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 all the status codes JobLevel, JobType, JobStatus.
20 - Document dynamic DNS
23 Testing to do: (painful)
24 - that ALL console command line options work and are always implemented
25 - blocksize recognition code.
26 - Test if rewind at end of tape waits for tape to rewind.
28 - Test not zeroing Autochanger slot when it is wrong.
29 - Test of last block is correct in JobMedia when splitting file
31 - Test recycling and purging (code changed in db_find_next_volume and
33 - Figure out how to use ssh or stunnel to protect Bacula communications.
36 - Backup on havana gave:
38 The defined Storage resources are:
41 Select Storage resource (1-2): 2
42 Connecting to Storage daemon DDS-3 at havana:9103
44 havana-sd Version: 1.31 (08 Jul 2003)
45 Daemon started 10-Jul-2003 19:06, 0 Jobs run.
46 Device /tmp is not open.
47 Device /dev/nst0 is mounted with Volume "msu-8Jul03"
48 Device is being initialized.
49 Total Bytes Read=4,629,249,084,728,272,628 Blocks Read=0 Bytes/block=134,735,604
50 Positioned at File=0 Block=0
51 Full Backup job msu-8Jul03 using Volume "msu-backup.2003-07-10_19.06.42" on device /dev/nst0
52 Files=0 Bytes=0 Bytes/sec=0
53 FDReadSeqNo=6 in_msg=6 out_msg=6 fd=7
56 - Do full check the command line args in update (e.g. VolStatus ...).
57 - Maybe remove multiple simultaneous devices code in SD.
58 - Increment DB version prior to releasing.
59 - Turn off FULL_DEBUG prior to releasing.
60 - On Windows with very long path names, it may be impossible to create
61 a file (and thus restore it) because the total length is too long.
62 We must cd into the directory then create the file without the
64 - Move JobFiles and JobBytes to SD rather than FD -- more correct.
65 - lstat() is not going to work on Win32 for testing date.
66 - Implement a Recycle command
67 - Something is not right in last block of fill command.
68 - Implement List Volume Job=xxx or List scheduled volumes or Status Director
69 - Check if Incremental is working correctly when it looks for the previous Job
71 - Add FileSet to command line arguments for restore.
75 - Implement a Recycle or Prune Current Volume (as from Nic Bellamy),
76 possibly implement Prune Oldest Volume.
77 - Allow multiple Storage specifications (or multiple names on
78 a single Storage specification) in the Job record. Thus a job
79 can be backed up to a number of storage devices.
80 - Add client name to cram-md5 challenge so Director can immediately
81 verify if it is the correct client.
82 - Implement ClientRunBeforeJob and ClientRunAfterJob.
83 - Implement new alist in FileSet scanning.
84 - Add JobLevel in FD status (but make sure it is defined).
85 - Audit all UA commands to ensure that we always prompt where possible.
86 - Restrict characters permitted in a Resource name, and don't permit
88 - Prohibit backing up archive device (findlib/find_one.c:128)
89 - Make | and < work on FD side.
90 - Check Jmsg in bnet, may not work, must dup bsock.
91 - Suppress Job Name in Jmsg for console
92 - Create Pools that are referenced in a Run statement at startup if possible.
93 - Use runbeforejob to unload, then reload a volume previously used,
94 then the next job run gets an error reading the drive.
95 - Make bootstrap filename unique.
96 - Test a second language e.g. french.
97 - Start working on Base jobs.
98 - Make "make binary-release" work from any directory.
99 - Unsaved Flag in Job record (use JobMissingFiles).
100 - Base Flag in Job record.
101 - Implement UnsavedFiles DB record.
102 - Implement argc/argv for daemon command line scanning using table driven
104 - Implement table driven single argc/argv scanner to pickup all arguments.
105 Much like xxx_conf.c scan table.
106 keyword, handler(store_routine), store_address, code, flags, default.
107 - From Phil Stracchino:
108 It would probably be a per-client option, and would be called
109 something like, say, "Automatically purge obsoleted jobs". What it
110 would do is, when you successfully complete a Differential backup of a
111 client, it would automatically purge all Incremental backups for that
112 client that are rendered redundant by that Differential. Likewise,
113 when a Full backup on a client completed, it would automatically purge
114 all Differential and Incremental jobs obsoleted by that Full backup.
115 This would let people minimize the number of tapes they're keeping on
116 hand without having to master the art of retention times.
117 - Implement new serialize subroutines
118 send(socket, "string", &Vol, "uint32", &i, NULL)
119 - Scratch Pool where the volumes can be re-assigned to any Pool.
120 - Implement a M_SECURITY message class.
121 - Implement forward spacing block/file: position_device(bsr) --
122 just before read_block_from_device();
123 - When doing a Backup send all attributes back to the Director, who
124 would then figure out what files have been deleted.
125 - Currently in mount.c:236 the SD simply creates a Volume. It should have
126 explicit permission to do so. It should also mark the tape in error
127 if there is an error.
128 - Make sure all restore counters are working correctly in the FD.
129 - SD Bytes Read is wrong.
130 - Configure mtx-changer to have correct path to mtx.
131 - Look at ALL higher level routines that call block.c to be sure
132 they don't expect something in errmsg.
133 - Investigate doing RAW backup of Win32 partition.
134 - Add JobName= to VerifyToCatalog so that all verifies can be done at the end.
135 - Add thread specific data to hold the jcr -- send error messages from
136 low level routines by accessing it and using Jmsg().
137 - Cancel waiting for Client connect in SD if FD goes away.
139 - Add Progress command that periodically reports the progress of
141 - One block was orphaned in the SD probably after cancel.
142 - Add all command line arguments to "update", e.g. slot=nn volStatus=append, ...
145 - Examine Bare Metal restore problem (a FD crash exists somewhere ...).
146 - Implement timeout in response() when it should come quickly.
147 - Implement console @echo command.
148 - Implement a Slot priority (loaded/not loaded).
149 - Implement "vacation" Incremental only saves.
150 - Implement single pane restore (much like the Gftp panes).
151 - Implement Automatic Mount even in operator wait.
152 - Implement create "FileSet"?
153 - Implement Release Device in the Job resource to unmount a drive.
154 - Implement Acquire Device in the Job resource to mount a drive,
155 be sure this works with admin jobs so that the user can get
156 prompted to insert the correct tape. Possibly some way to say to
157 run the job but don't save the files.
158 - Implement all command line args on run.
159 - Implement command line "restore" args.
160 - Implement "restore current select=no"
161 - Fix watchdog pthread crash on Win32 (this is pthread_kill() Cygwin bug)
162 - Implement "scratch pool" where tapes are defined and can be
163 taken by any pool that needs them.
164 - Implement restore "current system", but take all files without
165 doing selection tree -- so that jobs without File records can
167 - Implement disk spooling. Two parts: 1. Spool to disk then
168 immediately to tape to speed up tape operations. 2. Spool to
169 disk only when the tape is full, then when a tape is hung move
171 - Implement a relocatable bacula.spec
172 - Implement dump/print label to UA
173 - Add prefixlinks to where or not where absolute links to FD.
174 - Issue message to mount a new tape before the rewind.
175 - Simplified client job initiation for portables.
176 - If SD cannot open a drive, make it periodically retry.
177 - Implement LabelTemplate (at least first cut).
178 - Add more of the config info to the tape label.
179 - Implement a Mount Command and an Unmount Command where
180 the users could specify a system command to be performed
181 to do the mount, after which Bacula could attempt to
182 read the device. This is for Removeable media such as a CDROM.
183 - Most likely, this mount command would be invoked explicitly
184 by the user using the current Console "mount" and "unmount"
185 commands -- the Storage Daemon would do the right thing
186 depending on the exact nature of the device.
187 - As with tape drives, when Bacula wanted a new removable
188 disk mounted, it would unmount the old one, and send a message
189 to the user, who would then use "mount" as described above
190 once he had actually inserted the disk.
192 - Make some way so that if a machine is skipped because it is not up
193 that Bacula will continue retrying for a specified period of time --
195 - If tape is marked read-only, then try opening it read-only rather than
196 failing, and remember that it cannot be written.
197 - Refine SD waiting output:
198 Device is being positioned
199 > Device is being positioned for append
200 > Device is being positioned to file x
202 - Figure out some way to estimate output size and to avoid splitting
203 a backup across two Volumes -- this could be useful for writing CDROMs
204 where you really prefer not to have it split -- not serious.
205 - Add RunBeforeJob and RunAfterJob to the Client program.
206 - Have SD compute MD5 or SHA1 and compare to what FD computes.
207 - Make VolumeToCatalog calculate an MD5 or SHA1 from the
208 actual data on the Volume and compare it.
209 - Implement FileOptions (see end of this document)
210 - Implement Bacula plugins -- design API
211 - Make bcopy read through bad tape records.
212 - Fix read_record to handle multiple sessions.
213 - Program files (i.e. execute a program to read/write files).
214 Pass read date of last backup, size of file last time.
215 - Add Signature type to File DB record.
216 - CD into subdirectory when open()ing files for backup to
217 speed up things. Test with testfind().
218 - Priority job to go to top of list.
219 - Why are save/restore of device different sizes (sparse?) Yup! Fix it.
220 - Implement some way for the Console to dynamically create a job.
221 - Restore to a particular time -- e.g. before date, after date.
222 - Solaris -I on tar for include list
223 - Need a verbose mode in restore, perhaps to bsr.
224 - bscan without -v is too quiet -- perhaps show jobs.
225 - Add code to reject whole blocks if not wanted on restore.
226 - Check if we can increase Bacula FD priorty in Win2000
227 - Make sure the MaxVolFiles is fully implemented in SD
228 - Check if both CatalogFiles and UseCatalog are set to SD.
229 - Need return status on read_cb() from read_records(). Need multiple
230 records -- one per Job, maybe a JCR or some other structure with
231 a block and a record.
232 - Figure out how to do a bare metal Windows restore
233 - Possibly add email to Watchdog if drive is unmounted too
234 long and a job is waiting on the drive.
235 - Use read_record.c in SD code.
236 - Restore program that errors in SD due to no tape reports
237 OK incorrectly in output.
238 - After unmount, if restore job started, ask to mount.
239 - Convert all %x substitution variables, which are hard to remember
240 and read to %(variable-name). Idea from TMDA.
241 - Remove NextId for SQLite. Optimize.
242 - Move all SQL statements into a single location.
243 - Add UA rc and history files.
244 - put termcap (used by console) in ./configure and
245 allow -with-termcap-dir.
246 - Enhance time and size scanning routines.
247 - Fix Autoprune for Volumes to respect need for full save.
248 - Fix Win32 config file definition name on /install
249 - Compare tape to Client files (attributes, or attributes and data)
250 - Make all database Ids 64 bit.
251 - Write an applet for Linux.
252 - Add estimate to Console commands
253 - Implement new daemon communications protocol.
254 - Allow console commands to detach or run in background.
255 - Fix status delay on storage daemon during rewind.
256 - Add SD message variables to control operator wait time
257 - Maximum Operator Wait
258 - Minimum Message Interval
259 - Maximum Message Interval
260 - Send Operator message when cannot read tape label.
261 - Verify level=Volume (scan only), level=Data (compare of data to file).
262 Verify level=Catalog, level=InitCatalog
264 - Add keyword search to show command in Console.
265 - Events : tape has more than xxx bytes.
266 - Complete code in Bacula Resources -- this will permit
267 reading a new config file at any time.
268 - Handle ctl-c in Console
269 - Implement script driven addition of File daemon to config files.
270 - Think about how to make Bacula work better with File (non-tape) archives.
271 - Write Unix emulator for Windows.
272 - Put memory utilization in Status output of each daemon
273 if full status requested or if some level of debug on.
274 - Make database type selectable by .conf files i.e. at runtime
275 - Set flag for uname -a. Add to Volume label.
276 - Implement throttled work queue.
277 - Restore files modified after date
278 - Restore file modified before date
279 - Restore -- do nothing but show what would happen
280 - SET LD_RUN_PATH=$HOME/mysql/lib/mysql
281 - Implement Restore FileSet=
282 - Create a protocol.h and protocol.c where all protocol messages
284 - Remove duplicate fields from jcr (e.g. jcr.level and jcr.jr.Level, ...).
285 - Timout a job or terminate if link goes down, or reopen link and query.
286 - Concept of precious tapes (cannot be reused).
287 - Make bcopy copy with a single tape drive.
288 - Permit changing ownership during restore.
291 > My suggestion: Add a feature on the systray menu-icon menu to request
292 > an immediate backup now. This would be useful for laptop users who may
293 > not be on the network when the regular scheduled backup is run.
295 > My wife's suggestion: Add a setting to the win32 client to allow it to
296 > shut down the machine after backup is complete (after, of course,
297 > displaying a "System will shut down in one minute, click here to cancel"
298 > warning dialog). This would be useful for sites that want user
299 > woorkstations to be shut down overnight to save power.
302 - Autolabel should be specified by DIR instead of SD.
305 - AutoScan (check checksum of tape)
306 - Format command = "format /dev/nst0"
310 - Seek resolution (usually corresponds to buffer size)
311 - EODErrorCode=ENOSPC or code
312 - Partial Read error code
313 - Partial write error code
314 - Nonformatted read error
315 - Nonformatted write error
316 - WriteProtected error
320 - IgnoreCloseErrors=yes
330 - FD sends unsaved file list to Director at end of job (see
332 - File daemon should build list of files skipped, and then
333 at end of save retry and report any errors.
334 - Write a Storage daemon that uses pipes and
335 standard Unix programs to write to the tape.
337 - Need something that monitors the JCR queue and
338 times out jobs by asking the deamons where they are.
339 - Enhance Jmsg code to permit buffering and saving to disk.
340 - device driver = "xxxx" for drives.
342 - Ensure that /dev/null works
343 - Need report class for messages. Perhaps
344 report resource where report=group of messages
345 - enhance scan_attrib and rename scan_jobtype, and
346 fill in code for "since" option
347 - Director needs a time after which the report status is sent
348 anyway -- or better yet, a retry time for the job.
349 - Don't reschedule a job if previous incarnation is still running.
350 - Some way to automatically backup everything is needed????
351 - Need a structure for pending actions:
353 - termination status (part of buffered msgs?)
355 Read, Write, Clean, Delete
356 - Login to Bacula; Bacula users with different permissions:
357 owner, group, user, quotas
358 - Store info on each file system type (probably in the job header on tape.
359 This could be the output of df; or perhaps some sort of /etc/mtab record.
362 - Design at hierarchial storage for Bacula. Migration and Clone.
363 - Implement FSM (File System Modules).
364 - Audit M_ error codes to ensure they are correct and consistent.
365 - Add variable break characters to lex analyzer.
366 Either a bit mask or a string of chars so that
367 the caller can change the break characters.
368 - Make a single T_BREAK to replace T_COMMA, etc.
369 - Ensure that File daemon and Storage daemon can
370 continue a save if the Director goes down (this
371 is NOT currently the case). Must detect socket error,
372 buffer messages for later.
373 - Enhance time/duration input to allow multiple qualifiers e.g. 3d2h
374 - Add ability to backup to two Storage devices (two SD sessions) at
375 the same time -- e.g. onsite, offsite.
376 - Add the ability to consolidate old backup sets (basically do a restore
377 to tape and appropriately update the catalog). Compress Volume sets.
378 Might need to spool via file is only one drive is available.
379 - Compress or consolidate Volumes of old possibly deleted files. Perhaps
380 someway to do so with every volume that has less than x% valid
384 Migration: Move a backup from one Volume to another
385 Clone: Copy a backup -- two Volumes
387 Bacula Migration is based on Jobs (apparently Networker is file by file).
389 Migration triggered by:
393 Highwater mark (keep total size)
397 Bacula Projects Roadmap
399 last update 8 May 2003
401 Item 1: Multiple simultaneous Jobs. (done)
402 Done -- Restore part needs better implementation to work correctly
403 Also, it needs considerable testing
405 What: Permit multiple simultaneous jobs in Bacula.
407 Why: An enterprise level solution needs to go fast without the
408 need for the system administrator to carefully tweak
409 timing. Based on the benchmarks, during a full
410 backup, NetWorker typically hit 10 times the bandwidth to
411 the tape compared to Bacula--largely. This is probably due to
412 running parallel jobs and multi-threaded filling of buffers
413 and writing them to tape. This should also make things work
414 better when you have a mix of fast and slow machines backing
417 Notes: Bacula was designed to run multiple simultaneous jobs. Thus
418 implementing this is a matter of some small cleanups and
422 Item 2: Make the Storage daemon use intermediate file storage to buffer data.
423 Deferred -- not necessary yet -- possibly implement with Migration.
425 What: If data is coming into the SD too fast, buffer it to
426 disk if the user has configured this option.
428 Why: This would be nice, especially if it more or less falls out
429 when implementing (1) above. If not, it probably should not
430 be given a high priority because fundamentally the backup time
431 is limited by the tape bandwidth. Even though you may finish a
432 client job quicker by spilling to disk, you still have to
433 eventually get it onto tape. If intermediate disk buffering
434 allows us to improve write bandwidth to tape, it may make
437 Notes: Whether or not this is implemented will depend upon performance
438 testing after item 1 is implemented.
441 Item 3: Write the bscan program -- also write a bcopy program.
444 What: Write a program that reads a Bacula tape and puts all the
445 appropriate data into the catalog. This allows recovery
446 from a tape that is no longer in the database, or it allows
447 re-creation of a database if lost.
449 Why: This is a fundamental robustness and disaster recovery tool
450 which will increase the comfort level of a sysadmin
451 considering adopting Bacula.
453 Notes: A skeleton of this program already exists, but much work
454 needs to be done. Implementing this will also make apparent
455 any deficiencies in the current Bacula tape format.
458 Item 4: Implement Base jobs.
460 What: A base job is sort of like a Full save except that you
461 will want the FileSet to contain only files that are unlikely
462 to change in the future (i.e. a snapshot of most of your
463 system after installing it). After the base job has been run,
464 when you are doing a Full save, you can specify to exclude
465 all files saved by the base job that have not been modified.
467 Why: This is something none of the competition does, as far as we know
468 (except BackupPC, which is a Perl program that saves to disk
469 only). It is big win for the user, it makes Bacula stand out
470 as offering a unique optimization that immediately saves time
473 Notes: Big savings in tape usage. Will require more resources because
474 the e. DIR must send FD a list of files/attribs, and the FD must
475 search the list and compare it for each file to be saved.
478 Item 5: Implement Label templates
481 What: This is a mechanism whereby Bacula can automatically create
482 a tape label for new tapes according to a detailed specification
483 provided by the user.
485 Why: It is a major convenience item for folks who use automated label
488 Notes: Bacula already has a working form of automatic tape label
489 creation, but it is very crude. The design for the complete
490 tape labeling project is already documented in the manual.
493 Item 6: Write a regression script.
494 Done -- Continue to expand its testing.
496 What: This is an automatic script that runs and tests as many features
497 of Bacula as possible. The output is compared to previous
498 versions of Bacula and any differences are reported.
500 Why: This is an enormous help in preventing introduction of new
501 errors in parts of the program that already work correctly.
503 Notes: This probably should be ranked higher, it's something the typical
504 user doesn't see. Depending on how it's implemented, it may
505 make sense to defer it until the archival tape format and
506 user interface mature.
509 Item 7: GUI for interactive restore
510 Item 8: GUI for interactive backup
512 What: The current interactive restore is implemented with a tty
513 interface. It would be much nicer to be able to "see" the
514 list of files backed up in typical GUI tree format.
515 The same mechanism could also be used for creating
516 ad-hoc backup FileSets (item 8).
518 Why: Ease of use -- especially for the end user.
520 Notes: Rather than implementing in Gtk, we probably should go directly
521 for a Browser implementation, even if doing so meant the
522 capability wouldn't be available until much later. Not only
523 is there the question of Windows sites, most
524 Solaris/HP/IRIX, etc, shops can't currently run Gtk programs
525 without installing lots of stuff admins are very wary about.
526 Real sysadmins will always use the command line anyway, and
527 the user who's doing an interactive restore or backup of his
528 own files will in most cases be on a Windows machine running
532 Item 9: Add SSL to daemon communications.
534 What: This provides for secure communications between the daemons.
536 Why: This would allow doing backup across the Internet without
537 privacy concerns (or with much less concern).
539 Notes: The vast majority of near term potential users will be backing up
540 a single site over a LAN and, correctly or not, they probably
541 won't be concerned with security, at least not enough to go to
542 the trouble to set up keys, etc. to screw things down. We suspect
543 that many users genuinely interested in multi-site backup
544 already run some form of VPN software in their internetwork
545 connections, and are willing to delegate security to that layer.
548 Item 10: Define definitive tape format.
551 What: Define that definitive tape format that will not change
552 for the next millennium.
554 Why: Stability, security.
556 Notes: See notes for item 11 below.
559 Item 11: New daemon communication protocol.
561 What: The current daemon to daemon protocol is basically an ASCII
562 printf() and sending the buffer. On the receiving end, the
563 buffer is sscanf()ed to unpack it. The new scheme would
564 be a binary format that allows quick packing and unpacking
565 of any data type with named fields.
567 Why: Using binary packing would be faster. Named fields will permit
568 error checking to ensure that what is sent is what the
569 receiver really wants.
571 Notes: These are internal improvements in the interest of the
572 long-term stability and evolution of the program. On the one
573 hand, the sooner they're done, the less code we have to rip
574 up when the time comes to install them. On the other hand, they
575 don't bring an immediately perceptible benefit to potential
576 users. Item 10 and possibly item 11 should be deferred until Bacula
577 is well established with a growing user community more or
578 less happy with the feature set. At that time, it will make a
579 good "next generation" upgrade in the interest of data
585 ======================================================
587 It is somewhat like a Full save becomes an incremental since
588 the Base job (or jobs) plus other non-base files.
590 - New BaseFile table that contains:
591 JobId, BaseJobId, FileId (from Base).
592 i.e. for each base file that exists but is not saved because
593 it has not changed, the File daemon sends the JobId, BaseId,
594 and FileId back to the Director who creates the DB entry.
595 - To initiate a Base save, the Director sends the FD
596 the FileId, and full filename for each file in the Base.
597 - When the FD finds a Base file, he requests the Director to
598 send him the full File entry (stat packet plus MD5), or
599 conversely, the FD sends it to the Director and the Director
600 says yes or no. This can be quite rapid if the FileId is kept
601 by the FD for each Base Filename.
602 - It is probably better to have the comparison done by the FD
603 despite the fact that the File entry must be sent across the
605 - An alternative would be to send the FD the whole File entry
606 from the start. The disadvantage is that it requires a lot of
607 space. The advantage is that it requires less communications
609 - The Job record must be updated to indicate that one or more
611 - At end of Job, FD returns:
612 1. Count of base files/bytes not written to tape (i.e. matches)
613 2. Count of base file that were saved i.e. had changed.
614 - No tape record would be written for a Base file that matches, in the
615 same way that no tape record is written for Incremental jobs where
616 the file is not saved because it is unchanged.
617 - On a restore, all the Base file records must explicitly be
618 found from the BaseFile tape. I.e. for each Full save that is marked
619 to have one or more Base Jobs, search the BaseFile for all occurrences
621 - An optimization might be to make the BaseFile have:
627 This would avoid the need to explicitly fetch each File record for
628 the Base job. The Base Job record will be fetched to get the
629 VolSessionId and VolSessionTime.
630 =========================================================
634 =============================================================
636 Request For Comments For File Backup Options
639 Subject: File Backup Options
642 A few days ago, a Bacula user who is backing up to file volumes and
643 using compression asked if it was possible to suppress compressing
644 all .gz files since it was a waste of CPU time. Although Bacula
645 currently permits using different options (compression, ...) on
646 a directory by directory basis, it cannot do it on a file by
647 file basis, which is clearly what was desired.
649 Proposed Implementation:
650 To solve this problem, I propose the following:
652 - Add a new Director resource type called FileOptions.
654 - The FileOptions resource will have records for all
655 options that can currently be specified on the Include record
656 (in a FileSet). Examples below.
658 - The FileOptions resource will permit an exclude option as well
659 as a number of additional options.
661 - The heart of the FileOptions resource is the ability to
662 supply any number of ApplyTo records which specify POSIX
663 regular expressions. These ApplyTo regular expressions are
664 applied to the fully qualified filename (path and all). If
665 one matches, then the FileOptions will be used.
667 - When an ApplyTo specification matches an included file, the
668 options specified in the FileOptions resource will override
669 the default options specified on the Include record.
671 - Include records will be modified to permit referencing one or
672 more FileOptions resources. The FileOptions will be used
673 in the order listed on the Include record and the first
674 one that matches will be applied.
676 - Options (or specifications) currently supplied on the Include
677 record will be deprecated (i.e. removed in a later version a
678 year or so from now).
680 - The Exclude record will be deprecated as the same functionality
681 can be obtained by using an Exclude = yes in the FileOptions.
684 The following records can appear in the FileOptions resource. An
685 asterisk preceding the name indicates a feature not currently
689 - Compression= (GZIP, ...)
690 - Signature= (MD5, SHA1, ...)
692 - OneFs= (yes/no) - remain on one filesystem
693 - Recurse= (yes/no) - recurse into subdirectories
694 - Sparse= (yes/no) - do sparse file backup
695 - *Exclude= (yes/no) - exclude file from being saved
696 - *Reader= (filename) - external read (backup) program
697 - *Plugin= (filename) - read/write plugin module
700 - verify= (ipnougsamc5) - verify options
703 - replace= (always/ifnewer/ifolder/never) - replace options currently
705 - *Writer= (filename) - external write (restore) program
709 Currently options specifying compression, MD5 signatures, recursion,
710 ... of a FileSet are supplied on the Include record. These will now
711 all be collected into a FileOptions resource, which will be
712 specified on the Include in place of the options. Multiple FileOptions
713 may be specified. Since the FileOptions contain regular expressions
714 that are applied to the full filename, this will give the ability
715 to specify backup options on a file by file basis to whatever level
724 Include = compression=GZIP signature=MD5 {
733 Include = FileOptions=Opts {
744 That's a lot more to do the same thing, but it gives the ability to
745 apply options on a file by file basis. For example, suppose you
746 want to compress all files but not any file with extensions .gz or .Z.
747 You could do so as follows:
751 Include = FileOptions=NoCompress FileOptions=Opts {
759 ApplyTo = /*.?*/ # matches all files
764 # Note multiple ApplyTos are ORed
765 ApplyTo = /*.gz/ # matches .gz files */
766 ApplyTo = /*.Z/ # matches .Z files */
769 Now, since the NoCompress FileOptions is specified first on the
770 Include line, any *.gz or *.Z file will have an MD5 signature computed,
771 but will not be compressed. For all other files, the NoCompress will not
772 match, so the Opts options will be used which will include GZIP
776 - Is it necessary to provide some means of ANDing regular expressions
777 and negation? (not currently planned)
779 e.g. ApplyTo = /*.gz/ && !/big.gz/
781 - I see that Networker has a "null" module which, if specified, does not
782 backup the file, but does make an record of the file in the catalog
783 so that the catalog will reflect an exact picture of the filesystem.
784 The result is that the file can be "seen" when "browsing" the save
785 sets, but it cannot be restored.
787 Is this really useful? Should it be implemented in Bacula?
790 After implementing the above, the user will be able to specify
791 on a file by file basis (using regular expressions) what options are
792 applied for the backup.
795 =============================================
797 ==========================================================
799 For each Incremental job that is run, there may be files that
800 were found but not saved because they were locked (this applies
801 only to Windows). Such a system could send back to the Director
802 a list of Unsaved files.
804 - New UnSavedFiles table that contains:
808 - Then in the next Incremental job, the list of Unsaved Files will be
809 feed to the FD, who will ensure that they are explicitly chosen even
810 if standard date/time check would not have selected them.
811 =============================================================
814 Done: (see kernsdone for more)
815 - Heartbeat between daemons.
816 - Fix Dir heartbeat in restore and verify vol. Be sure to make
817 bnet_recv() ignore BNET_HEARTBEAT.
818 - Implement HEART_BEAT while SD waiting for tapes.
819 - Include RunBeforeJob and RunAfterJob output in the message
821 - Change M_INFO to M_RESTORED for all restored files.
822 - Fix command prompt in gnome-console by checking on Ready.
823 - Merge SQLite, MySQL, and Rel spec into a single file.
824 - Fix config of "console"
825 - Check if cancel works with FD (fixed).
826 - Properly configure console and gconsole (currently for source not
827 configured for installation).
828 - Error labeling tape from console gets Jmsg error because of no Job.
829 - Test and implement get_pint and get_yesno.
830 - Implement global with DB name and add to btraceback.gdb
831 - Remove subsysdir from conf files (used only in autostart scripts).
833 rufus-dir: Max configured use duration exceeded. Marking Volume "MatouBackup" as Used.
834 rufus-sd: Volume "" previously written, moving to end of data.
835 rufus-sd: Matou.2003-05-10_10.39.18 Error: I canot write on this volume because:
836 The number of files mismatch! Volume=1 Catalog=0
837 rufus-sd: Matou.2003-05-10_10.39.18 Error: askdir.c:155 NULL Volume name. This shouldn't happen!!!
838 - Shell character expansion is failing occassionally.
839 - Add a section to the doc on Manual cycling of Volumes.
840 - Check if Job/File retentions apply to multivolume jobs.
841 - Fix missing casette in autoloader during read:
842 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
843 14-May-2003 14:41 undef-sd: 3301 Issuing autochanger "loaded" command.
844 14-May-2003 14:41 undef-sd: 3302 Issuing autochanger "unload" command.
845 14-May-2003 14:42 undef-sd: 3303 Issuing autochanger "load slot 1" command.
846 14-May-2003 14:42 undef-sd: 3304 Autochanger "load slot 1" status is OK.
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 Warning: acquire.c:106 Volume name mismatch. Wanted TestVolume0005 got TestVolume0009
852 14-May-2003 14:42 undef-sd: 3301 Issuing autochanger "loaded" command.
853 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
854 14-May-2003 14:42 undef-sd: 3301 Issuing autochanger "loaded" command.
855 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".
856 14-May-2003 14:42 undef-dir: Bacula 1.31 (12May03): 14-May-2003 14:42
857 - Fix problem reported by Christopher McCurdy <cmccurdy@eecis.udel.edu>
858 xeon-fd: Could not stat c:/Documents and Settings/All
859 Users/Application Data/Humc:\Documents and Settings\All User98_AIX.kbf:
860 ERR=No such file or directory
862 - The following Re-read last block at EOT failed. ERR=block.c:523 Read zero bytes on device /dev/nrsa0.
863 undef-sd: block.c:523 Read zero bytes on device /dev/nrsa0.
864 apparently masks the standard EOM message.
865 - BSD (probably) does not have strtoll()
866 - BSD does not have ioctl() MTEOM
867 - BSD defines a number of MT_xxx variables which conflict
868 with those defined by Bacula.
869 - Make default duration days if no qualifier (e.g. s) is specified.
870 - BSDI fix finding gcc version
871 - When the FD errs (e.g. disk full) have a more graceful shutdown.
872 - Make sure Bacula prunes/purges canceled and failed jobs too and all
873 jobs with zero JobFiles.
874 - Implement Volume name checking.
875 - Document what characters can go into Volume names.
876 - Getting the following on all directories on Win32
877 19-May-2003 01:14 tibs-fd: Could not access c:/cygwin/home/kern/rxvt: ERR=Permission denied
878 - Cancellation caused JobMedia error:
879 babylon5-dir: Last FULL backup time not found. Doing FULL backup.
880 babylon5-dir: Start Backup JobId 416, Job=Zocalo_Save.2003-05-19_02.15.06
881 babylon5-sd: End of media on Volume VXA-V17-Inc-001 Bytes=31,982,900,672 Blocks=495,781.
882 babylon5-sd: Job Zocalo_Save.2003-05-19_02.15.06 waiting. Cannot find any appendable volumes.
883 babylon5-sd: Someone woke me up, but I cannot find any appendable volumes
884 for Job=Zocalo_Save.2003-05-19_02.15.06.
885 babylon5-sd: Zocalo_Save.2003-05-19_02.15.06 Fatal error: Job
886 Zocalo_Save.2003-05-19_02.15.06 canceled while waiting for mount on
887 Storage Device "Ecrix_VXA-1".
888 babylon5-sd: Zocalo_Save.2003-05-19_02.15.06 Fatal error: Cannot fixup device
889 error. Job Zocalo_Save.2003-05-19_02.15.06 canceled while waiting for
890 mount on Storage Device "Ecrix_VXA-1".
892 babylon5-dir: Zocalo_Save.2003-05-19_02.15.06 Error: Catalog error creating
893 JobMedia record. sql_create.c:125 Create JobMedia failed. Record already
895 babylon5-sd: Zocalo_Save.2003-05-19_02.15.06 Error: Error creating JobMedia
896 record: 1991 Update JobMedia error
898 babylon5-sd: Zocalo_Save.2003-05-19_02.15.06 Error: askdir.c:158 NULL Volume
899 name. This shouldn't happen!!!
900 zocalo-fd: Zocalo_Save.2003-05-19_02.15.06 Error: bnet.c:310 Write error
901 sending to Storage daemon:babylon5:9103: ERR=Broken pipe
902 - Volume names with spaces get jammed into the catalog with 0x1
903 i.e. the SD bashes the Volume but they are not unbased by Dir.
904 jerom-dir: MonthlySave.2003-05-10_17.12.01 Error: Unable to get Media
905 record for Volume Tape^A1: ERR=sql_get.c:788 Media record for Volume
907 jerom-sd: MonthlySave.2003-05-10_17.12.01 Error: Error updating Volume
908 Info: 1991 Catalog Request failed: sql_get.c:788 Media record for Volume
910 - ChangeServiceConfig2A does not exist on WinNT (ADVAPI32.DLL).
911 - Fix "access not allowed" for backup of files on WinXP.
912 - Check for existence of all new Win32 API's. See LoadLibrary in
914 - Count errors during restore and print them in the Job report.
915 - Bug: fix access problems on files restored on WinXP.
916 - Put system type returned by FD into catalog.
917 - Finish WIN32_DATA stream code (bextract, check if can handle stream)
918 - Make SD keep track of Files, Bytes during restore.
919 - If you enter the userid by hand for restore, you get:
920 Enter JobId(s), comma separated, to restore: 74
921 You have selected the following JobId: 74
922 Building directory tree for JobId 74 ...
923 134645140 items inserted into the tree and marked for extraction.
924 - Add SDWriteSeqNo to SD, and probably Read on FD side.
925 - If bootstrap is non-zero for restore, do not show JobId in the
926 OK to run? (yes/mod/no): list.
927 - When all cassettes in magazine are used, got:
928 22-May-2003 18:24 undef-sd: 3304 Autochanger "load slot 1" status is OK.
929 22-May-2003 18:24 undef-sd: NightlySave.2003-05-22_14.08.16 Warning: mount.c:245 Director wanted Volume "TestVolume0009".
930 Current Volume "TestVolume0005" not acceptable because:
931 1998 Volume "TestVolume0005" not Append or Recycle.
932 22-May-2003 18:24 undef-sd: NightlySave.2003-05-22_14.08.16 Error: Autochanger Volume "TestVolume0009" not found in slot 1.
933 Setting slot to zero in catalog.
934 22-May-2003 18:24 undef-sd: Please mount Volume "TestVolume0009" on Storage Device "ARCHIVE 4586" for Job NightlySave.2003-05-22_14
936 Use "mount" command to release Job.
937 22-May-2003 19:24 undef-sd: Please mount Volume "TestVolume0009" on Storage Device "ARCHIVE 4586" for Job NightlySave.2003-05-22_14
939 Use "mount" command to release Job.
940 - Don't zero the Slot when the wrong volume is found -- simply ask
942 - Implement MTIOCERRSTAT on FreeBSD to clear tape error conditions.
943 - Shell expansion fails for working_directory in SD from time to time.
944 - File the Automatically selected: xxx
945 to say Automatically selected Pool: xxx
946 - Default duration with no qualifier is sec should be 1 day
947 - zap sd_auth_key in SD after FD connection.
948 - Find a solution for the multiple FileSet problem (when it is changed). Add date?
949 - Look at Python for a Bacula scripting language -- www.python.org
950 - When Marking a file in Restore that is a hard link, also
951 mark the link so that the data will be reloaded.
952 - Emergency restore info:
954 - Backup working directory
956 - Why don't we get an error message from Win32 FD when bootstrap
957 file cannot be created for restore command?
958 - Fix Win2000 error with no messages during startup.
959 - Make restore more robust in counting error and not immediately bailing
960 out. Also print error message once, but try to continue.
961 - Add code to check that blocks are sequential on restore.
962 - Remove "rufus" and such references from regress.
963 - No READLINE_SRC if found in alternate directory.
964 - If ./btape is called without /dev, assume argument is a Storage resource name.
965 - Find general solution for sscanf size problems (as well as sprintf. Do at run time?
966 - Bytes restored is wrong.
967 - The "List last 20 Jobs run" doesnt work correctly in restore.
968 It doesnt show the last 20 jobs , but some older ones.
969 - Fix Verify VolumeToCatalog to use BSRs -- it is broken.
970 - Implement Release Storage=xxx
971 - Fix restore on Win95/98
972 - Remove the Jmsg() in sql_find.c:102 or only print on hard error.
973 - Implement FileSet VolIndex -- done, but must update old records.
974 - Check this below from Phil.
975 This was SD reported data rather than FD data!
976 > When the job was done, Bacula reported 11084 files restored:
979 > Job: Zocalo_Restore.2003-06-05_16.42.01
981 > Start time: 05-Jun-2003 16:42
982 > End time: 06-Jun-2003 01:21
983 > Files Restored: 11,084
984 > Bytes Restored: 65,474,772
986 > FD termination status: OK
987 > Termination: Restore OK
989 > when it should probably have reported 11084 files scanned, 250 restored.
990 > The bytes restored count looks about right.
992 - Should Bacula make an Append tape as Purged when purging?
993 - Use switch() in backup.c and restore.c in FD instead of giant if statement.
994 - If during a restore, a hard linked file already exists (on option), delete
995 the file and re-link it. This is to avoid the possibility that the
996 user had re-linked the file between the backup and the restore.
997 Do lstat() to see if it is already properly linked.
998 Same for symlinked file.
999 Make sure ifnewer, ifolder, never, ... apply correctly.
1000 - Flag so that no connect does not error, and Reschedule a job.
1001 - Implement "Reschedule OnError=yes interval=nnn times=xxx"
1002 - That restoring a hard link that already exists works correctly.
1004 - Make Pool resource handle Counter resources.
1005 - Fix first block number after label to be zero instead of 1 (reset after label).
1006 - Grep for Backup OK in regression script.
1007 - Do NOT reuse same JobId if tape written.
1008 - Implement non-blocking writes and bsock->terminate in heartbeat
1009 thread, or set it in status.c cancel (used pthread_kill() instead of
1011 - Add restore to specific date.
1012 - Instrument use_count on DEVICE packets and ensure that the device is
1013 being close()ed at the appropriate time.
1014 - Test long path names (>64 chars) in Windows -- crashes FD?
1015 - Implement fast block rejection: match_bsr_block().
1016 - Complain if record dropped in bnet_recv because too long.
1017 - Test multiple simultaneous Volumes
1018 - Document recycling algorithm.
1019 - Make Restore report an error if FD or SD term codes are not OK.
1020 - To link with mysqlclient_r may require -lssl -lcrypto
1021 - Document Heart beat code
1022 - Non-fatal errors are not correct counting attribs.c:277
1023 - Check that Block number in JobMedia are correct.
1024 - The bsr for Dan's job has file indexes covering the whole range rather
1025 than only the range contained on the volume.
1026 Constrain FileIndex to be within range for Volume.
1027 - Pass prefix_links to FD.
1028 - Fix restore list of volumes if Volume not selected.
1029 - Complete (or turn off) the command line code in update.
1030 - Testing Tibs job erred and hung director on Storage resource. This was
1031 because there were a whole pile of jobs hanging around in the SD
1032 waiting for a connection from the FD that was never coming.
1033 - Possibly update all client records at startup.
1034 - Listing of Win32 restored files has ??????? for user.
1035 - In Win portable restore the directory is not create
1036 27-Jun-2003 16:52 tibs-fd: kernsrestore.2003-06-27_16.52.20 Error:
1037 create_file.c:175 Could not create
1038 /tmp/bacula-restores/cygwin/home/kern/bacula/k/src/dird/dird_conf.o: 0
1039 ERR=The system cannot find the path specified.
1040 - Finish Windows implementation (add setting of correct type on restore,
1041 add Portable Data Format flag).
1042 - Document portable=yes
1043 - Document c:/working directory better than /working directory.
1044 - Document run "yes".
1045 - Document update volume: jobid, current, before, all
1046 - Document that bscan does not work with multiple simultaneous jobs.
1047 - Update Automatic Volume Labeling in disk.wml
1048 - Figure out how to handle DHCP IP addresses -- use dynamic DNS.
1049 - Add OS, GCC version to traceback output.