\bf Run Before Job} except that the
1507 program is run on the client machine. The same restrictions apply to
1508 Unix systems as noted above for the {\bf RunScript}.
1510 \item [Client Run After Job = \lt{}command\gt{}]
1511 \index[dir]{Client Run After Job}
1512 \index[dir]{Directive!Client Run After Job}
1513 The specified {\bf command} is run on the client machine as soon
1514 as data spooling is complete in order to allow restarting applications
1515 on the client as soon as possible. .
1517 Note, please see the notes above in {\bf RunScript}
1518 concerning Windows clients.
1520 \item [Rerun Failed Levels = \lt{}yes\vb{}no\gt{}]
1521 \index[dir]{Rerun Failed Levels}
1522 \index[dir]{Directive!Rerun Failed Levels}
1523 If this directive is set to {\bf yes} (default no), and Bacula detects that
1524 a previous job at a higher level (i.e. Full or Differential) has failed,
1525 the current job level will be upgraded to the higher level. This is
1526 particularly useful for Laptops where they may often be unreachable, and if
1527 a prior Full save has failed, you wish the very next backup to be a Full
1528 save rather than whatever level it is started as.
1530 There are several points that must be taken into account when using this
1531 directive: first, a failed job is defined as one that has not terminated
1532 normally, which includes any running job of the same name (you need to
1533 ensure that two jobs of the same name do not run simultaneously);
1534 secondly, the {\bf Ignore FileSet Changes} directive is not considered
1535 when checking for failed levels, which means that any FileSet change will
1538 \item [Spool Data = \lt{}yes\vb{}no\gt{}]
1539 \index[dir]{Spool Data}
1540 \index[dir]{Directive!Spool Data}
1542 If this directive is set to {\bf yes} (default no), the Storage daemon will
1543 be requested to spool the data for this Job to disk rather than write it
1544 directly to the Volume (normally a tape).
1546 Thus the data is written in large blocks to the Volume rather than small
1547 blocks. This directive is particularly useful when running multiple
1548 simultaneous backups to tape. Once all the data arrives or the spool
1549 files' maximum sizes are reached, the data will be despooled and written
1552 Spooling data prevents interleaving date from several job and reduces or
1553 eliminates tape drive stop and start commonly known as "shoe-shine".
1555 We don't recommend using this option if you are writing to a disk file
1556 using this option will probably just slow down the backup jobs.
1558 NOTE: When this directive is set to yes, Spool Attributes is also
1559 automatically set to yes.
1561 \item [SpoolData=yes\vb{}no]
1562 \index[dir]{SpoolData}
1563 \index[dir]{Directive!SpoolData}
1564 tells Bacula to request the Storage daemon to spool data to a disk file
1565 before writing it to the Volume (normally a tape).
1567 \item [Spool Attributes = \lt{}yes\vb{}no\gt{}]
1568 \index[dir]{Spool Attributes}
1569 \index[dir]{Directive!Spool Attributes}
1571 \index[general]{slow}
1572 \index[dir]{Backups!slow}
1573 \index[general]{Backups!slow}
1574 The default is set to {\bf no}, which means that the File attributes are
1575 sent by the Storage daemon to the Director as they are stored on tape.
1576 However, if you want to avoid the possibility that database updates will
1577 slow down writing to the tape, you may want to set the value to {\bf
1578 yes}, in which case the Storage daemon will buffer the File attributes
1579 and Storage coordinates to a temporary file in the Working Directory,
1580 then when writing the Job data to the tape is completed, the attributes
1581 and storage coordinates will be sent to the Director.
1583 NOTE: When Spool Data is set to yes, Spool Attributes is also
1584 automatically set to yes.
1586 \item [SpoolSize={\it bytes}]
1587 \index[dir]{SpoolSize}
1588 \index[dir]{Directive!SpoolSize}
1589 where the bytes specify the maximum spool size for this job.
1590 The default is take from Device Maximum Spool Size limit.
1591 This directive is available only in Bacula version 2.3.5 or
1595 \item [Where = \lt{}directory\gt{}]
1597 \index[dir]{Directive!Where}
1598 This directive applies only to a Restore job and specifies a prefix to
1599 the directory name of all files being restored. This permits files to
1600 be restored in a different location from which they were saved. If {\bf
1601 Where} is not specified or is set to backslash ({\bf /}), the files will
1602 be restored to their original location. By default, we have set {\bf
1603 Where} in the example configuration files to be {\bf
1604 /tmp/bacula-restores}. This is to prevent accidental overwriting of
1607 \item [Add Prefix = \lt{}directory\gt{}]
1608 \label{confaddprefix}
1609 \index[dir]{AddPrefix}
1610 \index[dir]{Directive!AddPrefix}
1611 This directive applies only to a Restore job and specifies a prefix to the
1612 directory name of all files being restored. This will use \ilink{File
1613 Relocation}{filerelocation} feature implemented in Bacula 2.1.8 or later.
1615 \item [Add Suffix = \lt{}extention\gt{}]
1616 \index[dir]{AddSuffix}
1617 \index[dir]{Directive!AddSuffix}
1618 This directive applies only to a Restore job and specifies a suffix to all
1619 files being restored. This will use \ilink{File Relocation}{filerelocation}
1620 feature implemented in Bacula 2.1.8 or later.
1622 Using \texttt{Add Suffix=.old}, \texttt{/etc/passwd} will be restored to
1623 \texttt{/etc/passwsd.old}
1625 \item [Strip Prefix = \lt{}directory\gt{}]
1626 \index[dir]{StripPrefix}
1627 \index[dir]{Directive!StripPrefix}
1628 This directive applies only to a Restore job and specifies a prefix to remove
1629 from the directory name of all files being restored. This will use the
1630 \ilink{File Relocation}{filerelocation} feature implemented in Bacula 2.1.8
1633 Using \texttt{Strip Prefix=/etc}, \texttt{/etc/passwd} will be restored to
1636 Under Windows, if you want to restore \texttt{c:/files} to \texttt{d:/files},
1644 \item [RegexWhere = \lt{}expressions\gt{}]
1645 \index[dir]{RegexWhere}
1646 \index[dir]{Directive!RegexWhere}
1647 This directive applies only to a Restore job and specifies a regex filename
1648 manipulation of all files being restored. This will use \ilink{File
1649 Relocation}{filerelocation} feature implemented in Bacula 2.1.8 or later.
1651 For more informations about how use this option, see
1652 \ilink{this}{useregexwhere}.
1654 \item [Replace = \lt{}replace-option\gt{}]
1655 \index[dir]{Replace}
1656 \index[dir]{Directive!Replace}
1657 This directive applies only to a Restore job and specifies what happens
1658 when Bacula wants to restore a file or directory that already exists.
1659 You have the following options for {\bf replace-option}:
1665 when the file to be restored already exists, it is deleted and then
1666 replaced by the copy that was backed up. This is the default value.
1669 \index[dir]{ifnewer}
1670 if the backed up file (on tape) is newer than the existing file, the
1671 existing file is deleted and replaced by the back up.
1674 \index[dir]{ifolder}
1675 if the backed up file (on tape) is older than the existing file, the
1676 existing file is deleted and replaced by the back up.
1680 if the backed up file already exists, Bacula skips restoring this file.
1683 \item [Prefix Links=\lt{}yes\vb{}no\gt{}]
1684 \index[dir]{Prefix Links}
1685 \index[dir]{Directive!Prefix Links}
1686 If a {\bf Where} path prefix is specified for a recovery job, apply it
1687 to absolute links as well. The default is {\bf No}. When set to {\bf
1688 Yes} then while restoring files to an alternate directory, any absolute
1689 soft links will also be modified to point to the new alternate
1690 directory. Normally this is what is desired -- i.e. everything is self
1691 consistent. However, if you wish to later move the files to their
1692 original locations, all files linked with absolute names will be broken.
1694 \item [Maximum Concurrent Jobs = \lt{}number\gt{}]
1695 \index[dir]{Maximum Concurrent Jobs}
1696 \index[dir]{Directive!Maximum Concurrent Jobs}
1697 where \lt{}number\gt{} is the maximum number of Jobs from the current
1698 Job resource that can run concurrently. Note, this directive limits
1699 only Jobs with the same name as the resource in which it appears. Any
1700 other restrictions on the maximum concurrent jobs such as in the
1701 Director, Client, or Storage resources will also apply in addition to
1702 the limit specified here. The default is set to 1, but you may set it
1703 to a larger number. We strongly recommend that you read the WARNING
1704 documented under \ilink{ Maximum Concurrent Jobs}{DirMaxConJobs} in the
1705 Director's resource.
1707 \item [Reschedule On Error = \lt{}yes\vb{}no\gt{}]
1708 \index[dir]{Reschedule On Error}
1709 \index[dir]{Directive!Reschedule On Error}
1710 If this directive is enabled, and the job terminates in error, the job
1711 will be rescheduled as determined by the {\bf Reschedule Interval} and
1712 {\bf Reschedule Times} directives. If you cancel the job, it will not
1713 be rescheduled. The default is {\bf no} (i.e. the job will not be
1716 This specification can be useful for portables, laptops, or other
1717 machines that are not always connected to the network or switched on.
1719 \item [Reschedule Interval = \lt{}time-specification\gt{}]
1720 \index[dir]{Reschedule Interval}
1721 \index[dir]{Directive!Reschedule Interval}
1722 If you have specified {\bf Reschedule On Error = yes} and the job
1723 terminates in error, it will be rescheduled after the interval of time
1724 specified by {\bf time-specification}. See \ilink{the time
1725 specification formats}{Time} in the Configure chapter for details of
1726 time specifications. If no interval is specified, the job will not be
1727 rescheduled on error.
1729 \item [Reschedule Times = \lt{}count\gt{}]
1730 \index[dir]{Reschedule Times}
1731 \index[dir]{Directive!Reschedule Times}
1732 This directive specifies the maximum number of times to reschedule the
1733 job. If it is set to zero (the default) the job will be rescheduled an
1734 indefinite number of times.
1736 \item [Allow Duplicate Jobs = \lt{}yes\vb{}no\gt{}]
1737 \index[general]{Allow Duplicate Jobs}
1739 \begin{figure}[htbp]
1741 \includegraphics[width=13cm]{\idir duplicate-real.eps}
1742 \caption{Allow Duplicate Jobs usage}
1743 \label{fig:allowduplicatejobs}
1746 A duplicate job in the sense we use it here means a second or subsequent job
1747 with the same name starts. This happens most frequently when the first job
1748 runs longer than expected because no tapes are available.
1750 If this directive is enabled duplicate jobs will be run. If
1751 the directive is set to {\bf no} (default) then only one job of a given name
1752 may run at one time, and the action that Bacula takes to ensure only
1753 one job runs is determined by the other directives (see below).
1755 If {\bf Allow Duplicate Jobs} is set to {\bf no} and two jobs
1756 are present and none of the three directives given below permit
1757 cancelling a job, then the current job (the second one started)
1760 \item [Allow Higher Duplicates = \lt{}yes\vb{}no\gt{}]
1761 \index[general]{Allow Higher Duplicates}
1762 This directive was implemented in version 5.0.0, but does not work
1763 as expected. If used, it should always be set to no. In later versions
1764 of Bacula the directive is disabled (disregarded).
1767 \item [Cancel Lower Level Duplicates = \lt{}yes\vb{}no\gt{}]
1768 \index[general]{Cancel Lower Level Duplicates}
1769 If \textbf{Allow Duplicates Jobs} is set to \textbf{no} and this
1770 directive is set to \textbf{yes}, Bacula will choose between duplicated
1771 jobs the one with the highest level. For example, it will cancel a
1772 previous Incremental to run a Full backup. It works only for Backup
1773 jobs. The default is \texttt{no}. If the levels of the duplicated
1774 jobs are the same, nothing is done and the other
1775 Cancel XXX Duplicate directives will be examined.
1777 \item [Cancel Queued Duplicates = \lt{}yes\vb{}no\gt{}]
1778 \index[general]{Cancel Queued Duplicates}
1779 If {\bf Allow Duplicate Jobs} is set to {\bf no} and
1780 if this directive is set to {\bf yes} any job that is
1781 already queued to run but not yet running will be canceled.
1782 The default is {\bf no}.
1784 \item[Cancel Running Duplicates = \lt{}yes\vb{}no\gt{}]
1785 \index[general]{Cancel Running Duplicates}
1786 If {\bf Allow Duplicate Jobs} is set to {\bf no} and
1787 if this directive is set to {\bf yes} any job that is already running
1788 will be canceled. The default is {\bf no}.
1791 %%\item[DuplicateJobProximity = \lt{}time-specification\gt{}]
1792 %%\index[general]{Duplicate Job Proximity}
1793 %% This directive permits to determine if two jobs are really duplicated.
1794 %% If the first one is running for long time, this is probably not a good
1795 %% idea to cancel it.
1797 \item [Run = \lt{}job-name\gt{}]
1799 \index[dir]{Directive!Run}
1800 \index[dir]{Clone a Job}
1801 The Run directive (not to be confused with the Run option in a
1802 Schedule) allows you to start other jobs or to clone jobs. By using the
1803 cloning keywords (see below), you can backup
1804 the same data (or almost the same data) to two or more drives
1805 at the same time. The {\bf job-name} is normally the same name
1806 as the current Job resource (thus creating a clone). However, it
1807 may be any Job name, so one job may start other related jobs.
1809 The part after the equal sign must be enclosed in double quotes,
1810 and can contain any string or set of options (overrides) that you
1811 can specify when entering the Run command from the console. For
1812 example {\bf storage=DDS-4 ...}. In addition, there are two special
1813 keywords that permit you to clone the current job. They are {\bf level=\%l}
1814 and {\bf since=\%s}. The \%l in the level keyword permits
1815 entering the actual level of the current job and the \%s in the since
1816 keyword permits putting the same time for comparison as used on the
1817 current job. Note, in the case of the since keyword, the \%s must be
1818 enclosed in double quotes, and thus they must be preceded by a backslash
1819 since they are already inside quotes. For example:
1822 run = "Nightly-backup level=%l since=\"%s\" storage=DDS-4"
1825 A cloned job will not start additional clones, so it is not
1826 possible to recurse.
1828 Please note that all cloned jobs, as specified in the Run directives are
1829 submitted for running before the original job is run (while it is being
1830 initialized). This means that any clone job will actually start before
1831 the original job, and may even block the original job from starting
1832 until the original job finishes unless you allow multiple simultaneous
1833 jobs. Even if you set a lower priority on the clone job, if no other
1834 jobs are running, it will start before the original job.
1836 If you are trying to prioritize jobs by using the clone feature (Run
1837 directive), you will find it much easier to do using a RunScript
1838 resource, or a RunBeforeJob directive.
1841 \item [Priority = \lt{}number\gt{}]
1842 \index[dir]{Priority}
1843 \index[dir]{Directive!Priority}
1844 This directive permits you to control the order in which your jobs will
1845 be run by specifying a positive non-zero number. The higher the number,
1846 the lower the job priority. Assuming you are not running concurrent jobs,
1847 all queued jobs of priority 1 will run before queued jobs of priority 2
1848 and so on, regardless of the original scheduling order.
1850 The priority only affects waiting jobs that are queued to run, not jobs
1851 that are already running. If one or more jobs of priority 2 are already
1852 running, and a new job is scheduled with priority 1, the currently
1853 running priority 2 jobs must complete before the priority 1 job is
1854 run, unless Allow Mixed Priority is set.
1856 The default priority is 10.
1858 If you want to run concurrent jobs you should
1859 keep these points in mind:
1862 \item See \ilink{Running Concurrent Jobs}{ConcurrentJobs} on how to setup
1865 \item Bacula concurrently runs jobs of only one priority at a time. It
1866 will not simultaneously run a priority 1 and a priority 2 job.
1868 \item If Bacula is running a priority 2 job and a new priority 1 job is
1869 scheduled, it will wait until the running priority 2 job terminates even
1870 if the Maximum Concurrent Jobs settings would otherwise allow two jobs
1871 to run simultaneously.
1873 \item Suppose that bacula is running a priority 2 job and a new priority 1
1874 job is scheduled and queued waiting for the running priority 2 job to
1875 terminate. If you then start a second priority 2 job, the waiting
1876 priority 1 job will prevent the new priority 2 job from running
1877 concurrently with the running priority 2 job. That is: as long as there
1878 is a higher priority job waiting to run, no new lower priority jobs will
1879 start even if the Maximum Concurrent Jobs settings would normally allow
1880 them to run. This ensures that higher priority jobs will be run as soon
1884 If you have several jobs of different priority, it may not best to start
1885 them at exactly the same time, because Bacula must examine them one at a
1886 time. If by Bacula starts a lower priority job first, then it will run
1887 before your high priority jobs. If you experience this problem, you may
1888 avoid it by starting any higher priority jobs a few seconds before lower
1889 priority ones. This insures that Bacula will examine the jobs in the
1890 correct order, and that your priority scheme will be respected.
1892 \label{AllowMixedPriority}
1893 \item [Allow Mixed Priority = \lt{}yes\vb{}no\gt{}]
1894 \index[dir]{Allow Mixed Priority}
1895 This directive is only implemented in version 2.5 and later. When
1896 set to {\bf yes} (default {\bf no}), this job may run even if lower
1897 priority jobs are already running. This means a high priority job
1898 will not have to wait for other jobs to finish before starting.
1899 The scheduler will only mix priorities when all running jobs have
1902 Note that only higher priority jobs will start early. Suppose the
1903 director will allow two concurrent jobs, and that two jobs with
1904 priority 10 are running, with two more in the queue. If a job with
1905 priority 5 is added to the queue, it will be run as soon as one of
1906 the running jobs finishes. However, new priority 10 jobs will not
1907 be run until the priority 5 job has finished.
1909 \label{WritePartAfterJob}
1910 \item [Write Part After Job = \lt{}yes\vb{}no\gt{}]
1911 \index[dir]{Write Part After Job}
1912 \index[dir]{Directive!Write Part After Job}
1913 This directive is only implemented in version 1.37 and later.
1914 If this directive is set to {\bf yes} (default {\bf no}), a new part file
1915 will be created after the job is finished.
1917 It should be set to {\bf yes} when writing to devices that require mount
1918 (for example DVD), so you are sure that the current part, containing
1919 this job's data, is written to the device, and that no data is left in
1920 the temporary file on the hard disk. However, on some media, like DVD+R
1921 and DVD-R, a lot of space (about 10Mb) is lost every time a part is
1922 written. So, if you run several jobs each after another, you could set
1923 this directive to {\bf no} for all jobs, except the last one, to avoid
1924 wasting too much space, but to ensure that the data is written to the
1925 medium when all jobs are finished.
1927 This directive is ignored with tape and FIFO devices.
1931 The following is an example of a valid Job resource definition:
1938 Level = Incremental # default
1940 FileSet="Minou Full Set"
1943 Schedule = "MinouWeeklyCycle"
1949 \section{The JobDefs Resource}
1950 \label{JobDefsResource}
1951 \index[general]{JobDefs Resource}
1952 \index[general]{Resource!JobDefs}
1954 The JobDefs resource permits all the same directives that can appear in a Job
1955 resource. However, a JobDefs resource does not create a Job, rather it can be
1956 referenced within a Job to provide defaults for that Job. This permits you to
1957 concisely define several nearly identical Jobs, each one referencing a JobDefs
1958 resource which contains the defaults. Only the changes from the defaults need to
1959 be mentioned in each Job.
1961 \section{The Schedule Resource}
1962 \label{ScheduleResource}
1963 \index[general]{Resource!Schedule}
1964 \index[general]{Schedule Resource}
1966 The Schedule resource provides a means of automatically scheduling a Job as
1967 well as the ability to override the default Level, Pool, Storage and Messages
1968 resources. If a Schedule resource is not referenced in a Job, the Job can only
1969 be run manually. In general, you specify an action to be taken and when.
1974 \index[dir]{Schedule}
1975 \index[dir]{Directive!Schedule}
1976 Start of the Schedule directives. No {\bf Schedule} resource is
1977 required, but you will need at least one if you want Jobs to be
1978 automatically started.
1980 \item [Name = \lt{}name\gt{}]
1982 \index[dir]{Directive!Name}
1983 The name of the schedule being defined. The Name directive is required.
1985 \item [Run = \lt{}Job-overrides\gt{} \lt{}Date-time-specification\gt{}]
1987 \index[dir]{Directive!Run}
1988 The Run directive defines when a Job is to be run, and what overrides if
1989 any to apply. You may specify multiple {\bf run} directives within a
1990 {\bf Schedule} resource. If you do, they will all be applied (i.e.
1991 multiple schedules). If you have two {\bf Run} directives that start at
1992 the same time, two Jobs will start at the same time (well, within one
1993 second of each other).
1995 The {\bf Job-overrides} permit overriding the Level, the Storage, the
1996 Messages, and the Pool specifications provided in the Job resource. In
1997 addition, the FullPool, the IncrementalPool, and the DifferentialPool
1998 specifications permit overriding the Pool specification according to
1999 what backup Job Level is in effect.
2001 By the use of overrides, you may customize a particular Job. For
2002 example, you may specify a Messages override for your Incremental
2003 backups that outputs messages to a log file, but for your weekly or
2004 monthly Full backups, you may send the output by email by using a
2005 different Messages override.
2007 {\bf Job-overrides} are specified as: {\bf keyword=value} where the
2008 keyword is Level, Storage, Messages, Pool, FullPool, DifferentialPool,
2009 or IncrementalPool, and the {\bf value} is as defined on the respective
2010 directive formats for the Job resource. You may specify multiple {\bf
2011 Job-overrides} on one {\bf Run} directive by separating them with one or
2012 more spaces or by separating them with a trailing comma. For example:
2018 \index[dir]{Directive!Level}
2019 is all files in the FileSet whether or not they have changed.
2021 \item [Level=Incremental]
2023 \index[dir]{Directive!Level}
2024 is all files that have changed since the last backup.
2028 \index[dir]{Directive!Pool}
2029 specifies to use the Pool named {\bf Weekly}.
2031 \item [Storage=DLT\_Drive]
2032 \index[dir]{Storage}
2033 \index[dir]{Directive!Storage}
2034 specifies to use {\bf DLT\_Drive} for the storage device.
2036 \item [Messages=Verbose]
2037 \index[dir]{Messages}
2038 \index[dir]{Directive!Messages}
2039 specifies to use the {\bf Verbose} message resource for the Job.
2041 \item [FullPool=Full]
2042 \index[dir]{FullPool}
2043 \index[dir]{Directive!FullPool}
2044 specifies to use the Pool named {\bf Full} if the job is a full backup, or
2046 upgraded from another type to a full backup.
2048 \item [DifferentialPool=Differential]
2049 \index[dir]{DifferentialPool}
2050 \index[dir]{Directive!DifferentialPool}
2051 specifies to use the Pool named {\bf Differential} if the job is a
2052 differential backup.
2054 \item [IncrementalPool=Incremental]
2055 \index[dir]{IncrementalPool}
2056 \index[dir]{Directive!IncrementalPool}
2057 specifies to use the Pool named {\bf Incremental} if the job is an
2061 \item [WritePartAfterJob=yes\vb{}no]
2062 \index[dir]{WritePartAfterJob}
2063 \index[dir]{Directive!WritePartAfterJob}
2064 tells Bacula to request the Storage daemon to write the current part
2065 file to the device when the job is finished (see \ilink{Write Part After
2066 Job directive in the Job resource}{WritePartAfterJob}). Please note,
2067 this directive is implemented only in version 1.37 and later. The
2068 default is yes. We strongly recommend that you keep this set to yes
2069 otherwise, when the last job has finished one part will remain in the
2070 spool file and restore may or may not work.
2074 {\bf Date-time-specification} determines when the Job is to be run. The
2075 specification is a repetition, and as a default Bacula is set to run a job at
2076 the beginning of the hour of every hour of every day of every week of every
2077 month of every year. This is not normally what you want, so you must specify
2078 or limit when you want the job to run. Any specification given is assumed to
2079 be repetitive in nature and will serve to override or limit the default
2080 repetition. This is done by specifying masks or times for the hour, day of the
2081 month, day of the week, week of the month, week of the year, and month when
2082 you want the job to run. By specifying one or more of the above, you can
2083 define a schedule to repeat at almost any frequency you want.
2085 Basically, you must supply a {\bf month}, {\bf day}, {\bf hour}, and {\bf
2086 minute} the Job is to be run. Of these four items to be specified, {\bf day}
2087 is special in that you may either specify a day of the month such as 1, 2,
2088 ... 31, or you may specify a day of the week such as Monday, Tuesday, ...
2089 Sunday. Finally, you may also specify a week qualifier to restrict the
2090 schedule to the first, second, third, fourth, or fifth week of the month.
2092 For example, if you specify only a day of the week, such as {\bf Tuesday} the
2093 Job will be run every hour of every Tuesday of every Month. That is the {\bf
2094 month} and {\bf hour} remain set to the defaults of every month and all
2097 Note, by default with no other specification, your job will run at the
2098 beginning of every hour. If you wish your job to run more than once in any
2099 given hour, you will need to specify multiple {\bf run} specifications each
2100 with a different minute.
2102 The date/time to run the Job can be specified in the following way in
2109 <week-keyword> = 1st | 2nd | 3rd | 4th | 5th | first |
2110 second | third | fourth | fifth
2111 <wday-keyword> = sun | mon | tue | wed | thu | fri | sat |
2112 sunday | monday | tuesday | wednesday |
2113 thursday | friday | saturday
2114 <week-of-year-keyword> = w00 | w01 | ... w52 | w53
2115 <month-keyword> = jan | feb | mar | apr | may | jun | jul |
2116 aug | sep | oct | nov | dec | january |
2117 february | ... | december
2118 <daily-keyword> = daily
2119 <weekly-keyword> = weekly
2120 <monthly-keyword> = monthly
2121 <hourly-keyword> = hourly
2122 <digit> = 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 0
2123 <number> = <digit> | <digit><number>
2124 <12hour> = 0 | 1 | 2 | ... 12
2125 <hour> = 0 | 1 | 2 | ... 23
2126 <minute> = 0 | 1 | 2 | ... 59
2127 <day> = 1 | 2 | ... 31
2128 <time> = <hour>:<minute> |
2129 <12hour>:<minute>am |
2131 <time-spec> = <at-keyword> <time> |
2133 <date-keyword> = <void-keyword> <weekly-keyword>
2134 <day-range> = <day>-<day>
2135 <month-range> = <month-keyword>-<month-keyword>
2136 <wday-range> = <wday-keyword>-<wday-keyword>
2137 <range> = <day-range> | <month-range> |
2139 <date> = <date-keyword> | <day> | <range>
2140 <date-spec> = <date> | <date-spec>
2141 <day-spec> = <day> | <wday-keyword> |
2142 <day> | <wday-range> |
2143 <week-keyword> <wday-keyword> |
2144 <week-keyword> <wday-range> |
2146 <month-spec> = <month-keyword> | <month-range> |
2148 <date-time-spec> = <month-spec> <day-spec> <time-spec>
2154 Note, the Week of Year specification wnn follows the ISO standard definition
2155 of the week of the year, where Week 1 is the week in which the first Thursday
2156 of the year occurs, or alternatively, the week which contains the 4th of
2157 January. Weeks are numbered w01 to w53. w00 for Bacula is the week that
2158 precedes the first ISO week (i.e. has the first few days of the year if any
2159 occur before Thursday). w00 is not defined by the ISO specification. A week
2160 starts with Monday and ends with Sunday.
2162 According to the NIST (US National Institute of Standards and Technology),
2163 12am and 12pm are ambiguous and can be defined to anything. However,
2164 12:01am is the same as 00:01 and 12:01pm is the same as 12:01, so Bacula
2165 defines 12am as 00:00 (midnight) and 12pm as 12:00 (noon). You can avoid
2166 this abiguity (confusion) by using 24 hour time specifications (i.e. no
2167 am/pm). This is the definition in Bacula version 2.0.3 and later.
2169 An example schedule resource that is named {\bf WeeklyCycle} and runs a job
2170 with level full each Sunday at 2:05am and an incremental job Monday through
2171 Saturday at 2:05am is:
2176 Name = "WeeklyCycle"
2177 Run = Level=Full sun at 2:05
2178 Run = Level=Incremental mon-sat at 2:05
2183 An example of a possible monthly cycle is as follows:
2188 Name = "MonthlyCycle"
2189 Run = Level=Full Pool=Monthly 1st sun at 2:05
2190 Run = Level=Differential 2nd-5th sun at 2:05
2191 Run = Level=Incremental Pool=Daily mon-sat at 2:05
2196 The first of every month:
2202 Run = Level=Full on 1 at 2:05
2203 Run = Level=Incremental on 2-31 at 2:05
2214 Run = Level=Full hourly at 0:05
2215 Run = Level=Full hourly at 0:15
2216 Run = Level=Full hourly at 0:25
2217 Run = Level=Full hourly at 0:35
2218 Run = Level=Full hourly at 0:45
2219 Run = Level=Full hourly at 0:55
2224 \section{Technical Notes on Schedules}
2225 \index[general]{Schedules!Technical Notes on}
2226 \index[general]{Technical Notes on Schedules}
2228 Internally Bacula keeps a schedule as a bit mask. There are six masks and a
2229 minute field to each schedule. The masks are hour, day of the month (mday),
2230 month, day of the week (wday), week of the month (wom), and week of the year
2231 (woy). The schedule is initialized to have the bits of each of these masks
2232 set, which means that at the beginning of every hour, the job will run. When
2233 you specify a month for the first time, the mask will be cleared and the bit
2234 corresponding to your selected month will be selected. If you specify a second
2235 month, the bit corresponding to it will also be added to the mask. Thus when
2236 Bacula checks the masks to see if the bits are set corresponding to the
2237 current time, your job will run only in the two months you have set. Likewise,
2238 if you set a time (hour), the hour mask will be cleared, and the hour you
2239 specify will be set in the bit mask and the minutes will be stored in the
2242 For any schedule you have defined, you can see how these bits are set by doing
2243 a {\bf show schedules} command in the Console program. Please note that the
2244 bit mask is zero based, and Sunday is the first day of the week (bit zero).
2248 \section{The Client Resource}
2249 \label{ClientResource2}
2250 \index[general]{Resource!Client}
2251 \index[general]{Client Resource}
2253 The Client resource defines the attributes of the Clients that are served by
2254 this Director; that is the machines that are to be backed up. You will need
2255 one Client resource definition for each machine to be backed up.
2259 \item [Client (or FileDaemon)]
2260 \index[dir]{Client (or FileDaemon)}
2261 \index[dir]{Directive!Client (or FileDaemon)}
2262 Start of the Client directives.
2264 \item [Name = \lt{}name\gt{}]
2266 \index[dir]{Directive!Name}
2267 The client name which will be used in the Job resource directive or in the
2268 console run command. This directive is required.
2270 \item [Address = \lt{}address\gt{}]
2271 \index[dir]{Address}
2272 \index[dir]{Directive!FD Address}
2273 \index[dir]{File Daemon Address}
2274 \index[dir]{Client Address}
2275 Where the address is a host name, a fully qualified domain name, or a
2276 network address in dotted quad notation for a Bacula File server daemon.
2277 This directive is required.
2279 \item [FD Port = \lt{}port-number\gt{}]
2280 \index[dir]{FD Port}
2281 \index[dir]{Directive!FD Port}
2282 Where the port is a port number at which the Bacula File server daemon can
2283 be contacted. The default is 9102.
2285 \item [Catalog = \lt{}Catalog-resource-name\gt{}]
2286 \index[dir]{Catalog}
2287 \index[dir]{Directive!Catalog}
2288 This specifies the name of the catalog resource to be used for this Client.
2289 This directive is required.
2291 \item [Password = \lt{}password\gt{}]
2292 \index[dir]{Password}
2293 \index[dir]{Directive!Password}
2294 This is the password to be used when establishing a connection with the File
2295 services, so the Client configuration file on the machine to be backed up
2296 must have the same password defined for this Director. This directive is
2297 required. If you have either {\bf /dev/random} {\bf bc} on your machine,
2298 Bacula will generate a random password during the configuration process,
2299 otherwise it will be left blank.
2301 The password is plain text. It is not generated through any special
2302 process, but it is preferable for security reasons to make the text
2305 \label{FileRetention}
2306 \item [File Retention = \lt{}time-period-specification\gt{}]
2307 \label{FileRetention}
2308 \index[dir]{File Retention}
2309 \index[dir]{Directive!File Retention}
2310 The File Retention directive defines the length of time that Bacula will
2311 keep File records in the Catalog database after the End time of the
2312 Job corresponding to the File records.
2313 When this time period expires, and if
2314 {\bf AutoPrune} is set to {\bf yes} Bacula will prune (remove) File records
2315 that are older than the specified File Retention period. Note, this affects
2316 only records in the catalog database. It does not affect your archive
2319 File records may actually be retained for a shorter period than you specify
2320 on this directive if you specify either a shorter {\bf Job Retention} or a
2321 shorter {\bf Volume Retention} period. The shortest retention period of the
2322 three takes precedence. The time may be expressed in seconds, minutes,
2323 hours, days, weeks, months, quarters, or years. See the
2324 \ilink{ Configuration chapter}{Time} of this manual for
2325 additional details of time specification.
2327 The default is 60 days.
2329 \label{JobRetention}
2330 \item [Job Retention = \lt{}time-period-specification\gt{}]
2331 \label{JobRetention}
2332 \index[dir]{Job Retention}
2333 \index[dir]{Directive!Job Retention}
2334 The Job Retention directive defines the length of time that Bacula will keep
2335 Job records in the Catalog database after the Job End time. When
2336 this time period expires, and if {\bf AutoPrune} is set to {\bf yes}
2337 Bacula will prune (remove) Job records that are older than the specified
2338 File Retention period. As with the other retention periods, this
2339 affects only records in the catalog and not data in your archive backup.
2341 If a Job record is selected for pruning, all associated File and JobMedia
2342 records will also be pruned regardless of the File Retention period set.
2343 As a consequence, you normally will set the File retention period to be
2344 less than the Job retention period. The Job retention period can actually
2345 be less than the value you specify here if you set the {\bf Volume
2346 Retention} directive in the Pool resource to a smaller duration. This is
2347 because the Job retention period and the Volume retention period are
2348 independently applied, so the smaller of the two takes precedence.
2350 The Job retention period is specified as seconds, minutes, hours, days,
2351 weeks, months, quarters, or years. See the
2352 \ilink{ Configuration chapter}{Time} of this manual for
2353 additional details of time specification.
2355 The default is 180 days.
2358 \item [AutoPrune = \lt{}yes\vb{}no\gt{}]
2359 \index[dir]{AutoPrune}
2360 \index[dir]{Directive!AutoPrune}
2361 If AutoPrune is set to {\bf yes} (default), Bacula (version 1.20 or greater)
2362 will automatically apply the File retention period and the Job retention
2363 period for the Client at the end of the Job. If you set {\bf AutoPrune = no},
2364 pruning will not be done, and your Catalog will grow in size each time you
2365 run a Job. Pruning affects only information in the catalog and not data
2366 stored in the backup archives (on Volumes).
2368 \item [Maximum Concurrent Jobs = \lt{}number\gt{}]
2369 \index[dir]{Maximum Concurrent Jobs}
2370 \index[dir]{Directive!Maximum Concurrent Jobs}
2371 where \lt{}number\gt{} is the maximum number of Jobs with the current Client
2372 that can run concurrently. Note, this directive limits only Jobs for Clients
2373 with the same name as the resource in which it appears. Any other
2374 restrictions on the maximum concurrent jobs such as in the Director, Job, or
2375 Storage resources will also apply in addition to any limit specified here.
2376 The default is set to 1, but you may set it to a larger number.
2378 \item [Maximum Bandwidth Per Job = \lt{}speed\gt{}]
2379 \index[dir]{Maximum Bandwidth Per Job}
2380 \index[dir]{Directive!Maximum Bandwidth Per Job}
2382 The speed parameter specifies the maximum allowed bandwidth that a job may use
2383 when started for this Client. The speed parameter should be specified in
2384 k/s, Kb/s, m/s or Mb/s.
2386 \item [Priority = \lt{}number\gt{}]
2387 \index[dir]{Priority}
2388 \index[dir]{Directive!Priority}
2389 The number specifies the priority of this client relative to other clients
2390 that the Director is processing simultaneously. The priority can range from
2391 1 to 1000. The clients are ordered such that the smaller number priorities
2392 are performed first (not currently implemented).
2395 The following is an example of a valid Client resource definition:
2401 FDAddress = minimatou
2403 Password = very_good
2408 \section{The Storage Resource}
2409 \label{StorageResource2}
2410 \index[general]{Resource!Storage}
2411 \index[general]{Storage Resource}
2413 The Storage resource defines which Storage daemons are available for use by
2419 \index[dir]{Storage}
2420 \index[dir]{Directive!Storage}
2421 Start of the Storage resources. At least one storage resource must be
2424 \item [Name = \lt{}name\gt{}]
2426 \index[dir]{Directive!Name}
2427 The name of the storage resource. This name appears on the Storage directive
2428 specified in the Job resource and is required.
2430 \item [Address = \lt{}address\gt{}]
2431 \index[dir]{Address}
2432 \index[dir]{Directive!SD Address}
2433 \index[dir]{Storage daemon Address}
2434 Where the address is a host name, a {\bf fully qualified domain name}, or an
2435 {\bf IP address}. Please note that the \lt{}address\gt{} as specified here
2436 will be transmitted to the File daemon who will then use it to contact the
2437 Storage daemon. Hence, it is {\bf not}, a good idea to use {\bf localhost} as
2438 the name but rather a fully qualified machine name or an IP address. This
2439 directive is required.
2441 \item [SD Port = \lt{}port\gt{}]
2442 \index[dir]{SD Port}
2443 \index[dir]{Directive!SD Port}
2444 Where port is the port to use to contact the storage daemon for information
2445 and to start jobs. This same port number must appear in the Storage resource
2446 of the Storage daemon's configuration file. The default is 9103.
2448 \item [Password = \lt{}password\gt{}]
2449 \index[dir]{Password}
2450 \index[dir]{Directive!Password}
2451 This is the password to be used when establishing a connection with the
2452 Storage services. This same password also must appear in the Director
2453 resource of the Storage daemon's configuration file. This directive is
2454 required. If you have either {\bf /dev/random} {\bf bc} on your machine,
2455 Bacula will generate a random password during the configuration process,
2456 otherwise it will be left blank.
2458 The password is plain text. It is not generated through any special
2459 process, but it is preferable for security reasons to use random text.
2461 \item [Device = \lt{}device-name\gt{}]
2463 \index[dir]{Directive!Device}
2464 This directive specifies the Storage daemon's name of the device
2465 resource to be used for the storage. If you are using an Autochanger,
2466 the name specified here should be the name of the Storage daemon's
2467 Autochanger resource rather than the name of an individual device. This
2468 name is not the physical device name, but the logical device name as
2469 defined on the {\bf Name} directive contained in the {\bf Device} or the
2470 {\bf Autochanger} resource definition of the {\bf Storage daemon}
2471 configuration file. You can specify any name you would like (even the
2472 device name if you prefer) up to a maximum of 127 characters in length.
2473 The physical device name associated with this device is specified in the
2474 {\bf Storage daemon} configuration file (as {\bf Archive Device}).
2475 Please take care not to define two different Storage resource directives
2476 in the Director that point to the same Device in the Storage daemon.
2477 Doing so may cause the Storage daemon to block (or hang) attempting to
2478 open the same device that is already open. This directive is required.
2481 \item [Media Type = \lt{}MediaType\gt{}]
2482 \index[dir]{Media Type}
2483 \index[dir]{Directive!Media Type}
2484 This directive specifies the Media Type to be used to store the data.
2485 This is an arbitrary string of characters up to 127 maximum that you
2486 define. It can be anything you want. However, it is best to make it
2487 descriptive of the storage media (e.g. File, DAT, "HP DLT8000", 8mm,
2488 ...). In addition, it is essential that you make the {\bf Media Type}
2489 specification unique for each storage media type. If you have two DDS-4
2490 drives that have incompatible formats, or if you have a DDS-4 drive and
2491 a DDS-4 autochanger, you almost certainly should specify different {\bf
2492 Media Types}. During a restore, assuming a {\bf DDS-4} Media Type is
2493 associated with the Job, Bacula can decide to use any Storage daemon
2494 that supports Media Type {\bf DDS-4} and on any drive that supports it.
2496 If you are writing to disk Volumes, you must make doubly sure that each
2497 Device resource defined in the Storage daemon (and hence in the
2498 Director's conf file) has a unique media type. Otherwise for Bacula
2499 versions 1.38 and older, your restores may not work because Bacula
2500 will assume that you can mount any Media Type with the same name on
2501 any Device associated with that Media Type. This is possible with
2502 tape drives, but with disk drives, unless you are very clever you
2503 cannot mount a Volume in any directory -- this can be done by creating
2504 an appropriate soft link.
2506 Currently Bacula permits only a single Media Type per Storage
2507 and Device definition. Consequently, if
2508 you have a drive that supports more than one Media Type, you can
2509 give a unique string to Volumes with different intrinsic Media
2510 Type (Media Type = DDS-3-4 for DDS-3 and DDS-4 types), but then
2511 those volumes will only be mounted on drives indicated with the
2512 dual type (DDS-3-4).
2514 If you want to tie Bacula to using a single Storage daemon or drive, you
2515 must specify a unique Media Type for that drive. This is an important
2516 point that should be carefully understood. Note, this applies equally
2517 to Disk Volumes. If you define more than one disk Device resource in
2518 your Storage daemon's conf file, the Volumes on those two devices are in
2519 fact incompatible because one can not be mounted on the other device
2520 since they are found in different directories. For this reason, you
2521 probably should use two different Media Types for your two disk Devices
2522 (even though you might think of them as both being File types). You can
2523 find more on this subject in the \ilink{Basic Volume
2524 Management}{DiskChapter} chapter of this manual.
2526 The {\bf MediaType} specified in the Director's Storage resource, {\bf
2527 must} correspond to the {\bf Media Type} specified in the {\bf Device}
2528 resource of the {\bf Storage daemon} configuration file. This directive
2529 is required, and it is used by the Director and the Storage daemon to
2530 ensure that a Volume automatically selected from the Pool corresponds to
2531 the physical device. If a Storage daemon handles multiple devices (e.g.
2532 will write to various file Volumes on different partitions), this
2533 directive allows you to specify exactly which device.
2535 As mentioned above, the value specified in the Director's Storage
2536 resource must agree with the value specified in the Device resource in
2537 the {\bf Storage daemon's} configuration file. It is also an additional
2538 check so that you don't try to write data for a DLT onto an 8mm device.
2540 \label{Autochanger1}
2541 \item [Autochanger = \lt{}yes\vb{}no\gt{}]
2542 \index[dir]{Autochanger}
2543 \index[dir]{Directive!Autochanger}
2544 If you specify {\bf yes} for this command (the default is {\bf no}),
2545 when you use the {\bf label} command or the {\bf add} command to create
2546 a new Volume, {\bf Bacula} will also request the Autochanger Slot
2547 number. This simplifies creating database entries for Volumes in an
2548 autochanger. If you forget to specify the Slot, the autochanger will
2549 not be used. However, you may modify the Slot associated with a Volume
2550 at any time by using the {\bf update volume} or {\bf update slots}
2551 command in the console program. When {\bf autochanger} is enabled, the
2552 algorithm used by Bacula to search for available volumes will be
2553 modified to consider only Volumes that are known to be in the
2554 autochanger's magazine. If no {\bf in changer} volume is found, Bacula
2555 will attempt recycling, pruning, ..., and if still no volume is found,
2556 Bacula will search for any volume whether or not in the magazine. By
2557 privileging in changer volumes, this procedure minimizes operator
2558 intervention. The default is {\bf no}.
2560 For the autochanger to be used, you must also specify {\bf Autochanger =
2561 yes} in the \ilink{Device Resource}{Autochanger} in the Storage daemon's
2562 configuration file as well as other important Storage daemon
2563 configuration information. Please consult the \ilink{Using
2564 Autochangers}{AutochangersChapter} manual of this chapter for the
2565 details of using autochangers.
2567 \item [Maximum Concurrent Jobs = \lt{}number\gt{}]
2568 \index[dir]{Maximum Concurrent Jobs}
2569 \index[dir]{Directive!Maximum Concurrent Jobs}
2570 where \lt{}number\gt{} is the maximum number of Jobs with the current
2571 Storage resource that can run concurrently. Note, this directive limits
2572 only Jobs for Jobs using this Storage daemon. Any other restrictions on
2573 the maximum concurrent jobs such as in the Director, Job, or Client
2574 resources will also apply in addition to any limit specified here. The
2575 default is set to 1, but you may set it to a larger number. However, if
2576 you set the Storage daemon's number of concurrent jobs greater than one,
2577 we recommend that you read the waring documented under \ilink{Maximum
2578 Concurrent Jobs}{DirMaxConJobs} in the Director's resource or simply
2579 turn data spooling on as documented in the \ilink{Data
2580 Spooling}{SpoolingChapter} chapter of this manual.
2582 \item [AllowCompression = \lt{}yes\vb{}no\gt{}]
2583 \label{AllowCompression}
2584 \index[dir]{AllowCompression}
2585 \index[dir]{Directive!AllowCompression}
2587 This directive is optional, and if you specify {\bf No} (the default is {\bf
2588 Yes}), it will cause backups jobs running on this storage resource to run
2589 without client File Daemon compression. This effectively overrides
2590 compression options in FileSets used by jobs which use this storage
2593 \item [Heartbeat Interval = \lt{}time-interval\gt{}]
2594 \index[dir]{Heartbeat Interval}
2595 \index[dir]{Directive!Heartbeat}
2596 This directive is optional and if specified will cause the Director to
2597 set a keepalive interval (heartbeat) in seconds on each of the sockets
2598 it opens for the Storage resource. This value will override any
2599 specified at the Director level. It is implemented only on systems
2600 (Linux, ...) that provide the {\bf setsockopt} TCP\_KEEPIDLE function.
2601 The default value is zero, which means no change is made to the socket.
2605 The following is an example of a valid Storage resource definition:
2609 # Definition of tape storage device
2613 Password = storage_password # password for Storage daemon
2614 Device = "HP DLT 80" # same as Device in Storage daemon
2615 Media Type = DLT8000 # same as MediaType in Storage daemon
2620 \section{The Pool Resource}
2621 \label{PoolResource}
2622 \index[general]{Resource!Pool}
2623 \index[general]{Pool Resource}
2625 The Pool resource defines the set of storage Volumes (tapes or files) to be
2626 used by Bacula to write the data. By configuring different Pools, you can
2627 determine which set of Volumes (media) receives the backup data. This permits,
2628 for example, to store all full backup data on one set of Volumes and all
2629 incremental backups on another set of Volumes. Alternatively, you could assign
2630 a different set of Volumes to each machine that you backup. This is most
2631 easily done by defining multiple Pools.
2633 Another important aspect of a Pool is that it contains the default attributes
2634 (Maximum Jobs, Retention Period, Recycle flag, ...) that will be given to a
2635 Volume when it is created. This avoids the need for you to answer a large
2636 number of questions when labeling a new Volume. Each of these attributes can
2637 later be changed on a Volume by Volume basis using the {\bf update} command in
2638 the console program. Note that you must explicitly specify which Pool Bacula
2639 is to use with each Job. Bacula will not automatically search for the correct
2642 Most often in Bacula installations all backups for all machines (Clients) go
2643 to a single set of Volumes. In this case, you will probably only use the {\bf
2644 Default} Pool. If your backup strategy calls for you to mount a different tape
2645 each day, you will probably want to define a separate Pool for each day. For
2646 more information on this subject, please see the
2647 \ilink{Backup Strategies}{StrategiesChapter} chapter of this
2651 To use a Pool, there are three distinct steps. First the Pool must be defined
2652 in the Director's configuration file. Then the Pool must be written to the
2653 Catalog database. This is done automatically by the Director each time that it
2654 starts, or alternatively can be done using the {\bf create} command in the
2655 console program. Finally, if you change the Pool definition in the Director's
2656 configuration file and restart Bacula, the pool will be updated alternatively
2657 you can use the {\bf update pool} console command to refresh the database
2658 image. It is this database image rather than the Director's resource image
2659 that is used for the default Volume attributes. Note, for the pool to be
2660 automatically created or updated, it must be explicitly referenced by a Job
2663 Next the physical media must be labeled. The labeling can either be done with
2664 the {\bf label} command in the {\bf console} program or using the {\bf btape}
2665 program. The preferred method is to use the {\bf label} command in the {\bf
2668 Finally, you must add Volume names (and their attributes) to the Pool. For
2669 Volumes to be used by Bacula they must be of the same {\bf Media Type} as the
2670 archive device specified for the job (i.e. if you are going to back up to a
2671 DLT device, the Pool must have DLT volumes defined since 8mm volumes cannot be
2672 mounted on a DLT drive). The {\bf Media Type} has particular importance if you
2673 are backing up to files. When running a Job, you must explicitly specify which
2674 Pool to use. Bacula will then automatically select the next Volume to use from
2675 the Pool, but it will ensure that the {\bf Media Type} of any Volume selected
2676 from the Pool is identical to that required by the Storage resource you have
2677 specified for the Job.
2679 If you use the {\bf label} command in the console program to label the
2680 Volumes, they will automatically be added to the Pool, so this last step is
2681 not normally required.
2683 It is also possible to add Volumes to the database without explicitly labeling
2684 the physical volume. This is done with the {\bf add} console command.
2686 As previously mentioned, each time Bacula starts, it scans all the Pools
2687 associated with each Catalog, and if the database record does not already
2688 exist, it will be created from the Pool Resource definition. {\bf Bacula}
2689 probably should do an {\bf update pool} if you change the Pool definition, but
2690 currently, you must do this manually using the {\bf update pool} command in
2691 the Console program.
2693 The Pool Resource defined in the Director's configuration file
2694 (bacula-dir.conf) may contain the following directives:
2700 \index[dir]{Directive!Pool}
2701 Start of the Pool resource. There must be at least one Pool resource
2705 \item [Name = \lt{}name\gt{}]
2707 \index[dir]{Directive!Name}
2708 The name of the pool. For most applications, you will use the default
2709 pool name {\bf Default}. This directive is required.
2712 \item [Maximum Volumes = \lt{}number\gt{}]
2713 \index[dir]{Maximum Volumes}
2714 \index[dir]{Directive!Maximum Volumes}
2715 This directive specifies the maximum number of volumes (tapes or files)
2716 contained in the pool. This directive is optional, if omitted or set to
2717 zero, any number of volumes will be permitted. In general, this
2718 directive is useful for Autochangers where there is a fixed number of
2719 Volumes, or for File storage where you wish to ensure that the backups
2720 made to disk files do not become too numerous or consume too much space.
2722 \item [Pool Type = \lt{}type\gt{}]
2723 \index[dir]{Pool Type}
2724 \index[dir]{Directive!Pool Type}
2725 This directive defines the pool type, which corresponds to the type of
2726 Job being run. It is required and may be one of the following:
2736 Note, only Backup is current implemented.
2738 \item [Storage = \lt{}storage-resource-name\gt{}]
2739 \index[dir]{Storage}
2740 \index[dir]{Directive!Storage}
2741 The Storage directive defines the name of the storage services where you
2742 want to backup the FileSet data. For additional details, see the
2743 \ilink{Storage Resource Chapter}{StorageResource2} of this manual.
2744 The Storage resource may also be specified in the Job resource,
2745 but the value, if any, in the Pool resource overrides any value
2746 in the Job. This Storage resource definition is not required by either
2747 the Job resource or in the Pool, but it must be specified in
2748 one or the other. If not configuration error will result.
2750 \item [Use Volume Once = \lt{}yes\vb{}no\gt{}]
2751 \index[dir]{Use Volume Once}
2752 \index[dir]{Directive!Use Volume Once}
2753 This directive if set to {\bf yes} specifies that each volume is to be
2754 used only once. This is most useful when the Media is a file and you
2755 want a new file for each backup that is done. The default is {\bf no}
2756 (i.e. use volume any number of times). This directive will most likely
2757 be phased out (deprecated), so you are recommended to use {\bf Maximum
2758 Volume Jobs = 1} instead.
2760 The value defined by this directive in the bacula-dir.conf file is the
2761 default value used when a Volume is created. Once the volume is
2762 created, changing the value in the bacula-dir.conf file will not change
2763 what is stored for the Volume. To change the value for an existing
2764 Volume you must use the {\bf update} command in the Console.
2766 Please see the notes below under {\bf Maximum Volume Jobs} concerning
2767 using this directive with multiple simultaneous jobs.
2769 \item [Maximum Volume Jobs = \lt{}positive-integer\gt{}]
2770 \index[dir]{Maximum Volume Jobs}
2771 \index[dir]{Directive!Maximum Volume Jobs}
2772 This directive specifies the maximum number of Jobs that can be written
2773 to the Volume. If you specify zero (the default), there is no limit.
2774 Otherwise, when the number of Jobs backed up to the Volume equals {\bf
2775 positive-integer} the Volume will be marked {\bf Used}. When the Volume
2776 is marked {\bf Used} it can no longer be used for appending Jobs, much
2777 like the {\bf Full} status but it can be recycled if recycling is
2778 enabled, and thus used again. By setting {\bf MaximumVolumeJobs} to
2779 one, you get the same effect as setting {\bf UseVolumeOnce = yes}.
2781 The value defined by this directive in the bacula-dir.conf
2782 file is the default value used when a Volume is created. Once the volume is
2783 created, changing the value in the bacula-dir.conf file will not change what
2784 is stored for the Volume. To change the value for an existing Volume you
2785 must use the {\bf update} command in the Console.
2787 If you are running multiple simultaneous jobs, this directive may not
2788 work correctly because when a drive is reserved for a job, this
2789 directive is not taken into account, so multiple jobs may try to
2790 start writing to the Volume. At some point, when the Media record is
2791 updated, multiple simultaneous jobs may fail since the Volume can no
2794 \item [Maximum Volume Files = \lt{}positive-integer\gt{}]
2795 \index[dir]{Maximum Volume Files}
2796 \index[dir]{Directive!Maximum Volume Files}
2797 This directive specifies the maximum number of files that can be written
2798 to the Volume. If you specify zero (the default), there is no limit.
2799 Otherwise, when the number of files written to the Volume equals {\bf
2800 positive-integer} the Volume will be marked {\bf Used}. When the Volume
2801 is marked {\bf Used} it can no longer be used for appending Jobs, much
2802 like the {\bf Full} status but it can be recycled if recycling is
2803 enabled and thus used again. This value is checked and the {\bf Used}
2804 status is set only at the end of a job that writes to the particular
2807 The value defined by this directive in the bacula-dir.conf file is the
2808 default value used when a Volume is created. Once the volume is
2809 created, changing the value in the bacula-dir.conf file will not change
2810 what is stored for the Volume. To change the value for an existing
2811 Volume you must use the {\bf update} command in the Console.
2813 \item [Maximum Volume Bytes = \lt{}size\gt{}]
2814 \index[dir]{Maximum Volume Bytes}
2815 \index[dir]{Directive!Maximum Volume Bytes}
2816 This directive specifies the maximum number of bytes that can be written
2817 to the Volume. If you specify zero (the default), there is no limit
2818 except the physical size of the Volume. Otherwise, when the number of
2819 bytes written to the Volume equals {\bf size} the Volume will be marked
2820 {\bf Used}. When the Volume is marked {\bf Used} it can no longer be
2821 used for appending Jobs, much like the {\bf Full} status but it can be
2822 recycled if recycling is enabled, and thus the Volume can be re-used
2823 after recycling. This value is checked and the {\bf Used} status set
2824 while the job is writing to the particular volume.
2826 This directive is particularly useful for restricting the size
2827 of disk volumes, and will work correctly even in the case of
2828 multiple simultaneous jobs writing to the volume.
2830 The value defined by this directive in the bacula-dir.conf file is the
2831 default value used when a Volume is created. Once the volume is
2832 created, changing the value in the bacula-dir.conf file will not change
2833 what is stored for the Volume. To change the value for an existing
2834 Volume you must use the {\bf update} command in the Console.
2836 \item [Volume Use Duration = \lt{}time-period-specification\gt{}]
2837 \index[dir]{Volume Use Duration}
2838 \index[dir]{Directive!Volume Use Duration}
2839 The Volume Use Duration directive defines the time period that the
2840 Volume can be written beginning from the time of first data write to the
2841 Volume. If the time-period specified is zero (the default), the Volume
2842 can be written indefinitely. Otherwise, the next time a job
2843 runs that wants to access this Volume, and the time period from the
2844 first write to the volume (the first Job written) exceeds the
2845 time-period-specification, the Volume will be marked {\bf Used}, which
2846 means that no more Jobs can be appended to the Volume, but it may be
2847 recycled if recycling is enabled. Using the command {\bf
2848 status dir} applies algorithms similar to running jobs, so
2849 during such a command, the Volume status may also be changed.
2851 recycled, it will be available for use again.
2853 You might use this directive, for example, if you have a Volume used for
2854 Incremental backups, and Volumes used for Weekly Full backups. Once the
2855 Full backup is done, you will want to use a different Incremental
2856 Volume. This can be accomplished by setting the Volume Use Duration for
2857 the Incremental Volume to six days. I.e. it will be used for the 6
2858 days following a Full save, then a different Incremental volume will be
2859 used. Be careful about setting the duration to short periods such as 23
2860 hours, or you might experience problems of Bacula waiting for a tape
2861 over the weekend only to complete the backups Monday morning when an
2862 operator mounts a new tape.
2864 The use duration is checked and the {\bf Used} status is set only at the
2865 end of a job that writes to the particular volume, which means that even
2866 though the use duration may have expired, the catalog entry will not be
2867 updated until the next job that uses this volume is run. This
2868 directive is not intended to be used to limit volume sizes
2869 and will not work correctly (i.e. will fail jobs) if the use
2870 duration expires while multiple simultaneous jobs are writing
2873 Please note that the value defined by this directive in the bacula-dir.conf
2874 file is the default value used when a Volume is created. Once the volume is
2875 created, changing the value in the bacula-dir.conf file will not change what
2876 is stored for the Volume. To change the value for an existing Volume you
2878 \ilink{\bf update volume}{UpdateCommand} command in the Console.
2880 \item [Catalog Files = \lt{}yes\vb{}no\gt{}]
2881 \index[dir]{Catalog Files}
2882 \index[dir]{Directive!Catalog Files}
2883 This directive defines whether or not you want the names of the files
2884 that were saved to be put into the catalog. The default is {\bf yes}.
2885 The advantage of specifying {\bf Catalog Files = No} is that you will
2886 have a significantly smaller Catalog database. The disadvantage is that
2887 you will not be able to produce a Catalog listing of the files backed up
2888 for each Job (this is often called Browsing). Also, without the File
2889 entries in the catalog, you will not be able to use the Console {\bf
2890 restore} command nor any other command that references File entries.
2892 \label{PoolAutoPrune}
2893 \item [AutoPrune = \lt{}yes\vb{}no\gt{}]
2894 \index[dir]{AutoPrune}
2895 \index[dir]{Directive!AutoPrune}
2896 If AutoPrune is set to {\bf yes} (default), Bacula (version 1.20 or
2897 greater) will automatically apply the Volume Retention period when new
2898 Volume is needed and no appendable Volumes exist in the Pool. Volume
2899 pruning causes expired Jobs (older than the {\bf Volume Retention}
2900 period) to be deleted from the Catalog and permits possible recycling of
2903 \label{VolRetention}
2904 \item [Volume Retention = \lt{}time-period-specification\gt{}]
2905 \index[dir]{Volume Retention}
2906 \index[dir]{Directive!Volume Retention}
2907 The Volume Retention directive defines the length of time that {\bf
2908 Bacula} will keep records associated with the Volume in
2909 the Catalog database after the End time of each Job written to the
2910 Volume. When this time period expires, and if {\bf AutoPrune} is set to
2911 {\bf yes} Bacula may prune (remove) Job records that are older than the
2912 specified Volume Retention period if it is necessary to free up a
2913 Volume. Recycling will not occur until it is absolutely necessary to
2914 free up a volume (i.e. no other writable volume exists).
2915 All File records associated with pruned Jobs are also
2916 pruned. The time may be specified as seconds, minutes, hours, days,
2917 weeks, months, quarters, or years. The {\bf Volume Retention} is
2918 applied independently of the {\bf Job Retention} and the {\bf File
2919 Retention} periods defined in the Client resource. This means that all
2920 the retentions periods are applied in turn and that the shorter period
2921 is the one that effectively takes precedence. Note, that when the {\bf
2922 Volume Retention} period has been reached, and it is necessary to obtain
2923 a new volume, Bacula will prune both the Job and the File records. This
2924 pruning could also occur during a {\bf status dir} command because it
2925 uses similar algorithms for finding the next available Volume.
2927 It is important to know that when the Volume Retention period expires,
2928 Bacula does not automatically recycle a Volume. It attempts to keep the
2929 Volume data intact as long as possible before over writing the Volume.
2931 By defining multiple Pools with different Volume Retention periods, you
2932 may effectively have a set of tapes that is recycled weekly, another
2933 Pool of tapes that is recycled monthly and so on. However, one must
2934 keep in mind that if your {\bf Volume Retention} period is too short, it
2935 may prune the last valid Full backup, and hence until the next Full
2936 backup is done, you will not have a complete backup of your system, and
2937 in addition, the next Incremental or Differential backup will be
2938 promoted to a Full backup. As a consequence, the minimum {\bf Volume
2939 Retention} period should be at twice the interval of your Full backups.
2940 This means that if you do a Full backup once a month, the minimum Volume
2941 retention period should be two months.
2943 The default Volume retention period is 365 days, and either the default
2944 or the value defined by this directive in the bacula-dir.conf file is
2945 the default value used when a Volume is created. Once the volume is
2946 created, changing the value in the bacula-dir.conf file will not change
2947 what is stored for the Volume. To change the value for an existing
2948 Volume you must use the {\bf update} command in the Console.
2950 \item [Action On Purge = \lt{Truncate}]
2951 \index[dir]{actiononpurge}
2953 This directive \textbf{ActionOnPurge=Truncate} instructs Bacula to truncate the
2954 volume when it is purged with the \texttt{purge volume action=truncate}
2955 command. It is useful to prevent disk based volumes from consuming too much
2961 Action On Purge = Truncate
2966 You can schedule the truncate operation at the end of your CatalogBackup job
2967 like in this example:
2971 Name = CatalogBackup
2976 Console = "purge volume action=all allpools storage=File"
2981 \label{PoolScratchPool}
2982 \item [ScratchPool = \lt{}pool-resource-name\gt{}]
2983 \index[dir]{ScrachPool}
2984 \index[dir]{Directive!ScrachPool}
2985 This directive permits to specify a dedicate \textsl{Scratch} for the
2986 current pool. This pool will replace the special pool named \textsl{Scrach}
2987 for volume selection. For more information about \textsl{Scratch} see
2988 \ilink{Scratch Pool}{TheScratchPool} section of this manual. This is useful
2989 when using multiple storage sharing the same mediatype or when you want to
2990 dedicate volumes to a particular set of pool.
2992 \label{PoolRecyclePool}
2993 \item [RecyclePool = \lt{}pool-resource-name\gt{}]
2994 \index[dir]{RecyclePool}
2995 \index[dir]{Directive!RecyclePool}
2996 This directive defines to which pool
2997 the Volume will be placed (moved) when it is recycled. Without
2998 this directive, a Volume will remain in the same pool when it is
2999 recycled. With this directive, it can be moved automatically to any
3000 existing pool during a recycle. This directive is probably most
3001 useful when defined in the Scratch pool, so that volumes will
3002 be recycled back into the Scratch pool. For more on the see the
3003 \ilink{Scratch Pool}{TheScratchPool} section of this manual.
3005 Although this directive is called RecyclePool, the Volume in
3006 question is actually moved from its current pool to the one
3007 you specify on this directive when Bacula prunes the Volume and
3008 discovers that there are no records left in the catalog and hence
3009 marks it as {\bf Purged}.
3013 \item [Recycle = \lt{}yes\vb{}no\gt{}]
3014 \index[dir]{Recycle}
3015 \index[dir]{Directive!Recycle}
3016 This directive specifies whether or not Purged Volumes may be recycled.
3017 If it is set to {\bf yes} (default) and Bacula needs a volume but finds
3018 none that are appendable, it will search for and recycle (reuse) Purged
3019 Volumes (i.e. volumes with all the Jobs and Files expired and thus
3020 deleted from the Catalog). If the Volume is recycled, all previous data
3021 written to that Volume will be overwritten. If Recycle is set to {\bf
3022 no}, the Volume will not be recycled, and hence, the data will remain
3023 valid. If you want to reuse (re-write) the Volume, and the recycle flag
3024 is no (0 in the catalog), you may manually set the recycle flag (update
3025 command) for a Volume to be reused.
3027 Please note that the value defined by this directive in the
3028 bacula-dir.conf file is the default value used when a Volume is created.
3029 Once the volume is created, changing the value in the bacula-dir.conf
3030 file will not change what is stored for the Volume. To change the value
3031 for an existing Volume you must use the {\bf update} command in the
3034 When all Job and File records have been pruned or purged from the
3035 catalog for a particular Volume, if that Volume is marked as
3036 Append, Full, Used, or Error, it will then be marked as Purged. Only
3037 Volumes marked as Purged will be considered to be converted to the
3038 Recycled state if the {\bf Recycle} directive is set to {\bf yes}.
3041 \label{RecycleOldest}
3042 \item [Recycle Oldest Volume = \lt{}yes\vb{}no\gt{}]
3043 \index[dir]{Recycle Oldest Volume}
3044 \index[dir]{Directive!Recycle Oldest Volume}
3045 This directive instructs the Director to search for the oldest used
3046 Volume in the Pool when another Volume is requested by the Storage
3047 daemon and none are available. The catalog is then {\bf pruned}
3048 respecting the retention periods of all Files and Jobs written to this
3049 Volume. If all Jobs are pruned (i.e. the volume is Purged), then the
3050 Volume is recycled and will be used as the next Volume to be written.
3051 This directive respects any Job, File, or Volume retention periods that
3052 you may have specified, and as such it is {\bf much} better to use this
3053 directive than the Purge Oldest Volume.
3055 This directive can be useful if you have a fixed number of Volumes in the
3056 Pool and you want to cycle through them and you have specified the correct
3059 However, if you use this directive and have only one
3060 Volume in the Pool, you will immediately recycle your Volume if you fill
3061 it and Bacula needs another one. Thus your backup will be totally invalid.
3062 Please use this directive with care. The default is {\bf no}.
3064 \label{RecycleCurrent}
3066 \item [Recycle Current Volume = \lt{}yes\vb{}no\gt{}]
3067 \index[dir]{Recycle Current Volume}
3068 \index[dir]{Directive!Recycle Current Volume}
3069 If Bacula needs a new Volume, this directive instructs Bacula to Prune
3070 the volume respecting the Job and File retention periods. If all Jobs
3071 are pruned (i.e. the volume is Purged), then the Volume is recycled and
3072 will be used as the next Volume to be written. This directive respects
3073 any Job, File, or Volume retention periods that you may have specified,
3074 and thus it is {\bf much} better to use it rather than the Purge Oldest
3077 This directive can be useful if you have: a fixed number of Volumes in
3078 the Pool, you want to cycle through them, and you have specified
3079 retention periods that prune Volumes before you have cycled through the
3082 However, if you use this directive and have only one Volume in the Pool,
3083 you will immediately recycle your Volume if you fill it and Bacula needs
3084 another one. Thus your backup will be totally invalid. Please use this
3085 directive with care. The default is {\bf no}.
3089 \item [Purge Oldest Volume = \lt{}yes\vb{}no\gt{}]
3090 \index[dir]{Purge Oldest Volume}
3091 \index[dir]{Directive!Purge Oldest Volume}
3092 This directive instructs the Director to search for the oldest used
3093 Volume in the Pool when another Volume is requested by the Storage
3094 daemon and none are available. The catalog is then {\bf purged}
3095 irrespective of retention periods of all Files and Jobs written to this
3096 Volume. The Volume is then recycled and will be used as the next Volume
3097 to be written. This directive overrides any Job, File, or Volume
3098 retention periods that you may have specified.
3100 This directive can be useful if you have a fixed number of Volumes in
3101 the Pool and you want to cycle through them and reusing the oldest one
3102 when all Volumes are full, but you don't want to worry about setting
3103 proper retention periods. However, by using this option you risk losing
3106 Please be aware that {\bf Purge Oldest Volume} disregards all retention
3107 periods. If you have only a single Volume defined and you turn this
3108 variable on, that Volume will always be immediately overwritten when it
3109 fills! So at a minimum, ensure that you have a decent number of Volumes
3110 in your Pool before running any jobs. If you want retention periods to
3111 apply do not use this directive. To specify a retention period, use the
3112 {\bf Volume Retention} directive (see above).
3114 We {\bf highly} recommend against using this directive, because it is
3115 sure that some day, Bacula will recycle a Volume that contains current
3116 data. The default is {\bf no}.
3118 \item [File Retention = \lt{}time-period-specification\gt{}]
3119 \index[dir]{File Retention}
3120 \index[dir]{Directive!File Retention}
3121 The File Retention directive defines the length of time that Bacula will
3122 keep File records in the Catalog database after the End time of the
3123 Job corresponding to the File records.
3125 This directive takes precedence over Client directives of the same name. For
3126 example, you can decide to increase Retention times for Archive or OffSite
3129 Note, this affects only records in the catalog database. It does not affect
3130 your archive backups.
3132 For more information see Client documentation about
3133 \ilink{FileRetention}{FileRetention}
3135 \item [Job Retention = \lt{}time-period-specification\gt{}]
3136 \index[dir]{Job Retention}
3137 \index[dir]{Directive!Job Retention}
3139 The Job Retention directive defines the length of time that Bacula will keep
3140 Job records in the Catalog database after the Job End time. As with the
3141 other retention periods, this affects only records in the catalog and not
3142 data in your archive backup.
3144 This directive takes precedence over Client directives of the same name.
3145 For example, you can decide to increase Retention times for Archive or
3148 For more information see Client side documentation
3149 \ilink{JobRetention}{JobRetention}
3151 \item [Cleaning Prefix = \lt{}string\gt{}]
3152 \index[dir]{Cleaning Prefix}
3153 \index[dir]{Directive!Cleaning Prefix}
3154 This directive defines a prefix string, which if it matches the
3155 beginning of a Volume name during labeling of a Volume, the Volume will
3156 be defined with the VolStatus set to {\bf Cleaning} and thus Bacula will
3157 never attempt to use this tape. This is primarily for use with
3158 autochangers that accept barcodes where the convention is that barcodes
3159 beginning with {\bf CLN} are treated as cleaning tapes.
3162 \item [Label Format = \lt{}format\gt{}]
3163 \index[dir]{Label Format}
3164 \index[dir]{Directive!Label Format}
3165 This directive specifies the format of the labels contained in this
3166 pool. The format directive is used as a sort of template to create new
3167 Volume names during automatic Volume labeling.
3169 The {\bf format} should be specified in double quotes, and consists of
3170 letters, numbers and the special characters hyphen ({\bf -}), underscore
3171 ({\bf \_}), colon ({\bf :}), and period ({\bf .}), which are the legal
3172 characters for a Volume name. The {\bf format} should be enclosed in
3175 In addition, the format may contain a number of variable expansion
3176 characters which will be expanded by a complex algorithm allowing you to
3177 create Volume names of many different formats. In all cases, the
3178 expansion process must resolve to the set of characters noted above that
3179 are legal Volume names. Generally, these variable expansion characters
3180 begin with a dollar sign ({\bf \$}) or a left bracket ({\bf [}). If you
3181 specify variable expansion characters, you should always enclose the
3182 format with double quote characters ({\bf "}). For more details on
3183 variable expansion, please see the \ilink{Variable
3184 Expansion}{VarsChapter} Chapter of this manual.
3186 If no variable expansion characters are found in the string, the Volume
3187 name will be formed from the {\bf format} string appended with the
3188 a unique number that increases. If you do not remove volumes from the
3189 pool, this number should be the number of volumes plus one, but this
3190 is not guaranteed. The unique number will be edited as four
3191 digits with leading zeros. For example, with a {\bf Label Format =
3192 "File-"}, the first volumes will be named {\bf File-0001}, {\bf
3195 With the exception of Job specific variables, you can test your {\bf
3196 LabelFormat} by using the \ilink{ var command}{var} the Console Chapter
3199 In almost all cases, you should enclose the format specification (part
3200 after the equal sign) in double quotes. Please note that this directive
3201 is deprecated and is replaced in version 1.37 and greater with a Python
3202 script for creating volume names.
3206 In order for a Pool to be used during a Backup Job, the Pool must have at
3207 least one Volume associated with it. Volumes are created for a Pool using
3208 the {\bf label} or the {\bf add} commands in the {\bf Bacula Console},
3209 program. In addition to adding Volumes to the Pool (i.e. putting the
3210 Volume names in the Catalog database), the physical Volume must be labeled
3211 with a valid Bacula software volume label before {\bf Bacula} will accept
3212 the Volume. This will be automatically done if you use the {\bf label}
3213 command. Bacula can automatically label Volumes if instructed to do so,
3214 but this feature is not yet fully implemented.
3216 The following is an example of a valid Pool resource definition:
3228 \subsection{The Scratch Pool}
3229 \label{TheScratchPool}
3230 \index[general]{Scratch Pool}
3231 In general, you can give your Pools any name you wish, but there is one
3232 important restriction: the Pool named {\bf Scratch}, if it exists behaves
3233 like a scratch pool of Volumes in that when Bacula needs a new Volume for
3234 writing and it cannot find one, it will look in the Scratch pool, and if
3235 it finds an available Volume, it will move it out of the Scratch pool into
3236 the Pool currently being used by the job.
3239 \section{The Catalog Resource}
3240 \label{CatalogResource}
3241 \index[general]{Resource!Catalog}
3242 \index[general]{Catalog Resource}
3244 The Catalog Resource defines what catalog to use for the current job.
3245 Currently, Bacula can only handle a single database server (SQLite, MySQL,
3246 PostgreSQL) that is defined when configuring {\bf Bacula}. However, there
3247 may be as many Catalogs (databases) defined as you wish. For example, you
3248 may want each Client to have its own Catalog database, or you may want
3249 backup jobs to use one database and verify or restore jobs to use another
3252 Since SQLite is compiled in, it always runs on the same machine
3253 as the Director and the database must be directly accessible (mounted) from
3254 the Director. However, since both MySQL and PostgreSQL are networked
3255 databases, they may reside either on the same machine as the Director
3256 or on a different machine on the network. See below for more details.
3261 \index[dir]{Catalog}
3262 \index[dir]{Directive!Catalog}
3263 Start of the Catalog resource. At least one Catalog resource must be
3267 \item [Name = \lt{}name\gt{}]
3269 \index[dir]{Directive!Name}
3270 The name of the Catalog. No necessary relation to the database server
3271 name. This name will be specified in the Client resource directive
3272 indicating that all catalog data for that Client is maintained in this
3273 Catalog. This directive is required.
3275 \item [password = \lt{}password\gt{}]
3276 \index[dir]{password}
3277 \index[dir]{Directive!password}
3278 This specifies the password to use when logging into the database. This
3279 directive is required.
3281 \item [DB Name = \lt{}name\gt{}]
3282 \index[dir]{DB Name}
3283 \index[dir]{Directive!DB Name}
3284 This specifies the name of the database. If you use multiple catalogs
3285 (databases), you specify which one here. If you are using an external
3286 database server rather than the internal one, you must specify a name
3287 that is known to the server (i.e. you explicitly created the Bacula
3288 tables using this name. This directive is required.
3290 \item [user = \lt{}user\gt{}]
3292 \index[dir]{Directive!user}
3293 This specifies what user name to use to log into the database. This
3294 directive is required.
3296 \item [DB Socket = \lt{}socket-name\gt{}]
3297 \index[dir]{DB Socket}
3298 \index[dir]{Directive!DB Socket}
3299 This is the name of a socket to use on the local host to connect to the
3300 database. This directive is used only by MySQL and is ignored by SQLite.
3301 Normally, if neither {\bf DB Socket} or {\bf DB Address} are specified, MySQL
3302 will use the default socket. If the DB Socket is specified, the
3303 MySQL server must reside on the same machine as the Director.
3305 \item [DB Address = \lt{}address\gt{}]
3306 \index[dir]{DB Address}
3307 \index[dir]{Directive!DB Address}
3308 This is the host address of the database server. Normally, you would specify
3309 this instead of {\bf DB Socket} if the database server is on another machine.
3310 In that case, you will also specify {\bf DB Port}. This directive is used
3311 only by MySQL and PostgreSQL and is ignored by SQLite if provided.
3312 This directive is optional.
3314 \item [DB Port = \lt{}port\gt{}]
3315 \index[dir]{DB Port}
3316 \index[dir]{Directive!DB Port}
3317 This defines the port to be used in conjunction with {\bf DB Address} to
3318 access the database if it is on another machine. This directive is used only
3319 by MySQL and PostgreSQL and is ignored by SQLite if provided. This
3320 directive is optional.
3322 %% \item [Multiple Connections = \lt{}yes\vb{}no\gt{}]
3323 %% \index[dir]{Multiple Connections}
3324 %% \index[dir]{Directive!Multiple Connections}
3325 %% By default, this directive is set to no. In that case, each job that uses
3327 %% same Catalog will use a single connection to the catalog. It will be shared,
3328 %% and Bacula will allow only one Job at a time to communicate. If you set this
3329 %% directive to yes, Bacula will permit multiple connections to the database,
3330 %% and the database must be multi-thread capable. For SQLite and PostgreSQL,
3331 %% this is no problem. For MySQL, you must be *very* careful to have the
3332 %% multi-thread version of the client library loaded on your system. When this
3333 %% directive is set yes, each Job will have a separate connection to the
3334 %% database, and the database will control the interaction between the
3336 %% Jobs. This can significantly speed up the database operations if you are
3337 %% running multiple simultaneous jobs. In addition, for SQLite and PostgreSQL,
3338 %% Bacula will automatically enable transactions. This can significantly speed
3339 %% up insertion of attributes in the database either for a single Job or
3340 %% multiple simultaneous Jobs.
3342 %% This directive has not been tested. Please test carefully before running it
3343 %% in production and report back your results.
3347 The following is an example of a valid Catalog resource definition:
3356 password = "" # no password = no security
3361 or for a Catalog on another machine:
3371 DB Address = remote.acme.com
3377 \section{The Messages Resource}
3378 \label{MessagesResource2}
3379 \index[general]{Resource!Messages}
3380 \index[general]{Messages Resource}
3382 For the details of the Messages Resource, please see the
3383 \ilink{Messages Resource Chapter}{MessagesChapter} of this
3386 \section{The Console Resource}
3387 \label{ConsoleResource1}
3388 \index[general]{Console Resource}
3389 \index[general]{Resource!Console}
3391 As of Bacula version 1.33 and higher, there are three different kinds of
3392 consoles, which the administrator or user can use to interact with the
3393 Director. These three kinds of consoles comprise three different security
3397 \item The first console type is an {\bf anonymous} or {\bf default} console,
3398 which has full privileges. There is no console resource necessary for
3399 this type since the password is specified in the Director's resource and
3400 consequently such consoles do not have a name as defined on a {\bf Name
3401 =} directive. This is the kind of console that was initially
3402 implemented in versions prior to 1.33 and remains valid. Typically you
3403 would use it only for administrators.
3405 \item The second type of console, and new to version 1.33 and higher is a
3406 "named" console defined within a Console resource in both the Director's
3407 configuration file and in the Console's configuration file. Both the
3408 names and the passwords in these two entries must match much as is the
3409 case for Client programs.
3411 This second type of console begins with absolutely no privileges except
3412 those explicitly specified in the Director's Console resource. Thus you
3413 can have multiple Consoles with different names and passwords, sort of
3414 like multiple users, each with different privileges. As a default,
3415 these consoles can do absolutely nothing -- no commands whatsoever. You
3416 give them privileges or rather access to commands and resources by
3417 specifying access control lists in the Director's Console resource. The
3418 ACLs are specified by a directive followed by a list of access names.
3419 Examples of this are shown below.
3421 \item The third type of console is similar to the above mentioned one in that
3422 it requires a Console resource definition in both the Director and the
3423 Console. In addition, if the console name, provided on the {\bf Name =}
3424 directive, is the same as a Client name, that console is permitted to
3425 use the {\bf SetIP} command to change the Address directive in the
3426 Director's client resource to the IP address of the Console. This
3427 permits portables or other machines using DHCP (non-fixed IP addresses)
3428 to "notify" the Director of their current IP address.
3431 The Console resource is optional and need not be specified. The following
3432 directives are permitted within the Director's configuration resource:
3436 \item [Name = \lt{}name\gt{}]
3438 \index[dir]{Directive!Name}
3439 The name of the console. This name must match the name specified in the
3440 Console's configuration resource (much as is the case with Client
3443 \item [Password = \lt{}password\gt{}]
3444 \index[dir]{Password}
3445 \index[dir]{Directive!Password}
3446 Specifies the password that must be supplied for a named Bacula Console
3447 to be authorized. The same password must appear in the {\bf Console}
3448 resource of the Console configuration file. For added security, the
3449 password is never actually passed across the network but rather a
3450 challenge response hash code created with the password. This directive
3451 is required. If you have either {\bf /dev/random} {\bf bc} on your
3452 machine, Bacula will generate a random password during the configuration
3453 process, otherwise it will be left blank.
3455 The password is plain text. It is not generated through any special
3456 process. However, it is preferable for security reasons to choose
3459 \item [JobACL = \lt{}name-list\gt{}]
3461 \index[dir]{Directive!JobACL}
3462 This directive is used to specify a list of Job resource names that can
3463 be accessed by the console. Without this directive, the console cannot
3464 access any of the Director's Job resources. Multiple Job resource names
3465 may be specified by separating them with commas, and/or by specifying
3466 multiple JobACL directives. For example, the directive may be specified
3471 JobACL = kernsave, "Backup client 1", "Backup client 2"
3472 JobACL = "RestoreFiles"
3477 With the above specification, the console can access the Director's resources
3478 for the four jobs named on the JobACL directives, but for no others.
3480 \item [ClientACL = \lt{}name-list\gt{}]
3481 \index[dir]{ClientACL}
3482 \index[dir]{Directive!ClientACL}
3483 This directive is used to specify a list of Client resource names that can
3485 accessed by the console.
3487 \item [StorageACL = \lt{}name-list\gt{}]
3488 \index[dir]{StorageACL}
3489 \index[dir]{Directive!StorageACL}
3490 This directive is used to specify a list of Storage resource names that can
3491 be accessed by the console.
3493 \item [ScheduleACL = \lt{}name-list\gt{}]
3494 \index[dir]{ScheduleACL}
3495 \index[dir]{Directive!ScheduleACL}
3496 This directive is used to specify a list of Schedule resource names that can
3497 be accessed by the console.
3499 \item [PoolACL = \lt{}name-list\gt{}]
3500 \index[dir]{PoolACL}
3501 \index[dir]{Directive!PoolACL}
3502 This directive is used to specify a list of Pool resource names that can be
3503 accessed by the console.
3505 \item [FileSetACL = \lt{}name-list\gt{}]
3506 \index[dir]{FileSetACL}
3507 \index[dir]{Directive!FileSetACL}
3508 This directive is used to specify a list of FileSet resource names that
3509 can be accessed by the console.
3511 \item [CatalogACL = \lt{}name-list\gt{}]
3512 \index[dir]{CatalogACL}
3513 \index[dir]{Directive!CatalogACL}
3514 This directive is used to specify a list of Catalog resource names that
3515 can be accessed by the console.
3517 \item [CommandACL = \lt{}name-list\gt{}]
3518 \index[dir]{CommandACL}
3519 \index[dir]{Directive!CommandACL}
3520 This directive is used to specify a list of of console commands that can
3521 be executed by the console.
3523 \item [WhereACL = \lt{}string\gt{}]
3524 \index[dir]{WhereACL}
3525 \index[dir]{Directive!WhereACL}
3526 This directive permits you to specify where a restricted console
3527 can restore files. If this directive is not specified, only the
3528 default restore location is permitted (normally {\bf
3529 /tmp/bacula-restores}. If {\bf *all*} is specified any path the
3530 user enters will be accepted (not very secure), any other
3531 value specified (there may be multiple WhereACL directives) will
3532 restrict the user to use that path. For example, on a Unix system,
3533 if you specify "/", the file will be restored to the original
3534 location. This directive is untested.
3538 Aside from Director resource names and console command names, the special
3539 keyword {\bf *all*} can be specified in any of the above access control lists.
3540 When this keyword is present, any resource or command name (which ever is
3541 appropriate) will be accepted. For an example configuration file, please see
3543 \ilink{Console Configuration}{ConsoleConfChapter} chapter of this
3546 \section{The Counter Resource}
3547 \label{CounterResource}
3548 \index[general]{Resource!Counter}
3549 \index[general]{Counter Resource}
3551 The Counter Resource defines a counter variable that can be accessed by
3552 variable expansion used for creating Volume labels with the {\bf LabelFormat}
3554 \ilink{LabelFormat}{Label} directive in this chapter for more
3560 \index[dir]{Counter}
3561 \index[dir]{Directive!Counter}
3562 Start of the Counter resource. Counter directives are optional.
3564 \item [Name = \lt{}name\gt{}]
3566 \index[dir]{Directive!Name}
3567 The name of the Counter. This is the name you will use in the variable
3568 expansion to reference the counter value.
3570 \item [Minimum = \lt{}integer\gt{}]
3571 \index[dir]{Minimum}
3572 \index[dir]{Directive!Minimum}
3573 This specifies the minimum value that the counter can have. It also becomes
3574 the default. If not supplied, zero is assumed.
3576 \item [Maximum = \lt{}integer\gt{}]
3577 \index[dir]{Maximum}
3578 \index[dir]{Directive!Maximum}
3579 \index[dir]{Directive!Maximum}
3580 This is the maximum value value that the counter can have. If not specified
3581 or set to zero, the counter can have a maximum value of 2,147,483,648 (2 to
3582 the 31 power). When the counter is incremented past this value, it is reset
3585 \item [*WrapCounter = \lt{}counter-name\gt{}]
3586 \index[dir]{*WrapCounter}
3587 \index[dir]{Directive!*WrapCounter}
3588 If this value is specified, when the counter is incremented past the
3590 and thus reset to the minimum, the counter specified on the {\bf WrapCounter}
3591 is incremented. (This is not currently implemented).
3593 \item [Catalog = \lt{}catalog-name\gt{}]
3594 \index[dir]{Catalog}
3595 \index[dir]{Directive!Catalog}
3596 If this directive is specified, the counter and its values will be saved in
3597 the specified catalog. If this directive is not present, the counter will be
3598 redefined each time that Bacula is started.
3601 \section{Example Director Configuration File}
3602 \label{SampleDirectorConfiguration}
3603 \index[general]{File!Example Director Configuration}
3604 \index[general]{Example Director Configuration File}
3606 An example Director configuration file might be the following:
3611 # Default Bacula Director Configuration file
3613 # The only thing that MUST be changed is to add one or more
3614 # file or directory names in the Include directive of the
3617 # For Bacula release 1.15 (5 March 2002) -- redhat
3619 # You might also want to change the default email address
3620 # from root to your address. See the "mail" and "operator"
3621 # directives in the Messages resource.
3623 Director { # define myself
3625 QueryFile = "/home/kern/bacula/bin/query.sql"
3626 WorkingDirectory = "/home/kern/bacula/bin/working"
3627 PidDirectory = "/home/kern/bacula/bin/working"
3628 Password = "XkSfzu/Cf/wX4L8Zh4G4/yhCbpLcz3YVdmVoQvU3EyF/"
3630 # Define the backup Job
3632 Name = "NightlySave"
3634 Level = Incremental # default
3637 Schedule = "WeeklyCycle"
3647 Where = /tmp/bacula-restores
3653 # List of files to be backed up
3657 Options { signature=SHA1}
3659 # Put your list of files here, one per line or include an
3660 # external list with:
3664 # Note: / backs up everything
3669 # When to do the backups
3671 Name = "WeeklyCycle"
3672 Run = level=Full sun at 2:05
3673 Run = level=Incremental mon-sat at 2:05
3675 # Client (File Services) to backup
3680 Password = "MQk6lVinz4GG2hdIZk1dsKE/LxMZGo6znMHiD7t7vzF+"
3681 File Retention = 60d # sixty day file retention
3682 Job Retention = 1y # 1 year Job retention
3683 AutoPrune = yes # Auto apply retention periods
3685 # Definition of DLT tape storage device
3689 Password = "jMeWZvfikUHvt3kzKVVPpQ0ccmV6emPnF2cPYFdhLApQ"
3690 Device = "HP DLT 80" # same as Device in Storage daemon
3691 Media Type = DLT8000 # same as MediaType in Storage daemon
3693 # Definition for a DLT autochanger device
3697 Password = "jMeWZvfikUHvt3kzKVVPpQ0ccmV6emPnF2cPYFdhLApQ"
3698 Device = "Autochanger" # same as Device in Storage daemon
3699 Media Type = DLT-8000 # Different from DLTDrive
3702 # Definition of DDS tape storage device
3706 Password = "jMeWZvfikUHvt3kzKVVPpQ0ccmV6emPnF2cPYFdhLApQ"
3707 Device = SDT-10000 # same as Device in Storage daemon
3708 Media Type = DDS-4 # same as MediaType in Storage daemon
3710 # Definition of 8mm tape storage device
3714 Password = "jMeWZvfikUHvt3kzKVVPpQ0ccmV6emPnF2cPYFdhLApQ"
3715 Device = "Exabyte 8mm"
3718 # Definition of file storage device
3722 Password = "jMeWZvfikUHvt3kzKVVPpQ0ccmV6emPnF2cPYFdhLApQ"
3723 Device = FileStorage
3726 # Generic catalog service
3729 dbname = bacula; user = bacula; password = ""
3731 # Reasonable message delivery -- send most everything to
3732 # the email address and to the console
3735 mail = root@localhost = all, !skipped, !terminate
3736 operator = root@localhost = mount
3737 console = all, !skipped, !saved
3740 # Default pool definition
3748 # Restricted console used by tray-monitor to get the status of the director
3752 Password = "GN0uRo7PTUmlMbqrJ2Gr1p0fk0HQJTxwnFyE4WSST3MWZseR"
3753 CommandACL = status, .status