8 Data encryption Meno Abels (see projects)
9 Communication encryption Meno Abels
10 Version 1.35 Kern (see below)
11 ========================================================
13 1.35 Items to do for release:
14 - Backspace to beginning of line (conio) does not erase first char.
15 - Doc to do unmount before removing magazine.
16 - Detect fixed tape block mode during positioning by looking at
17 block numbers in btape "test". Possibly adjust in Bacula.
18 - Fix possible bug in update volumefrompool (email 9/28 from
20 - Look into failing Excludes in Win32
22 - Document a get out of jail procedure if everything breaks if
23 you lost/broke the Catalog -- do the same for "I know my
24 file is there how do I get it back?".
29 - Look at patches/bacula_db.b2z postgresql that loops during restore.
31 - Add delete JobId to regression.
32 - Add bscan to four-concurrent-jobs regression.
33 - Add IPv6 to regression
34 - Perhaps add read/write programs and/or plugins to FileSets.
35 - How to handle backing up portables ...
36 - Add "Rerun failed levels = yes/no" to Job resource.
37 - Add some sort of guaranteed Interval for upgrading jobs.
39 Documentation to do: (any release a little bit at a time)
40 - Alternative to static linking "ldd prog" save all binaries listed,
41 restore them and point LD_LIBRARY_PATH to them.
42 - Document add "</dev/null >/dev/null 2>&1" to the bacula-fd command line
43 - Document query file format.
44 - Add more documentation for bsr files.
45 - Document problems with Verify and pruning.
46 - Document how to use multiple databases.
47 - VXA drives have a "cleaning required"
48 indicator, but Exabyte recommends preventive cleaning after every 75
51 In this context, it should be noted that Exabyte has a command-line
52 vxatool utility available for free download. (The current version is
53 vxatool-3.72.) It can get diagnostic info, read, write and erase tapes,
54 test the drive, unload tapes, change drive settings, flash new firmware,
56 Of particular interest in this context is that vxatool <device> -i will
57 report, among other details, the time since last cleaning in tape motion
58 minutes. This information can be retrieved (and settings changed, for
59 that matter) through the generic-SCSI device even when Bacula has the
60 regular tape device locked. (Needless to say, I don't recommend
61 changing tape settings while a job is running.)
62 - Lookup HP cleaning recommendations.
63 - Lookup HP tape replacement recommendations (see trouble shooting autochanger)
64 - Add more obvious documentation of @ for conf files.
65 - Document doing table repair
68 Testing to do: (painful)
71 For 1.37 Testing/Documentation:
72 - Implement a Client filename case fold option (default = fold
74 - Fix list volumes to output volume retention in some other
75 units, perhaps via a directive.
76 - If opening a tape in read/write mode fails attempt to open
77 it in read-only mode, and mark the tape for read only.
78 - Add a read-only mode to the mount option.
79 - Add "Allow multiple connections" in Catalog resource to open a new
80 database connection for each job.
81 - Allow Simultaneous Priorities = yes => run up to Max concurrent jobs even
82 with multiple priorities.
83 - Fix find_device in stored/dircmd.c:462 (see code)
84 - Add db check test to regression. Test each function like delete,
86 - If you use restore replace=never, the directory attributes for
87 non-existent directories will not be restored properly.
92 mt status and tapeinfo output are attached. Default blocksize was
93 originally 512 bytes. I ran the following test on my machine:
95 - set defblksize to 512
96 - started bacula, labelled tape, ran job
97 - job completes with the following in log
98 SD termination status: OK
99 Termination: Backup OK
100 - restore attempts with blocksize set to 0 or 512 result in the familiar
101 "Volume data error! Wanted block-id BB02, got . Buffer discarded."
104 # tapeinfo -f /dev/sg0
105 Product Type: Tape Drive
107 Product ID: 'TSL-11000 '
109 Attached Changer: Yes
110 SerialNumber: '0010000443'
117 Medium Type: Not Loaded
122 DataDeCompEnabled: yes
128 # mt -f /dev/nst0 status
130 File number=0, block number=132, partition=0.
131 Tape block size 512 bytes. Density code 0x26 (DDS-4 or QIC-4GB).
132 Soft error count since last status=0
133 General status bits on (1010000):
139 - see lzma401.zip in others directory for new compression
141 - Minimal autochanger handling in Bacula and in btape.
142 - Look into how tar does not save sockets and the possiblity of
143 not saving them in Bacula (Martin Simmons reported this).
144 - Add All Local Partitions = yes to new style saves.
145 - localmounts=`awk '/ext/ { print $2 }' /proc/mounts` # or whatever
146 find $localmounts -xdev -type s -ls
147 - Fix restore jobs so that multiple jobs can run if they
148 are not using the same tape(s).
149 - Allow the user to select JobType for manual pruning/purging.
150 - Look at adding Client run command that will use the
151 port opened by the client.
152 - bscan does not put first of two volumes back with all info in
154 - Implement the FreeBSD nodump flag in chflags.
155 - Figure out how to make named console messages go only to that
156 console and to the non-restricted console (new console class?).
157 - Make restricted console prompt for password if *ask* is set or
158 perhaps if password is undefined.
159 - Implement "from ISO-date/time every x hours/days/weeks/months" in
161 ==== from Marc Schoechlin
162 - the help-command should be more verbose
163 (it should explain the paramters of the different
165 -> it´s time-comsuming to consult the manual anytime
166 you need a special parameter
167 -> maybe it´s more easy to maintain this, if the
168 descriptions of that commands are outsourced to
170 - the cd-command should allow complete paths
171 i.e. cd /foo/bar/foo/bar
172 -> if a customer mails me the path to a certain file,
173 it´s faster to enter the specified directory
174 - if the password is not configured in bconsole.conf
175 you should be asked for it.
176 -> sometimes you like to do restore on a customer-machine
177 which shouldn´t know the password for bacula.
178 -> adding the password to the file favours admins
179 to forget to remove the password after usage
181 the protection of that file is less important
182 - long-listed-output of commands should be scrollable
183 like the unix more/less-command does
184 -> if someone runs 200 and more machines, the lists could
185 be a little long and complex
186 - command-output should be shown column by column
187 to reduce scrolling and to increase clarity
189 - lsmark should list the selected files with full
191 - wildcards for selecting and file and directories would be nice
192 - any actions should be interuptable with STRG+C
193 - command-expansion would be pretty cool
195 - Compaction of Disk space by "migrating" Volumes that have pruned
196 Jobs (what criteria? size, #jobs, time).
197 - Add prune all command
198 - Document fact that purge can destroy a part of a restore by purging
199 one volume while others remain valid -- perhaps mark Jobs.
200 - Add multiple-media-types.txt
201 - look at mxt-changer.html
202 - Make ? do a help command (no return needed).
203 - Implement restore directory.
204 - Document streams and how to implement them.
205 - Possibly implement "Ensure Full Backup = yes" looks for a failed full backup
206 and upgrades the current backup if one exists.
207 - Check that barcode reading and update slots scan works.
208 - Try not to re-backup a file if a new hard link is added.
209 - Add feature to backup hard links only, but not the data.
210 - Fix stream handling to be simpler.
211 - Add Priority and Bootstrap to Run a Job.
212 - Eliminate Restore "Run Restore Job" prompt by allowing new "run command
214 - Remove View FileSet button from Run a Job dialog.
215 - Don't count higher level non-saved directories in Files expected
217 - Handle prompt for restore job at end of Restore command.
218 - Add display of total selected files to Restore window.
219 - Add tree pane to left of window.
220 - Add progress meter.
221 - Max wait time or max run time causes seg fault -- see runtime-bug.txt
222 - Document writing to a CD/DVD with Bacula.
223 - Add a "base" package to the window installer for pthreadsVCE.dll
224 which is needed by all packages.
225 - Add message to user to check for fixed block size when the forward
226 space test fails in btape.
227 - When unmarking a directory check if all files below are unmarked and
228 then remove the + flag -- in the restore tree.
229 - Possibly implement: Action = Unmount Device="TapeDrive1" in Admin jobs.
230 - Setup lrrd graphs: (http://www.linpro.no/projects/lrrd/) Mike Acar.
231 - Revisit the question of multiple Volumes (disk) on a single device.
232 - Add a block copy option to bcopy.
233 - Investigate adding Mac Resource Forks.
234 - Finish work on Gnome restore GUI.
235 - Fix "llist jobid=xx" where no fileset or client exists.
237 it seems to be complaining about 12:00pm which should be a valid 12
238 hour time. I changed the time to 11:59am and everything works fine.
239 Also 12:00am works fine. 0:00pm also works (which I don't think
240 should). None of the values 12:00pm - 12:59pm work for that matter.
241 - Add level to estimate command.
242 - For each job type (Admin, Restore, ...) require only the really necessary
244 - Fix option 2 of restore -- list where file is backed up -- require Client,
245 then list last 20 backups.
246 - Add all pools in Dir conf to DB also update them to catch changed
247 LabelFormats and such.
248 - Pass Director resource name as an option to the Console.
249 - Add a "batch" mode to the Console (no unsolicited queries, ...).
250 - Add a .list all files in the restore tree (probably also a list all files)
251 Do both a long and short form.
252 - Allow browsing the catalog to see all versions of a file (with
253 stat data on each file).
254 - Restore attributes of directory if replace=never set but directory
256 - Use SHA1 on authentication if possible.
257 - See comtest-xxx.zip for Windows code to talk to USB.
258 - Make btape accept Device Names in addition to Archive names.
259 - Add Events and Perl scripting.
260 - Add John's appended files:
261 Appended = { /files/server/logs/http/*log }
262 and such files would be treated as follows.On a FULL backup, they would
263 be backed up like any other file.On an INCREMENTAL backup, where a
264 previous INCREMENTAL or FULL was already in thecatalogue and the length
265 of the file wasgreater than the length of the last backup, only thedata
266 added since the last backup will be dumped.On an INCREMENTAL backup, if
267 the length of the file is less than thelength of the file with the same
268 name last backed up, the completefile is dumped.On Windows systems, with
269 creation date of files, we can be evensmarter about this and not count
270 entirely upon the length.On a restore, the full and all incrementals
271 since it will beapplied in sequence to restore the file.
272 - Add a regression test for dbcheck.
273 - Add disk seeking on restore. - Allow
274 for optional cancelling of SD and FD in case DIR
275 gets a fatal error. Requested by Jesse Guardiani <jesse@wingnet.net>
276 - Add "limit=n" for "list jobs"
277 - Check new HAVE_WIN32 open bits.
278 - Check if the tape has moved before writing.
279 - Handling removable disks -- see below:
280 - Keep track of tape use time, and report when cleaning is necessary.
281 - Add FromClient and ToClient keywords on restore command (or
282 BackupClient RestoreClient).
283 - Implement a JobSet, which groups any number of jobs. If the
284 JobSet is started, all the jobs are started together.
285 Allow Pool, Level, and Schedule overrides.
286 - Enhance cancel to timeout BSOCK packets after a specific delay.
287 - Do scheduling by UTC using gmtime_r() in run_conf, scheduler, and
288 ua_status.!!! Thanks to Alan Brown for this tip.
289 - Look at updating Volume Jobs so that Max Volume Jobs = 1 will work
290 correctly for multiple simultaneous jobs.
291 - Correct code so that FileSet MD5 is calculated for < and | filename
293 - Implement the Media record flag that indicates that the Volume does disk
295 - Implement VolAddr, which is used when Volume is addressed like a disk,
296 and form it from VolFile and VolBlock.
297 - Make multiple restore jobs for multiple media types specifying
298 the proper storage type.
299 - Implement MediaType keyword in bsr?
300 - Fix fast block rejection (stored/read_record.c:118). It passes a null
301 pointer (rec) to try_repositioning().
302 - Look at extracting Win data from BackupRead.
303 - Implement RestoreJobRetention? Maybe better "JobRetention" in a Job,
304 which would take precidence over the Catalog "JobRetention".
305 - Implement Label Format in Add and Label console commands.
306 - Possibly up network buffers to 65K. Put on variable.
307 - Put email tape request delays on one or more variables. User wants
308 to cancel the job after a certain time interval. Maximum Mount Wait?
309 - Job, Client, Device, Pool, or Volume?
310 Is it possible to make this a directive which is *optional* in multiple
311 resources, like Level? If so, I think I'd make it an optional directive
312 in Job, Client, and Pool, with precedence such that Job overrides Client
313 which in turn overrides Pool.
315 - Spooling ideas taken from Volker Sauer's and other's emails:
316 > IMHO job spooling should be turned on
322 > where and 2) overrides 1) and 3) is independent.
324 Yes, this is the minimum that I think is necessary.
328 > It should be switched by job, because the job that backs up the machine
329 > with the bacula-sd on doesn't need spooling.
330 > It should be switched by schedule, because for full-backups I don't need
331 > spooling, so I can switch it off (because the network faster then the
334 True, with the exception that if you have enough disk spool space,
335 and you want to run concurrent jobs, spooling can eliminate the block
336 interleaving restore inefficiencies.
338 > And you should be able to turn it of by sd for sd-machines with low disk
339 > capacity or if you just don't need or want this feature.
342 > - definitly the possibility for multipe spool direcories
344 Having multiple directories is no problem -- having different maximum
345 sizes creates specification problems. At some point, I will probably
346 have a common SD pool of spool directories as well as a set of
347 private spool directories for each device. The first implementation
348 will be a set of private spool directories for each device since
349 managing a global pool with a bunch of threads writing into the same
350 directory is *much* more complicated and prone to error.
352 > - the ability to spool parts of a backup (not the whole client)
354 This may change in the future, but for the moment, it will spool
355 either to a job high water mark, or until the directory is full
356 (reaches max spool size or I/O error). It will then write to tape,
357 truncate the spool file, and begin spooling again.
359 > - spooling while writing to tape
361 Not within a job, but yes, if you run concurrent jobs -- each is a
362 different thread. Within a job could be a feature, but *much* later.
364 > - parallel spooling (like parallel jobs/ concurrent jobs) of clients
366 Yes, this is one of my main motivations for doing it (aside from
367 eliminating tape "shoe shine" during incremental backups.
369 > - flushing a backup that only went to disk (like amflush in amanda)
371 This will be a future feature, since spooling is different from backing
372 up to disk. The future feature will be "migration" which will move a job
373 from one backup Volume to another.
375 - New Storage specifications:
376 - Want to write to multiple storage devices simultaneously
377 - Want to write to multiple storage devices sequentially (in one job)
378 - Want to read/write simultaneously
379 - Key is MediaType -- it must match
381 Passed to SD as a sort of BSR record called Storage Specification
385 MediaType -> Next MediaType
387 Device -> Next Device
389 Allow multiple Storage specifications
397 Allow Multiple Pool specifications (note, Pool currently
399 Allow Multiple MediaType specifications in Dir conf
400 Allow Multiple Device specifications in Dir conf
401 Perhaps keep this in a single SSR
402 Tie a Volume to a specific device by using a MediaType that
403 is contained in only one device.
404 In SD allow Device to have Multiple MediaTypes
407 - Look at www.nu2.nu/pebuilder as a helper for full windows
409 Ideas from Jerry Scharf:
410 First let's point out some big pluses that bacula has for this
412 more importantly it's active. Thank you so much for that
413 even more important, it's not flaky
414 it has an open access catalog, opening many possibilities
415 it's pushing toward heterogeneous systems capability
417 Macintosh file client
418 macs are an interesting niche, but I fear a server is a rathole
419 working bare iron recovery for windows
420 the option for inc/diff backups not reset on fileset revision
421 a) use both change and inode update time against base time
422 b) do the full catalog check (expensive but accurate)
423 sizing guide (how much system is needed to back up N systems/files)
424 consultants on using bacula in building a disaster recovery system
426 or how to get at fancy things that one could do with bacula
427 logwatch code for bacula logs (or similar)
428 linux distro inclusion of bacula (brings good and bad, but necessary)
429 win2k/XP server capability (icky but you asked)
430 support for Oracle database ??
432 - Look at adding SQL server and Exchange support for Windows.
433 - Restore: Enter Filename: 'C:/Documents and Settings/Comercial/My
434 Documents/MOP/formulário de registro BELAS ARTES.doc' causes Bacula to
436 - Each DVD-RAM disk would be a volume, just like each tape is
437 a volume. It's a 4.7GB media with random access, but there's nothing about
438 it that I can see that makes it so different than a tape from bacula's
439 perspective. Why couldn't I back up to a bare floppy as a volume (ignoring
441 - Make dev->file and dev->block_num signed integers so that -1 can
442 be an invalid value which happens with BSR.
443 - Create VolAddr for disk files in place of VolFile and VolBlock. This
444 is needed to properly specify ranges.
445 - Print bsmtp output to job report so that problems will be seen.
446 - Pass the number of files to be restored to the FD for reporting
447 - Add progress of files/bytes to SD and FD.
448 - Don't continue Restore if no files selected.
449 - Print warning message if FileId > 4 billion
450 - do a "messages" before the first prompt in Console
451 - Add a date and time stamp at the beginning of every line in the
452 Job report (Volker Sauer).
453 - Client does not show busy during Estimate command.
454 - Implement Console mtx commands.
455 - Add a default DB password to MySQL.
456 GRANT all privileges ON bacula.* TO bacula@localhost IDENTIFIED BY
459 - Implement a Mount Command and an Unmount Command where
460 the users could specify a system command to be performed
461 to do the mount, after which Bacula could attempt to
462 read the device. This is for Removeable media such as a CDROM.
463 - Most likely, this mount command would be invoked explicitly
464 by the user using the current Console "mount" and "unmount"
465 commands -- the Storage Daemon would do the right thing
466 depending on the exact nature of the device.
467 - As with tape drives, when Bacula wanted a new removable
468 disk mounted, it would unmount the old one, and send a message
469 to the user, who would then use "mount" as described above
470 once he had actually inserted the disk.
471 - Implement dump/print label to UA
472 - Spool to disk only when the tape is full, then when a tape is hung move
474 - Scratch Pool where the volumes can be re-assigned to any Pool.
475 - bextract is sending everything to the log file ****FIXME****
476 - Add Progress command that periodically reports the progress of
478 - Restrict characters permitted in a Resource name, and don't permit
480 - Allow multiple Storage specifications (or multiple names on
481 a single Storage specification) in the Job record. Thus a job
482 can be backed up to a number of storage devices.
483 - Implement some way for the File daemon to contact the Director
484 to start a job or pass its DHCP obtained IP number.
485 - Implement multiple Consoles.
486 - Implement a query tape prompt/replace feature for a console
488 Two jobs ready to go, first one blocked waiting for media
489 Cancel 2nd job ("waiting execution" one)
492 - Copy console @ code to gnome2-console
493 - Make AES the only encryption algorithm see
494 http://csrc.nist.gov/CryptoToolkit/aes/). It's
495 an officially adopted standard, has survived peer
496 review, and provides keys up to 256 bits.
497 - Think about how space could be freed up on a tape -- perhaps this
498 is a Merge or Compact feature that is needed.
499 - Modify FileSet, did not upgrade the current Increment job, but
500 waited for the next job to be upgraded.
501 - Take a careful look at SetACL http://setacl.sourceforge.net
502 - Implement a where command for the tree telling where a file
504 - Take a careful look at Level for the estimate command, maybe make
505 it a command line option.
506 - Add Volume name to "I cannot write on this volume because"
507 - Make tree walk routines like cd, ls, ... more user friendly
508 by handling spaces better.
509 - Write your PID file and chown root:wheel before drop.
510 - Make sure there is no symlink in a file before creating a
512 - Look at mktemp or mkstemp(3).
513 mktemp and mkstemp create files with predictable names too. That's
514 not the vulnerability. The vulnerability is in creating files without
515 using the O_EXCL flag, which means "only create this file if it doesn't
516 exist, including if the file is a dangling symlink."
518 It is *NOT* enough to do the equivalent of
520 if doesn't exist $filename
521 then create $filename
523 because between the test and the create another process could have
524 gotten the CPU and created the file. You must use atomic functions
525 (those that don't get interrupted by other processes) and O_EXCL is
526 the only way for this particular example.
527 - Automatically create pools, but instead of looking for what
528 in in Job records, walk through the pool resources.
529 - Check and double check tree code, why does it take so long?
530 - Add device name to "Current Volume not acceptable because ..."
531 - Make sure that Bacula rechecks the tape after the 20 min wait.
532 - Set IO_NOWAIT on Bacula TCP/IP packets.
533 - Try doing a raw partition backup and restore by mounting a
536 Yes, it would allow to highly automatic the request for new tapes. If a
537 tape is empty, bacula reads the barcodes (native or simulated), and if
538 an unused tape is found, it runs the label command with all the
539 necessary parameters.
541 By the way can bacula automatically "move" an empty/purged volume say
542 in the "short" pool to the "long" pool if this pool runs out of volume
544 - Eliminate orphaned jobs: dbcheck, normal pruning, delete job command.
545 Hm. Well, there are the remaining orphaned job records:
547 | 105 | Llioness Save | 0000-00-00 00:00:00 | B | D | 0 | 0 | f |
548 | 110 | Llioness Save | 0000-00-00 00:00:00 | B | I | 0 | 0 | f |
549 | 115 | Llioness Save | 2003-09-10 02:22:03 | B | I | 0 | 0 | A |
550 | 128 | Catalog Save | 2003-09-11 03:53:32 | B | I | 0 | 0 | C |
551 | 131 | Catalog Save | 0000-00-00 00:00:00 | B | I | 0 | 0 | f |
553 As you can see, three of the five are failures. I already deleted the
554 one restore and one other failure using the by-client option. Deciding
555 what is an orphaned job is a tricky problem though, I agree. All these
556 records have or had 0 files/ 0 bytes, except for the restore. With no
557 files, of course, I don't know of the job ever actually becomes
558 associated with a Volume.
560 (I'm not sure if this is documented anywhere -- what are the meanings of
561 all the possible JobStatus codes?)
563 Looking at my database, it appears to me as though all the "orphaned"
564 jobs fit into one of two categories:
566 1) The Job record has a StartTime but no EndTime, and the job is not
569 2) The Job record has an EndTime, indicating that it completed, but
570 it has no associated JobMedia record.
573 This does suggest an approach. If failed jobs (or jobs that, for some
574 other reason, write no files) are associated with a volume via a
575 JobMedia record, then they should be purged when the associated volume
576 is purged. I see two ways to handle jobs that are NOT associated with a
579 1) purge them automatically whenever any volume is manually purged;
581 2) add an option to the purge command to manually purge all jobs with
582 no associated volume.
584 I think Restore jobs also fall into category 2 above .... so one might
585 want to make that "The Job record has an EndTime,, but no associated
586 JobMedia record, and is not a Restore job."
587 - make "btape /tmp" work.
588 - Make sure a rescheduled job is properly reported by status.
589 - Walk through the Pool records rather than the Job records
590 in dird.c to create/update pools.
591 - What to do about "list files job=xxx".
592 - When job rescheduled, status gives is waiting for Client Rufus
593 to connect to Storage File. Dir needs to inform SD that job
595 - Make Dmsg look at global before calling subroutine.
596 - Enable trace output at runtime for Win32
597 - Available volumes for autochangers (see patrick@baanboard.com 3 Sep 03
598 and 4 Sep) scan slots.
599 - Get and test MySQL 4.0
600 - Do a complete audit of all pthreads_mutex, cond, ... to ensure that
601 any that are dynamically initialized are destroyed when no longer used.
602 - Look at how fuser works and /proc/PID/fd that is how Nic found the
603 file descriptor leak in Bacula.
604 - Implement WrapCounters in Counters.
605 - Turn on SIGHUP in dird.c and test.
606 - Use system dependent calls to get more precise info on tape errors.
607 - Add heartbeat from FD to SD if hb interval expires.
608 - Suppress read error on blank tape when doing a label.
609 - Can we dynamically change FileSets?
610 - If pool specified to label command and Label Format is specified,
611 automatically generate the Volume name.
612 - Why can't SQL do the filename sort for restore?
613 - Look at libkse (man kse) for FreeBSD threading.
614 - Look into Microsoft Volume Shadowcopy Service VSS for backing
615 up system state components (Active Directory, System Volume, ...)
616 - Add ExhautiveRestoreSearch
617 - Look at the possibility of loading only the necessary
618 data into the restore tree (i.e. do it one directory at a
619 time as the user walks through the tree).
620 - Possibly use the hash code if the user selects all for a restore command.
621 - Orphaned Dir buffer at parse_conf.c:373 => store_dir
622 - Fix "restore all" to bypass building the tree.
623 - Prohibit backing up archive device (findlib/find_one.c:128)
624 - Implement Release Device in the Job resource to unmount a drive.
625 - Implement Acquire Device in the Job resource to mount a drive,
626 be sure this works with admin jobs so that the user can get
627 prompted to insert the correct tape. Possibly some way to say to
628 run the job but don't save the files.
629 - Implement FileOptions (see end of this document)
630 - Make things like list where a file is saved case independent for
633 - Bacula needs to propagate SD errors.
634 > > cluster-dir: Start Backup JobId 252, Job=REUTERS.2003-08-11_15.04.12
635 > > prod4-sd: REUTERS.2003-08-11_15.04.12 Error: Write error on device
636 /dev/nst0. ERR=Input/output error.
637 > > prod4-sd: REUTERS.2003-08-11_15.04.12 Error: Re-read of last block failed.
638 Last block=5162 Current block=5164.
639 > > prod4-sd: End of medium on Volume "REU007" Bytes=16,303,521,933
641 - Use autochanger to handle multiple devices.
642 - Add SuSE install doc to list.
643 - Check and rechedk "Invalid block number"
644 - Make bextract release the drive properly between tapes
645 so that an autochanger can be made to work.
646 - User wants to NOT backup up certain big files (email files).
647 - Maybe remove multiple simultaneous devices code in SD.
648 - On Windows with very long path names, it may be impossible to create
649 a file (and thus restore it) because the total length is too long.
650 We must cd into the directory then create the file without the
652 - lstat() is not going to work on Win32 for testing date.
653 - Implement a Recycle command
654 - Add client name to cram-md5 challenge so Director can immediately
655 verify if it is the correct client.
656 - Add JobLevel in FD status (but make sure it is defined).
657 - Audit all UA commands to ensure that we always prompt where possible.
658 - Check Jmsg in bnet, may not work, must dup bsock.
659 - Suppress Job Name in Jmsg for console
660 - Create Pools that are referenced in a Run statement at startup if possible.
661 - Use runbeforejob to unload, then reload a volume previously used,
662 then the next job run gets an error reading the drive.
663 - Make bootstrap filename unique.
664 - Test a second language e.g. french.
665 - Start working on Base jobs.
666 - Make "make binary-release" work from any directory.
667 - Implement UnsavedFiles DB record.
668 - Implement argc/argv for daemon command line scanning using table driven
670 - Implement table driven single argc/argv scanner to pickup all arguments.
671 Much like xxx_conf.c scan table.
672 keyword, handler(store_routine), store_address, code, flags, default.
673 - From Phil Stracchino:
674 It would probably be a per-client option, and would be called
675 something like, say, "Automatically purge obsoleted jobs". What it
676 would do is, when you successfully complete a Differential backup of a
677 client, it would automatically purge all Incremental backups for that
678 client that are rendered redundant by that Differential. Likewise,
679 when a Full backup on a client completed, it would automatically purge
680 all Differential and Incremental jobs obsoleted by that Full backup.
681 This would let people minimize the number of tapes they're keeping on
682 hand without having to master the art of retention times.
683 - Implement a M_SECURITY message class.
684 - When doing a Backup send all attributes back to the Director, who
685 would then figure out what files have been deleted.
686 - Currently in mount.c:236 the SD simply creates a Volume. It should have
687 explicit permission to do so. It should also mark the tape in error
688 if there is an error.
689 - Make sure all restore counters are working correctly in the FD.
690 - SD Bytes Read is wrong.
691 - Look at ALL higher level routines that call block.c to be sure
692 they don't expect something in errmsg.
693 - Investigate doing RAW backup of Win32 partition.
694 - Add thread specific data to hold the jcr -- send error messages from
695 low level routines by accessing it and using Jmsg().
696 - Cancel waiting for Client connect in SD if FD goes away.
698 - Examine Bare Metal restore problem (a FD crash exists somewhere ...).
699 - Implement timeout in response() when it should come quickly.
700 - Implement console @echo command.
701 - Implement a Slot priority (loaded/not loaded).
702 - Implement "vacation" Incremental only saves.
703 - Implement single pane restore (much like the Gftp panes).
704 - Implement Automatic Mount even in operator wait.
705 - Implement create "FileSet"?
706 - Fix watchdog pthread crash on Win32 (this is pthread_kill() Cygwin bug)
707 - Implement "scratch pool" where tapes are defined and can be
708 taken by any pool that needs them.
709 - Implement restore "current system", but take all files without
710 doing selection tree -- so that jobs without File records can
712 - Add prefixlinks to where or not where absolute links to FD.
713 - Issue message to mount a new tape before the rewind.
714 - Simplified client job initiation for portables.
715 - If SD cannot open a drive, make it periodically retry.
716 - Add more of the config info to the tape label.
718 - If tape is marked read-only, then try opening it read-only rather than
719 failing, and remember that it cannot be written.
720 - Refine SD waiting output:
721 Device is being positioned
722 > Device is being positioned for append
723 > Device is being positioned to file x
725 - Figure out some way to estimate output size and to avoid splitting
726 a backup across two Volumes -- this could be useful for writing CDROMs
727 where you really prefer not to have it split -- not serious.
728 - Have SD compute MD5 or SHA1 and compare to what FD computes.
729 - Make VolumeToCatalog calculate an MD5 or SHA1 from the
730 actual data on the Volume and compare it.
731 - Implement Bacula plugins -- design API
732 - Make bcopy read through bad tape records.
733 - Program files (i.e. execute a program to read/write files).
734 Pass read date of last backup, size of file last time.
735 - Add Signature type to File DB record.
736 - CD into subdirectory when open()ing files for backup to
737 speed up things. Test with testfind().
738 - Priority job to go to top of list.
739 - Why are save/restore of device different sizes (sparse?) Yup! Fix it.
740 - Implement some way for the Console to dynamically create a job.
741 - Restore to a particular time -- e.g. before date, after date.
742 - Solaris -I on tar for include list
743 - Need a verbose mode in restore, perhaps to bsr.
744 - bscan without -v is too quiet -- perhaps show jobs.
745 - Add code to reject whole blocks if not wanted on restore.
746 - Check if we can increase Bacula FD priorty in Win2000
747 - Make sure the MaxVolFiles is fully implemented in SD
748 - Check if both CatalogFiles and UseCatalog are set to SD.
749 - Figure out how to do a bare metal Windows restore
750 - Possibly add email to Watchdog if drive is unmounted too
751 long and a job is waiting on the drive.
752 - Restore program that errs in SD due to no tape, reports
753 OK incorrectly in output.
754 - After unmount, if restore job started, ask to mount.
755 - Convert all %x substitution variables, which are hard to remember
756 and read to %(variable-name). Idea from TMDA.
757 - Remove NextId for SQLite. Optimize.
758 - Move all SQL statements into a single location.
759 - Add UA rc and history files.
760 - put termcap (used by console) in ./configure and
761 allow -with-termcap-dir.
762 - Fix Autoprune for Volumes to respect need for full save.
763 - Fix Win32 config file definition name on /install
764 - Compare tape to Client files (attributes, or attributes and data)
765 - Make all database Ids 64 bit.
766 - Write an applet for Linux.
767 - Allow console commands to detach or run in background.
768 - Fix status delay on storage daemon during rewind.
769 - Add SD message variables to control operator wait time
770 - Maximum Operator Wait
771 - Minimum Message Interval
772 - Maximum Message Interval
773 - Send Operator message when cannot read tape label.
774 - Verify level=Volume (scan only), level=Data (compare of data to file).
775 Verify level=Catalog, level=InitCatalog
777 - Add keyword search to show command in Console.
778 - Events : tape has more than xxx bytes.
779 - Complete code in Bacula Resources -- this will permit
780 reading a new config file at any time.
781 - Handle ctl-c in Console
782 - Implement script driven addition of File daemon to config files.
783 - Think about how to make Bacula work better with File (non-tape) archives.
784 - Write Unix emulator for Windows.
785 - Put memory utilization in Status output of each daemon
786 if full status requested or if some level of debug on.
787 - Make database type selectable by .conf files i.e. at runtime
788 - Set flag for uname -a. Add to Volume label.
789 - Restore files modified after date
790 - SET LD_RUN_PATH=$HOME/mysql/lib/mysql
791 - Remove duplicate fields from jcr (e.g. jcr.level and jcr.jr.Level, ...).
792 - Timout a job or terminate if link goes down, or reopen link and query.
793 - Concept of precious tapes (cannot be reused).
794 - Make bcopy copy with a single tape drive.
795 - Permit changing ownership during restore.
798 > My suggestion: Add a feature on the systray menu-icon menu to request
799 > an immediate backup now. This would be useful for laptop users who may
800 > not be on the network when the regular scheduled backup is run.
802 > My wife's suggestion: Add a setting to the win32 client to allow it to
803 > shut down the machine after backup is complete (after, of course,
804 > displaying a "System will shut down in one minute, click here to cancel"
805 > warning dialog). This would be useful for sites that want user
806 > woorkstations to be shut down overnight to save power.
809 - Autolabel should be specified by DIR instead of SD.
812 - AutoScan (check checksum of tape)
813 - Format command = "format /dev/nst0"
817 - Seek resolution (usually corresponds to buffer size)
818 - EODErrorCode=ENOSPC or code
819 - Partial Read error code
820 - Partial write error code
821 - Nonformatted read error
822 - Nonformatted write error
823 - WriteProtected error
827 - IgnoreCloseErrors=yes
837 - FD sends unsaved file list to Director at end of job (see
839 - File daemon should build list of files skipped, and then
840 at end of save retry and report any errors.
841 - Write a Storage daemon that uses pipes and
842 standard Unix programs to write to the tape.
844 - Need something that monitors the JCR queue and
845 times out jobs by asking the deamons where they are.
846 - Enhance Jmsg code to permit buffering and saving to disk.
847 - device driver = "xxxx" for drives.
849 - Ensure that /dev/null works
850 - Need report class for messages. Perhaps
851 report resource where report=group of messages
852 - enhance scan_attrib and rename scan_jobtype, and
853 fill in code for "since" option
854 - Director needs a time after which the report status is sent
855 anyway -- or better yet, a retry time for the job.
856 - Don't reschedule a job if previous incarnation is still running.
857 - Some way to automatically backup everything is needed????
858 - Need a structure for pending actions:
860 - termination status (part of buffered msgs?)
862 Read, Write, Clean, Delete
863 - Login to Bacula; Bacula users with different permissions:
864 owner, group, user, quotas
865 - Store info on each file system type (probably in the job header on tape.
866 This could be the output of df; or perhaps some sort of /etc/mtab record.
868 ========= ideas ===============
869 From: "Jerry K. Schieffer" <jerry@skylinetechnology.com>
870 To: <kern@sibbald.com>
871 Subject: RE: [Bacula-users] future large programming jobs
872 Date: Thu, 26 Feb 2004 11:34:54 -0600
874 I noticed the subject thread and thought I would offer the following
875 merely as sources of ideas, i.e. something to think about, not even as
876 strong as a request. In my former life (before retiring) I often
877 dealt with backups and storage management issues/products as a
878 developer and as a consultant. I am currently migrating my personal
879 network from amanda to bacula specifically because of the ability to
880 cross media boundaries during storing backups.
881 Are you familiar with the commercial product called ADSM (I think IBM
882 now sells it under the Tivoli label)? It has a couple of interesting
883 ideas that may apply to the following topics.
885 1. Migration: Consider that when you need to restore a system, there
886 may be pressure to hurry. If all the information for a single client
887 can eventually end up on the same media (and in chronological order),
888 the restore is facillitated by not having to search past information
889 from other clients. ADSM has the concept of "client affinity" that
890 may be associated with it's storage pools. It seems to me that this
891 concept (as an optional feature) might fit in your architecture for
894 ADSM also has the concept of defining one or more storage pools as
895 "copy pools" (almost mirrors, but only in the sense of contents).
896 These pools provide the ability to have duplicte data stored both
897 onsite and offsite. The copy process can be scheduled to be handled
898 by their storage manager during periods when there is no backup
899 activity. Again, the migration process might be a place to consider
900 implementing something like this.
903 > It strikes me that it would be very nice to be able to do things
905 > have the Job(s) backing up the machines run, and once they have all
906 > completed, start a migration job to copy the data from disks Volumes
908 > a tape library and then to offsite storage. Maybe this can already
910 > done with some careful scheduling and Job prioritzation; the events
911 > mechanism described below would probably make it very easy.
913 This is the goal. In the first step (before events), you simply
915 the Migration to tape later.
917 2. Base jobs: In ADSM, each copy of each stored file is tracked in
918 the database. Once a file (unique by path and metadata such as dates,
919 size, ownership, etc.) is in a copy pool, no more copies are made. In
920 other words, when you start ADSM, it begins like your concept of a
921 base job. After that it is in the "incremental" mode. You can
922 configure the number of "generations" of files to be retained, plus a
923 retention date after which even old generations are purged. The
924 database tracks the contents of media and projects the percentage of
925 each volume that is valid. When the valid content of a volume drops
926 below a configured percentage, the valid data are migrated to another
927 volume and the old volume is marked as empty. Note, this requires
928 ADSM to have an idea of the contents of a client, i.e. marking the
929 database when an existing file was deleted, but this would solve your
930 issue of restoring a client without restoring deleted files.
932 This is pretty far from what bacula now does, but if you are going to
933 rip things up for Base jobs,.....
934 Also, the benefits of this are huge for very large shops, especially
935 with media robots, but are a pain for shops with manual media
939 > Base jobs sound pretty useful, but I'm not dying for them.
941 Nobody is dying for them, but when you see what it does, you will die
944 3. Restoring deleted files: Since I think my comments in (2) above
945 have low probability of implementation, I'll also suggest that you
946 could approach the issue of deleted files by a mechanism of having the
947 fd report to the dir, a list of all files on the client for every
948 backup job. The dir could note in the database entry for each file
949 the date that the file was seen. Then if a restore as of date X takes
950 place, only files that exist from before X until after X would be
951 restored. Probably the major cost here is the extra date container in
952 each row of the files table.
954 Thanks for "listening". I hope some of this helps. If you want to
955 contact me, please send me an email - I read some but not all of the
956 mailing list traffic and might miss a reply there.
958 Please accept my compliments for bacula. It is doing a great job for
959 me!! I sympathize with you in the need to wrestle with excelence in
960 execution vs. excelence in feature inclusion.
965 ==============================
968 - Design at hierarchial storage for Bacula. Migration and Clone.
969 - Implement FSM (File System Modules).
970 - Audit M_ error codes to ensure they are correct and consistent.
971 - Add variable break characters to lex analyzer.
972 Either a bit mask or a string of chars so that
973 the caller can change the break characters.
974 - Make a single T_BREAK to replace T_COMMA, etc.
975 - Ensure that File daemon and Storage daemon can
976 continue a save if the Director goes down (this
977 is NOT currently the case). Must detect socket error,
978 buffer messages for later.
979 - Enhance time/duration input to allow multiple qualifiers e.g. 3d2h
980 - Add ability to backup to two Storage devices (two SD sessions) at
981 the same time -- e.g. onsite, offsite.
982 - Add the ability to consolidate old backup sets (basically do a restore
983 to tape and appropriately update the catalog). Compress Volume sets.
984 Might need to spool via file is only one drive is available.
985 - Compress or consolidate Volumes of old possibly deleted files. Perhaps
986 someway to do so with every volume that has less than x% valid
990 Migration: Move a backup from one Volume to another
991 Clone: Copy a backup -- two Volumes
993 Bacula Migration is based on Jobs (apparently Networker is file by file).
995 Migration triggered by:
999 Highwater mark (keep total size)
1004 ======================================================
1006 It is somewhat like a Full save becomes an incremental since
1007 the Base job (or jobs) plus other non-base files.
1009 - A Base backup is same as Full backup, just different type.
1010 - New BaseFiles table that contains:
1012 BaseJobId - Base JobId referenced for this FileId (needed ???)
1013 JobId - JobId currently running
1014 FileId - File not backed up, exists in Base Job
1015 FileIndex - FileIndex from Base Job.
1016 i.e. for each base file that exists but is not saved because
1017 it has not changed, the File daemon sends the JobId, BaseId,
1018 FileId, FileIndex back to the Director who creates the DB entry.
1019 - To initiate a Base save, the Director sends the FD
1020 the FileId, and full filename for each file in the Base.
1021 - When the FD finds a Base file, he requests the Director to
1022 send him the full File entry (stat packet plus MD5), or
1023 conversely, the FD sends it to the Director and the Director
1024 says yes or no. This can be quite rapid if the FileId is kept
1025 by the FD for each Base Filename.
1026 - It is probably better to have the comparison done by the FD
1027 despite the fact that the File entry must be sent across the
1029 - An alternative would be to send the FD the whole File entry
1030 from the start. The disadvantage is that it requires a lot of
1031 space. The advantage is that it requires less communications
1033 - The Job record must be updated to indicate that one or more
1035 - At end of Job, FD returns:
1036 1. Count of base files/bytes not written to tape (i.e. matches)
1037 2. Count of base file that were saved i.e. had changed.
1038 - No tape record would be written for a Base file that matches, in the
1039 same way that no tape record is written for Incremental jobs where
1040 the file is not saved because it is unchanged.
1041 - On a restore, all the Base file records must explicitly be
1042 found from the BaseFile tape. I.e. for each Full save that is marked
1043 to have one or more Base Jobs, search the BaseFile for all occurrences
1045 - An optimization might be to make the BaseFile have:
1051 This would avoid the need to explicitly fetch each File record for
1052 the Base job. The Base Job record will be fetched to get the
1053 VolSessionId and VolSessionTime.
1054 =========================================================
1057 ==========================================================
1059 For each Incremental job that is run, there may be files that
1060 were found but not saved because they were locked (this applies
1061 only to Windows). Such a system could send back to the Director
1062 a list of Unsaved files.
1064 - New UnSavedFiles table that contains:
1068 - Then in the next Incremental job, the list of Unsaved Files will be
1069 feed to the FD, who will ensure that they are explicitly chosen even
1070 if standard date/time check would not have selected them.
1071 =============================================================
1074 Done: (see kernsdone for more)
1077 - John's Full save failed with 1.32c FD and 1.31 Dir no FD status,
1078 and no error message.
1079 - Add fd and st as Console keywords.
1080 - Recycling volume with a Slot requires an operator intervention:
1081 rufus-dir: Start Backup JobId 18, Job=kernsave.2003-11-01_21.23.52
1082 rufus-dir: Pruned 1 Job on Volume Vol01 from catalog.
1083 rufus-dir: There are no Jobs associated with Volume Vol01. Marking it purged.
1084 rufus-dir: Recycled volume "Vol01"
1085 rufus-sd: Please mount Volume "Vol01" on Storage Device "DDS-4" for Job kernsave.2003-11-01_21.23.52
1086 Use "mount" command to release Job.
1087 - Implement Dan's bacula script (email of 26 Oct).
1088 - Add JobName= to VerifyToCatalog so that all verifies can be done at the end.
1089 - Edit the Client/Storage name into authentication failure messages.
1090 - Fix packet too big problem. This is most likely a Windows TCP stack
1092 - Implement ClientRunBeforeJob and ClientRunAfterJob.
1093 - Implement forward spacing block/file: position_device(bsr) --
1094 just before read_block_from_device();
1097 Multiple drive autochanger data: see Alan Brown
1098 mtx -f xxx unloadStorage Element 1 is Already Full(drive 0 was empty)
1099 Unloading Data Transfer Element into Storage Element 1...source Element
1100 Address 480 is Empty
1102 (drive 0 was empty and so was slot 1)
1103 > mtx -f xxx load 15 0
1104 no response, just returns to the command prompt when complete.
1105 > mtx -f xxx status Storage Changer /dev/changer:2 Drives, 60 Slots ( 2 Import/Export )
1106 Data Transfer Element 0:Full (Storage Element 15 Loaded):VolumeTag = HX001
1107 Data Transfer Element 1:Empty
1108 Storage Element 1:Empty
1109 Storage Element 2:Full :VolumeTag=HX002
1110 Storage Element 3:Full :VolumeTag=HX003
1111 Storage Element 4:Full :VolumeTag=HX004
1112 Storage Element 5:Full :VolumeTag=HX005
1113 Storage Element 6:Full :VolumeTag=HX006
1114 Storage Element 7:Full :VolumeTag=HX007
1115 Storage Element 8:Full :VolumeTag=HX008
1116 Storage Element 9:Full :VolumeTag=HX009
1117 Storage Element 10:Full :VolumeTag=HX010
1118 Storage Element 11:Empty
1119 Storage Element 12:Empty
1120 Storage Element 13:Empty
1121 Storage Element 14:Empty
1122 Storage Element 15:Empty
1123 Storage Element 16:Empty....
1124 Storage Element 28:Empty
1125 Storage Element 29:Full :VolumeTag=CLNU01L1
1126 Storage Element 30:Empty....
1127 Storage Element 57:Empty
1128 Storage Element 58:Full :VolumeTag=NEX261L2
1129 Storage Element 59 IMPORT/EXPORT:Empty
1130 Storage Element 60 IMPORT/EXPORT:Empty
1132 Unloading Data Transfer Element into Storage Element 15...done
1134 (just to verify it remembers where it came from, however it can be
1135 overrriden with mtx unload {slotnumber} to go to any storage slot.)
1137 There needs to be a table of drive # to devices somewhere - If there are
1138 multiple changers or drives there may not be a 1:1 correspondance between
1139 changer drive number and system device name - and depending on the way the
1140 drives are hooked up to scsi busses, they may not be linearly numbered
1141 from an offset point either.something like
1143 Autochanger drives = 2
1144 Autochanger drive 0 = /dev/nst1
1145 Autochanger drive 1 = /dev/nst2
1146 IMHO, it would be _safest_ to use explicit mtx unload commands at all
1147 times, not just for multidrive changers. For a 1 drive changer, that's
1153 MTX's manpage (1.2.15):
1154 unload [<slotnum>] [ <drivenum> ]
1155 Unloads media from drive <drivenum> into slot
1156 <slotnum>. If <drivenum> is omitted, defaults to
1157 drive 0 (as do all commands). If <slotnum> is
1158 omitted, defaults to the slot that the drive was
1159 loaded from. Note that there's currently no way
1160 to say 'unload drive 1's media to the slot it
1161 came from', other than to explicitly use that
1162 slot number as the destination.AB
1168 undef# camcontrol devlist
1169 <WANGTEK 51000 SCSI M74H 12B3> at scbus0 target 2 lun 0 (pass0,sa0)
1170 <ARCHIVE 4586XX 28887-XXX 4BGD> at scbus0 target 4 lun 0 (pass1,sa1)
1171 <ARCHIVE 4586XX 28887-XXX 4BGD> at scbus0 target 4 lun 1 (pass2)
1173 tapeinfo -f /dev/sg0 with a bad tape in drive 1:
1174 [kern@rufus mtx-1.2.17kes]$ ./tapeinfo -f /dev/sg0
1175 Product Type: Tape Drive
1177 Product ID: 'C5713A '
1179 Attached Changer: No
1180 TapeAlert[3]: Hard Error: Uncorrectable read/write error.
1181 TapeAlert[20]: Clean Now: The tape drive neads cleaning NOW.
1188 Medium Type: Not Loaded
1191 DataCompEnabled: yes
1192 DataCompCapable: yes
1193 DataDeCompEnabled: yes
1200 Handling removable disks
1202 From: Karl Cunningham <karlc@keckec.com>
1204 My backups are only to hard disk these days, in removable bays. This is my
1205 idea of how a backup to hard disk would work more smoothly. Some of these
1206 things Bacula does already, but I mention them for completeness. If others
1207 have better ways to do this, I'd like to hear about it.
1209 1. Accommodate several disks, rotated similar to how tapes are. Identified
1210 by partition volume ID or perhaps by the name of a subdirectory.
1211 2. Abort & notify the admin if the wrong disk is in the bay.
1212 3. Write backups to different subdirectories for each machine to be backed
1214 4. Volumes (files) get created as needed in the proper subdirectory, one
1216 5. When a disk is recycled, remove or zero all old backup files. This is
1217 important as the disk being recycled may be close to full. This may be
1218 better done manually since the backup files for many machines may be
1219 scattered in many subdirectories.
1224 - To pass Include 1 or two letter commands
1225 I Name Include name - first record
1226 B Name Base name - repeat
1229 E zzz Exclude expression (wild card)
1231 D "reader" Reader program
1232 T "writer" Writer program
1233 O Options In current commpressed format
1234 (compression, signature, onefs, recurse, sparse,
1235 replace, verify options, ...)
1237 B BaseName Start second option set any letter
1240 F Number Number of filenames to follow
1245 F Number Number of filenames to follow
1248 - Add regression of btape "fill"
1250 - Implement "label slots=1,2... pool=yyy barcodes"
1251 - Mark a non-removable disk Volume in error rather than asking sys op to
1253 - Implement Fixed storage LabelFormat test.
1254 - Add reporting in attr despooling.
1255 - Figure out how to use ssh or stunnel to protect Bacula communications.
1256 Add Dan's work to manual See ssl.txt
1259 - Implement fast tree insert (doubly linked list)
1260 - Disallow using Internal database
1261 - Feedback while the restore tree is being built.
1262 - Add detection of Whitebox release in ./configure
1263 - Add InChanger to "list media"
1264 - Implement second cut of SIGHUP in Dir
1265 (Implement resources on a single pointer)
1266 - Add regular expressions to FileSets (needs testing).
1267 - Test new despooling error recovery code when partition fills.
1268 - Sort Scheduled jobs status listing by start time.
1269 - Add priority to Scheduled jobs status listing.
1270 - Look at Win32 inc problem when new directory added files not saved???
1271 - Require resource names to be unique.
1272 - Make sure that all errors from libacl are printed.
1273 - Make FD run on Win95 if possible (not really possible).
1274 - Fix GROUP BYs in all SQL (done by Dan -- thanks)
1275 - Make entering multiple Storage names in Dir illegal.
1276 - Make btape release the drive during the "test" append.
1278 You have selected the following JobIds: 739,856,797
1279 Building directory tree for JobId 739 ...
1280 Building directory tree for JobId 856 ...
1281 Building directory tree for JobId 797 ...3
1282 Jobs inserted into the tree.
1283 - Fix errno handling in win32 compat routines.
1284 - Add better error codes to run_program (10000+)
1285 - Revisit and revise Disaster Recovery (fix SCSI and RAID
1287 - When passwords do not match, print message that points the
1289 - Do tape alerts -- see tapealert.txt
1290 - Find memory leaks in address_conf.c
1291 - Network order of port and ip addresses.
1292 - SetIP in dird/fd_cmds.c is broken.
1293 - don't use inet_addr() it is deprecated.
1294 - Fix code in filed.c for init_bsock();
1295 - Fix code in console.c for init_bsock();
1296 - Solve the termcap.h problem on Solaris configure.
1297 - Make Verify jobs require exclusive use of Volume as Restore
1299 - Fix restore ++++ that get intermingled with "Building directory tree"
1300 - Fix doc in dirdconf.html for @, where the file must
1301 have not a list of files, but
1304 - Test Win32 errno handling.
1305 - port 1.34.6-block.patch to 1.35
1306 - Look at Chris' patch for bscan bug.
1307 - Fix error handling in spooling both data and attribute.
1308 - Implement Ignore FileSet Change.
1309 - Doc new duration time input editing.
1310 - Bacula rescue CDROM implement isolinux
1311 - Make sure Qmsgs are dequeued by FD and SD.
1312 - Check if ACLs allocated at dird_conf.c:1214 are being properly
1314 - Test/doc Tape Alerts
1315 - Doc dbcheck eliminate orphaned clients.
1316 - Doc Phil's new delete job jobid scanning code.
1317 - Document that console commands can be abbreviated.
1318 - Doc update AllFromVol
1319 - Doc -p option in stored
1320 - New IP address specification is used as follows:
1321 [sdaddresses|diraddresses|fdaddresses] = { [[ip|ipv4|ipv6] = {
1322 [[addr|port] = [^ ]+[\n;]+] }] }
1324 so it could look for example like this:
1325 SDaddresses = { ip = {
1326 addr = 1.2.3.4; port = 1205; }
1328 addr = 1.2.3.4; port = http; }
1341 addr = 201:220:222::2
1344 addr = bluedot.thun.net
1347 as a consequence, you can now specify multiple IP addresses and
1348 ports to be used. In the case of a server, it will listen on
1349 all those that you specify. In the case of connecting to the server,
1350 Bacula will attempt connecting to one at a time until it succeeds.
1351 And, in a few other special cases, Bacula will use only the first
1354 The default port numbers are still the same and the services and hosts
1355 are also resolved by name. So now you could use the real names for the
1358 An ip section will allow resolution to either an ipv4 or an ipv6 address.
1359 An ipv4 section forces the resolution to be only ipv4, and an ipv6 section
1360 forces the resolution to be only ipv6.
1361 - Fix silly restriction requiring Include { Options { xxx } } to be
1363 - Restore c: with a prefix into /prefix/c/ to prevent c: and d:
1364 files with the same name from overwritting each other.
1365 - Add "Multiple connections = yes/no" to catalog resource.
1366 - Add new DCR calling sequences everywhere in SD. This will permit
1367 simultaneous use of multiple devices by a single job.
1368 - Fix bscan so that it releases the drive when requesting a new tape.
1369 - List verify options for DiskToCatalog in doc.
1370 - Turn on transactions if multiple connections are on in DB.
1371 - Fix storage overrides in Run directives.