4 Documentation to do: (any release a little bit at a time)
5 - Document running a test version.
6 - Document query file format.
7 - Document static linking
8 - Document problems with Verify and pruning.
9 - Document how to use multiple databases.
10 - Add a section to the doc on Manual cycling of Volumes.
13 Testing to do: (painful)
14 - that ALL console command line options work and are always implemented
15 - blocksize recognition code.
16 - multiple simultaneous Volumes
17 - Test if rewind at end of tape waits for tape to rewind.
19 - Figure out how to use ssh or stunnel to protect Bacula communications.
22 - Check if Job/File retentions apply to multivolume jobs.
23 - Add Progress command that periodically reports the progress of
25 - Implement "Reschedule OnError=yes interval=nnn times=xxx"
26 - Shell character expansion is failing occassionally.
27 - One block was orphaned in the SD probably after cancel.
30 rufus-dir: Max configured use duration exceeded. Marking Volume "MatouBackup" as Used.
31 rufus-sd: Volume "" previously written, moving to end of data.
32 rufus-sd: Matou.2003-05-10_10.39.18 Error: I canot write on this volume because:
33 The number of files mismatch! Volume=1 Catalog=0
34 rufus-sd: Matou.2003-05-10_10.39.18 Error: askdir.c:155 NULL Volume name. This shouldn't happen!!!
36 - Fix "access not allowed" for backup of files on WinXP.
37 - Check for existence of all new Win32 API's. See LoadLibrary in
39 - Fix problem reported by Christopher McCurdy <cmccurdy@eecis.udel.edu>
40 xeon-fd: Could not stat c:/Documents and Settings/All
41 Users/Application Data/Humc:\Documents and Settings\All User98_AIX.kbf:
42 ERR=No such file or directory
43 - Implement argv/argk in place of sscanf in the daemon protocol.
44 - Examine Bare Metal restore problem (a FD crash exists somewhere ...).
45 - Test multiple simultaneous Volumes
46 - Document FInclude ...
47 - Implement timeout in response() when it should come quickly.
48 - Implement console @echo command.
49 - Bug: fix access problems on files restored on WinXP.
50 - Implement a Slot priority (loaded/not loaded).
51 - Implement "vacation" Incremental only saves.
52 - Implement single pane restore (much like the Gftp panes).
53 - Implement Automatic Mount even in operator wait.
54 - Implement create "FileSet"?
55 - Implement Release Device in the Job resource to unmount a drive.
56 - Implement Acquire Device in the Job resource to mount a drive,
57 be sure this works with admin jobs so that the user can get
58 prompted to insert the correct tape. Possibly some way to say to
59 run the job but don't save the files.
60 - Implement all command line args on run.
61 - Implement command line "restore" args.
62 - Implement "restore current select=no"
63 - Fix watchdog pthread crash on Win32 (this is pthread_kill() Cygwin bug)
64 - Implement "scratch pool" where tapes are defined and can be
65 taken by any pool that needs them.
66 - Implement restore "current system", but take all files without
67 doing selection tree -- so that jobs without File records can
69 - Make | and < work on FD side.
70 - Pass prefix_links to FD.
71 - Implement a M_SECURITY message class.
72 - Implement disk spooling. Two parts: 1. Spool to disk then
73 immediately to tape to speed up tape operations. 2. Spool to
74 disk only when the tape is full, then when a tape is hung move
76 - From Phil Stracchino:
77 It would probably be a per-client option, and would be called
78 something like, say, "Automatically purge obsoleted jobs". What it
79 would do is, when you successfully complete a Differential backup of a
80 client, it would automatically purge all Incremental backups for that
81 client that are rendered redundant by that Differential. Likewise,
82 when a Full backup on a client completed, it would automatically purge
83 all Differential and Incremental jobs obsoleted by that Full backup.
84 This would let people minimize the number of tapes they're keeping on
85 hand without having to master the art of retention times.
86 - Implement a relocatable bacula.spec
87 - Allow multiple Storage specifications (or multiple names on
88 a single Storage specification) in the Job record. Thus a job
89 can be backed up to a number of storage devices.
90 - Implement dump/print label to UA
91 - Add prefixlinks to where or not where absolute links to FD.
92 - Look at Python for a Bacula scripting language -- www.python.org
93 - Issue message to mount a new tape before the rewind.
94 - Simplified client job initiation for portables.
95 - If SD cannot open a drive, make it periodically retry.
96 - Implement LabelTemplate (at least first cut).
97 - Add more of the config info to the tape label.
98 - Implement a Mount Command and an Unmount Command where
99 the users could specify a system command to be performed
100 to do the mount, after which Bacula could attempt to
101 read the device. This is for Removeable media such as a CDROM.
102 - Most likely, this mount command would be invoked explicitly
103 by the user using the current Console "mount" and "unmount"
104 commands -- the Storage Daemon would do the right thing
105 depending on the exact nature of the device.
106 - As with tape drives, when Bacula wanted a new removable
107 disk mounted, it would unmount the old one, and send a message
108 to the user, who would then use "mount" as described above
109 once he had actually inserted the disk.
111 - Make some way so that if a machine is skipped because it is not up
112 that Bacula will continue retrying for a specified period of time --
114 - If tape is marked read-only, then try opening it read-only rather than
115 failing, and remember that it cannot be written.
116 - Refine SD waiting output:
117 Device is being positioned
118 > Device is being positioned for append
119 > Device is being positioned to file x
121 - Figure out some way to estimate output size and to avoid splitting
122 a backup across two Volumes -- this could be useful for writing CDROMs
123 where you really prefer not to have it split -- not serious.
124 - Add RunBeforeJob and RunAfterJob to the Client program.
125 - Have SD compute MD5 or SHA1 and compare to what FD computes.
126 - Make VolumeToCatalog calculate an MD5 or SHA1 from the
127 actual data on the Volume and compare it.
128 - Implement FileOptions (see end of this document)
129 - Implement Bacula plugins -- design API
130 - Make bcopy read through bad tape records.
131 - Fix read_record to handle multiple sessions.
132 - Program files (i.e. execute a program to read/write files).
133 Pass read date of last backup, size of file last time.
134 - Add Signature type to File DB record.
135 - Make Restore report an error if FD or SD term codes are not OK.
136 - CD into subdirectory when open()ing files for backup to
137 speed up things. Test with testfind().
138 - Priority job to go to top of list.
139 - Find out why Full saves run slower and slower (hashing?)
140 - Why are save/restore of device different sizes (sparse?) Yup! Fix it.
141 - Implement some way for the Console to dynamically create a job.
142 - Restore to a particular time -- e.g. before date, after date.
143 - Solaris -I on tar for include list
144 - Prohibit backing up archive device (findlib/find_one.c:128)
145 - Need a verbose mode in restore, perhaps to bsr.
146 - bscan without -v is too quiet -- perhaps show jobs.
147 - Add code to reject whole blocks if not wanted on restore.
148 - Start working on Base jobs.
149 - Check if we can increase Bacula FD priorty in Win2000
150 - Make sure the MaxVolFiles is fully implemented in SD
151 - Check if both CatalogFiles and UseCatalog are set to SD.
152 - Need return status on read_cb() from read_records(). Need multiple
153 records -- one per Job, maybe a JCR or some other structure with
154 a block and a record.
155 - Figure out how to do a bare metal Windows restore
156 - Put system type returned by FD into catalog.
157 - Possibly add email to Watchdog if drive is unmounted too
158 long and a job is waiting on the drive.
159 - Use read_record.c in SD code.
160 - Why don't we get an error message from Win32 FD when bootstrap
161 file cannot be created for restore command?
162 - When Marking a file in Restore that is a hard link, also
163 mark the link so that the data will be reloaded.
164 - Restore program that errors in SD due to no tape reports
165 OK incorrectly in output.
166 - After unmount, if restore job started, ask to mount.
167 - Convert all %x substitution variables, which are hard to remember
168 and read to %(variable-name). Idea from TMDA.
169 - Add JobLevel in FD status (but make sure it is defined).
170 - Make Pool resource handle Counter resources.
171 - Remove NextId for SQLite. Optimize.
172 - Move all SQL statements into a single location.
173 - Add UA rc and history files.
174 - put termcap (used by console) in ./configure and
175 allow -with-termcap-dir.
176 - Enhance time and size scanning routines.
177 - Fix Autoprune for Volumes to respect need for full save.
178 - Fix Win32 config file definition name on /install
179 - No READLINE_SRC if found in alternate directory.
180 - Test a second language e.g. french.
181 - Compare tape to Client files (attributes, or attributes and data)
182 - Make all database Ids 64 bit.
183 - Write an applet for Linux.
184 - Add estimate to Console commands
185 - Find solution to blank filename (i.e. path only) problem.
186 - Implement new daemon communications protocol.
187 - Remove PoolId from Job table, it exists in Media.
188 - Allow console commands to detach or run in background.
189 - Fix status delay on storage daemon during rewind.
190 - Add SD message variables to control operator wait time
191 - Maximum Operator Wait
192 - Minimum Message Interval
193 - Maximum Message Interval
194 - Send Operator message when cannot read tape label.
195 - Verify level=Volume (scan only), level=Data (compare of data to file).
196 Verify level=Catalog, level=InitCatalog
198 - Add keyword search to show command in Console.
199 - Fix Win2000 error with no messages during startup.
200 - Events : tape has more than xxx bytes.
201 - Restrict characters permitted in a Resource name.
202 - Complete code in Bacula Resources -- this will permit
203 reading a new config file at any time.
204 - Handle ctl-c in Console
205 - Implement script driven addition of File daemon to config files.
206 - Think about how to make Bacula work better with File (non-tape) archives.
207 - Write Unix emulator for Windows.
208 - Implement new serialize subroutines
209 send(socket, "string", &Vol, "uint32", &i, NULL)
210 - Audit all UA commands to ensure that we always prompt where possible.
211 - If ./btape is called without /dev, assume argument is a Storage resource name.
212 - Put memory utilization in Status output of each daemon
213 if full status requested or if some level of debug on.
214 - Make database type selectable by .conf files i.e. at runtime
215 - Set flag for uname -a. Add to Volume label.
216 - Implement throttled work queue.
217 - Check for EOT at ENOSPC or EIO or ENXIO (unix Pc)
218 - Restore files modified after date
219 - Restore file modified before date
220 - Emergency restore info:
222 - Backup working directory
224 - Restore -- do nothing but show what would happen
225 - SET LD_RUN_PATH=$HOME/mysql/lib/mysql
226 - Implement Restore FileSet=
227 - Create a protocol.h and protocol.c where all protocol messages
229 - Remove duplicate fields from jcr (e.g. jcr.level and jcr.jr.Level, ...).
230 - Timout a job or terminate if link goes down, or reopen link and query.
231 - Find general solution for sscanf size problems (as well
232 as sprintf. Do at run time?
233 - Concept of precious tapes (cannot be reused).
234 - Make bcopy copy with a single tape drive.
235 - Permit changing ownership during restore.
237 - Autolabel should be specified by DIR instead of SD.
238 - Find out how to get the system tape block limits, e.g.:
239 Apr 22 21:22:10 polymatou kernel: st1: Block limits 1 - 245760 bytes.
240 Apr 22 21:22:10 polymatou kernel: st0: Block limits 2 - 16777214 bytes.
243 - AutoScan (check checksum of tape)
244 - Format command = "format /dev/nst0"
248 - Seek resolution (usually corresponds to buffer size)
249 - EODErrorCode=ENOSPC or code
250 - Partial Read error code
251 - Partial write error code
252 - Nonformatted read error
253 - Nonformatted write error
254 - WriteProtected error
258 - IgnoreCloseErrors=yes
268 - FD sends unsaved file list to Director at end of job (see
270 - File daemon should build list of files skipped, and then
271 at end of save retry and report any errors.
272 - Write a Storage daemon that uses pipes and
273 standard Unix programs to write to the tape.
275 - Need something that monitors the JCR queue and
276 times out jobs by asking the deamons where they are.
277 - Enhance Jmsg code to permit buffering and saving to disk.
278 - device driver = "xxxx" for drives.
279 - restart: paranoid: read label fsf to
280 eom read append block, and go
281 super-paranoid: read label, read all files
282 in between, read append block, and go
283 verify: backspace, read append block, and go
284 permissive: same as above but frees drive
285 if tape is not valid.
287 - Ensure that /dev/null works
288 - Need report class for messages. Perhaps
289 report resource where report=group of messages
290 - enhance scan_attrib and rename scan_jobtype, and
291 fill in code for "since" option
292 - Director needs a time after which the report status is sent
293 anyway -- or better yet, a retry time for the job.
294 Don't reschedule a job if previous incarnation is still running.
295 - Some way to automatically backup everything is needed????
296 - Need a structure for pending actions:
298 - termination status (part of buffered msgs?)
300 Read, Write, Clean, Delete
301 - Login to Bacula; Bacula users with different permissions:
302 owner, group, user, quotas
303 - Store info on each file system type (probably in the job header on tape.
304 This could be the output of df; or perhaps some sort of /etc/mtab record.
307 - Design at hierarchial storage for Bacula. Migration and Clone.
308 - Implement FSM (File System Modules).
309 - Identify unchanged or "system" files and save them to a
310 special tape thus removing them from the standard
311 backup FileSet -- BASE backup.
312 - Audit M_ error codes to ensure they are correct and consistent.
313 - Add variable break characters to lex analyzer.
314 Either a bit mask or a string of chars so that
315 the caller can change the break characters.
316 - Make a single T_BREAK to replace T_COMMA, etc.
317 - Ensure that File daemon and Storage daemon can
318 continue a save if the Director goes down (this
319 is NOT currently the case). Must detect socket error,
320 buffer messages for later.
321 - Enhance time/duration input to allow multiple qualifiers e.g. 3d2h
322 - Add ability to backup to two Storage devices (two SD sessions) at
323 the same time -- e.g. onsite, offsite.
324 - Add the ability to consolidate old backup sets (basically do a restore
325 to tape and appropriately update the catalog). Compress Volume sets.
326 Might need to spool via file is only one drive is available.
327 - Compress or consolidate Volumes of old possibly deleted files. Perhaps
328 someway to do so with every volume that has less than x% valid
332 Migration: Move a backup from one Volume to another
333 Clone: Copy a backup -- two Volumes
335 Bacula Migration is based on Jobs (apparently Networker is file by file).
337 Migration triggered by:
341 Highwater mark (keep total size)
345 Bacula Projects Roadmap
347 last update 8 May 2003
349 Item 1: Multiple simultaneous Jobs. (done)
350 Done -- Restore part needs better implementation to work correctly
351 Also, it needs considerable testing
353 What: Permit multiple simultaneous jobs in Bacula.
355 Why: An enterprise level solution needs to go fast without the
356 need for the system administrator to carefully tweak
357 timing. Based on the benchmarks, during a full
358 backup, NetWorker typically hit 10 times the bandwidth to
359 the tape compared to Bacula--largely. This is probably due to
360 running parallel jobs and multi-threaded filling of buffers
361 and writing them to tape. This should also make things work
362 better when you have a mix of fast and slow machines backing
365 Notes: Bacula was designed to run multiple simultaneous jobs. Thus
366 implementing this is a matter of some small cleanups and
370 Item 2: Make the Storage daemon use intermediate file storage to buffer data.
371 Deferred -- not necessary yet.
373 What: If data is coming into the SD too fast, buffer it to
374 disk if the user has configured this option.
376 Why: This would be nice, especially if it more or less falls out
377 when implementing (1) above. If not, it probably should not
378 be given a high priority because fundamentally the backup time
379 is limited by the tape bandwidth. Even though you may finish a
380 client job quicker by spilling to disk, you still have to
381 eventually get it onto tape. If intermediate disk buffering
382 allows us to improve write bandwidth to tape, it may make
385 Notes: Whether or not this is implemented will depend upon performance
386 testing after item 1 is implemented.
389 Item 3: Write the bscan program -- also write a bcopy program.
392 What: Write a program that reads a Bacula tape and puts all the
393 appropriate data into the catalog. This allows recovery
394 from a tape that is no longer in the database, or it allows
395 re-creation of a database if lost.
397 Why: This is a fundamental robustness and disaster recovery tool
398 which will increase the comfort level of a sysadmin
399 considering adopting Bacula.
401 Notes: A skeleton of this program already exists, but much work
402 needs to be done. Implementing this will also make apparent
403 any deficiencies in the current Bacula tape format.
406 Item 4: Implement Base jobs.
408 What: A base job is sort of like a Full save except that you
409 will want the FileSet to contain only files that are unlikely
410 to change in the future (i.e. a snapshot of most of your
411 system after installing it). After the base job has been run,
412 when you are doing a Full save, you can specify to exclude
413 all files saved by the base job that have not been modified.
415 Why: This is something none of the competition does, as far as we know
416 (except BackupPC, which is a Perl program that saves to disk
417 only). It is big win for the user, it makes Bacula stand out
418 as offering a unique optimization that immediately saves time
421 Notes: Big savings in tape usage. Will require more resources because
422 the e. DIR must send FD a list of files/attribs, and the FD must
423 search the list and compare it for each file to be saved.
426 Item 5: Implement Label templates
428 What: This is a mechanism whereby Bacula can automatically create
429 a tape label for new tapes according to a detailed specification
430 provided by the user.
432 Why: It is a major convenience item for folks who use automated label
435 Notes: Bacula already has a working form of automatic tape label
436 creation, but it is very crude. The design for the complete
437 tape labeling project is already documented in the manual.
440 Item 6: Write a regression script.
441 Done -- Continue to expand its testing.
443 What: This is an automatic script that runs and tests as many features
444 of Bacula as possible. The output is compared to previous
445 versions of Bacula and any differences are reported.
447 Why: This is an enormous help in preventing introduction of new
448 errors in parts of the program that already work correctly.
450 Notes: This probably should be ranked higher, it's something the typical
451 user doesn't see. Depending on how it's implemented, it may
452 make sense to defer it until the archival tape format and
453 user interface mature.
456 Item 7: GUI for interactive restore
457 Item 8: GUI for interactive backup
459 What: The current interactive restore is implemented with a tty
460 interface. It would be much nicer to be able to "see" the
461 list of files backed up in typical GUI tree format.
462 The same mechanism could also be used for creating
463 ad-hoc backup FileSets (item 8).
465 Why: Ease of use -- especially for the end user.
467 Notes: Rather than implementing in Gtk, we probably should go directly
468 for a Browser implementation, even if doing so meant the
469 capability wouldn't be available until much later. Not only
470 is there the question of Windows sites, most
471 Solaris/HP/IRIX, etc, shops can't currently run Gtk programs
472 without installing lots of stuff admins are very wary about.
473 Real sysadmins will always use the command line anyway, and
474 the user who's doing an interactive restore or backup of his
475 own files will in most cases be on a Windows machine running
479 Item 9: Add SSL to daemon communications.
480 Inprogress as of version 1.31.
482 What: This provides for secure communications between the daemons.
484 Why: This would allow doing backup across the Internet without
485 privacy concerns (or with much less concern).
487 Notes: The vast majority of near term potential users will be backing up
488 a single site over a LAN and, correctly or not, they probably
489 won't be concerned with security, at least not enough to go to
490 the trouble to set up keys, etc. to screw things down. We suspect
491 that many users genuinely interested in multi-site backup
492 already run some form of VPN software in their internetwork
493 connections, and are willing to delegate security to that layer.
496 Item 10: Define definitive tape format.
499 What: Define that definitive tape format that will not change
500 for the next millennium.
502 Why: Stability, security.
504 Notes: See notes for item 11 below.
507 Item 11: New daemon communication protocol.
509 What: The current daemon to daemon protocol is basically an ASCII
510 printf() and sending the buffer. On the receiving end, the
511 buffer is sscanf()ed to unpack it. The new scheme would
512 be a binary format that allows quick packing and unpacking
513 of any data type with named fields.
515 Why: Using binary packing would be faster. Named fields will permit
516 error checking to ensure that what is sent is what the
517 receiver really wants.
519 Notes: These are internal improvements in the interest of the
520 long-term stability and evolution of the program. On the one
521 hand, the sooner they're done, the less code we have to rip
522 up when the time comes to install them. On the other hand, they
523 don't bring an immediately perceptible benefit to potential
524 users. Item 10 and possibly item 11 should be deferred until Bacula
525 is well established with a growing user community more or
526 less happy with the feature set. At that time, it will make a
527 good "next generation" upgrade in the interest of data
533 ======================================================
535 It is somewhat like a Full save becomes an incremental since
536 the Base job (or jobs) plus other non-base files.
538 - New BaseFile table that contains:
539 JobId, BaseJobId, FileId (from Base).
540 i.e. for each base file that exists but is not saved because
541 it has not changed, the File daemon sends the JobId, BaseId,
542 and FileId back to the Director who creates the DB entry.
543 - To initiate a Base save, the Director sends the FD
544 the FileId, and full filename for each file in the Base.
545 - When the FD finds a Base file, he requests the Director to
546 send him the full File entry (stat packet plus MD5), or
547 conversely, the FD sends it to the Director and the Director
548 says yes or no. This can be quite rapid if the FileId is kept
549 by the FD for each Base Filename.
550 - It is probably better to have the comparison done by the FD
551 despite the fact that the File entry must be sent across the
553 - An alternative would be to send the FD the whole File entry
554 from the start. The disadvantage is that it requires a lot of
555 space. The advantage is that it requires less communications
557 - The Job record must be updated to indicate that one or more
559 - At end of Job, FD returns:
560 1. Count of base files/bytes not written to tape (i.e. matches)
561 2. Count of base file that were saved i.e. had changed.
562 - No tape record would be written for a Base file that matches, in the
563 same way that no tape record is written for Incremental jobs where
564 the file is not saved because it is unchanged.
565 - On a restore, all the Base file records must explicitly be
566 found from the BaseFile tape. I.e. for each Full save that is marked
567 to have one or more Base Jobs, search the BaseFile for all occurrences
569 - An optimization might be to make the BaseFile have:
575 This would avoid the need to explicitly fetch each File record for
576 the Base job. The Base Job record will be fetched to get the
577 VolSessionId and VolSessionTime.
578 =========================================================
582 =============================================================
584 Request For Comments For File Backup Options
587 Subject: File Backup Options
590 A few days ago, a Bacula user who is backing up to file volumes and
591 using compression asked if it was possible to suppress compressing
592 all .gz files since it was a waste of CPU time. Although Bacula
593 currently permits using different options (compression, ...) on
594 a directory by directory basis, it cannot do it on a file by
595 file basis, which is clearly what was desired.
597 Proposed Implementation:
598 To solve this problem, I propose the following:
600 - Add a new Director resource type called FileOptions.
602 - The FileOptions resource will have records for all
603 options that can currently be specified on the Include record
604 (in a FileSet). Examples below.
606 - The FileOptions resource will permit an exclude option as well
607 as a number of additional options.
609 - The heart of the FileOptions resource is the ability to
610 supply any number of ApplyTo records which specify POSIX
611 regular expressions. These ApplyTo regular expressions are
612 applied to the fully qualified filename (path and all). If
613 one matches, then the FileOptions will be used.
615 - When an ApplyTo specification matches an included file, the
616 options specified in the FileOptions resource will override
617 the default options specified on the Include record.
619 - Include records will be modified to permit referencing one or
620 more FileOptions resources. The FileOptions will be used
621 in the order listed on the Include record and the first
622 one that matches will be applied.
624 - Options (or specifications) currently supplied on the Include
625 record will be deprecated (i.e. removed in a later version a
626 year or so from now).
628 - The Exclude record will be deprecated as the same functionality
629 can be obtained by using an Exclude = yes in the FileOptions.
632 The following records can appear in the FileOptions resource. An
633 asterisk preceding the name indicates a feature not currently
637 - Compression= (GZIP, ...)
638 - Signature= (MD5, SHA1, ...)
640 - OneFs= (yes/no) - remain on one filesystem
641 - Recurse= (yes/no) - recurse into subdirectories
642 - Sparse= (yes/no) - do sparse file backup
643 - *Exclude= (yes/no) - exclude file from being saved
644 - *Reader= (filename) - external read (backup) program
645 - *Plugin= (filename) - read/write plugin module
648 - verify= (ipnougsamc5) - verify options
651 - replace= (always/ifnewer/ifolder/never) - replace options currently
653 - *Writer= (filename) - external write (restore) program
657 Currently options specifying compression, MD5 signatures, recursion,
658 ... of a FileSet are supplied on the Include record. These will now
659 all be collected into a FileOptions resource, which will be
660 specified on the Include in place of the options. Multiple FileOptions
661 may be specified. Since the FileOptions contain regular expressions
662 that are applied to the full filename, this will give the ability
663 to specify backup options on a file by file basis to whatever level
672 Include = compression=GZIP signature=MD5 {
681 Include = FileOptions=Opts {
692 That's a lot more to do the same thing, but it gives the ability to
693 apply options on a file by file basis. For example, suppose you
694 want to compress all files but not any file with extensions .gz or .Z.
695 You could do so as follows:
699 Include = FileOptions=NoCompress FileOptions=Opts {
707 ApplyTo = /*.?*/ # matches all files
712 # Note multiple ApplyTos are ORed
713 ApplyTo = /*.gz/ # matches .gz files */
714 ApplyTo = /*.Z/ # matches .Z files */
717 Now, since the NoCompress FileOptions is specified first on the
718 Include line, any *.gz or *.Z file will have an MD5 signature computed,
719 but will not be compressed. For all other files, the NoCompress will not
720 match, so the Opts options will be used which will include GZIP
724 - Is it necessary to provide some means of ANDing regular expressions
725 and negation? (not currently planned)
727 e.g. ApplyTo = /*.gz/ && !/big.gz/
729 - I see that Networker has a "null" module which, if specified, does not
730 backup the file, but does make an record of the file in the catalog
731 so that the catalog will reflect an exact picture of the filesystem.
732 The result is that the file can be "seen" when "browsing" the save
733 sets, but it cannot be restored.
735 Is this really useful? Should it be implemented in Bacula?
738 After implementing the above, the user will be able to specify
739 on a file by file basis (using regular expressions) what options are
740 applied for the backup.
743 =============================================
745 ==========================================================
747 For each Incremental job that is run, there may be files that
748 were found but not saved because they were locked (this applies
749 only to Windows). Such a system could send back to the Director
750 a list of Unsaved files.
752 - New UnSavedFiles table that contains:
756 - Then in the next Incremental job, the list of Unsaved Files will be
757 feed to the FD, who will ensure that they are explicitly chosen even
758 if standard date/time check would not have selected them.
759 =============================================================
762 Done: (see kernsdone for more)
763 - Heartbeat between daemons.
764 - Fix Dir heartbeat in restore and verify vol. Be sure to make
765 bnet_recv() ignore BNET_HEARTBEAT.
766 - Implement HEART_BEAT while SD waiting for tapes.
767 - Include RunBeforeJob and RunAfterJob output in the message
769 - Change M_INFO to M_RESTORED for all restored files.
770 - Fix command prompt in gnome-console by checking on Ready.
771 - Merge SQLite, MySQL, and Rel spec into a single file.
772 - Fix config of "console"
773 - Check if cancel works with FD (fixed).
774 - Properly configure console and gconsole (currently for source not
775 configured for installation).
776 - Error labeling tape from console gets Jmsg error because of no Job.
777 - Test and implement get_pint and get_yesno.
778 - Implement global with DB name and add to btraceback.gdb
779 - Remove subsysdir from conf files (used only in autostart scripts).