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 - For FreeBSD typical /dev/nrsa0 and for mtx /dev/pass1
11 - VXA drives have a "cleaning required"
12 indicator, but Exabyte recommends preventive cleaning after every 75
14 - Lookup HP cleaning recommendations.
15 - Lookup HP tape replacement recommendations (see trouble shooting autochanger)
16 - Document FInclude ...
17 - Document need to add "-u root" to most of MySQL script calls
18 (./create_mys... ./make_my...).
19 - Document c:/working directory better than /working directory.
20 - Document all the status codes JobLevel, JobType, JobStatus.
21 - Document update volume: jobid, current, before, all
23 - Document that bscan does not work with multiple simultaneous jobs.
26 Testing to do: (painful)
27 - that ALL console command line options work and are always implemented
28 - blocksize recognition code.
29 - Test if rewind at end of tape waits for tape to rewind.
31 - Test not zeroing Autochanger slot when it is wrong.
32 - Test of last block is correct in JobMedia when splitting file
34 - Test recycling and purging (code changed in db_find_next_volume and
36 - Figure out how to use ssh or stunnel to protect Bacula communications.
39 - In Win portable restore the directory is not create
40 27-Jun-2003 16:52 tibs-fd: kernsrestore.2003-06-27_16.52.20 Error:
41 create_file.c:175 Could not create
42 /tmp/bacula-restores/cygwin/home/kern/bacula/k/src/dird/dird_conf.o: 0
43 ERR=The system cannot find the path specified.
44 - Finish Windows implementation (add setting of correct type on restore,
45 add Portable Data Format flag).
46 - Remove multiple simultaneous devices code in SD.
47 - Check that Block number in JobMedia are correct.
48 - Increment DB version prior to releasing.
49 - Turn off FULL_DEBUG prior to releasing.
50 - On Windows with very long path names, it may be impossible to create
51 a file (and thus restore it) because the total length is too long.
52 We must cd into the directory then create the file without the
54 - Move JobFiles and JobBytes to SD rather than FD -- more correct.
55 - Add client name to cram-md5 challenge so Director can immediately
56 verify if it is the correct client.
57 - lstat() is not going to work on Win32 for testing date.
58 - Implement a Recycle command
59 - Something is not right in last block of fill command.
60 - Implement List Volume Job=xxx or List scheduled volumes or
62 - Check if Incremental is working correctly when it looks for the previous Job
64 - Add next Volume to be used to status output.
65 - The bsr for Dan's job has file indexes covering the whole range rather
66 than only the range contained on the volume.
67 Constrain FileIndex to be within range for Volume.
68 - Pass prefix_links to FD.
72 - Implement new alist in FileSet scanning.
73 - Add JobLevel in FD status (but make sure it is defined).
74 - Audit all UA commands to ensure that we always prompt where possible.
75 - Restrict characters permitted in a Resource name, and don't permit
77 - Prohibit backing up archive device (findlib/find_one.c:128)
78 - Make | and < work on FD side.
79 - Check Jmsg in bnet, may not work, must dup bsock.
80 - Suppress Job Name in Jmsg for console
81 - Create Pools that are referenced in a Run statement at startup if possible.
82 - Use runbeforejob to unload, then reload a volume previously used,
83 then the next job run gets an error reading the drive.
84 - Make bootstrap filename unique.
85 - Test a second language e.g. french.
86 - Start working on Base jobs.
87 - Make "make binary-release" work from any directory.
88 - Unsaved Flag in Job record (use JobMissingFiles).
89 - Base Flag in Job record.
90 - Implement UnsavedFiles DB record.
91 - Implement argc/argv for daemon command line scanning using table driven
93 - Implement table driven single argc/argv scanner to pickup all arguments.
94 Much like xxx_conf.c scan table.
95 keyword, handler(store_routine), store_address, code, flags, default.
96 - From Phil Stracchino:
97 It would probably be a per-client option, and would be called
98 something like, say, "Automatically purge obsoleted jobs". What it
99 would do is, when you successfully complete a Differential backup of a
100 client, it would automatically purge all Incremental backups for that
101 client that are rendered redundant by that Differential. Likewise,
102 when a Full backup on a client completed, it would automatically purge
103 all Differential and Incremental jobs obsoleted by that Full backup.
104 This would let people minimize the number of tapes they're keeping on
105 hand without having to master the art of retention times.
106 - Implement new serialize subroutines
107 send(socket, "string", &Vol, "uint32", &i, NULL)
108 - Scratch Pool where the volumes can be re-assigned to any Pool.
109 - Implement a M_SECURITY message class.
110 - Implement forward spacing block/file: position_device(bsr) --
111 just before read_block_from_device();
112 - When doing a Backup send all attributes back to the Director, who
113 would then figure out what files have been deleted.
114 - Currently in mount.c:236 the SD simply creates a Volume. It should have
115 explicit permission to do so. It should also mark the tape in error
116 if there is an error.
117 - Make sure all restore counters are working correctly in the FD.
118 - SD Bytes Read is wrong.
119 - Configure mtx-changer to have correct path to mtx.
120 - Look at ALL higher level routines that call block.c to be sure
121 they don't expect something in errmsg.
122 - Investigate doing RAW backup of Win32 partition.
123 - Add JobName= to VerifyToCatalog so that all verifies can be done at the end.
124 - Add thread specific data to hold the jcr -- send error messages from
125 low level routines by accessing it and using Jmsg().
126 - Cancel waiting for Client connect in SD if FD goes away.
127 - Testing Tibs job erred and hung director on Storage resource. This was
128 because there were a whole pile of jobs hanging around in the SD
129 waiting for a connection from the FD that was never coming.
130 - Possibly update all client records at startup.
132 - Add Progress command that periodically reports the progress of
134 - One block was orphaned in the SD probably after cancel.
135 - Add all command line arguments to "update", e.g. slot=nn volStatus=append, ...
138 - Examine Bare Metal restore problem (a FD crash exists somewhere ...).
139 - Implement timeout in response() when it should come quickly.
140 - Implement console @echo command.
141 - Implement a Slot priority (loaded/not loaded).
142 - Implement "vacation" Incremental only saves.
143 - Implement single pane restore (much like the Gftp panes).
144 - Implement Automatic Mount even in operator wait.
145 - Implement create "FileSet"?
146 - Implement Release Device in the Job resource to unmount a drive.
147 - Implement Acquire Device in the Job resource to mount a drive,
148 be sure this works with admin jobs so that the user can get
149 prompted to insert the correct tape. Possibly some way to say to
150 run the job but don't save the files.
151 - Implement all command line args on run.
152 - Implement command line "restore" args.
153 - Implement "restore current select=no"
154 - Fix watchdog pthread crash on Win32 (this is pthread_kill() Cygwin bug)
155 - Implement "scratch pool" where tapes are defined and can be
156 taken by any pool that needs them.
157 - Implement restore "current system", but take all files without
158 doing selection tree -- so that jobs without File records can
160 - Implement disk spooling. Two parts: 1. Spool to disk then
161 immediately to tape to speed up tape operations. 2. Spool to
162 disk only when the tape is full, then when a tape is hung move
164 - Implement a relocatable bacula.spec
165 - Allow multiple Storage specifications (or multiple names on
166 a single Storage specification) in the Job record. Thus a job
167 can be backed up to a number of storage devices.
168 - Implement dump/print label to UA
169 - Add prefixlinks to where or not where absolute links to FD.
170 - Issue message to mount a new tape before the rewind.
171 - Simplified client job initiation for portables.
172 - If SD cannot open a drive, make it periodically retry.
173 - Implement LabelTemplate (at least first cut).
174 - Add more of the config info to the tape label.
175 - Implement a Mount Command and an Unmount Command where
176 the users could specify a system command to be performed
177 to do the mount, after which Bacula could attempt to
178 read the device. This is for Removeable media such as a CDROM.
179 - Most likely, this mount command would be invoked explicitly
180 by the user using the current Console "mount" and "unmount"
181 commands -- the Storage Daemon would do the right thing
182 depending on the exact nature of the device.
183 - As with tape drives, when Bacula wanted a new removable
184 disk mounted, it would unmount the old one, and send a message
185 to the user, who would then use "mount" as described above
186 once he had actually inserted the disk.
188 - Make some way so that if a machine is skipped because it is not up
189 that Bacula will continue retrying for a specified period of time --
191 - If tape is marked read-only, then try opening it read-only rather than
192 failing, and remember that it cannot be written.
193 - Refine SD waiting output:
194 Device is being positioned
195 > Device is being positioned for append
196 > Device is being positioned to file x
198 - Figure out some way to estimate output size and to avoid splitting
199 a backup across two Volumes -- this could be useful for writing CDROMs
200 where you really prefer not to have it split -- not serious.
201 - Add RunBeforeJob and RunAfterJob to the Client program.
202 - Have SD compute MD5 or SHA1 and compare to what FD computes.
203 - Make VolumeToCatalog calculate an MD5 or SHA1 from the
204 actual data on the Volume and compare it.
205 - Implement FileOptions (see end of this document)
206 - Implement Bacula plugins -- design API
207 - Make bcopy read through bad tape records.
208 - Fix read_record to handle multiple sessions.
209 - Program files (i.e. execute a program to read/write files).
210 Pass read date of last backup, size of file last time.
211 - Add Signature type to File DB record.
212 - CD into subdirectory when open()ing files for backup to
213 speed up things. Test with testfind().
214 - Priority job to go to top of list.
215 - Why are save/restore of device different sizes (sparse?) Yup! Fix it.
216 - Implement some way for the Console to dynamically create a job.
217 - Restore to a particular time -- e.g. before date, after date.
218 - Solaris -I on tar for include list
219 - Need a verbose mode in restore, perhaps to bsr.
220 - bscan without -v is too quiet -- perhaps show jobs.
221 - Add code to reject whole blocks if not wanted on restore.
222 - Check if we can increase Bacula FD priorty in Win2000
223 - Make sure the MaxVolFiles is fully implemented in SD
224 - Check if both CatalogFiles and UseCatalog are set to SD.
225 - Need return status on read_cb() from read_records(). Need multiple
226 records -- one per Job, maybe a JCR or some other structure with
227 a block and a record.
228 - Figure out how to do a bare metal Windows restore
229 - Possibly add email to Watchdog if drive is unmounted too
230 long and a job is waiting on the drive.
231 - Use read_record.c in SD code.
232 - Restore program that errors in SD due to no tape reports
233 OK incorrectly in output.
234 - After unmount, if restore job started, ask to mount.
235 - Convert all %x substitution variables, which are hard to remember
236 and read to %(variable-name). Idea from TMDA.
237 - Remove NextId for SQLite. Optimize.
238 - Move all SQL statements into a single location.
239 - Add UA rc and history files.
240 - put termcap (used by console) in ./configure and
241 allow -with-termcap-dir.
242 - Enhance time and size scanning routines.
243 - Fix Autoprune for Volumes to respect need for full save.
244 - Fix Win32 config file definition name on /install
245 - Compare tape to Client files (attributes, or attributes and data)
246 - Make all database Ids 64 bit.
247 - Write an applet for Linux.
248 - Add estimate to Console commands
249 - Implement new daemon communications protocol.
250 - Allow console commands to detach or run in background.
251 - Fix status delay on storage daemon during rewind.
252 - Add SD message variables to control operator wait time
253 - Maximum Operator Wait
254 - Minimum Message Interval
255 - Maximum Message Interval
256 - Send Operator message when cannot read tape label.
257 - Verify level=Volume (scan only), level=Data (compare of data to file).
258 Verify level=Catalog, level=InitCatalog
260 - Add keyword search to show command in Console.
261 - Events : tape has more than xxx bytes.
262 - Complete code in Bacula Resources -- this will permit
263 reading a new config file at any time.
264 - Handle ctl-c in Console
265 - Implement script driven addition of File daemon to config files.
266 - Think about how to make Bacula work better with File (non-tape) archives.
267 - Write Unix emulator for Windows.
268 - Put memory utilization in Status output of each daemon
269 if full status requested or if some level of debug on.
270 - Make database type selectable by .conf files i.e. at runtime
271 - Set flag for uname -a. Add to Volume label.
272 - Implement throttled work queue.
273 - Restore files modified after date
274 - Restore file modified before date
275 - Restore -- do nothing but show what would happen
276 - SET LD_RUN_PATH=$HOME/mysql/lib/mysql
277 - Implement Restore FileSet=
278 - Create a protocol.h and protocol.c where all protocol messages
280 - Remove duplicate fields from jcr (e.g. jcr.level and jcr.jr.Level, ...).
281 - Timout a job or terminate if link goes down, or reopen link and query.
282 - Concept of precious tapes (cannot be reused).
283 - Make bcopy copy with a single tape drive.
284 - Permit changing ownership during restore.
287 > My suggestion: Add a feature on the systray menu-icon menu to request
288 > an immediate backup now. This would be useful for laptop users who may
289 > not be on the network when the regular scheduled backup is run.
291 > My wife's suggestion: Add a setting to the win32 client to allow it to
292 > shut down the machine after backup is complete (after, of course,
293 > displaying a "System will shut down in one minute, click here to cancel"
294 > warning dialog). This would be useful for sites that want user
295 > woorkstations to be shut down overnight to save power.
298 - Autolabel should be specified by DIR instead of SD.
301 - AutoScan (check checksum of tape)
302 - Format command = "format /dev/nst0"
306 - Seek resolution (usually corresponds to buffer size)
307 - EODErrorCode=ENOSPC or code
308 - Partial Read error code
309 - Partial write error code
310 - Nonformatted read error
311 - Nonformatted write error
312 - WriteProtected error
316 - IgnoreCloseErrors=yes
326 - FD sends unsaved file list to Director at end of job (see
328 - File daemon should build list of files skipped, and then
329 at end of save retry and report any errors.
330 - Write a Storage daemon that uses pipes and
331 standard Unix programs to write to the tape.
333 - Need something that monitors the JCR queue and
334 times out jobs by asking the deamons where they are.
335 - Enhance Jmsg code to permit buffering and saving to disk.
336 - device driver = "xxxx" for drives.
338 - Ensure that /dev/null works
339 - Need report class for messages. Perhaps
340 report resource where report=group of messages
341 - enhance scan_attrib and rename scan_jobtype, and
342 fill in code for "since" option
343 - Director needs a time after which the report status is sent
344 anyway -- or better yet, a retry time for the job.
345 - Don't reschedule a job if previous incarnation is still running.
346 - Some way to automatically backup everything is needed????
347 - Need a structure for pending actions:
349 - termination status (part of buffered msgs?)
351 Read, Write, Clean, Delete
352 - Login to Bacula; Bacula users with different permissions:
353 owner, group, user, quotas
354 - Store info on each file system type (probably in the job header on tape.
355 This could be the output of df; or perhaps some sort of /etc/mtab record.
358 - Design at hierarchial storage for Bacula. Migration and Clone.
359 - Implement FSM (File System Modules).
360 - Audit M_ error codes to ensure they are correct and consistent.
361 - Add variable break characters to lex analyzer.
362 Either a bit mask or a string of chars so that
363 the caller can change the break characters.
364 - Make a single T_BREAK to replace T_COMMA, etc.
365 - Ensure that File daemon and Storage daemon can
366 continue a save if the Director goes down (this
367 is NOT currently the case). Must detect socket error,
368 buffer messages for later.
369 - Enhance time/duration input to allow multiple qualifiers e.g. 3d2h
370 - Add ability to backup to two Storage devices (two SD sessions) at
371 the same time -- e.g. onsite, offsite.
372 - Add the ability to consolidate old backup sets (basically do a restore
373 to tape and appropriately update the catalog). Compress Volume sets.
374 Might need to spool via file is only one drive is available.
375 - Compress or consolidate Volumes of old possibly deleted files. Perhaps
376 someway to do so with every volume that has less than x% valid
380 Migration: Move a backup from one Volume to another
381 Clone: Copy a backup -- two Volumes
383 Bacula Migration is based on Jobs (apparently Networker is file by file).
385 Migration triggered by:
389 Highwater mark (keep total size)
393 Bacula Projects Roadmap
395 last update 8 May 2003
397 Item 1: Multiple simultaneous Jobs. (done)
398 Done -- Restore part needs better implementation to work correctly
399 Also, it needs considerable testing
401 What: Permit multiple simultaneous jobs in Bacula.
403 Why: An enterprise level solution needs to go fast without the
404 need for the system administrator to carefully tweak
405 timing. Based on the benchmarks, during a full
406 backup, NetWorker typically hit 10 times the bandwidth to
407 the tape compared to Bacula--largely. This is probably due to
408 running parallel jobs and multi-threaded filling of buffers
409 and writing them to tape. This should also make things work
410 better when you have a mix of fast and slow machines backing
413 Notes: Bacula was designed to run multiple simultaneous jobs. Thus
414 implementing this is a matter of some small cleanups and
418 Item 2: Make the Storage daemon use intermediate file storage to buffer data.
419 Deferred -- not necessary yet -- possibly implement with Migration.
421 What: If data is coming into the SD too fast, buffer it to
422 disk if the user has configured this option.
424 Why: This would be nice, especially if it more or less falls out
425 when implementing (1) above. If not, it probably should not
426 be given a high priority because fundamentally the backup time
427 is limited by the tape bandwidth. Even though you may finish a
428 client job quicker by spilling to disk, you still have to
429 eventually get it onto tape. If intermediate disk buffering
430 allows us to improve write bandwidth to tape, it may make
433 Notes: Whether or not this is implemented will depend upon performance
434 testing after item 1 is implemented.
437 Item 3: Write the bscan program -- also write a bcopy program.
440 What: Write a program that reads a Bacula tape and puts all the
441 appropriate data into the catalog. This allows recovery
442 from a tape that is no longer in the database, or it allows
443 re-creation of a database if lost.
445 Why: This is a fundamental robustness and disaster recovery tool
446 which will increase the comfort level of a sysadmin
447 considering adopting Bacula.
449 Notes: A skeleton of this program already exists, but much work
450 needs to be done. Implementing this will also make apparent
451 any deficiencies in the current Bacula tape format.
454 Item 4: Implement Base jobs.
456 What: A base job is sort of like a Full save except that you
457 will want the FileSet to contain only files that are unlikely
458 to change in the future (i.e. a snapshot of most of your
459 system after installing it). After the base job has been run,
460 when you are doing a Full save, you can specify to exclude
461 all files saved by the base job that have not been modified.
463 Why: This is something none of the competition does, as far as we know
464 (except BackupPC, which is a Perl program that saves to disk
465 only). It is big win for the user, it makes Bacula stand out
466 as offering a unique optimization that immediately saves time
469 Notes: Big savings in tape usage. Will require more resources because
470 the e. DIR must send FD a list of files/attribs, and the FD must
471 search the list and compare it for each file to be saved.
474 Item 5: Implement Label templates
477 What: This is a mechanism whereby Bacula can automatically create
478 a tape label for new tapes according to a detailed specification
479 provided by the user.
481 Why: It is a major convenience item for folks who use automated label
484 Notes: Bacula already has a working form of automatic tape label
485 creation, but it is very crude. The design for the complete
486 tape labeling project is already documented in the manual.
489 Item 6: Write a regression script.
490 Done -- Continue to expand its testing.
492 What: This is an automatic script that runs and tests as many features
493 of Bacula as possible. The output is compared to previous
494 versions of Bacula and any differences are reported.
496 Why: This is an enormous help in preventing introduction of new
497 errors in parts of the program that already work correctly.
499 Notes: This probably should be ranked higher, it's something the typical
500 user doesn't see. Depending on how it's implemented, it may
501 make sense to defer it until the archival tape format and
502 user interface mature.
505 Item 7: GUI for interactive restore
506 Item 8: GUI for interactive backup
508 What: The current interactive restore is implemented with a tty
509 interface. It would be much nicer to be able to "see" the
510 list of files backed up in typical GUI tree format.
511 The same mechanism could also be used for creating
512 ad-hoc backup FileSets (item 8).
514 Why: Ease of use -- especially for the end user.
516 Notes: Rather than implementing in Gtk, we probably should go directly
517 for a Browser implementation, even if doing so meant the
518 capability wouldn't be available until much later. Not only
519 is there the question of Windows sites, most
520 Solaris/HP/IRIX, etc, shops can't currently run Gtk programs
521 without installing lots of stuff admins are very wary about.
522 Real sysadmins will always use the command line anyway, and
523 the user who's doing an interactive restore or backup of his
524 own files will in most cases be on a Windows machine running
528 Item 9: Add SSL to daemon communications.
530 What: This provides for secure communications between the daemons.
532 Why: This would allow doing backup across the Internet without
533 privacy concerns (or with much less concern).
535 Notes: The vast majority of near term potential users will be backing up
536 a single site over a LAN and, correctly or not, they probably
537 won't be concerned with security, at least not enough to go to
538 the trouble to set up keys, etc. to screw things down. We suspect
539 that many users genuinely interested in multi-site backup
540 already run some form of VPN software in their internetwork
541 connections, and are willing to delegate security to that layer.
544 Item 10: Define definitive tape format.
547 What: Define that definitive tape format that will not change
548 for the next millennium.
550 Why: Stability, security.
552 Notes: See notes for item 11 below.
555 Item 11: New daemon communication protocol.
557 What: The current daemon to daemon protocol is basically an ASCII
558 printf() and sending the buffer. On the receiving end, the
559 buffer is sscanf()ed to unpack it. The new scheme would
560 be a binary format that allows quick packing and unpacking
561 of any data type with named fields.
563 Why: Using binary packing would be faster. Named fields will permit
564 error checking to ensure that what is sent is what the
565 receiver really wants.
567 Notes: These are internal improvements in the interest of the
568 long-term stability and evolution of the program. On the one
569 hand, the sooner they're done, the less code we have to rip
570 up when the time comes to install them. On the other hand, they
571 don't bring an immediately perceptible benefit to potential
572 users. Item 10 and possibly item 11 should be deferred until Bacula
573 is well established with a growing user community more or
574 less happy with the feature set. At that time, it will make a
575 good "next generation" upgrade in the interest of data
581 ======================================================
583 It is somewhat like a Full save becomes an incremental since
584 the Base job (or jobs) plus other non-base files.
586 - New BaseFile table that contains:
587 JobId, BaseJobId, FileId (from Base).
588 i.e. for each base file that exists but is not saved because
589 it has not changed, the File daemon sends the JobId, BaseId,
590 and FileId back to the Director who creates the DB entry.
591 - To initiate a Base save, the Director sends the FD
592 the FileId, and full filename for each file in the Base.
593 - When the FD finds a Base file, he requests the Director to
594 send him the full File entry (stat packet plus MD5), or
595 conversely, the FD sends it to the Director and the Director
596 says yes or no. This can be quite rapid if the FileId is kept
597 by the FD for each Base Filename.
598 - It is probably better to have the comparison done by the FD
599 despite the fact that the File entry must be sent across the
601 - An alternative would be to send the FD the whole File entry
602 from the start. The disadvantage is that it requires a lot of
603 space. The advantage is that it requires less communications
605 - The Job record must be updated to indicate that one or more
607 - At end of Job, FD returns:
608 1. Count of base files/bytes not written to tape (i.e. matches)
609 2. Count of base file that were saved i.e. had changed.
610 - No tape record would be written for a Base file that matches, in the
611 same way that no tape record is written for Incremental jobs where
612 the file is not saved because it is unchanged.
613 - On a restore, all the Base file records must explicitly be
614 found from the BaseFile tape. I.e. for each Full save that is marked
615 to have one or more Base Jobs, search the BaseFile for all occurrences
617 - An optimization might be to make the BaseFile have:
623 This would avoid the need to explicitly fetch each File record for
624 the Base job. The Base Job record will be fetched to get the
625 VolSessionId and VolSessionTime.
626 =========================================================
630 =============================================================
632 Request For Comments For File Backup Options
635 Subject: File Backup Options
638 A few days ago, a Bacula user who is backing up to file volumes and
639 using compression asked if it was possible to suppress compressing
640 all .gz files since it was a waste of CPU time. Although Bacula
641 currently permits using different options (compression, ...) on
642 a directory by directory basis, it cannot do it on a file by
643 file basis, which is clearly what was desired.
645 Proposed Implementation:
646 To solve this problem, I propose the following:
648 - Add a new Director resource type called FileOptions.
650 - The FileOptions resource will have records for all
651 options that can currently be specified on the Include record
652 (in a FileSet). Examples below.
654 - The FileOptions resource will permit an exclude option as well
655 as a number of additional options.
657 - The heart of the FileOptions resource is the ability to
658 supply any number of ApplyTo records which specify POSIX
659 regular expressions. These ApplyTo regular expressions are
660 applied to the fully qualified filename (path and all). If
661 one matches, then the FileOptions will be used.
663 - When an ApplyTo specification matches an included file, the
664 options specified in the FileOptions resource will override
665 the default options specified on the Include record.
667 - Include records will be modified to permit referencing one or
668 more FileOptions resources. The FileOptions will be used
669 in the order listed on the Include record and the first
670 one that matches will be applied.
672 - Options (or specifications) currently supplied on the Include
673 record will be deprecated (i.e. removed in a later version a
674 year or so from now).
676 - The Exclude record will be deprecated as the same functionality
677 can be obtained by using an Exclude = yes in the FileOptions.
680 The following records can appear in the FileOptions resource. An
681 asterisk preceding the name indicates a feature not currently
685 - Compression= (GZIP, ...)
686 - Signature= (MD5, SHA1, ...)
688 - OneFs= (yes/no) - remain on one filesystem
689 - Recurse= (yes/no) - recurse into subdirectories
690 - Sparse= (yes/no) - do sparse file backup
691 - *Exclude= (yes/no) - exclude file from being saved
692 - *Reader= (filename) - external read (backup) program
693 - *Plugin= (filename) - read/write plugin module
696 - verify= (ipnougsamc5) - verify options
699 - replace= (always/ifnewer/ifolder/never) - replace options currently
701 - *Writer= (filename) - external write (restore) program
705 Currently options specifying compression, MD5 signatures, recursion,
706 ... of a FileSet are supplied on the Include record. These will now
707 all be collected into a FileOptions resource, which will be
708 specified on the Include in place of the options. Multiple FileOptions
709 may be specified. Since the FileOptions contain regular expressions
710 that are applied to the full filename, this will give the ability
711 to specify backup options on a file by file basis to whatever level
720 Include = compression=GZIP signature=MD5 {
729 Include = FileOptions=Opts {
740 That's a lot more to do the same thing, but it gives the ability to
741 apply options on a file by file basis. For example, suppose you
742 want to compress all files but not any file with extensions .gz or .Z.
743 You could do so as follows:
747 Include = FileOptions=NoCompress FileOptions=Opts {
755 ApplyTo = /*.?*/ # matches all files
760 # Note multiple ApplyTos are ORed
761 ApplyTo = /*.gz/ # matches .gz files */
762 ApplyTo = /*.Z/ # matches .Z files */
765 Now, since the NoCompress FileOptions is specified first on the
766 Include line, any *.gz or *.Z file will have an MD5 signature computed,
767 but will not be compressed. For all other files, the NoCompress will not
768 match, so the Opts options will be used which will include GZIP
772 - Is it necessary to provide some means of ANDing regular expressions
773 and negation? (not currently planned)
775 e.g. ApplyTo = /*.gz/ && !/big.gz/
777 - I see that Networker has a "null" module which, if specified, does not
778 backup the file, but does make an record of the file in the catalog
779 so that the catalog will reflect an exact picture of the filesystem.
780 The result is that the file can be "seen" when "browsing" the save
781 sets, but it cannot be restored.
783 Is this really useful? Should it be implemented in Bacula?
786 After implementing the above, the user will be able to specify
787 on a file by file basis (using regular expressions) what options are
788 applied for the backup.
791 =============================================
793 ==========================================================
795 For each Incremental job that is run, there may be files that
796 were found but not saved because they were locked (this applies
797 only to Windows). Such a system could send back to the Director
798 a list of Unsaved files.
800 - New UnSavedFiles table that contains:
804 - Then in the next Incremental job, the list of Unsaved Files will be
805 feed to the FD, who will ensure that they are explicitly chosen even
806 if standard date/time check would not have selected them.
807 =============================================================
810 Done: (see kernsdone for more)
811 - Heartbeat between daemons.
812 - Fix Dir heartbeat in restore and verify vol. Be sure to make
813 bnet_recv() ignore BNET_HEARTBEAT.
814 - Implement HEART_BEAT while SD waiting for tapes.
815 - Include RunBeforeJob and RunAfterJob output in the message
817 - Change M_INFO to M_RESTORED for all restored files.
818 - Fix command prompt in gnome-console by checking on Ready.
819 - Merge SQLite, MySQL, and Rel spec into a single file.
820 - Fix config of "console"
821 - Check if cancel works with FD (fixed).
822 - Properly configure console and gconsole (currently for source not
823 configured for installation).
824 - Error labeling tape from console gets Jmsg error because of no Job.
825 - Test and implement get_pint and get_yesno.
826 - Implement global with DB name and add to btraceback.gdb
827 - Remove subsysdir from conf files (used only in autostart scripts).
829 rufus-dir: Max configured use duration exceeded. Marking Volume "MatouBackup" as Used.
830 rufus-sd: Volume "" previously written, moving to end of data.
831 rufus-sd: Matou.2003-05-10_10.39.18 Error: I canot write on this volume because:
832 The number of files mismatch! Volume=1 Catalog=0
833 rufus-sd: Matou.2003-05-10_10.39.18 Error: askdir.c:155 NULL Volume name. This shouldn't happen!!!
834 - Shell character expansion is failing occassionally.
835 - Add a section to the doc on Manual cycling of Volumes.
836 - Check if Job/File retentions apply to multivolume jobs.
837 - Fix missing casette in autoloader during read:
838 14-May-2003 14:41 undef-sd: RestoreFiles.2003-05-14_14.41.00 Warning: acquire.c:106 Volume name mismatch. Wanted TestVolume0005 got TestVolume0010
839 14-May-2003 14:41 undef-sd: 3301 Issuing autochanger "loaded" command.
840 14-May-2003 14:41 undef-sd: 3302 Issuing autochanger "unload" command.
841 14-May-2003 14:42 undef-sd: 3303 Issuing autochanger "load slot 1" command.
842 14-May-2003 14:42 undef-sd: 3304 Autochanger "load slot 1" status is OK.
843 14-May-2003 14:42 undef-sd: RestoreFiles.2003-05-14_14.41.00 Warning: acquire.c:106 Volume name mismatch. Wanted TestVolume0005 got TestVolume0009
844 14-May-2003 14:42 undef-sd: 3301 Issuing autochanger "loaded" command.
845 14-May-2003 14:42 undef-sd: RestoreFiles.2003-05-14_14.41.00 Warning: acquire.c:106 Volume name mismatch. Wanted TestVolume0005 got TestVolume0009
846 14-May-2003 14:42 undef-sd: 3301 Issuing autochanger "loaded" command.
847 14-May-2003 14:42 undef-sd: RestoreFiles.2003-05-14_14.41.00 Warning: acquire.c:106 Volume name mismatch. Wanted TestVolume0005 got TestVolume0009
848 14-May-2003 14:42 undef-sd: 3301 Issuing autochanger "loaded" command.
849 14-May-2003 14:42 undef-sd: RestoreFiles.2003-05-14_14.41.00 Warning: acquire.c:106 Volume name mismatch. Wanted TestVolume0005 got TestVolume0009
850 14-May-2003 14:42 undef-sd: 3301 Issuing autochanger "loaded" command.
851 14-May-2003 14:42 undef-sd: RestoreFiles.2003-05-14_14.41.00 Fatal error: acquire.c:129 Too many errors trying to mount device "/dev/nrsa0".
852 14-May-2003 14:42 undef-dir: Bacula 1.31 (12May03): 14-May-2003 14:42
853 - Fix problem reported by Christopher McCurdy <cmccurdy@eecis.udel.edu>
854 xeon-fd: Could not stat c:/Documents and Settings/All
855 Users/Application Data/Humc:\Documents and Settings\All User98_AIX.kbf:
856 ERR=No such file or directory
858 - The following Re-read last block at EOT failed. ERR=block.c:523 Read zero bytes on device /dev/nrsa0.
859 undef-sd: block.c:523 Read zero bytes on device /dev/nrsa0.
860 apparently masks the standard EOM message.
861 - BSD (probably) does not have strtoll()
862 - BSD does not have ioctl() MTEOM
863 - BSD defines a number of MT_xxx variables which conflict
864 with those defined by Bacula.
865 - Make default duration days if no qualifier (e.g. s) is specified.
866 - BSDI fix finding gcc version
867 - When the FD errs (e.g. disk full) have a more graceful shutdown.
868 - Make sure Bacula prunes/purges canceled and failed jobs too and all
869 jobs with zero JobFiles.
870 - Implement Volume name checking.
871 - Document what characters can go into Volume names.
872 - Getting the following on all directories on Win32
873 19-May-2003 01:14 tibs-fd: Could not access c:/cygwin/home/kern/rxvt: ERR=Permission denied
874 - Cancellation caused JobMedia error:
875 babylon5-dir: Last FULL backup time not found. Doing FULL backup.
876 babylon5-dir: Start Backup JobId 416, Job=Zocalo_Save.2003-05-19_02.15.06
877 babylon5-sd: End of media on Volume VXA-V17-Inc-001 Bytes=31,982,900,672 Blocks=495,781.
878 babylon5-sd: Job Zocalo_Save.2003-05-19_02.15.06 waiting. Cannot find any appendable volumes.
879 babylon5-sd: Someone woke me up, but I cannot find any appendable volumes
880 for Job=Zocalo_Save.2003-05-19_02.15.06.
881 babylon5-sd: Zocalo_Save.2003-05-19_02.15.06 Fatal error: Job
882 Zocalo_Save.2003-05-19_02.15.06 canceled while waiting for mount on
883 Storage Device "Ecrix_VXA-1".
884 babylon5-sd: Zocalo_Save.2003-05-19_02.15.06 Fatal error: Cannot fixup device
885 error. Job Zocalo_Save.2003-05-19_02.15.06 canceled while waiting for
886 mount on Storage Device "Ecrix_VXA-1".
888 babylon5-dir: Zocalo_Save.2003-05-19_02.15.06 Error: Catalog error creating
889 JobMedia record. sql_create.c:125 Create JobMedia failed. Record already
891 babylon5-sd: Zocalo_Save.2003-05-19_02.15.06 Error: Error creating JobMedia
892 record: 1991 Update JobMedia error
894 babylon5-sd: Zocalo_Save.2003-05-19_02.15.06 Error: askdir.c:158 NULL Volume
895 name. This shouldn't happen!!!
896 zocalo-fd: Zocalo_Save.2003-05-19_02.15.06 Error: bnet.c:310 Write error
897 sending to Storage daemon:babylon5:9103: ERR=Broken pipe
898 - Volume names with spaces get jammed into the catalog with 0x1
899 i.e. the SD bashes the Volume but they are not unbased by Dir.
900 jerom-dir: MonthlySave.2003-05-10_17.12.01 Error: Unable to get Media
901 record for Volume Tape^A1: ERR=sql_get.c:788 Media record for Volume
903 jerom-sd: MonthlySave.2003-05-10_17.12.01 Error: Error updating Volume
904 Info: 1991 Catalog Request failed: sql_get.c:788 Media record for Volume
906 - ChangeServiceConfig2A does not exist on WinNT (ADVAPI32.DLL).
907 - Fix "access not allowed" for backup of files on WinXP.
908 - Check for existence of all new Win32 API's. See LoadLibrary in
910 - Count errors during restore and print them in the Job report.
911 - Bug: fix access problems on files restored on WinXP.
912 - Put system type returned by FD into catalog.
913 - Finish WIN32_DATA stream code (bextract, check if can handle stream)
914 - Make SD keep track of Files, Bytes during restore.
915 - If you enter the userid by hand for restore, you get:
916 Enter JobId(s), comma separated, to restore: 74
917 You have selected the following JobId: 74
918 Building directory tree for JobId 74 ...
919 134645140 items inserted into the tree and marked for extraction.
920 - Add SDWriteSeqNo to SD, and probably Read on FD side.
921 - If bootstrap is non-zero for restore, do not show JobId in the
922 OK to run? (yes/mod/no): list.
923 - When all cassettes in magazine are used, got:
924 22-May-2003 18:24 undef-sd: 3304 Autochanger "load slot 1" status is OK.
925 22-May-2003 18:24 undef-sd: NightlySave.2003-05-22_14.08.16 Warning: mount.c:245 Director wanted Volume "TestVolume0009".
926 Current Volume "TestVolume0005" not acceptable because:
927 1998 Volume "TestVolume0005" not Append or Recycle.
928 22-May-2003 18:24 undef-sd: NightlySave.2003-05-22_14.08.16 Error: Autochanger Volume "TestVolume0009" not found in slot 1.
929 Setting slot to zero in catalog.
930 22-May-2003 18:24 undef-sd: Please mount Volume "TestVolume0009" on Storage Device "ARCHIVE 4586" for Job NightlySave.2003-05-22_14
932 Use "mount" command to release Job.
933 22-May-2003 19:24 undef-sd: Please mount Volume "TestVolume0009" on Storage Device "ARCHIVE 4586" for Job NightlySave.2003-05-22_14
935 Use "mount" command to release Job.
936 - Don't zero the Slot when the wrong volume is found -- simply ask
938 - Implement MTIOCERRSTAT on FreeBSD to clear tape error conditions.
939 - Shell expansion fails for working_directory in SD from time to time.
940 - File the Automatically selected: xxx
941 to say Automatically selected Pool: xxx
942 - Default duration with no qualifier is sec should be 1 day
943 - zap sd_auth_key in SD after FD connection.
944 - Find a solution for the multiple FileSet problem (when it is changed). Add date?
945 - Look at Python for a Bacula scripting language -- www.python.org
946 - When Marking a file in Restore that is a hard link, also
947 mark the link so that the data will be reloaded.
948 - Emergency restore info:
950 - Backup working directory
952 - Why don't we get an error message from Win32 FD when bootstrap
953 file cannot be created for restore command?
954 - Fix Win2000 error with no messages during startup.
955 - Make restore more robust in counting error and not immediately bailing
956 out. Also print error message once, but try to continue.
957 - Add code to check that blocks are sequential on restore.
958 - Remove "rufus" and such references from regress.
959 - No READLINE_SRC if found in alternate directory.
960 - If ./btape is called without /dev, assume argument is a Storage resource name.
961 - Find general solution for sscanf size problems (as well as sprintf. Do at run time?
962 - Bytes restored is wrong.
963 - The "List last 20 Jobs run" doesnt work correctly in restore.
964 It doesnt show the last 20 jobs , but some older ones.
965 - Fix Verify VolumeToCatalog to use BSRs -- it is broken.
966 - Implement Release Storage=xxx
967 - Fix restore on Win95/98
968 - Remove the Jmsg() in sql_find.c:102 or only print on hard error.
969 - Implement FileSet VolIndex -- done, but must update old records.
970 - Check this below from Phil.
971 This was SD reported data rather than FD data!
972 > When the job was done, Bacula reported 11084 files restored:
975 > Job: Zocalo_Restore.2003-06-05_16.42.01
977 > Start time: 05-Jun-2003 16:42
978 > End time: 06-Jun-2003 01:21
979 > Files Restored: 11,084
980 > Bytes Restored: 65,474,772
982 > FD termination status: OK
983 > Termination: Restore OK
985 > when it should probably have reported 11084 files scanned, 250 restored.
986 > The bytes restored count looks about right.
988 - Should Bacula make an Append tape as Purged when purging?
989 - Use switch() in backup.c and restore.c in FD instead of giant if statement.
990 - If during a restore, a hard linked file already exists (on option), delete
991 the file and re-link it. This is to avoid the possibility that the
992 user had re-linked the file between the backup and the restore.
993 Do lstat() to see if it is already properly linked.
994 Same for symlinked file.
995 Make sure ifnewer, ifolder, never, ... apply correctly.
996 - Flag so that no connect does not error, and Reschedule a job.
997 - Implement "Reschedule OnError=yes interval=nnn times=xxx"
998 - That restoring a hard link that already exists works correctly.
1000 - Make Pool resource handle Counter resources.
1001 - Fix first block number after label to be zero instead of 1 (reset after label).
1002 - Grep for Backup OK in regression script.
1003 - Do NOT reuse same JobId if tape written.
1004 - Implement non-blocking writes and bsock->terminate in heartbeat
1005 thread, or set it in status.c cancel (used pthread_kill() instead of
1007 - Add restore to specific date.
1008 - Instrument use_count on DEVICE packets and ensure that the device is
1009 being close()ed at the appropriate time.
1010 - Test long path names (>64 chars) in Windows -- crashes FD?
1011 - Implement fast block rejection: match_bsr_block().
1012 - Complain if record dropped in bnet_recv because too long.
1013 - Test multiple simultaneous Volumes
1014 - Document recycling algorithm.
1015 - Make Restore report an error if FD or SD term codes are not OK.
1016 - To link with mysqlclient_r may require -lssl -lcrypto
1017 - Document Heart beat code
1018 - Non-fatal errors are not correct counting attribs.c:277