1 Technical notes on version 1.37
8 - Ensure that the drive is closed before calling the
9 mtx-changer script so that the script can access the drive.
10 - Add drive name to reserved Volume list printout in SD.
12 - Fix bug in acquire.c that incorrectly reported volume
14 - Add additional debug code and messages in reserve.c
15 - Eliminate unwanted warning message in reserve.c
17 Changes to 1.37.36 release 22Aug05:
19 - Landon's fix for NLS detection
20 - Eliminate incorrect compiler warning on FreeBSD.
21 - Move Win32 errno message build into berrno constructor.
22 - Minor Win32 build tweaks
24 - A number of minor build Win32 fixes.
25 - Remove a PostQuitMessage() as suggested by Thorsten so that
26 BartPE restore can work correctly.
27 - Fix for create JobMedia so that VolIndex remains valid even
28 during a delete Job or pruning -- bug 402.
29 - Minor tweak for Win32 build.
31 - Win32 fix -- remove debug O_NONBLOCK code.
32 - Fix bug 399 -- make_catalog_backup does not work for sqlite3
33 - Implement unloading a volume in a different drive if it
34 is needed in the current drive.
35 - Implement search for unused autochanger drive.
36 - Implement search for exact Volume in reservation before
38 - Fix picking up drive in Dir so that it is not done in
40 - Eliminate double check on "loaded" for autochanger.
42 - Start coding better reservation algorithm
43 - Always look for slot for label command.
44 - Add more debug code for autochangers.
45 - Apply fix from Stephan Leemburg <sleemburg@jvc.nl> for
46 improper scanning of schedule resource:
47 Run = Level=Full Pool=Catalog daily at 1:20
48 - Apply patch from Chris Lee <labmonkey42@gmail.com> for
49 adding --enable-build-dird --enable-build-stored.
50 - Tweak datadir definition in configure.in
52 - Fix bug that missed drive=nn specification.
53 - Eliminate nonblocking kludge in heartbeat of FD as
54 it caused high CPU usage.
55 - Pickup loaded slot when doing open() of tape drive.
56 - Make autochanger reservation code go through full list
57 first pass until exact match found.
59 - Fix how FileSet is saved in job record to correct continual
61 - Make datadir print on config.out listing.
62 - Move get FileSet record up in backup init to eliminate continual
63 Full save seen by Peter Sjoberg.
64 - Add VolumeName to read-only Python variables.
65 - Add VolumePurged event for Python.
66 - Suppress /dev/ Filesystem change prohibited INFO messages.
67 - Do not delete FT_RAW files before restore (allows FIFOs
68 to be used for restore).
70 - Add drive specification to mount, unmount, release, label,
71 and relabel for Autochangers. Note Dir<->SD protocol has
76 - Disable parts of NLS as the configure does not work here.
77 - In job backup init (backup.c) define definitive Job level and
78 since time, *then* apply Job Pool override selection if any.
80 - Modified bconsole script so that it is improperly
81 installed, it will refuse to execute. This avoids
86 - Apply David's ACL fix to src/filed/acl.c
88 - Apply patches sent by David Duchscher <kreios@gmail.com> for
89 making ACLs work on MacOS X and FreeBSD.
91 - Apply patch in bug#397 that improved configure
92 - find readline under $with_readline/include/readline
93 - no libutil under Solaris
94 - no need for -ldl under Solaris
95 - Make reservation system single threaded during the
96 search to avoid two threads competing for the same
98 - Correct a return code in find_suitable_device_for_job()
99 Possibly cause of "busy writing to another volume".
101 - Modify open() for tape so nonblocking really works.
102 - Use fcntl() to reset blocking status rather than close()
103 and reopen the drive.
104 - Make sure dev->open() is always called so that any change
105 in read/write permissions will occur.
106 - Open drives initially in daemon in read-only mode.
107 - Ensure that each time the VolHdr.VolumeName is zapped
108 or changed that free_volume() is called on the old name.
112 - Require 5 arguments to mtx-changer except list and slots
113 - Turn -EPIPE status returns from bpipe to ETIME
114 - Include Slot in SD status output
115 - Do not term_dev() during initialization in SD if the device
116 could not be opened. In the case of a tape drive, there may
117 be no tape in the drive.
121 - Correct PostgreSQL database scripts as suggested by a user.
122 - Add additional info to FATAL message generated when a device
123 is busy writing to another volume.
124 - Suppress an inappropriate NULL Volume name message after a cancel.
125 - Correct a warning message in reserve.c
127 - Apply user's patch to make mutiple modifiers for times
129 - Make read_dev_volume_label() handle ANSI/IBM labels
130 correctly -- ie space over any label at the beginning
133 - Make ANSI/IBM writing of HDR1/2 labels ignore any
134 errors if at end of tape.
135 - Apply Martin's patch to improve Python detection in
137 - Temporarily turn off disk seeking until I find the
138 cause of the problem.
140 - Add OSF1 patch supplied by user.
141 - Use number of files selected from write_bsr() only
142 if it is not defined.
143 - Explicitly seek to end of file when getting size for
145 - Correctly set EndBlock position in JobMedia record
147 - Remove unnecessary set StartBlock in bscan. Caused bscan
150 - Modify mtx-changer to wait a maximum of 300 seconds.
151 - Do restart of failed jobs only for Backups job types.
152 - A number of DVD updates from Nicolas.
154 - Turn off old service helper code in Win32.
155 - Correct Messages bug found by Phil in stored.
157 - Complete (almost) documentation of 1.38.
158 - Add error messages for error conditions with VSS.
159 - Fix additional problems with VSS backup that I introduced.
162 - Correct compiler complaints in wx-console and tray-monitor.
163 - Correct VSS problems recognizing c:
164 - Add VSS before job status
165 - Fix output of status from being one big line.
166 - Change cd xx; make to cd xx && make as suggested by Phil.
167 - Cleanup projects file
168 - Remove unnecessary casting of FF_PKT in filed.
169 - Apply Thorsten's bugfix for vss_generic.cpp
170 - Add check for df path for dvd_freespace
171 - Use df to get space used on DVD.
172 - Change sense of flag indicating erase DVD or not before writing.
173 - Fix bpipe so that it never modifies the result pointer.
174 - Replace more dev_name by print_name().
175 - Rewrite edit_device_codes_dev() so it does not overwrite the
177 - Update printing of labels (for bls) so that critical information
178 is printed (Job name and timestamp).
179 - Cleanup old spool files when starting the SD.
180 - Modify vss.cpp to allow C: as a path name. Otherwise VSS
181 doesn't work when only a drive name is given.
182 - Modify vss.c to eliminate double / in filenames.
183 - Update doc -- particularly the restore chapter.
185 - Make all files in working directory have .xxx at end.
186 - Work on DVD writing.
187 - Fix keepatime bug (bugs database).
188 - Move Python variables from Job to Bacula. They are
189 DirName, Version, ConfigFile, and WorkingDir
190 - Fix delete of bootstrap to only occur on Bacula created
192 - Allow cancelling a Job name that is not active. It is
195 - Fix name space pollution by OpenSSL 0.9.8 reported by
196 Matthias Kurz -- applied his patch.
197 - Fix bpipe.c so that it does not modify results pointer.
198 ***FIXME*** calling sequence should be changed.
199 - Remove some remaining references to dev_name.
200 - Fix calls to mount_dev() and unmount_dev() to
201 correspond to returned value (bool instead of int).
202 - Try without success to make DVD writing work.
204 Changes to 1.37.30 released 16 July 05:
206 - Fix "dir" command scanning field misalignment in
208 - Switch to using the wxWidgets Unicode library.
209 - Include msvcr71.dll in distribution.
210 - Add VSS to status line in Win32 FD if enabled.
211 - Get VSS build scripts working with Thorsten's help.
212 - Unlink the bootstrap file after sending it to
214 - Remove sending include/exclude lists to the FD during
215 a restore -- deprecated code.
216 - Cleanup the bootstrap files in the FD.
220 - Remove old commented out code from configure.in
221 - Add baculavssfd.mak file for building VSS version of
223 - Correct date (year) in vss.cpp and vss_generic.cpp
225 - I finally found and squashed the elusive SD crash.
226 I needed to initialize the used volume list before
227 firing off the device initialization thread.
230 - Make sure that bpipe results are zapped even on
232 - Lots of documentation.
233 - Do not prune volume marked as append when needing a
235 - Print a warning message in SD if a non-used Volume
236 is specified and autolabel not turned on.
237 - Correct a bug in chksum.c concerning SHA1 signatures
238 (an * should have been & when checking for a bit flag).
239 - Print File:Block for all label records in label.c -- concerns
240 primarily bls when doing Job listings (-j).
241 - Correct is_volume_in_use() to return false if testing
242 on the same device where the Volume is already mounted.
243 - Define a init_done flag in the SD that is set when the
244 devices are initialized and make users connecting wait.
245 This prevents useless connect failure warning messages.
246 - Do additional device locking in ask_op_to_mount_volume()
247 to prevent race conditions with a user labeling a Volume
250 - Add a test for error return from bnet_wait... in heartbeat.c
251 in FD to avoid CPU loop.
252 - Implement TLS in gnome console and wx-console.
254 - Correct a NULL pointer reference in the mount command.
255 - Correct typo in Copyright
256 - Add detection of EOM for IBM drives (i.e. errno == ENOSPC)
258 - Remove temp file created in mtx-changer script.
259 - Make fsf_dev() into a class method.
261 - Modify mtx-changer.in script to return slot:barcode for
262 Volumes that are loaded in the drives.
263 - Correct some more places where dev->is_blocked() needs
264 to be checked in dircmd.c in SD.
267 - Add code to ensure that reserved but unused volumes
269 - Correct how Volumes are mounted and handled so that the SD
270 does not get stuck if multiple volumes are used (recycling,
272 - Correct bug where you could relabel a volume while it
273 was being acquired -- created chaos.
275 - Correct seg fault caused by open() calling sequence change.
277 - Add new rc-chio-changer script by Rudolf Cejka to
278 examples/autochangers
279 - Apply Rudolf's changes to bacula.in
280 - Expand the space from 8 to 10 characters in editing
281 file sizes for restore and dir of catalog, otherwise
282 GB sizes are truncated -- fixes bug report.
283 - Modify wx-console to know about 10 character widths.
284 - Allow decending into top level directory if "recurse=no"
285 is set. Fixes a bug report.
286 - Install pthreadVCE.dll when installing console or wx-console
287 on Win32 systems. Fixes bug report.
289 - Tweak dvd-writepart script to prevent door from opening/closing
291 - Remove GROUP BY in several PostgreSQL commands to prevent error.
293 - Ensure that < as first character of filename list is not treated
294 as a directory for restore.
295 - Add debug to heartbeat in FD as it seems to go into an
296 infinite loop from time to time during SD failure in DVD writing.
297 - Add more debug code to dvd writing.
298 - Attempt not to destroy existing fs on DVD.
300 - Detect device mounted for DVD and suppress be sure to
301 mount message after label.
302 - Set Cleaning tape status to "Cleaning" and force no
304 - Get DVD writing working with new standard Bacula open()
306 - Rename get_filename() to make more sense.
307 - Detect "is already mounted on" on mount command so to avoid
308 error if device is already mounted.
309 - Eliminated guess_name() code. It may be necessary to
311 - Eliminate seg fault from printing invalid results.
312 - Make dvd_write_part() bool.
315 - Attempt to fix DVD writing by eliminating a number of the
316 DVD subroutines to simplify.
317 - Modify DEVICE::open() to take dcr as first argument. This
318 will permit providing more info to DVD opening.
319 - Fix scanning for time/size items which in some cases
321 - Eliminate read_dvd_volume_label(). New code (not yet written)
322 *must* open dvd appropriately before calling
323 read_dev_volume_label.
324 - Modify open_first_part() open_next_part() to take DCR as
326 - Make label command from console work on DVDs.
327 - Make mount command from console work on DVDs.
328 Unmount does not work yet.
332 - Add Database vendor to CatalogRes tuple for Python.
334 - Implement DoesVolumeExist(Vol) for Python.
335 - Prevent python command from seg faulting if no arg given.
339 - Add set_mode method in DEVICE.
340 - Correct set_mode method in DEVICE
341 - Add more DVD debug info
343 - Check for incorrect duration and size modifiers in conf files.
345 - Make Version a tuple (version, build-date)
346 - Add CatalogRes tuple (DBName, Address, User, Password,
348 - Add Version, ConfigFile, and WorkingDir as Python attributes
350 - Implement code (principally for Win32) that on failure to
351 create a file, it will cd into the directory and attempt
352 to create the file using a relative path. This avoids creating
353 files with paths which fail on Win32.
354 - Fix parsing of times and sizes with decimal numbers.
355 - Make free_volume_list() in SD work if vol list is not
356 initialized (./bacula-sd -t).
358 - Add debug error printout when open() fails.
359 - If open() of DVD fails in mount.c, return false.
360 - Split open() code for DVD into separate subroutine in dev.c
362 Changes to 1.37.25 released on 20 Jun 05:
364 - Remove (cd gnome-console; make clean) from src/Makefile.in
365 - Fix bug where Storage daemon gets confused about what
366 tape is mounted. (one line of code was inadvertently
371 - DVD writing/reading seems to be mostly working.
372 - Set execute bits on dvd-freespace and dvd-writepart
373 - Make dvd-freespace use existing dummy file.
374 - Modify dvd-freespace to pickup size from Track Size:
376 - Add Date, Job, level to updates to .bsr file in
378 - Add debug info to dvd-freespace.in
379 - Fix hard coded bacula.sql in make_catalog_backup reported
381 - Make sure a verify volume to catalog never reports an error
382 if there are zero files to verify.
383 - Remove confusing debug info in filed/backup.c on network
385 - Make sure output from console is not sent to system log.
386 - Convert open_dev() into a class method.
387 - Change VolHdr.VolName to VolHdr.VolumeName.
388 - Add a flag in the device state word to indicate that
389 we found Media in the drive (DVD).
390 - Make mount_dev() and unmount_dev() return bool.
394 - Fix the same state variable problem in open_next_part().
395 - C++ify the SD code a bit more.
397 - Add more debug code and clarify debug code for DVDs.
398 - Do not save and restore state in open_first_part()
399 since state should be properly set after open_dev().
400 - Make default mandir /usr/share/man
401 - Install Bacula man page.
402 - Implement passing of FileSet Enable VSS to FD.
403 - Move main body of reserve drive code into subroutine
404 so that it can be called multiple times.
406 - Eliminate getpass() for Win32 builds in console as the
407 function does not exist.
409 - Correct some reservation problems in SD when no devices
411 - Start removing #ifdef HAVE_TLS by sneaky tricks.
412 - Begin implementation of TLS in wx-console
413 - Remove ignoring SIGCHLD from console.
414 - Rework the dlist binary search routines for implemenation
415 of the Volume reservation code -- make it more general.
416 - Strip double slashes // from Win32 filenames in an attempt
417 to resolve restore problems on some systems.
418 - Fix a minor bugs in the trace code that caused the first
419 line output to be lost.
420 - Implement a good first cut at adding Volume reservation code
421 to the storage daemon (in file reserve.c).
422 - Remove old unused code from the tree.c routines.
426 - Fix compile problems on Win32
427 - Start writing Volume reservation list (already exists, but
428 is not really very good).
429 - Implement attribute caching to put Signature into database
430 at the same time as the file attributes thus eliminating a
431 number of database accesses.
432 - Correct a reservation problem.
433 - Implement full Dir Storage use.
434 - Reduce a bit of TLS #ifdeffing.
440 - Add more documentation to mtx-changer.in
441 - Correct link to manual in authenticate.c in various
443 - Create a new src/stored/reserve.c file where the
444 Use Storage command is processed and drives are
446 - Modify src/stored/autochanger.c to keep track of each
447 Slot that is loaded for each device.
448 - Ensure that changer_command and changer_name are picked
449 up from Autochanger resource if not specified, and if
450 neither is specified, err.
452 - Fix bextract.c compile problem
454 - Make make distclean clean a bit better
456 - Remove old code in jcr.c
457 - Make testls release jcr chain when terminating.
459 - Implement Maximum Job Spool Size (actually DCR based)
461 - Use light weight non-recursive locking on jcr chain.
462 - Make JCR a class and implement inc_use_count() and
463 dec_use_count() methods that ensure that the jcr is
464 locked when inc/dec the use count.
465 - Remove the global jcr lock when traversing the jcr
467 - Use dlist to implement the jcr chain rather than hand
468 crafted next and prev links.
469 - Lock the jcr chain inside each function that modifies
474 - Fix compile problem of ua_restore.c on broken compilers.
475 - Apply patch from bug 326 to permit bacula status by any user.
476 - Fix bug 325 -- conversion of 12:30pm to 24hour time.
478 - Put Dmsg() on inside if() to avoid calling subroutine.
479 - Make restore.bsr have unique name.
480 - Allow user to define bsr filename on restore command line
481 with bootstrap=xxx.bsr
482 - Add limit=nnn to "list jobs" command.
483 - Remove old restore code that did not use .bsr file.
484 - unlink automatically generated bsr file.
485 - Cleanup heartbeat code so that duped fd is almost sure
486 to be released. Previously under certain conditions, the
487 memory was not released due to race conditions.
490 - Unify the reserve_device() for a single device into one subroutine.
492 - Modify wait during use_device to happen only after all devices
493 have been examined rather than in the reserve_device code.
494 - Correct updating count of number of Volumes in a pool.
498 - Add more debug to SD for Autochangers + status output.
499 - Add Scratch to PoolType in PostgreSQL make...tables and do not
500 permit NULL PoolTypes. Fix for bug 319 reported by Eric.
502 - Add quotes around filename in parse_config error message. Bug
505 - Change nested \include to \input so that sections are properly
506 included in the pdf manual -- update the Web site.
507 - Set reconnect flag in MySQL packet to 1 to ensure that connection
508 is re-established. MySQL 5 changed default to 0. Fixes bug report.
509 - Fix Scratch pool handling as reported in a bug by Eric Bollengier
510 by applying his patch.
511 - Remove delete job in favor of delete jobid.
512 - Add = NULL to configfile definitions as reported by Eric in a bug
514 - Update winbacula.nsi.in to reflect new manual file structure.
516 - Correct a minor build problem with wx-console.
517 - Add cancel() to Dir Python scripting.
518 - Re-correct bug in parse-config error handling.
519 - Reorganization of use_command in SD to permit
520 waiting and multiple drive autochanger support.
522 - Correct bug in parse_config error handling.
523 - Where ever possible mark a volume in error or not
525 - Fix bug in changing tape pools after first backup. Reported
527 - Enhance mtx-changer to use Working Directory as temp.
528 - Remove all but initial setup locking of Res in SD.
530 - Add Client OS type to Job report.
531 - Add version to manual
532 - Update the Web site to have a single page for
533 the documentation links.
535 Changes to 1.37.18 release 08May05:
537 - Correct attribute definition compile error in 1.37.17
538 - Correct inverted order of CreateTime and MD5 pointed out
539 by a user in a bug report in FileSet db routine causing
540 the MD5 to print in the output instead of the time/date.
544 - Implement cstrlen() in sql list routines.
545 - Implement caching of attributes to add the
546 signature so that only one DB call will be made
547 per file. Not yet turned on.
548 - Fix Win32 build for TLS.
549 - Optimize File pruning to eliminate one database call.
550 - Fix bug that prevented File pruning from working.
551 - Implement a cstrlen() which returns the character
552 length of a UTF-8 string.
554 - Move test for MaxStartDelay as suggested by Peter.
555 - Implement Python methods (I had to read the Python source
557 - Implement run() method in Director.
558 - Add Priority and Scheduled time to Job report.
559 - Add JobInit and JobRun events.
560 - Add Priority as Python read/write attribute to Job.
561 - Correct typo in bsmtp reported by Jo.
563 Changes to 1.37.16 released 05May05
565 - Make a few tls ifdef tweaks.
566 - Fix create_file.c Win32 problem pointed out by
568 - Fix really ugly bstrncpy() but found by Thorsten.
569 - Move winapi.h/c from findlib to lib for inclusion in
572 - Thorsten Engel finished his work on Win32 Unicode. We
573 now have a single executable that runs on all Win32 machines.
574 - Move job initialization code after job scheduling so that
575 Verify jobs check for the prior JobId after they are really
576 started rather than before.
577 - Fix lib/fnmatch.c so that it does proper testing before folding.
578 - More documentation -- at tls and ansi labels chapters.
579 - Fix fileset_convert.pl to handle empty Exclude statements.
580 - Turn regex back off in Win32
582 - Fix sign extension problem in lex.c that reads UTF-8
583 with Chinese characters incorrectly.
587 - Remove a few HAVE_TLS #ifdefs
588 - Implement final Python style interface. More implemention to
589 be done, but the interface should change little if at all.
591 - Make default no tls support. You must add
592 --with-openssl to get tls support.
594 - Update Web header to include google search in the search
595 box -- thanks to input from Michel Meyers.
596 - Fix md5sum so that it builds with the new openssl stuff.
597 - Take some enhancements to the md5sum test program proposed
600 - Make option 2 of restore work better (list last 20
601 locations for a file).
602 - Make SD re-open a device with the right permissions if
603 it was previously opened with something different. This
604 should allow reading read-only Volumes under all circumstances.
605 - Implement restore of a single directory.
606 - Apply Tru64 patch supplied by Pascal Pederiva <freebsd@paped.com>
607 - Apply Unicode fixes for Win32 from "Thorsten Engel"
608 <thorsten.engel@matrix-computer.com>
609 - More work on Python read feature in FD.
611 - Fix (hopefully) the ftello() overflow reported by Peter.
612 - Landon Fuller committed his TLS patch.
613 - Fixed two minor warnings in console.c with TLS turned off.
614 - Updated the Makefile.in for wx-windows and tray-monitor
615 to handle TLS. I can only build the tray-monitor.
616 - Fix bscan to open tape in read-only mode (actually fix
617 it so that it doesn't use the standard open routine).
618 - Correct what appears to be an error in setting the
619 return value in dvd.c
621 - Get FD Python running -- design Python backup interface.
622 - Fix seg fault in SD when referencing Alert Command.
623 - More documentation.
624 - Fix one more thing in Win32 build.
628 - Put in production here
632 - Fix SQLite and PostgreSQL table creation script syntax
634 - Fix new Python code to work for Director.
635 - Move lib/python.c to lib/pythonlib.c so that debug output
636 is easier to read (can distinguish lib from dird, ...).
637 - Cleanup Python build so that Python is not dragged
638 into programs that don't use it.
642 - Make Bacula build without Python
643 - Drop Status table in drop_ scripts -- bug 283.
645 - First cut of new Python implementation.
647 - Correct SuSE autostart installation directory name --
649 - Fix director crash if Name directive not supplied
651 - More cleanup of changing %u to %s for DB IDs.
652 - Replace dev_can_write(dev) with dev->can_write().
654 - Make fsr_dev() a method dev->fsr()
655 - Remove Created new FileSet message as it always comes out in
657 - eliminate dev_is_tape() in favor of method dev->is_tape()
658 - Turn on disk seeking during restore for Disk Volumes. This
659 required some changes to bscan -- they seem a bit suspect as
660 they are not symmetrical for tape/files.
661 - Remove some of the recursion in stored/parse_bsr.c
662 - Move rescue out to a new Bacula project.
664 - Update of web site to replace old released doc with LaTeX
666 - Replace logo in doc with new bat logo.
667 - Make a good number of updates to the manual.
668 - Implement Python scripting using a Python object.
669 This makes interfacing with Bacula cleaner (object
671 - Add Phil's Status table to the database -- used
672 for getting a text description of the status
674 - Modify FileSet so that an empty Options (no
675 pattern matching) with an exclude will exclude
677 - Modify FileSet so that no top level included File
678 item can be excluded by a matching pattern.
679 - Suppress Created new FileSet record in daemon output.
680 - Implement Python in FD.
681 - Turn off old bfile reader code -- to be replaced
684 - Remove more recursion in src/dird/bsr.c as pointed out
687 - Increase the index width of Filename and Path entries
688 to 255 chars -- suggestion of Meno Abels.
689 - Change remaining VARCHARS to TINYBLOBs.
690 - Remove recursion from is_volume_selected() following
691 bug report from Peter.
692 - Implement mostly correct handling of use_storage in the
693 SD with full lists, ...
695 - Reset NumVols in Pool record from database on every update
697 - Modify DB to support multiple simultaneous copies and
699 - Pass copy and stripe between DIR and SD and put into
700 the JobMedia DB record.
701 - Update and test SQLite and MySQL datebase creation and
703 - Implement version 9 of the DB.
705 - Convert more atoi to str_to_int64() for DB.
706 - Implement filling in NumVols by querying DB rather
707 than trying to keep track of it.
708 - Add storage name to string passed to in use storage=
709 - Fix newVolume() so that the Python script is always
711 - Fix handling of pool,PoolId, and storage in ua_output.
712 - Same fix in ua_status.c
713 - Remove required locking of resources
714 - Replace pthread_cond_signal() by pthread_cond_broadcast()
715 hoping to fix the /lib/tls hang problems (lost signal).
716 - Move resource locking seaching from parse_conf.c to res.c
718 - Modify end of volume handling so that fixup_... does not
719 redo what block.c has already done -- writing Vol info to
720 DIR. This fixes a bug with bad numbers of files on a tape
721 when it filled as reported by Peter.
722 - In release_device() do not update the DIR on the Volume
723 info if the the information was already written at the
726 - NOTE!!!! This version has a new DIR <--> SD protocol. Both
727 must be updated at the same time.
728 - Begin implementation of passing all the Storage and Device
729 possibilities to the SD for examination during the reserve
731 - Modify the reserve and acquire code in the SD to make a
732 job wait if the device is not available.
733 - Implement New Volume Each Job in DIR and pass to SD, not yet
735 - Remove init/update of the Device resource in DIR
736 - Remove passing PoolId to SD and back.
738 - Remove \a and -e from error echos in most Makefiles.
739 - Add more debug code when there are errors on the tape
740 to try to find Peter's tape problem.
741 - Add wait.c (oops forgot previously).
742 - Move all the reserve/acquire_device_for_read/append to have
743 only a DCR as the argument.
744 - Rework the reserve_device_for_append() in stored to wait
745 if the drive is not available. Note! This is a short
748 - Comment out Multiple Connections in the document.
749 - Move the P() and V() to subroutines so that they can be accessed
750 from class methods. The reference to strerror() caused problems.
751 - Implement new DEVICE class methods block() and unblock() that
752 do what was previously done in 3 lines of code.
753 - Implement wait_for_device(), which will wait for any device
754 to be released then return. This requires a new global mutex
755 and condition variable, and is implemented in src/stored/wait.c
756 - Change the code in reserve_device_for_read(), which previously
757 failed the job to use the new device wait code.
759 - Apply reschedule patch to 1.37 code.
760 - Add copyright to title page of manual so it is clear.
761 - Create patch for rescheduling problem found by Ludovic. Storage
762 pointers were lost during rescheduling.
763 - Attempt to fix 2.6 rescue disk -- failed!
764 - Start working on adding a wait routine in the SD.
765 - Cleanup some old invalid doc in watchdog.
766 - Convert a number of references to dev->dev_name to dev->print_name().
767 - Add new wait.c file to SD.
768 - Add a few more methods to DEVICE in SD to cleanup code a
769 bit -- implement a few of the methods.
771 - Fix more print_name()s for printing device name.
772 - Modify open_dev to try 10 times every 6 seconds to
773 open the device if it gets an I/O error (meaning no
774 volume mounted). This gives a bit of settling in time
775 for an autochanger and avoids spurious messages.
776 - Change all yes/no to yes|no in the manual.
777 - Fix win32 create_file.c typo.
778 - Fix a typo in an error message.
780 - Detect if fseeko exists with autoconf. If so, use it
782 - Remove old bacula-*.conf from examples directory (out
784 - Remove latex-fr index files from CVS.
785 - Rewrite code that stops reading the tape so that the
786 tape is marked at EOT, then once the work is done,
787 the EOT flag is removed.
788 - Flush output to file after every send in console.
789 - Make setting VolFiles to smaller number fatal.
790 - Disable Multiple Connections code.
791 - Add patch from user for NetBSD statvsfs() fix to
793 - Take more care with errors in acquire.c
794 - Don't run through dvd code in append.c if bad status
796 - Modify code so that an autochanger fault is fatal.
797 - Use dev->print_name() in more places.
798 - Implement dev->can_steal_lock() to simplify code.
799 - Make btape re-read first 10000 records on fill command.
800 - Check error return and fail job from fseeko and ftello
801 in spool.c. Don't let a -1 slip in as size.
805 - Apply NetBSD patch from kardel in bug 258.
807 - Add a second job and a second client to the default
808 bacula-dir.conf file.
809 - Remove old style Include/Excludes.
810 - Fix ANSI labels to put EOF1 and EOF2 after each file mark.
811 - Add Python to SD and FD.
813 - Implement IBM labels
814 - Implement EOF and EOV labels at the end of a volume.
815 - Fix a rather ugly problem with the PoolId not getting
816 passed correctly. Now the DIR passes the Pool name and
817 Media Type to the SD, who passes them back when requesting
818 the next Volume. The DIR then looks up the correct PoolId.
819 This takes more time, but always works, AND allows wild
820 card Media Types (i.e. the SD can decide).
821 - The DIR <==> SD protocol has changed.
825 - Fix scanf of PoolId in catreq to handle 64 bit Ids.
827 - Add new ua_update.c file and move update_cmd there.
828 - Modify "update slots" to obtain actual number of slots.
829 - Tweak autochanger code to handle new slots request.
830 - Modify autochanger code to lock/unlock around slots and
833 - Patch the FD so that it does not issue an error message if
834 it attempts to restore the permissions on a Win32 drive.
835 - Edit "Resource-name" (physical-name) for the device name
836 everywhere in the SD.
837 - Remove .linked.tex files in preparation for cutover to
838 using .tex in place of .wml.
840 - Copy latest config.sub and config.guess from autoconf.
841 - Try new way of identifying drives with:
842 "resource-name" (physical-name)
843 More work need to a complete conversion.
845 - Rework some of the autochanger data so that the DIR has
846 the number of drives.
847 - Modify the way the Device info is returned so that it comes
848 back as a special message type and can be sent anytime the
849 Device status changes.
850 - Copy the change name and changer command into the device
851 record if none is specified.
852 - Require the change command and changer name to be specified in
853 and AutoChanger resource.
854 - Force all the Media Type records of all devices in an Autochanger
857 - Add new "run" command to Job resource in DIR. This permits
858 cloning a job as many times as you want.
859 - Pass PoolId to SD on Query request. It is now used in the
860 Find_media catalog request.
861 - Reworked the Device resource in the DIR. Eliminated num_waiting
862 and use_count, but added max_writers, reserved, and PoolId.
863 - This DIR is nolonger compatible with previous SDs.
864 - Add since and cloned keywords to the Console run command
866 - Implemented store_alist_str() to allow multiple string items
867 to be specified in a .conf file.
868 - Added %s (since time) to Job code editing.
869 - Reworked reserving drives in the SD. It now does it much simpler
872 - Integrate HP-UX patch from Olivier Mehani <olivier.mehani@linbox.com>
873 - Fix FD job.c to test correctly for no level.
877 - Change Developers to Developer's Guide as requested by Michael.
878 - Fix developers link in manual
879 - Add additional dcr changes in SD to allow multiple dcrs.
881 - Fix a few problems with the MySQL table create in 1.37.
882 - Delete the new tables in the table delete files.
883 - Increase the number of items permitted in a conf table.
884 - Make Director loop over alternative Devices specified in the
885 Storage resource until one is reserved by SD.
886 - Fix storing of StorageId in Media records.
887 - Add AutoSelect = yes|no in bacula-sd.conf
888 - Add Autochanger support to Label command.
889 - Do not autoselect devices with autoselect set false
891 - Implement setting DIR Storage device to Autochanger
893 - Select first available device in Autochanger.
894 - Pass back actual device name used.
895 - Allow Query of AutoChanger.
896 - Modify Query to include name of AutoChanger if
897 Device belongs to one.
898 - Remove old Pool code in jobq.c
899 - Add Autoselect flag to query and DEVICE class (still
902 - Lock autochanger script when running.
903 - Mark Volume not InChanger if correct volume is not
905 - Corrected some typos in the make_xxx_tables.in files.
906 - Made preliminary split of pre-run and run code for each
907 job type. This will permit early opening of SD for reserving
909 - Add offline and autochanger fields to Device Query record.
910 - Correct pthread_mutex_init() for autochanger in SD.
911 - Tweak Makefile for LaTeX manual, plus add nav buttons.
913 - Clean up drive reservation system. Add more sanity checks.
914 - Implement a few more methods for the DEVICE class in SD.
915 - Add latex directories to make clean
916 - move DEV_BSIZE to B_DEV_BSIZE to avoid conflicts with
917 certain header files (FreeBSD).
919 - Fix an ASSERT that was triggering in stored/acquire.c
920 attempt to fix a bug report.
922 - Corrected SunOs to SunOS in btraceback (user submitted).
923 - Applied patch from Roger HÃ¥kansson <hson@ludd.luth.se>
924 to warn the user of defective AWKs during ./configure.
926 - Add some changes submitted by a user for HP client build.
927 Not all changes accepted.
928 - Rework code in filed/backup.c to ease #ifdefing and make
929 program flow more obvious.
930 - Split DVD code out of dev.c into dvd.c
931 - Tweak #ifdefing to add back all the performance measurement
932 #defines in version.h
933 - Put most of MTIOCGET code in a subroutine to simplify the
935 - Make clean remove old CVS files
936 - Remove unnecessary image files from Latex directory
937 - Implement remaining parts of Storage DB record and
938 its use in the Director.
940 FullMaxWaitTime, Differential Max Wait Time, and
941 Incremental Max Wait time in Job resource.
942 - Start work on SD Autochanger code.
944 - Add back JobId index for MySQL as default -- speeds up
946 - Add more database fields and fix the update scripts to
947 include the new items.
948 - Pass actual level to FD so that ClientRun editing can reflect
949 correct level -- ditto for job status. This makes the DIR
950 incompatible with older clients!
951 - Move jobq.c acquire resources to static subroutine so that
952 the code logic becomes clearer. This is in preparation for
953 actually using the new Device resources.
954 - Fix some lower case problems in sql_cmds.c reported by
956 - Correct a seg fault in the SD reported by a user. Occurred
957 only when a high debug level was set.
958 - Modify init_dev() in dev.c to take JCR as first arg so that
959 proper error messages can be reported in next item.
960 - Modify the query and use device SD commands to attempt to
961 open the device if it could not previously be opened.
962 - Correct error message for Could not reserve device.
963 - Correct some minor details with Autochanger resource in SD.
965 - Fix seg fault if debug level 900 set in SD.
966 - Truncate Win32 child return code to 8 bits.
967 - Remove some old lld's.
971 - Make another attempt at fixing the ClientRunXXX return code
972 bug on Win32 machines.
973 - Apply ua_status patch from Carsten Paeth <calle@calle.in-berlin.de>
974 which enforces console ACLs in the status command for Jobs.
976 - Fix Media LabelDate and FirstWritten to be correctly set.
977 - Fix deadlock in multiple simultaneous jobs.
978 - Fix tape "truncation"/"number of files" after restore bug.
980 - Ensure that correct error messages are returned when
981 reading an ANSI label.
983 - Modified ANSI label code to preserve any ANSI label
984 already found by skipping over it rather than rewriting
986 - Split the ANSI label code into ansi_label.c
987 - Do not let user relabel an ANSI labeled tape.
988 - Applied a patch for the console help command supplied
990 - Added some new dev methods. Most notably was
991 set_eof(), which handles setting all the dev variables
992 when an EOF is just read. This is now used most everywhere
995 - Added code to detect that no files were inserted into the
996 tree for a restore. If a specific JobId was specified, the
997 user has the option of restoring everything.
998 - More progress in implementing 64 bit DB Ids.
999 - Modified the daemon start messages for RH.
1000 - Implement update scripts for all database types.
1001 - First cut at implementing restore directory (it will not
1004 - OK, I think ANSI labels work.
1005 - Added Label Type = ANSI|IBM|Bacula to Device resource in SD.
1006 If this is set, it will force writing of the appropriate
1008 - Added Check Labels = yes|no to Device resource in SD. If this
1009 is set, Bacula will check for ANSI labels and accept them,
1010 otherwise, ANSI labels will not be accepted when the tape
1013 - Second cut ANSI labels.
1015 - Merge Preben's patch for ACLs and for Mac OS X resource forks.
1017 - Display more informative message when a device was not
1018 found or could not be opened.
1019 - Add the sqlite3 database scripts.
1020 - Add some patches for 1.36.1 (note, I have now prepared
1021 a 1.36.2 with all the patches and some new features --
1023 - Some minor doc updates.
1024 - Add Arno's baculareport.pl script to the examples directory.
1025 29Jan05 -- after vacation
1026 - Add support for SQLite3 (it seems to run at 1/2 the speed
1027 of SQLite2). Use --with-sqlite3 instead of --with-sqlite
1029 - Add target for running qemu to boot Rescue CDROM
1030 - Add code to support kernel 2.6 in Rescue CDROM -- does NOT yet
1032 - Implement ANSI labels -- not yet tested.
1033 This required changes to DB format. No upgrade script yet.
1034 Note, more work needed to modify "update" command to handle
1035 changing label types, also must restrict volume name lengths
1037 - Add new Device, Storage, and MediaType records to DB. No
1039 - Add MediaType to bsr file record types. Not yet used in SD.
1040 - Permit multiple device specifications in Storage resource in
1042 - Implement Device resources. Director requests Device resource
1043 info from SD on startup.
1044 - Note!!!! DIR->SD incompatible with previous versions.
1045 - Remove multiple Storage definitions in Job resource. One can
1046 still specify multiple Storage resources, but they all go into
1047 a single alist, and imply sending data to each Storage daemon
1049 - Implement Device query command between DIR and SD.
1050 - Allow DIR to "reserve" a Device. It will then be acquired
1051 when the FD connects to the SD.
1052 - Turn all DIR resources into classes, and implement a few class
1053 methods -- more to come.
1054 - Turn DEVICE in SD into a class, and implement a number of inline
1055 class methods -- more to come.
1056 - I had serious problems with ACL errors on my Laptop, and so had
1057 to add the following patch:
1060 /***** Do we really want to silently ignore errors from acl_get_file
1061 and acl_to_text? *****/
1065 - Added edit_int64()
1066 - Reworked and tested a bit the htable routines.
1067 - Major changes to SD acquire.c -- DIR can now reserve devices. Needs
1069 - Made a special state code for DVD -- this simplifies the logic
1070 of the code, but I probably broke it. Testing needed!!!!
1071 - Add AutoChanger resource to SD, but not yet used.
1075 - Integrate Preben 'Peppe' Guldberg <peppe@wielders.org>'s
1076 acl patch. Fix case where configured but no ACL exists.
1077 Rework calling arguments to be shorter and positioned
1078 more typically in Bacula usage.
1080 - Fix scripts/bacula.in to have awk on an environment variable
1081 and add comments for Solaris users.
1082 - Turn off inet_aton in src/lib/address_conf.c for Win32
1083 - Add new files to win32 build and eliminate a compiler warning.
1084 - Add sample DVD Device resource to bacula-sd.conf
1086 - Integrate Nicolas' patch for direct DVD support.
1088 - Fix fstype error returns.
1089 - Apply Preben's cleanup.patch which puts back much of the
1090 cleanup code in src/filed/restore.c
1092 - Apply all of Preben's patches, but revert to old backup.c
1093 and old restore.c in filed. Also turn off code in new
1094 acl.c because of errors. The new code, when fully implemented
1095 moves platform specific code into acl.c.
1096 One of the patches also implements WildFile and WildDir -- thanks.
1098 - Implement Python in the SD (no events yet though).
1099 - Fix some typos in the previous commit.
1101 - Enhance CDROM boot to include some documentation at boot time.
1102 - NOTE!!!!! The CDROM will not boot 2.6 kernels because the
1103 boot sequence has changed significantly. Updates to come
1105 - Add memtest option to CDROM boot.
1106 - Include Nicolas' changes to fix llist JobMedia records.
1107 - Make sure that ClientRunBefore/After messages from the program
1108 are terminated with a newline. Add strerror to output error
1110 - Return program exit status code in Win32.
1112 - Add memtest86 to Bacula Rescue disk
1113 - Enhance Rescue disk startup screen
1115 - Move some variables to eliminate Solaris 2.6 compiler warnings.
1116 - Fix the seg fault at the end of a job in the FD when using
1117 old style include/excludes.
1119 - Apply Preben's ACL patch.
1120 - Integrate Preben's restore patch.
1121 - Integrate Preben's verify teaks.
1122 - Fix doc/latex/Makefile to copy/remove .eps files when building
1123 html and web outputs.
1125 - Fix Bacula so that it does not exit if there is a syntax error
1126 in its conf file during a reload command. Bug 182.
1127 - Apply fixes suggested for old Solaris networking.
1129 - Apply Preben 'Peppe' Guldberg <peppe@wielders.org>
1130 three patches that clean up white space:
1131 ws.patch.02.strings:
1132 Breaks strings that span lines into concatenated strings. I am not sure
1133 if you like this one. Other code works with concatenated strings, though.
1134 ws.patch.03.trailing:
1135 This removes trailing whitespace. No changes resulted from this for
1137 ws.patch.04.leading:
1138 This replaces space runs at the start of line with tabs. No changes
1140 - Fix overriding storage specification to be done
1141 through a subroutine.
1142 - Fix autoconf so it runs with FC3.
1143 - Add Python4.3 to configure search paths.
1144 - Always copy and delete storage definitions into jcr.
1145 - Check that VolumeName supplied by Python is valid.
1148 - Fix undefined in non-Python build.
1149 - Update rescue disk to include mkinitrd
1150 - Fix umount_drives in rescue disk (only one arg to umount)
1151 - Ensure that if SD is manually set in Console, it is used.
1152 - Put generate_event on pointer and plug it in init. This
1153 permits using it in /lib
1154 - Correct despooling size reported to be Job specific rather
1155 than for the whole drive.
1157 - Fix bug 207. jcr use count off by one when manually
1159 - Remove FNMATCH test in configure.in and always use
1160 the one in our library to get the FN_CASEFOLD GNU
1161 extensions on all platforms.
1162 - While using the rescue CDROM after my computer would not
1163 boot, I realized that it would be very useful to have
1164 a umount_disks. So, it is not implemented, along with
1165 updates to the READMEs and some minor tweaks.
1166 - Moved mounting the CDROM in the rescue boot from /cdrom
1167 to /mnt/cdrom (more standard location).
1168 - Reboot in CDROM rescue should now work -- requires -d
1169 option (no write) to work.
1170 - Hopefully fixed all the IPV6/4 problems and buffer
1171 problems with networking in lib. Bugs 190 and 204.
1172 Cleaned up a lot of #ifdefing problems by using routines
1175 - Apply Preben 'Peppe' Guldberg <peppe@wielders.org>
1177 - Remove duplicate code from chksum.h (mentioned by Preben).
1179 - Integrate Tim Oberfoell <oberfoell@web.de> patch to ACLs
1180 to handle both the "standard" and "default" ACLs.
1182 - Integrated Preben 'Peppe' Guldberg <peppe@wielders.org>
1183 three cleanup patches (btest, verify, find).
1184 - Integrated Preben 'Peppe' Guldberg <peppe@wielders.org>
1185 three cleanup patches (backup, chksum, and verify)
1187 - Integrated Preben 'Peppe' Guldberg <peppe@wielders.org>
1188 patch to avoid doing MTIOCGET on OSes that do not support
1190 - Integrated Preben 'Peppe' Guldberg <peppe@wielders.org>
1191 patch to add filesystem type matching to FileSets in the
1193 - Integrated Preben 'Peppe' Guldberg <peppe@wielders.org>
1194 patch to add Mac OSX resource fork support (save/restore)
1195 to Bacula -- HFS Plus support.
1196 - Add FileSet to client Job listing query.
1198 - Integrated Preben 'Peppe' Guldberg <peppe@wielders.org>
1199 patch to backup directories skipped (due to no file system
1200 changes or no recursion), and to add a slash to the end
1201 of the directory name during the match process.
1202 - Implement Jamie ffolliott <jamieff@inline.net>
1203 patch to dird_conf.c that enables Multiple Connections and
1204 fixes a typo in show. The rest of his patch awaits my suggested
1207 - Implement run command in Python
1209 - Implement conversion of the manual, and some minor
1210 tweaks to the script tags.
1211 - Apply a patch supplied by Preben 'Peppe' Guldberg that implements
1212 ignore case in wild cards and regexes.
1213 - Fix a truncated line in the above patch due to my cut and paste.
1215 - Fix it so that the InChanger flag is only changed for Volumes
1217 - Add PIDOF configuration path and apply to bacula.in
1218 - Add user supplied patch to add inet_aton() of old Solaris
1220 - Require pools to match before allowing multiple simultaneous
1221 accesses to same storage resource.
1222 - Add patch supplied by Martin to correct buffer overrun in
1223 bsnprintf() with no library snprintf().
1225 - Apply user supplied patch that implements No Hard Links.
1226 - Document Python interface
1227 - Add hardlink keyword patch supplied by David R Bosso <dbosso@lsit.ucsb.edu>
1229 - Fix non-python prototypes in dummy routines.
1230 - Add python 2.3 to config search list (user submitted patch)
1231 - Add JobStatus to Python variables.
1233 - Add "python restart" command in Console.
1234 - Make built-in variables table driven.
1235 - First cut of Python Events for Bacula. Director only.
1236 StartJob, EndJob, NewVolume events.
1238 Version 1.36.1 Released 26Nov04:
1240 - Take Dan's fix to the fix_postgresql_tables (thanks Dan)
1241 - Increase Maximum Concurrent Jobs to 20 in SD and FD!
1242 - Fix improper handling of autochanger Volumes that are not
1243 marked InChanger. Bug 170.
1245 - Update authentication failure message to indicate possible
1246 problem with Maximum Concurrent Jobs. Add to doc too.
1247 - Commit PostgreSQL BIGINT fix -- apparently some people didn't
1249 - Do not use a slot unless it is InChanger in the autochanger
1250 code, otherwise autochanger gets upset not finding the Volume.
1251 Bug 170 (need fix above too).
1253 - Make authentication timeout compile time configurable.
1254 The value is set in src/baconfig.h
1255 - Fix removing items from watchdog queue, which apparently
1256 screwed up if there was more than one item.
1257 - Rework mediaformat part of manual to separate out old
1258 tape format for easier reading.
1259 - Add a lot of debug code to the authentication code.
1260 - Add seconds to the start/end times printed in the job report.
1262 - Fix mutex deadlock in dequeue of messages.
1263 - Add debug code to FD authorization.
1265 - Increase authentication timeouts to 10 minutes.
1267 - Fix cancel bug in FD on /lib/tls with zero pid in
1269 - Add date/time to all messages.
1270 - Make Qmsg use time message was queued rather than time
1272 - Indent job output two spaces.
1274 - Fix web page links for new manual.
1275 - Grant postgresql permission to cdimages.
1276 - Correct crash after "list nextvol" "list media" bug 160
1278 - Fix scripts/Makefile.in missing ;\ -- thanks Martin
1279 - A bit of work on btape to keep if from going into infinite
1280 loops when things do not work well, and to print a bit
1283 - JobDefs Storage resource completely overrode anything
1284 specified in Job resource -- bug 159 -- fixed.
1285 - Fix syntax of renaming postgresql per Dan.
1286 - Add working_directory to be /tmp for wx_console.
1288 - Allow both a JobId and a filename or list of files to be
1289 specified on a restore command line.
1290 - Save old mtx-changer before installing new one: bug 156
1291 - Fix errors in CDROM file pointed out by Scott.
1293 - Fix exepath when Bacula executed without path.
1294 - Move test for socket libraries for Solaris in configure.in
1295 before tcp wrapper tests so that linking works.
1296 - Add "make copy-static-fd" to makefile in CDROM file to
1297 copy existing static fd to CDROM rather than building it.
1300 - Fix ps command for OpenBSD
1301 - Rework the creation of indexes for SQL -- fall back
1302 to the old code, but document what can be added.
1304 - Add new Daemon message handler in default DIR conf.
1306 - Fix Seg Fault with -D100 in bpipe.c.
1307 - Fix Seg Fault in run specifying a JobId.
1308 - Make mail from daemon with a Messages Resource use
1309 the MailCommand with editing rather than the default
1311 - Replace Jmsg in dispatch_message() with Qmsg.
1312 - Make edit_job_codes handle NULL jcr.
1314 - Add M_ALERT class and put tape alerts into it.
1315 - Fix Verify count vs found by not double counting files
1316 that are split across files/Volumes.
1318 - Fix count returned from write_bsr_file() to handle multiple
1319 volumes. This fixes most cases of the Verify VolumeToCatalog.
1320 - Cleanup a bit the make clean for the rescue cdrom and remove
1321 unneeded files from the CVS.
1323 - Fixed acquiring a tape so that it does not block all acquires
1324 when operator intervention is needed.
1325 - Platform build script updates from Scott
1327 - Add patch to force Linux LD_ASSUME_KERNEL to avoid using the
1328 new /lib/tls. This is done in the startup scripts.
1329 - Modify mtx-changer so that it checks for ONLINE while
1331 - Modify make_postgresql_tables.in so that EndBlock is stored
1332 as a bigint. Prevents job failures when a disk volume is
1335 - Add grep ONLINE to wait_for_drive() in mtx-changer.in
1337 - Rebuild Scott's new configure