2 Release Notes for Bacula 1.37.14
4 Bacula code: Total files = 422 Total lines = 129,844 (*.h *.c *.in)
6 Note! The DB has been upgraded from version 8 to 9 and requres
10 - This version has a new DIR <--> SD protocol. Both must be
11 upgraded at the same time. Deamons are not compatible with
12 prior versions. All FDs must be upgraded.
13 - There is a new database format that is not compatible
14 with previous databases.
15 - SQLite3 support, but it seems to run at 1/2 to 1/4 the speed of
17 - Multiple drive autochanger support.
18 - Support for ANSI/IBM labels.
19 - New communications protocol between DIR and SD to reserve
21 - Python Event support has been added. See below for
22 configuration and details.
23 - DVD writing support, using parts, and a lot of new directives in
24 the Device resource of the Storage configuration file thanks
26 - Seven new options keywords in a FileSet resource:
27 ignorecase, fstype, hfsplussupport, wilddir, wildfile, regexdir,
28 and regexfile thanks to Pruben Guldberg). See below for details.
29 - Restore of all files for a Job or set of jobs even if the file
30 records have been removed from the catalog.
31 - Restore of a directory (non-recursive, i.e. only one level).
32 - Support for TLS (ssl) between all the daemon connections thanks
34 - Any Volume in the Pool named Scratch may be reassigned to any
35 other Pool when a new Volume is needed.
36 - You may clone a Job and thus write (almost) the same data
37 to multiple Volumes simultaneously (see below). Actually,
38 any Job may start any other Job with this mechanism, so it
39 also be used to group jobs.
40 - Unicode filename support for Win32 (thanks to Thorsten Engel)
41 - New manual format with an index (thanks to Karl Cunningham).
42 - New Web site format (thanks to Michael Scherer).
43 - Seeking on disk Volumes during restore.
46 - New Run directive in Job resource of DIR. It permits
47 cloning of jobs. To clone a copy of the current job, use
48 Run = "job-name level=%l since=\"%s\""
49 Note, job-name is normally the same name as the job that
50 is running but there is no restriction on what you put. If you
51 want to start the job by hand and use job overrides such as
52 storage=xxx, realize that the job will be started with the
53 default storage values not the overrides. The level=%l guarantees
54 that the chosen level of the job is the same, and the since=...
55 ensures that the job uses *exactly* the same time/date for incremental
56 and differential jobs. The since=... is ignored when level=Full.
57 A cloned job will not start additional clones, so it is not possible
59 - New Options keywords in a FileSet directive:
61 Will do a wild card match against directories (files will not
64 Will do a wild card match against files (directories will not
67 Will do a regular expression match against directories (files
70 Will do a regular expression match against files( directories
72 - IgnoreCase = yes | no
73 Will ignore case in wild card and regular expression matches.
74 This is handy for Windows where filename case is not significant.
76 where string is a filesystem type: ext2, jfs, ntfs, proc,
77 reiserfs, xfs, usbdevfs, sysfs, smbfs, iso9660. For ext3
78 systems, use ext2. You may have multiple fstype directives
79 and thus permit multiple filesystem types. If the type
80 specified on the fstype directive does not match the
81 filesystem for a particular directive, that directory will
82 not be backed up. This directive can be used to prevent
83 backing up non-local filesystems.
84 - HFS Plus Support = yes | no
85 If set, Mac OS X resource forks will be saved and restored.
86 - Label Type = ANSI | IBM | Bacula
87 Implemented in Director Pool resource and in SD Device resource.
88 If it is specified in the SD Device resource, it will take
89 precedence over the value passed from the Director to the SD.
90 IBM is not yet implemented.
91 - Check Labels = yes | no
92 Implemented in the SD Device resource. If you intend to read
93 ANSI or IBM labels, this *must* be set. Even if the volume
94 is not ANSI labeled, you can set this to yes, and Bacula will
96 - Scripts Directory = <directory> name. Defines the directory from
97 which Bacula scripts will be called for events. In fact, Bacula
98 appends this name to the standard Python list of search directories,
99 so the script could also be in any of the Python system directories.
100 - In FileSet, you can exclude backing up of hardlinks (if you have
101 a lot, it can be very expensive), by using:
103 in the Options section. Patch supplied by David R Bosso. Thanks.
104 - MaximumPartSize = bytes (SD, Device resource)
105 Defines the maximum part size.
106 - Requires Mount = Yes/No (SD, Device resource)
107 Defines if the device require to be mounted to be read, and if it
108 must be written in a special way. If it set, the following directives
109 must be defined in the same Device resource:
110 + Mount Point = directory
111 Directory where the device must be mounted.
112 + Mount Command = name-string
113 Command that must be executed to mount the device. Before the command
114 is executed, %a is replaced with the Archive Device, and %m with the
116 + Unmount Command = name-string
117 Command that must be executed to unmount the device. Before the
118 command is executed, %a is replaced with the Archive Device, and
119 %m with the Mount Point.
120 + Write Part Command = name-string
121 Command that must be executed to write a part to the device. Before
122 the command is executed, %a is replaced with the Archive Device, %m
123 with the Mount Point, %n with the current part number (0-based),
124 and %v with the current part filename.
125 + Free Space Command = name-string
126 Command that must be executed to check how much free space is left
127 on the device. Before the command is executed, %a is replaced with
128 the Archive Device, %m with the Mount Point, %n with the current part
129 number (0-based), and %v with the current part filename.
130 - Write Part After Job = Yes/No (DIR, Job Resource, and Schedule Resource)
131 If this directive is set to yes (default no), a new part file will be
132 created after the job is finished.
135 - "python restart" restarts the Python interpreter. Rather brutal, make
136 sure no Python scripts are running. This permits you to change
137 a Python script and get Bacula to use the new script.
140 - You must add --with-python=[DIR] to the configure command line
141 if you want Python support. Python 2.2 and 2.3 should be automatically
142 detected if in the standard place.
143 - With Python 2.2 version, the link of the Director gets a few linker
144 warnings due to the fact that Python pulls in some old non-secure
146 - With Python 2.3, there are a few compiler warnings.
147 - You must either create a new catalog database or upgrade your
148 old database (the PostgreSQL upgrade scripts not yet tested).
151 - 2 new scripts, dvd-writepart and dvd-freespace, in the scripts directory,
152 which are designed to be used as parameters to Write Part Command and
153 Free Space Command. They need the dvd+rw-tools to be installed
154 (http://fy.chalmers.se/~appro/linux/DVD+RW/).
155 - Part files support: File volumes can now be splitted in multiple
156 files, called "parts".
158 NOTE THE FOLLOWING IS NOW OUT OF DATE!!!!
159 - Python scripting support:
160 A Python script will be called at particular points or conditions
161 in Bacula called Events. The currently defined Events are called:
163 StartJob, EndJob, NewVolume
165 Where StartJob is called before the RunBeforeJob, EndJob is called
166 after RunAfterJob, and NewVolume, is called before all other
167 "methods" of obtaining a new Volume name, when one is needed.
169 The Python script of the same name as the Event name (but with a .py)
170 is called from the Scripts Directory (a directive defined in the
171 Director resource). Note, both the Filename, and the name of
172 the function in the file must correspond to the Event name.
174 Once the Python script gets control, it can have access to Bacula
179 The script is called with one argument, typically called jcr. This
180 argument *must* be used to access each bacula function. The
181 format of the call is slightly different for reading Bacula
182 variable and for writing bacula variables. See below.
184 Bacula variables can be read with:
186 jcr.get("Variable-name")
188 where jcr is the argument passed to the function, and Variable-name
189 is on of the following:
191 JobId, Client, Pool, Storage, Catalog, MediaType, NumVols, DirName,
192 Level, Type, Job, JobName, JobStatus
194 Bacula varibles can be set using Python keyword arguments:
196 jcr.set(VolumeName="xyz")
198 The two currently implemented writable "variables" are:
200 VolumeName and JobReport
202 It is possible to submit a Bacula run command with the following:
204 jcr.run("run kernsave client=Matou storage=File")
206 this function returns the JobId of the job that was started. If
207 there is an error, the return value is zero.
211 == File EndJob.py ===
215 jobid = jcr.get("JobId")
216 client = jcr.get("Client")
217 jcr.set(JobReport="EndJob output: JobId=%d Client=%s.\n" % (jobid, client))
219 startid = jcr.run("run kernsave")
220 print "Python started jobid=", startid
225 == File NewVolume.py ===
229 jobid = jcr.get("JobId")
230 print "JobId=", jobid
231 client = jcr.get("Client")
232 print "Client=" + client
233 numvol = jcr.get("NumVols");
234 print "NumVols=", numvol
235 jcr.set(JobReport="New Volume set for Job.\n")
236 jcr.set(VolumeName="TestA-001")