6 - !!! Cannot restore two jobs a the same time that were
7 written simultaneously unless they were totally spooled.
8 - Document cleaning up the spool files:
9 db, pid, state, bsr, mail, conmsg, spool
10 - Document the multiple-drive-changer.txt script.
11 - Pruning with Admin job.
12 - Does WildFile match against full name? Doc.
13 - %d and %v only valid on Director, not for ClientRunBefore/After.
14 - During tests with the 260 char fix code, I found one problem:
15 if the system "sees" a long path once, it seems to forget it's
16 working drive (e.g. c:\), which will lead to a problem during
17 the next job (create bootstrap file will fail). Here is the
18 workaround: specify absolute working and pid directory in
19 bacula-fd.conf (e.g. c:\bacula\working instead of
21 - Document techniques for restoring large numbers of files.
22 - Document setting my.cnf to big file usage.
23 - Add example of proper index output to doc. show index from File;
24 - Correct the Include syntax in the m4.xxx files in examples/conf
25 - Document JobStatus and Termination codes.
26 - Fix the error with the "DVI file can't be opened" while
27 building the French PDF.
28 - Document more DVD stuff
36 - Document all the little details of setting up certificates for
37 the Bacula data encryption code.
38 - Document more precisely how to use master keys -- especially
39 for disaster recovery.
42 - Migration from other vendors
46 - Backup conf/exe (all daemons)
47 - Backup up system state
48 - Detect state change of system (verify)
49 - Synthetic Full, Diff, Inc (Virtual, Reconstructed)
51 - Modules for Databases, Exchange, ...
52 - Novell NSS backup http://www.novell.com/coolsolutions/tools/18952.html
53 - Compliance norms that compare restored code hash code.
54 - When glibc crash, get address with
56 - How to sync remote offices.
58 http://www.microsoft.com/technet/itshowcase/content/exchbkup.mspx
61 Extract capability (#25)
62 Continued enhancement of bweb
63 Threshold triggered migration jobs (not currently in list, but will be
65 Client triggered backups
66 Complete rework of the scheduling system (not in list)
67 Performance and usage instrumentation (not in list)
68 See email of 21Aug2007 for details.
71 - Implement SDErrors (must return from SD)
72 - Implement USB keyboard support in rescue CD.
73 - Remove all install temp files in Win32 PLUGINSDIR.
74 - Audit retention periods to make sure everything is 64 bit.
75 - Use E'xxx' to escape PostgreSQL strings.
76 - No where in restore causes kaboom.
77 - Performance: multiple spool files for a single job.
78 - Performance: despool attributes when despooling data (problem
79 multiplexing Dir connection).
80 - Make restore use the in-use volume reservation algorithm.
81 - Look at mincore: http://insights.oetiker.ch/linux/fadvise.html
82 - Unicode input http://en.wikipedia.org/wiki/Byte_Order_Mark
83 - Add TLS to bat (should be done).
84 - When Pool specifies Storage command override does not work.
85 - Implement wait_for_sysop() message display in wait_for_device(), which
86 now prints warnings too often.
87 - Ensure that each device in an Autochanger has a different
89 - Add Catalog = to Pool resource so that pools will exist
90 in only one catalog -- currently Pools are "global".
91 - Look at sg_logs -a /dev/sg0 for getting soft errors.
92 - btape "test" command with Offline on Unmount = yes
94 This test is essential to Bacula.
96 I'm going to write one record in file 0,
97 two records in file 1,
98 and three records in file 2
100 02-Feb 11:00 btape: ABORTING due to ERROR in dev.c:715
101 dev.c:714 Bad call to rewind. Device "LTO" (/dev/nst0) not open
102 02-Feb 11:00 btape: Fatal Error because: Bacula interrupted by signal 11: Segmentation violation
103 Kaboom! btape, btape got signal 11. Attempting traceback.
105 - Encryption -- email from Landon
106 > The backup encryption algorithm is currently not configurable, and is
107 > set to AES_128_CBC in src/filed/backup.c. The encryption code
108 > supports a number of different ciphers (as well as adding arbitrary
109 > new ones) -- only a small bit of code would be required to map a
110 > configuration string value to a CRYPTO_CIPHER_* value, if anyone is
111 > interested in implementing this functionality.
113 - Why doesn't @"xxx abc" work in a conf file?
114 - Figure out some way to "automatically" backup conf changes.
115 - Add the OS version back to the Win32 client info.
116 - Restarted jobs have a NULL in the from field.
117 - Modify SD status command to indicate when the SD is writing
118 to a DVD (the device is not open -- see bug #732).
119 - Look at the possibility of adding "SET NAMES UTF8" for MySQL,
120 and possibly changing the blobs into varchar.
121 - Ensure that the SD re-reads the Media record if the JobFiles
122 does not match -- it may have been updated by another job.
123 - Look at moving the Storage directive from the Job to the
124 Pool in the default conf files.
126 - Test Volume compatibility between machine architectures
127 - Encryption documentation
128 - Wrong jobbytes with query 12 (todo)
129 - Bare-metal recovery Windows (todo)
134 - Access Mode = Read-Only, Read-Write, Unavailable, Destroyed, Offsite
136 - Maximum number of scratch volumes
138 - Next Pool (already have)
139 - Reclamation threshold
141 - Reuse delay (after all files purged from volume before it can be used)
142 - Copy Pool = xx, yyy (or multiple lines).
144 - Allow pool selection during restore.
146 - Average tape size from Eric
147 SELECT COALESCE(media_avg_size.volavg,0) * count(Media.MediaId) AS volmax, GROUP BY Media.MediaType, Media.PoolId, media_avg_size.volavg
148 count(Media.MediaId) AS volnum,
149 sum(Media.VolBytes) AS voltotal,
150 Media.PoolId AS PoolId,
151 Media.MediaType AS MediaType
153 LEFT JOIN (SELECT avg(Media.VolBytes) AS volavg,
154 Media.MediaType AS MediaType
156 WHERE Media.VolStatus = 'Full'
157 GROUP BY Media.MediaType
158 ) AS media_avg_size ON (Media.MediaType = media_avg_size.MediaType)
159 GROUP BY Media.MediaType, Media.PoolId, media_avg_size.volavg
163 - Add doc for bweb -- especially Installation
165 http://www.orangecrate.com/modules.php?name=News&file=article&sid=501
167 - Despool attributes in separate thread
170 - Check why restore repeatedly sends Rechdrs between
171 each data chunk -- according to James Harper 9Jan07.
174 - Full at least once a month, ...
175 - Cancel Inc if Diff/Full running
176 - More intelligent re-run
177 - New/deleted file backup
179 - Incremental backup -- rsync, Stow
183 - Look at mondo/mindi
184 - Don't restore Solaris Door files:
185 #define S_IFDOOR in st_mode.
186 see: http://docs.sun.com/app/docs/doc/816-5173/6mbb8ae23?a=view#indexterm-360
187 - Make Bacula by default not backup tmpfs, procfs, sysfs, ...
188 - Fix hardlinked immutable files when linking a second file, the
189 immutable flag must be removed prior to trying to link it.
190 - Implement Python event for backing up/restoring a file.
191 - Change dbcheck to tell users to use native tools for fixing
192 broken databases, and to ensure they have the proper indexes.
193 - add udev rules for Bacula devices.
194 - If a job terminates, the DIR connection can close before the
195 Volume info is updated, leaving the File count wrong.
196 - Look at why SIGPIPE during connection can cause seg fault in
197 writing the daemon message, when Dir dropped to bacula:bacula
198 - Look at zlib 32 => 64 problems.
199 - Possibly turn on St. Bernard code.
200 - Fix bextract to restore ACLs, or better yet, use common routines.
201 - Do we migrate appendable Volumes?
202 - Remove queue.c code.
203 - Print warning message if LANG environment variable does not specify
205 - New dot commands from Arno.
206 .show device=xxx lists information from one storage device, including
207 devices (I'm not even sure that information exists in the DIR...)
208 .move eject device=xxx mostly the same as 'unmount xxx' but perhaps with
209 better machine-readable output like "Ok" or "Error busy"
210 .move eject device=xxx toslot=yyy the same as above, but with a new
211 target slot. The catalog should be updated accordingly.
212 .move transfer device=xxx fromslot=yyy toslot=zzz
215 - Article: http://www.heise.de/open/news/meldung/83231
216 - Article: http://www.golem.de/0701/49756.html
217 - Article: http://lwn.net/Articles/209809/
218 - Article: http://www.onlamp.com/pub/a/onlamp/2004/01/09/bacula.html
219 - Article: http://www.linuxdevcenter.com/pub/a/linux/2005/04/07/bacula.html
220 - Article: http://www.osreviews.net/reviews/admin/bacula
221 - Article: http://www.debianhelp.co.uk/baculaweb.htm
223 - Wikis mentioning Bacula
224 http://wiki.finkproject.org/index.php/Admin:Backups
225 http://wiki.linuxquestions.org/wiki/Bacula
226 http://www.openpkg.org/product/packages/?package=bacula
227 http://www.iterating.com/products/Bacula
228 http://net-snmp.sourceforge.net/wiki/index.php/Net-snmp_extensions
229 http://www.section6.net/wiki/index.php/Using_Bacula_for_Tape_Backups
230 http://bacula.darwinports.com/
231 http://wiki.mandriva.com/en/Releases/Corporate/Server_4/Notes#Bacula
232 http://en.wikipedia.org/wiki/Bacula
235 http://www.devco.net/pubwiki/Bacula/
236 http://paramount.ind.wpi.edu/wiki/doku.php
237 http://gentoo-wiki.com/HOWTO_Backup
238 http://www.georglutz.de/wiki/Bacula
239 http://www.clarkconnect.com/wiki/index.php?title=Modules_-_LAN_Backup/Recovery
240 http://linuxwiki.de/Bacula (in German)
242 - Possibly allow SD to spool even if a tape is not mounted.
243 - It appears to me that you have run into some sort of race
244 condition where two threads want to use the same Volume and they
245 were both given access. Normally that is no problem. However,
246 one thread wanted the particular Volume in drive 0, but it was
247 loaded into drive 1 so it decided to unload it from drive 1 and
248 then loaded it into drive 0, while the second thread went on
249 thinking that the Volume could be used in drive 1 not realizing
250 that in between time, it was loaded in drive 0.
251 I'll look at the code to see if there is some way we can avoid
252 this kind of problem. Probably the best solution is to make the
253 first thread simply start using the Volume in drive 1 rather than
254 transferring it to drive 0.
255 - Fix re-read of last block to check if job has actually written
256 a block, and check if block was written by a different job
257 (i.e. multiple simultaneous jobs writing).
258 - Figure out how to configure query.sql. Suggestion to use m4:
259 == changequote.m4 ===
260 changequote(`[',`]')dnl
261 ==== query.sql.in ===
262 :List next 20 volumes to expire
264 Pool.Name AS PoolName,
269 [ FROM_UNIXTIME(UNIX_TIMESTAMP(Media.LastWritten) Media.VolRetention) AS Expire, ])dnl
271 [ media.lastwritten + interval '1 second' * media.volretention as expire, ])dnl
275 ON Media.PoolId=Pool.PoolId
276 WHERE Media.LastWritten>0
280 Command: m4 -DmySQL changequote.m4 query.sql.in >query.sql
282 The problem is that it requires m4, which is not present on all machines
284 - Given all the problems with FIFOs, I think the solution is to do something a
285 little different, though I will look at the code and see if there is not some
286 simple solution (i.e. some bug that was introduced). What might be a better
287 solution would be to use a FIFO as a sort of "key" to tell Bacula to read and
288 write data to a program rather than the FIFO. For example, suppose you
293 Then, I could imagine if you backup and restore this file with a direct
294 reference as is currently done for fifos, instead, during backup Bacula will
297 /home/kern/my-fifo.backup
299 and read the data that my-fifo.backup writes to stdout. For restore, Bacula
302 /home/kern/my-fifo.restore
304 and send the data backed up to stdout. These programs can either be an
305 executable or a shell script and they need only read/write to stdin/stdout.
307 I think this would give a lot of flexibility to the user without making any
308 significant changes to Bacula.
313 select FilenameId from Filename where Name='';
314 # Get list of all directories referenced in a Backup.
315 select Path.Path from Path,File where File.JobId=nnn and
316 File.FilenameId=(FilenameId-from-above) and File.PathId=Path.PathId
317 order by Path.Path ASC;
319 - Look into using Dart for testing
320 http://public.kitware.com/Dart/HTML/Index.shtml
322 - Look into replacing autotools with cmake
323 http://www.cmake.org/HTML/Index.html
325 === Migration from David ===
326 What I'd like to see:
329 Name = "<poolname>-migrate"
333 Migration Selection Type = LowestUtil | OldestVol | PoolOccupancy |
334 Client | PoolResidence | Volume | JobName | SQLquery
335 Migration Selection Pattern = "regexp"
336 Next Pool = <override>
339 There should be no need for a Level (migration is always Full, since you
340 don't calculate differential/incremental differences for migration),
341 Storage should be determined by the volume types in the pool, and Client
342 is really a selection issue. Migration should always occur to the
343 NextPool defined in the pool definition. If no nextpool is defined, the
344 job should end with a reason of "no place to go". If Next Pool statement
345 is present, we override the check in the pool definition and use the
348 Here's how I'd define Migration Selection Types:
351 Client -- Migrate data from selected client only. Migration Selection
352 Pattern regexp provides pattern to select client names, eg ^FS00* makes
353 all client names starting with FS00 eligible for migration.
355 Jobname -- Migration all jobs matching name. Migration Selection Pattern
356 regexp provides pattern to select jobnames existing in pool.
358 Volume -- Migrate all data on specified volumes. Migration Selection
359 Pattern regexp provides selection criteria for volumes to be migrated.
360 Volumes must exist in pool to be eligible for migration.
364 LowestUtil -- Identify the volume in the pool with the least data on it
365 and empty it. No Migration Selection Pattern required.
367 OldestVol -- Identify the LRU volume with data written, and empty it. No
368 Migration Selection Pattern required.
370 PoolOccupancy -- if pool occupancy exceeds <highmig>, migrate volumes
371 (starting with most full volumes) until pool occupancy drops below
372 <lowmig>. Pool highmig and lowmig values are in pool definition, no
373 Migration Selection Pattern required.
377 SQLQuery -- Migrate all jobuids returned by the supplied SQL query.
378 Migration Selection Pattern contains SQL query to execute; should return
379 a list of 1 or more jobuids to migrate.
381 PoolResidence -- Migrate data sitting in pool for longer than
382 PoolResidence value in pool definition. Migration Selection Pattern
383 optional; if specified, override value in pool definition (value in
387 [ possibly a Python event -- kes ]
389 - Mount on an Autochanger with no tape in the drive causes:
390 Automatically selected Storage: LTO-changer
391 Enter autochanger drive[0]: 0
392 3301 Issuing autochanger "loaded drive 0" command.
393 3302 Autochanger "loaded drive 0", result: nothing loaded.
394 3301 Issuing autochanger "loaded drive 0" command.
395 3302 Autochanger "loaded drive 0", result: nothing loaded.
396 3902 Cannot mount Volume on Storage Device "LTO-Drive1" (/dev/nst0) because:
397 Couldn't rewind device "LTO-Drive1" (/dev/nst0): ERR=dev.c:678 Rewind error on "LTO-Drive1" (/dev/nst0). ERR=No medium found.
398 3905 Device "LTO-Drive1" (/dev/nst0) open but no Bacula volume is mounted.
399 If this is not a blank tape, try unmounting and remounting the Volume.
400 - If Drive 0 is blocked, and drive 1 is set "Autoselect=no", drive 1 will
402 - Autochanger did not change volumes.
403 select * from Storage;
404 +-----------+-------------+-------------+
405 | StorageId | Name | AutoChanger |
406 +-----------+-------------+-------------+
407 | 1 | LTO-changer | 0 |
408 +-----------+-------------+-------------+
409 05-May 03:50 roxie-sd: 3302 Autochanger "loaded drive 0", result is Slot 11.
410 05-May 03:50 roxie-sd: Tibs.2006-05-05_03.05.02 Warning: Director wanted Volume "LT
411 Current Volume "LT0-002" not acceptable because:
412 1997 Volume "LT0-002" not in catalog.
413 05-May 03:50 roxie-sd: Tibs.2006-05-05_03.05.02 Error: Autochanger Volume "LT0-002"
414 Setting InChanger to zero in catalog.
415 05-May 03:50 roxie-dir: Tibs.2006-05-05_03.05.02 Error: Unable to get Media record
417 05-May 03:50 roxie-sd: Tibs.2006-05-05_03.05.02 Fatal error: Error getting Volume i
418 05-May 03:50 roxie-sd: Tibs.2006-05-05_03.05.02 Fatal error: Job 530 canceled.
419 05-May 03:50 roxie-sd: Tibs.2006-05-05_03.05.02 Fatal error: spool.c:249 Fatal appe
420 05-May 03:49 Tibs: Tibs.2006-05-05_03.05.02 Fatal error: c:\cygwin\home\kern\bacula
429 FirstWritten: 2006-05-05 03:11:54
430 LastWritten: 2006-05-05 03:50:23
431 LabelDate: 2005-12-26 16:52:40
442 VolRetention: 31,536,000
454 Note VolStatus is blank!!!!!
461 FirstWritten: 0000-00-00 00:00:00
462 LastWritten: 0000-00-00 00:00:00
463 LabelDate: 2005-12-26 16:52:40
474 VolRetention: 31,536,000
487 Automatically selected Storage: LTO-changer
488 Enter autochanger drive[0]: 0
489 3301 Issuing autochanger "loaded drive 0" command.
490 3302 Autochanger "loaded drive 0", result: nothing loaded.
491 3301 Issuing autochanger "loaded drive 0" command.
492 3302 Autochanger "loaded drive 0", result: nothing loaded.
493 3902 Cannot mount Volume on Storage Device "LTO-Drive1" (/dev/nst0) because:
494 Couldn't rewind device "LTO-Drive1" (/dev/nst0): ERR=dev.c:678 Rewind error on "LTO-Drive1" (/dev/nst0). ERR=No medium found.
496 3905 Device "LTO-Drive1" (/dev/nst0) open but no Bacula volume is mounted.
497 If this is not a blank tape, try unmounting and remounting the Volume.
499 - http://www.dwheeler.com/essays/commercial-floss.html
500 - Add VolumeLock to prevent all but lock holder (SD) from updating
501 the Volume data (with the exception of VolumeState).
502 - The btape fill command does not seem to use the Autochanger
503 - Make Windows installer default to system disk drive.
504 - Look at using ioctl(FIOBMAP, ...) on Linux, and
505 DeviceIoControl(..., FSCTL_QUERY_ALLOCATED_RANGES, ...) on
506 Win32 for sparse files.
507 http://www.flexhex.com/docs/articles/sparse-files.phtml
508 http://www.informatik.uni-frankfurt.de/~loizides/reiserfs/fibmap.html
509 - Directive: at <event> "command"
510 - Command: pycmd "command" generates "command" event. How to
511 attach to a specific job?
512 - Integrate Christopher's St. Bernard code.
513 - run_cmd() returns int should return JobId_t
514 - get_next_jobid_from_list() returns int should return JobId_t
515 - Document export LDFLAGS=-L/usr/lib64
516 - Don't attempt to restore from "Disabled" Volumes.
517 - Network error on Win32 should set Win32 error code.
518 - What happens when you rename a Disk Volume?
519 - Job retention period in a Pool (and hence Volume). The job would
521 - Look at -D_FORTIFY_SOURCE=2
522 - Add Win32 FileSet definition somewhere
523 - Look at fixing restore status stats in SD.
524 - Look at using ioctl(FIMAP) and FIGETBSZ for sparse files.
525 http://www.informatik.uni-frankfurt.de/~loizides/reiserfs/fibmap.html
526 - Implement a mode that says when a hard read error is
527 encountered, read many times (as it currently does), and if the
528 block cannot be read, skip to the next block, and try again. If
529 that fails, skip to the next file and try again, ...
531 create table LevelType (LevelType binary(1), LevelTypeLong tinyblob);
532 insert into LevelType (LevelType,LevelTypeLong) values
536 - Show files/second in client status output.
537 - Add a recursive mark command (rmark) to restore.
538 - "Minimum Job Interval = nnn" sets minimum interval between Jobs
539 of the same level and does not permit multiple simultaneous
540 running of that Job (i.e. lets any previous invocation finish
541 before doing Interval testing).
542 - Look at simplifying File exclusions.
543 - New directive "Delete purged Volumes"
544 - new pool XXX with ScratchPoolId = MyScratchPool's PoolId and
545 let it fill itself, and RecyclePoolId = XXX's PoolId so I can
546 see if it become stable and I just have to supervise
548 - If I want to remove this pool, I set RecyclePoolId = MyScratchPool's
549 PoolId, and when it is empty remove it.
550 - Figure out how to recycle Scratch volumes back to the Scratch Pool.
552 - Allow Check Labels to be used with Bacula labels.
553 - "Resuming" a failed backup (lost line for example) by using the
554 failed backup as a sort of "base" job.
556 - Email to the user when the tape is about to need changing x
557 days before it needs changing.
558 - Command to show next tape that will be used for a job even
559 if the job is not scheduled.
560 - From: Arunav Mandal <amandal@trolltech.com>
561 1. When jobs are running and bacula for some reason crashes or if I do a
562 restart it remembers and jobs it was running before it crashed or restarted
563 as of now I loose all jobs if I restart it.
565 2. When spooling and in the midway if client is disconnected for instance a
566 laptop bacula completely discard the spool. It will be nice if it can write
567 that spool to tape so there will be some backups for that client if not all.
569 3. We have around 150 clients machines it will be nice to have a option to
570 upgrade all the client machines bacula version automatically.
572 4. Atleast one connection should be reserved for the bconsole so at heavy load
573 I should connect to the director via bconsole which at sometimes I can't
575 5. Another most important feature that is missing, say at 10am I manually
576 started backup of client abc and it was a full backup since client abc has
577 no backup history and at 10.30am bacula again automatically started backup of
578 client abc as that was in the schedule. So now we have 2 multiple Full
579 backups of the same client and if we again try to start a full backup of
580 client backup abc bacula won't complain. That should be fixed.
582 - Fix bpipe.c so that it does not modify results pointer.
583 ***FIXME*** calling sequence should be changed.
584 - For Windows disaster recovery see http://unattended.sf.net/
585 - regardless of the retention period, Bacula will not prune the
586 last Full, Diff, or Inc File data until a month after the
587 retention period for the last Full backup that was done.
588 - update volume=xxx --- add status=Full
589 - Remove old spool files on startup.
590 - Exclude SD spool/working directory.
591 - Refuse to prune last valid Full backup. Same goes for Catalog.
593 - Make a callback when Rerun failed levels is called.
594 - Give Python program access to Scheduled jobs.
595 - Add setting Volume State via Python.
596 - Python script to save with Python, not save, save with Bacula.
597 - Python script to do backup.
599 - Change the Priority, Client, Storage, JobStatus (error)
600 at the start of a job.
601 - Why is SpoolDirectory = /home/bacula/spool; not reported
602 as an error when writing a DVD?
603 - Make bootstrap file handle multiple MediaTypes (SD)
604 - Remove all old Device resource code in Dir and code to pass it
605 back in SD -- better, rework it to pass back device statistics.
606 - Check locking of resources -- be sure to lock devices where previously
607 resources were locked.
608 - The last part is left in the spool dir.
611 - In restore don't compare byte count on a raw device -- directory
612 entry does not contain bytes.
616 MA = (last_MA * 3 + rate) / 4
617 rate = (bytes - last_bytes) / (runtime - last_runtime)
618 - Max Vols limit in Pool off by one?
619 - Implement Files/Bytes,... stats for restore job.
620 - Implement Total Bytes Written, ... for restore job.
621 - Despool attributes simultaneously with data in a separate
622 thread, rejoined at end of data spooling.
623 - 7. Implement new Console commands to allow offlining/reserving drives,
624 and possibly manipulating the autochanger (much asked for).
625 - Add start/end date editing in messages (%t %T, %e?) ...
626 - Add ClientDefs similar to JobDefs.
627 - Print more info when bextract -p accepts a bad block.
628 - Fix FD JobType to be set before RunBeforeJob in FD.
629 - Look at adding full Volume and Pool information to a Volume
630 label so that bscan can get *all* the info.
631 - If the user puts "Purge Oldest Volume = yes" or "Recycle Oldest Volume = yes"
632 and there is only one volume in the pool, refuse to do it -- otherwise
633 he fills the Volume, then immediately starts reusing it.
634 - Implement copies and stripes.
635 - Add history file to console.
636 - Each file on tape creates a JobMedia record. Peter has 4 million
637 files spread over 10000 tape files and four tapes. A restore takes
638 16 hours to build the restore list.
639 - Add and option to check if the file size changed during backup.
640 - Make sure SD deletes spool files on error exit.
641 - Delete old spool files when SD starts.
642 - When labeling tapes, if you enter 000026, Bacula uses
643 the tape index rather than the Volume name 000026.
644 - Add offline tape command to Bacula console.
646 Enter MediaId or Volume name: 32
647 Enter new Volume name: DLT-20Dec04
648 Automatically selected Pool: Default
649 Connecting to Storage daemon DLTDrive at 192.168.68.104:9103 ...
650 Sending relabel command from "DLT-28Jun03" to "DLT-20Dec04" ...
651 block.c:552 Write error at 0:0 on device /dev/nst0. ERR=Bad file descriptor.
652 Error writing final EOF to tape. This tape may not be readable.
653 dev.c:1207 ioctl MTWEOF error on /dev/nst0. ERR=Permission denied.
654 askdir.c:219 NULL Volume name. This shouldn't happen!!!
655 3912 Failed to label Volume: ERR=dev.c:1207 ioctl MTWEOF error on /dev/nst0. ERR=Permission denied.
656 Label command failed for Volume DLT-20Dec04.
657 Do not forget to mount the drive!!!
658 - Bug: if a job is manually scheduled to run later, it does not appear
659 in any status report and cannot be cancelled.
661 ==== Keeping track of deleted/new files ====
662 - To mark files as deleted, run essentially a Verify to disk, and
663 when a file is found missing (MarkId != JobId), then create
664 a new File record with FileIndex == -1. This could be done
665 by the FD at the same time as the backup.
667 My "trick" for keeping track of deletions is the following.
668 Assuming the user turns on this option, after all the files
669 have been backed up, but before the job has terminated, the
670 FD will make a pass through all the files and send their
671 names to the DIR (*exactly* the same as what a Verify job
672 currently does). This will probably be done at the same
673 time the files are being sent to the SD avoiding a second
674 pass. The DIR will then compare that to what is stored in
675 the catalog. Any files in the catalog but not in what the
676 FD sent will receive a catalog File entry that indicates
677 that at that point in time the file was deleted. This
678 either transmitted to the FD or simultaneously computed in
679 the FD, so that the FD can put a record on the tape that
680 indicates that the file has been deleted at this point.
681 A delete file entry could potentially be one with a FileIndex
682 of 0 or perhaps -1 (need to check if FileIndex is used for
683 some other thing as many of the Bacula fields are "overloaded"
686 During a restore, any file initially picked up by some
687 backup (Full, ...) then subsequently having a File entry
688 marked "delete" will be removed from the tree, so will not
689 be restored. If a file with the same name is later OK it
690 will be inserted in the tree -- this already happens. All
691 will be consistent except for possible changes during the
694 Since I'm on the subject, some of you may be wondering what
695 the utility of the in memory tree is if you are going to
696 restore everything (at least it comes up from time to time
697 on the list). Well, it is still *very* useful because it
698 allows only the last item found for a particular filename
699 (full path) to be entered into the tree, and thus if a file
700 is backed up 10 times, only the last copy will be restored.
701 I recently (last Friday) restored a complete directory, and
702 the Full and all the Differential and Incremental backups
703 spanned 3 Volumes. The first Volume was not even mounted
704 because all the files had been updated and hence backed up
705 since the Full backup was made. In this case, the tree
706 saved me a *lot* of time.
708 Make sure this information is stored on the tape too so
709 that it can be restored directly from the tape.
711 All the code (with the exception of formally generating and
712 saving the delete file entries) already exists in the Verify
713 Catalog command. It explicitly recognizes added/deleted files since
714 the last InitCatalog. It is more or less a "simple" matter of
715 taking that code and adapting it slightly to work for backups.
717 Comments from Martin Simmons (I think they are all covered):
718 Ok, that should cover the basics. There are few issues though:
720 - Restore will depend on the catalog. I think it is better to include the
721 extra data in the backup as well, so it can be seen by bscan and bextract.
723 - I'm not sure if it will preserve multiple hard links to the same inode. Or
724 maybe adding or removing links will cause the data to be dumped again?
726 - I'm not sure if it will handle renamed directories. Possibly it will work
727 by dumping the whole tree under a renamed directory?
729 - It remains to be seen how the backup performance of the DIR's will be
730 affected when comparing the catalog for a large filesystem.
734 How about introducing a Type = MgmtPolicy job type? That job type would
735 be responsible for scanning the Bacula environment looking for specific
736 conditions, and submitting the appropriate jobs for implementing said
740 Name = "Migration-Policy"
742 Policy Selection Job Type = Migrate
743 Scope = "<keyword> <operator> <regexp>"
744 Threshold = "<keyword> <operator> <regexp>"
745 Job Template = <template-name>
748 Where <keyword> is any legal job keyword, <operator> is a comparison
749 operator (=,<,>,!=, logical operators AND/OR/NOT) and <regexp> is a
750 appropriate regexp. I could see an argument for Scope and Threshold
751 being SQL queries if we want to support full flexibility. The
752 Migration-Policy job would then get scheduled as frequently as a site
753 felt necessary (suggested default: every 15 minutes).
758 Name = "Migration-Policy"
760 Policy Selection Job Type = Migration
762 Threshold = "Migration Selection Type = LowestUtil"
763 Job Template = "MigrationTemplate"
766 would select all pools for examination and generate a job based on
767 MigrationTemplate to automatically select the volume with the lowest
768 usage and migrate it's contents to the nextpool defined for that pool.
770 This policy abstraction would be really handy for adjusting the behavior
771 of Bacula according to site-selectable criteria (one thing that pops
772 into mind is Amanda's ability to automatically adjust backup levels
773 depending on various criteria).
779 - Add Pool/Storage override regression test.
780 - Add delete JobId to regression.
781 - Add a regression test for dbcheck.
782 - New test to add bscan to four-concurrent-jobs regression,
783 i.e. after the four-concurrent jobs zap the
784 database as is done in the bscan-test, then use bscan to
785 restore the database, do a restore and compare with the
787 - Add restore of specific JobId to regression (item 3
788 on the restore prompt)
789 - Add IPv6 to regression
790 - Add database test to regression. Test each function like delete,
793 - AntiVir can slow down backups on Win32 systems.
794 - Win32 systems with FAT32 can be much slower than NTFS for
795 more than 1000 files per directory.
799 - A HOLD command to stop all jobs from starting.
800 - A PAUSE command to pause all running jobs ==> release the
802 - Media Type = LTO,LTO-2,LTO-3
803 Media Type Read = LTO,LTO2,LTO3
804 Media Type Write = LTO2, LTO3
806 === From Carsten Menke <bootsy52@gmx.net>
808 Following is a list of what I think in the situations where I'm faced with,
809 could be a usefull enhancement to bacula, which I'm certain other users will
810 benefit from as well.
812 1. NextJob/NextJobs Directive within a Job Resource in the form of
813 NextJobs = job1,job2.
816 I currently solved the problem with running multiple jobs each after each
817 by setting the Max Wait Time for a job to 8 hours, and give
818 the jobs different Priorities. However, there scenarios where
819 1 Job is directly depending on another job, so if the former job fails,
820 the job after it needn't to be run
821 while maybe other jobs should run despite of that
824 A Backup Job and a Verify job, if the backup job fails there is no need to run
825 the verify job, as the backup job already failed. However, one may like
826 to backup the Catalog to disk despite of that the main backup job failed.
829 I see that this is related to the Event Handlers which are on the ToDo
830 list, also it is maybe a good idea to check for the return value and
831 execute different actions based on the return value
834 3. offline capability to bconsole
837 Currently I use a script which I execute within the last Job via the
838 RunAfterJob Directive, to release and eject the tape.
839 So I have to call bconsole "release=Storage-Name" and afterwards
840 mt -f /dev/nst0 eject to get the tape out.
842 If I have multiple Storage Devices, than these may not be /dev/nst0 and
843 I have to modify the script or call it with parameters etc.
844 This would actually not be needed, as everything is already defined
845 in bacula-sd.conf and if I can invoke bconsole with the
846 storage name via $1 in the script than I'm done and information is
849 4. %s for Storage Name added to the chars being substituted in "RunAfterJob"
853 For the reason mentioned in 3. to have the ability to call a
854 script with /scripts/foobar %s and in the script use $1
855 to pass the Storage Name to bconsole
857 5. Setting Volume State within a Job Resource
860 Instead of using "Maximum Volume Jobs" in the Pool Resource,
861 I would have the possibilty to define
862 in a Job Resource that after this certain job is run, the Volume State
863 should be set to "Volume State = Used", this give more flexibility (IMHO).
865 6. Localization of Bacula Messages
868 Unfortunatley many,many people I work with don't speak english very well.
869 So if at least the Reporting messages would be localized then they
870 would understand that they have to change the tape,etc. etc.
872 I volunteer to do the german translations, and if I can convince my wife also
873 french and Morre (western african language).
875 7. OK, this is evil, probably bound to security risks and maybe not possible
876 due to the design of bacula.
878 Implementation of Backtics ( `command` ) for shell comand execution to
879 the "Label Format" Directive.
883 Currently I have defined BACULA_DAY_OF_WEEK="day1|day2..." resulting in
884 Label Format = "HolyBackup-${BACULA_DAY_OF_WEEK[${WeekDay}]}". If I could
885 use backticks than I could use "Label Format = HolyBackup-`date +%A` to have
886 the localized name for the day of the week appended to the
887 format string. Then I have the tape labeled automatically with weekday
888 name in the correct language.
890 - Yes, that is surely the case. I probably should turn those into Warning
891 errors. In addition, you just made me think that it might not be bad to
892 add an option to check the file size after backing up the file and
893 report if it changes. This would be done as an option because it would
896 Kern, good idea. If you do do that, mention in the output: file
897 shrunk, or file expanded, just to make it obvious to the user
898 (without having to the refer to file size), just how the file size
901 Would this option be for all file, or just one file? Or a fileset?
902 - Make output from status use html table tags for nicely
903 presenting in a browser.
904 - Can one write tapes faster with 8192 byte block sizes?
905 - Document security problems with the same password for everyone in
906 rpm and Win32 releases.
907 - Browse generations of files.
908 - I've seen an error when my catalog's File table fills up. I
909 then have to recreate the File table with a larger maximum row
910 size. Relevant information is at
911 http://dev.mysql.com/doc/mysql/en/Full_table.html ; I think the
912 "Installing and Configuring MySQL" chapter should talk a bit
913 about this potential problem, and recommend a solution.
914 - For Solaris must use POSIX awk.
915 - Want speed of writing to tape while despooling.
916 - Supported autochanger:
924 Wangtek 6525ES (SCSI-1 QIC drive, 525MB), under Linux 2.4.something,
925 bacula 1.36.0/1 works with blocksize 16k INSIDE bacula-sd.conf.
926 - Add regex from http://www.pcre.org to Bacula for Win32.
927 - Use only shell tools no make in CDROM package.
928 - Include within include does it work?
929 - Implement a Pool of type Cleaning?
930 - Implement VolReadTime and VolWriteTime in SD
931 - Modify Backing up Your Database to include a bootstrap file.
932 - Think about making certain database errors fatal.
933 - Look at correcting the time jump in the scheduler for daylight
934 savings time changes.
935 - Add a "real" timer to network connections.
936 - Promote to Full = Time period
937 - Check dates entered by user for correctness (month/day/... ranges)
938 - Compress restore Volume listing by date and first file.
939 - Look at patches/bacula_db.b2z postgresql that loops during restore.
941 - Perhaps add read/write programs and/or plugins to FileSets.
942 - How to handle backing up portables ...
943 - Add some sort of guaranteed Interval for upgrading jobs.
944 - Can we write the state file after every job terminates? On Win32
945 the system crashes and the state file is not updated.
948 Documentation to do: (any release a little bit at a time)
949 - Doc to do unmount before removing magazine.
950 - Alternative to static linking "ldd prog" save all binaries listed,
951 restore them and point LD_LIBRARY_PATH to them.
952 - Document add "</dev/null >/dev/null 2>&1" to the bacula-fd command line
953 - Document query file format.
954 - Add more documentation for bsr files.
955 - Document problems with Verify and pruning.
956 - Document how to use multiple databases.
957 - VXA drives have a "cleaning required"
958 indicator, but Exabyte recommends preventive cleaning after every 75
961 In this context, it should be noted that Exabyte has a command-line
962 vxatool utility available for free download. (The current version is
963 vxatool-3.72.) It can get diagnostic info, read, write and erase tapes,
964 test the drive, unload tapes, change drive settings, flash new firmware,
966 Of particular interest in this context is that vxatool <device> -i will
967 report, among other details, the time since last cleaning in tape motion
968 minutes. This information can be retrieved (and settings changed, for
969 that matter) through the generic-SCSI device even when Bacula has the
970 regular tape device locked. (Needless to say, I don't recommend
971 changing tape settings while a job is running.)
972 - Lookup HP cleaning recommendations.
973 - Lookup HP tape replacement recommendations (see trouble shooting autochanger)
974 - Document doing table repair
977 ===================================
978 - Add macro expansions in JobDefs.
979 Run Before Job = "SomeFile %{Level} %{Client}"
980 Write Bootstrap="/some/dir/%{JobName}_%{Client}.bsr"
981 - Use non-blocking network I/O but if no data is available, use
983 - Use gather write() for network I/O.
984 - Autorestart on crash.
985 - Add bandwidth limiting.
986 - Add acks every once and a while from the SD to keep
987 the line from timing out.
988 - When an error in input occurs and conio beeps, you can back
989 up through the prompt.
990 - Detect fixed tape block mode during positioning by looking at
991 block numbers in btape "test". Possibly adjust in Bacula.
992 - Fix list volumes to output volume retention in some other
993 units, perhaps via a directive.
994 - Allow Simultaneous Priorities = yes => run up to Max concurrent jobs even
995 with multiple priorities.
996 - If you use restore replace=never, the directory attributes for
997 non-existent directories will not be restored properly.
999 - see lzma401.zip in others directory for new compression
1001 - Allow the user to select JobType for manual pruning/purging.
1002 - bscan does not put first of two volumes back with all info in
1004 - Implement the FreeBSD nodump flag in chflags.
1005 - Figure out how to make named console messages go only to that
1006 console and to the non-restricted console (new console class?).
1007 - Make restricted console prompt for password if *ask* is set or
1008 perhaps if password is undefined.
1009 - Implement "from ISO-date/time every x hours/days/weeks/months" in
1012 ==== from Marc Schoechlin
1013 - the help-command should be more verbose
1014 (it should explain the paramters of the different
1016 -> its time-comsuming to consult the manual anytime
1017 you need a special parameter
1018 -> maybe its more easy to maintain this, if the
1019 descriptions of that commands are outsourced to
1021 - the cd-command should allow complete paths
1022 i.e. cd /foo/bar/foo/bar
1023 -> if a customer mails me the path to a certain file,
1024 its faster to enter the specified directory
1025 - if the password is not configured in bconsole.conf
1026 you should be asked for it.
1027 -> sometimes you like to do restore on a customer-machine
1028 which shouldnt know the password for bacula.
1029 -> adding the password to the file favours admins
1030 to forget to remove the password after usage
1032 the protection of that file is less important
1033 - long-listed-output of commands should be scrollable
1034 like the unix more/less-command does
1035 -> if someone runs 200 and more machines, the lists could
1036 be a little long and complex
1037 - command-output should be shown column by column
1038 to reduce scrolling and to increase clarity
1040 - lsmark should list the selected files with full
1042 - wildcards for selecting and file and directories would be nice
1043 - any actions should be interuptable with STRG+C
1044 - command-expansion would be pretty cool
1046 - When the replace Never option is set, new directory permissions
1047 are not restored. See bug 213. To fix this requires creating a
1048 list of newly restored directories so that those directory
1049 permissions *can* be restored.
1050 - Add prune all command
1051 - Document fact that purge can destroy a part of a restore by purging
1052 one volume while others remain valid -- perhaps mark Jobs.
1053 - Add multiple-media-types.txt
1054 - look at mxt-changer.html
1055 - Make ? do a help command (no return needed).
1056 - Implement restore directory.
1057 - Document streams and how to implement them.
1058 - Try not to re-backup a file if a new hard link is added.
1059 - Add feature to backup hard links only, but not the data.
1060 - Fix stream handling to be simpler.
1061 - Add Priority and Bootstrap to Run a Job.
1062 - Eliminate Restore "Run Restore Job" prompt by allowing new "run command
1064 - Remove View FileSet button from Run a Job dialog.
1065 - Handle prompt for restore job at end of Restore command.
1066 - Add display of total selected files to Restore window.
1067 - Add tree pane to left of window.
1068 - Add progress meter.
1069 - Max wait time or max run time causes seg fault -- see runtime-bug.txt
1070 - Add message to user to check for fixed block size when the forward
1071 space test fails in btape.
1072 - When unmarking a directory check if all files below are unmarked and
1073 then remove the + flag -- in the restore tree.
1074 - Possibly implement: Action = Unmount Device="TapeDrive1" in Admin jobs.
1075 - Setup lrrd graphs: (http://www.linpro.no/projects/lrrd/) Mike Acar.
1076 - Revisit the question of multiple Volumes (disk) on a single device.
1077 - Add a block copy option to bcopy.
1078 - Finish work on Gnome restore GUI.
1079 - Fix "llist jobid=xx" where no fileset or client exists.
1080 - For each job type (Admin, Restore, ...) require only the really necessary
1081 fields.- Pass Director resource name as an option to the Console.
1082 - Add a "batch" mode to the Console (no unsolicited queries, ...).
1083 - Add a .list all files in the restore tree (probably also a list all files)
1084 Do both a long and short form.
1085 - Allow browsing the catalog to see all versions of a file (with
1086 stat data on each file).
1087 - Restore attributes of directory if replace=never set but directory
1089 - Use SHA1 on authentication if possible.
1090 - See comtest-xxx.zip for Windows code to talk to USB.
1091 - Add John's appended files:
1092 Appended = { /files/server/logs/http/*log }
1093 and such files would be treated as follows.On a FULL backup, they would
1094 be backed up like any other file.On an INCREMENTAL backup, where a
1095 previous INCREMENTAL or FULL was already in thecatalogue and the length
1096 of the file wasgreater than the length of the last backup, only thedata
1097 added since the last backup will be dumped.On an INCREMENTAL backup, if
1098 the length of the file is less than thelength of the file with the same
1099 name last backed up, the completefile is dumped.On Windows systems, with
1100 creation date of files, we can be evensmarter about this and not count
1101 entirely upon the length.On a restore, the full and all incrementals
1102 since it will beapplied in sequence to restore the file.
1103 - Check new HAVE_WIN32 open bits.
1104 - Check if the tape has moved before writing.
1105 - Handling removable disks -- see below:
1106 - Keep track of tape use time, and report when cleaning is necessary.
1107 - Add FromClient and ToClient keywords on restore command (or
1108 BackupClient RestoreClient).
1109 - Implement a JobSet, which groups any number of jobs. If the
1110 JobSet is started, all the jobs are started together.
1111 Allow Pool, Level, and Schedule overrides.
1112 - Enhance cancel to timeout BSOCK packets after a specific delay.
1113 - Do scheduling by UTC using gmtime_r() in run_conf, scheduler, and
1114 ua_status.!!! Thanks to Alan Brown for this tip.
1115 - Look at updating Volume Jobs so that Max Volume Jobs = 1 will work
1116 correctly for multiple simultaneous jobs.
1117 - Correct code so that FileSet MD5 is calculated for < and | filename
1119 - Implement the Media record flag that indicates that the Volume does disk
1121 - Implement VolAddr, which is used when Volume is addressed like a disk,
1122 and form it from VolFile and VolBlock.
1123 - Make multiple restore jobs for multiple media types specifying
1124 the proper storage type.
1125 - Fix fast block rejection (stored/read_record.c:118). It passes a null
1126 pointer (rec) to try_repositioning().
1127 - Look at extracting Win data from BackupRead.
1128 - Implement RestoreJobRetention? Maybe better "JobRetention" in a Job,
1129 which would take precidence over the Catalog "JobRetention".
1130 - Implement Label Format in Add and Label console commands.
1131 - Possibly up network buffers to 65K. Put on variable.
1132 - Put email tape request delays on one or more variables. User wants
1133 to cancel the job after a certain time interval. Maximum Mount Wait?
1134 - Job, Client, Device, Pool, or Volume?
1135 Is it possible to make this a directive which is *optional* in multiple
1136 resources, like Level? If so, I think I'd make it an optional directive
1137 in Job, Client, and Pool, with precedence such that Job overrides Client
1138 which in turn overrides Pool.
1140 - New Storage specifications:
1141 - Want to write to multiple storage devices simultaneously
1142 - Want to write to multiple storage devices sequentially (in one job)
1143 - Want to read/write simultaneously
1144 - Key is MediaType -- it must match
1146 Passed to SD as a sort of BSR record called Storage Specification
1150 MediaType -> Next MediaType
1152 Device -> Next Device
1154 Allow multiple Storage specifications
1162 Allow Multiple Pool specifications (note, Pool currently
1164 Allow Multiple MediaType specifications in Dir conf
1165 Allow Multiple Device specifications in Dir conf
1166 Perhaps keep this in a single SSR
1167 Tie a Volume to a specific device by using a MediaType that
1168 is contained in only one device.
1169 In SD allow Device to have Multiple MediaTypes
1171 - Ideas from Jerry Scharf:
1172 First let's point out some big pluses that bacula has for this
1174 more importantly it's active. Thank you so much for that
1175 even more important, it's not flaky
1176 it has an open access catalog, opening many possibilities
1177 it's pushing toward heterogeneous systems capability
1179 Macintosh file client
1180 macs are an interesting niche, but I fear a server is a rathole
1181 working bare iron recovery for windows
1182 the option for inc/diff backups not reset on fileset revision
1183 a) use both change and inode update time against base time
1184 b) do the full catalog check (expensive but accurate)
1185 sizing guide (how much system is needed to back up N systems/files)
1186 consultants on using bacula in building a disaster recovery system
1187 an integration guide
1188 or how to get at fancy things that one could do with bacula
1189 logwatch code for bacula logs (or similar)
1190 linux distro inclusion of bacula (brings good and bad, but necessary)
1191 win2k/XP server capability (icky but you asked)
1192 support for Oracle database ??
1194 - Look at adding SQL server and Exchange support for Windows.
1195 - Make dev->file and dev->block_num signed integers so that -1 can
1196 be an invalid value which happens with BSR.
1197 - Create VolAddr for disk files in place of VolFile and VolBlock. This
1198 is needed to properly specify ranges.
1199 - Add progress of files/bytes to SD and FD.
1200 - Print warning message if FileId > 4 billion
1201 - do a "messages" before the first prompt in Console
1202 - Client does not show busy during Estimate command.
1203 - Implement Console mtx commands.
1204 - Implement a Mount Command and an Unmount Command where
1205 the users could specify a system command to be performed
1206 to do the mount, after which Bacula could attempt to
1207 read the device. This is for Removeable media such as a CDROM.
1208 - Most likely, this mount command would be invoked explicitly
1209 by the user using the current Console "mount" and "unmount"
1210 commands -- the Storage Daemon would do the right thing
1211 depending on the exact nature of the device.
1212 - As with tape drives, when Bacula wanted a new removable
1213 disk mounted, it would unmount the old one, and send a message
1214 to the user, who would then use "mount" as described above
1215 once he had actually inserted the disk.
1216 - Implement dump/print label to UA
1217 - Spool to disk only when the tape is full, then when a tape is hung move
1219 - bextract is sending everything to the log file ****FIXME****
1220 - Allow multiple Storage specifications (or multiple names on
1221 a single Storage specification) in the Job record. Thus a job
1222 can be backed up to a number of storage devices.
1223 - Implement some way for the File daemon to contact the Director
1224 to start a job or pass its DHCP obtained IP number.
1225 - Implement a query tape prompt/replace feature for a console
1226 - Copy console @ code to gnome2-console
1227 - Make tree walk routines like cd, ls, ... more user friendly
1228 by handling spaces better.
1229 - Make sure that Bacula rechecks the tape after the 20 min wait.
1230 - Set IO_NOWAIT on Bacula TCP/IP packets.
1231 - Try doing a raw partition backup and restore by mounting a
1233 - From Lars Kellers:
1234 Yes, it would allow to highly automatic the request for new tapes. If a
1235 tape is empty, bacula reads the barcodes (native or simulated), and if
1236 an unused tape is found, it runs the label command with all the
1237 necessary parameters.
1239 By the way can bacula automatically "move" an empty/purged volume say
1240 in the "short" pool to the "long" pool if this pool runs out of volume
1242 - What to do about "list files job=xxx".
1243 - Look at how fuser works and /proc/PID/fd that is how Nic found the
1244 file descriptor leak in Bacula.
1245 - Implement WrapCounters in Counters.
1246 - Add heartbeat from FD to SD if hb interval expires.
1247 - Can we dynamically change FileSets?
1248 - If pool specified to label command and Label Format is specified,
1249 automatically generate the Volume name.
1250 - Why can't SQL do the filename sort for restore?
1251 - Add ExhautiveRestoreSearch
1252 - Look at the possibility of loading only the necessary
1253 data into the restore tree (i.e. do it one directory at a
1254 time as the user walks through the tree).
1255 - Possibly use the hash code if the user selects all for a restore command.
1256 - Fix "restore all" to bypass building the tree.
1257 - Prohibit backing up archive device (findlib/find_one.c:128)
1258 - Implement Release Device in the Job resource to unmount a drive.
1259 - Implement Acquire Device in the Job resource to mount a drive,
1260 be sure this works with admin jobs so that the user can get
1261 prompted to insert the correct tape. Possibly some way to say to
1262 run the job but don't save the files.
1263 - Make things like list where a file is saved case independent for
1265 - Use autochanger to handle multiple devices.
1266 - Implement a Recycle command
1267 - Start working on Base jobs.
1268 - Implement UnsavedFiles DB record.
1269 - From Phil Stracchino:
1270 It would probably be a per-client option, and would be called
1271 something like, say, "Automatically purge obsoleted jobs". What it
1272 would do is, when you successfully complete a Differential backup of a
1273 client, it would automatically purge all Incremental backups for that
1274 client that are rendered redundant by that Differential. Likewise,
1275 when a Full backup on a client completed, it would automatically purge
1276 all Differential and Incremental jobs obsoleted by that Full backup.
1277 This would let people minimize the number of tapes they're keeping on
1278 hand without having to master the art of retention times.
1279 - When doing a Backup send all attributes back to the Director, who
1280 would then figure out what files have been deleted.
1281 - Currently in mount.c:236 the SD simply creates a Volume. It should have
1282 explicit permission to do so. It should also mark the tape in error
1283 if there is an error.
1284 - Cancel waiting for Client connect in SD if FD goes away.
1286 - Implement timeout in response() when it should come quickly.
1287 - Implement a Slot priority (loaded/not loaded).
1288 - Implement "vacation" Incremental only saves.
1289 - Implement create "FileSet"?
1290 - Add prefixlinks to where or not where absolute links to FD.
1291 - Issue message to mount a new tape before the rewind.
1292 - Simplified client job initiation for portables.
1293 - If SD cannot open a drive, make it periodically retry.
1294 - Add more of the config info to the tape label.
1296 - Refine SD waiting output:
1297 Device is being positioned
1298 > Device is being positioned for append
1299 > Device is being positioned to file x
1301 - Figure out some way to estimate output size and to avoid splitting
1302 a backup across two Volumes -- this could be useful for writing CDROMs
1303 where you really prefer not to have it split -- not serious.
1304 - Have SD compute MD5 or SHA1 and compare to what FD computes.
1305 - Make VolumeToCatalog calculate an MD5 or SHA1 from the
1306 actual data on the Volume and compare it.
1307 - Implement Bacula plugins -- design API
1308 - Make bcopy read through bad tape records.
1309 - Program files (i.e. execute a program to read/write files).
1310 Pass read date of last backup, size of file last time.
1311 - Add Signature type to File DB record.
1312 - CD into subdirectory when open()ing files for backup to
1313 speed up things. Test with testfind().
1314 - Priority job to go to top of list.
1315 - Why are save/restore of device different sizes (sparse?) Yup! Fix it.
1316 - Implement some way for the Console to dynamically create a job.
1317 - Solaris -I on tar for include list
1318 - Need a verbose mode in restore, perhaps to bsr.
1319 - bscan without -v is too quiet -- perhaps show jobs.
1320 - Add code to reject whole blocks if not wanted on restore.
1321 - Check if we can increase Bacula FD priorty in Win2000
1322 - Make sure the MaxVolFiles is fully implemented in SD
1323 - Check if both CatalogFiles and UseCatalog are set to SD.
1324 - Possibly add email to Watchdog if drive is unmounted too
1325 long and a job is waiting on the drive.
1326 - After unmount, if restore job started, ask to mount.
1327 - Add UA rc and history files.
1328 - put termcap (used by console) in ./configure and
1329 allow -with-termcap-dir.
1330 - Fix Autoprune for Volumes to respect need for full save.
1331 - Compare tape to Client files (attributes, or attributes and data)
1332 - Make all database Ids 64 bit.
1333 - Allow console commands to detach or run in background.
1334 - Add SD message variables to control operator wait time
1335 - Maximum Operator Wait
1336 - Minimum Message Interval
1337 - Maximum Message Interval
1338 - Send Operator message when cannot read tape label.
1339 - Verify level=Volume (scan only), level=Data (compare of data to file).
1340 Verify level=Catalog, level=InitCatalog
1342 - Add keyword search to show command in Console.
1343 - Events : tape has more than xxx bytes.
1344 - Complete code in Bacula Resources -- this will permit
1345 reading a new config file at any time.
1346 - Handle ctl-c in Console
1347 - Implement script driven addition of File daemon to config files.
1348 - Think about how to make Bacula work better with File (non-tape) archives.
1349 - Write Unix emulator for Windows.
1350 - Put memory utilization in Status output of each daemon
1351 if full status requested or if some level of debug on.
1352 - Make database type selectable by .conf files i.e. at runtime
1353 - Set flag for uname -a. Add to Volume label.
1354 - Restore files modified after date
1355 - SET LD_RUN_PATH=$HOME/mysql/lib/mysql
1356 - Remove duplicate fields from jcr (e.g. jcr.level and jcr.jr.Level, ...).
1357 - Timout a job or terminate if link goes down, or reopen link and query.
1358 - Concept of precious tapes (cannot be reused).
1359 - Make bcopy copy with a single tape drive.
1360 - Permit changing ownership during restore.
1363 > My suggestion: Add a feature on the systray menu-icon menu to request
1364 > an immediate backup now. This would be useful for laptop users who may
1365 > not be on the network when the regular scheduled backup is run.
1367 > My wife's suggestion: Add a setting to the win32 client to allow it to
1368 > shut down the machine after backup is complete (after, of course,
1369 > displaying a "System will shut down in one minute, click here to cancel"
1370 > warning dialog). This would be useful for sites that want user
1371 > woorkstations to be shut down overnight to save power.
1374 - Autolabel should be specified by DIR instead of SD.
1376 - Add media capacity
1377 - AutoScan (check checksum of tape)
1378 - Format command = "format /dev/nst0"
1382 - Seek resolution (usually corresponds to buffer size)
1383 - EODErrorCode=ENOSPC or code
1384 - Partial Read error code
1385 - Partial write error code
1386 - Nonformatted read error
1387 - Nonformatted write error
1388 - WriteProtected error
1392 - IgnoreCloseErrors=yes
1402 - FD sends unsaved file list to Director at end of job (see
1404 - File daemon should build list of files skipped, and then
1405 at end of save retry and report any errors.
1406 - Write a Storage daemon that uses pipes and
1407 standard Unix programs to write to the tape.
1409 - Need something that monitors the JCR queue and
1410 times out jobs by asking the deamons where they are.
1411 - Enhance Jmsg code to permit buffering and saving to disk.
1412 - device driver = "xxxx" for drives.
1413 - Verify from Volume
1414 - Ensure that /dev/null works
1415 - Need report class for messages. Perhaps
1416 report resource where report=group of messages
1417 - enhance scan_attrib and rename scan_jobtype, and
1418 fill in code for "since" option
1419 - Director needs a time after which the report status is sent
1420 anyway -- or better yet, a retry time for the job.
1421 - Don't reschedule a job if previous incarnation is still running.
1422 - Some way to automatically backup everything is needed????
1423 - Need a structure for pending actions:
1425 - termination status (part of buffered msgs?)
1427 Read, Write, Clean, Delete
1428 - Login to Bacula; Bacula users with different permissions:
1429 owner, group, user, quotas
1430 - Store info on each file system type (probably in the job header on tape.
1431 This could be the output of df; or perhaps some sort of /etc/mtab record.
1433 ========= ideas ===============
1434 From: "Jerry K. Schieffer" <jerry@skylinetechnology.com>
1435 To: <kern@sibbald.com>
1436 Subject: RE: [Bacula-users] future large programming jobs
1437 Date: Thu, 26 Feb 2004 11:34:54 -0600
1439 I noticed the subject thread and thought I would offer the following
1440 merely as sources of ideas, i.e. something to think about, not even as
1441 strong as a request. In my former life (before retiring) I often
1442 dealt with backups and storage management issues/products as a
1443 developer and as a consultant. I am currently migrating my personal
1444 network from amanda to bacula specifically because of the ability to
1445 cross media boundaries during storing backups.
1446 Are you familiar with the commercial product called ADSM (I think IBM
1447 now sells it under the Tivoli label)? It has a couple of interesting
1448 ideas that may apply to the following topics.
1450 1. Migration: Consider that when you need to restore a system, there
1451 may be pressure to hurry. If all the information for a single client
1452 can eventually end up on the same media (and in chronological order),
1453 the restore is facillitated by not having to search past information
1454 from other clients. ADSM has the concept of "client affinity" that
1455 may be associated with it's storage pools. It seems to me that this
1456 concept (as an optional feature) might fit in your architecture for
1459 ADSM also has the concept of defining one or more storage pools as
1460 "copy pools" (almost mirrors, but only in the sense of contents).
1461 These pools provide the ability to have duplicte data stored both
1462 onsite and offsite. The copy process can be scheduled to be handled
1463 by their storage manager during periods when there is no backup
1464 activity. Again, the migration process might be a place to consider
1465 implementing something like this.
1468 > It strikes me that it would be very nice to be able to do things
1470 > have the Job(s) backing up the machines run, and once they have all
1471 > completed, start a migration job to copy the data from disks Volumes
1473 > a tape library and then to offsite storage. Maybe this can already
1475 > done with some careful scheduling and Job prioritzation; the events
1476 > mechanism described below would probably make it very easy.
1478 This is the goal. In the first step (before events), you simply
1480 the Migration to tape later.
1482 2. Base jobs: In ADSM, each copy of each stored file is tracked in
1483 the database. Once a file (unique by path and metadata such as dates,
1484 size, ownership, etc.) is in a copy pool, no more copies are made. In
1485 other words, when you start ADSM, it begins like your concept of a
1486 base job. After that it is in the "incremental" mode. You can
1487 configure the number of "generations" of files to be retained, plus a
1488 retention date after which even old generations are purged. The
1489 database tracks the contents of media and projects the percentage of
1490 each volume that is valid. When the valid content of a volume drops
1491 below a configured percentage, the valid data are migrated to another
1492 volume and the old volume is marked as empty. Note, this requires
1493 ADSM to have an idea of the contents of a client, i.e. marking the
1494 database when an existing file was deleted, but this would solve your
1495 issue of restoring a client without restoring deleted files.
1497 This is pretty far from what bacula now does, but if you are going to
1498 rip things up for Base jobs,.....
1499 Also, the benefits of this are huge for very large shops, especially
1500 with media robots, but are a pain for shops with manual media
1504 > Base jobs sound pretty useful, but I'm not dying for them.
1506 Nobody is dying for them, but when you see what it does, you will die
1509 3. Restoring deleted files: Since I think my comments in (2) above
1510 have low probability of implementation, I'll also suggest that you
1511 could approach the issue of deleted files by a mechanism of having the
1512 fd report to the dir, a list of all files on the client for every
1513 backup job. The dir could note in the database entry for each file
1514 the date that the file was seen. Then if a restore as of date X takes
1515 place, only files that exist from before X until after X would be
1516 restored. Probably the major cost here is the extra date container in
1517 each row of the files table.
1519 Thanks for "listening". I hope some of this helps. If you want to
1520 contact me, please send me an email - I read some but not all of the
1521 mailing list traffic and might miss a reply there.
1523 Please accept my compliments for bacula. It is doing a great job for
1524 me!! I sympathize with you in the need to wrestle with excelence in
1525 execution vs. excelence in feature inclusion.
1530 ==============================
1533 - Design at hierarchial storage for Bacula. Migration and Clone.
1534 - Implement FSM (File System Modules).
1535 - Audit M_ error codes to ensure they are correct and consistent.
1536 - Add variable break characters to lex analyzer.
1537 Either a bit mask or a string of chars so that
1538 the caller can change the break characters.
1539 - Make a single T_BREAK to replace T_COMMA, etc.
1540 - Ensure that File daemon and Storage daemon can
1541 continue a save if the Director goes down (this
1542 is NOT currently the case). Must detect socket error,
1543 buffer messages for later.
1544 - Enhance time/duration input to allow multiple qualifiers e.g. 3d2h
1545 - Add ability to backup to two Storage devices (two SD sessions) at
1546 the same time -- e.g. onsite, offsite.
1547 - Add the ability to consolidate old backup sets (basically do a restore
1548 to tape and appropriately update the catalog). Compress Volume sets.
1549 Might need to spool via file is only one drive is available.
1550 - Compress or consolidate Volumes of old possibly deleted files. Perhaps
1551 someway to do so with every volume that has less than x% valid
1555 Migration: Move a backup from one Volume to another
1556 Clone: Copy a backup -- two Volumes
1559 ======================================================
1561 It is somewhat like a Full save becomes an incremental since
1562 the Base job (or jobs) plus other non-base files.
1564 - A Base backup is same as Full backup, just different type.
1565 - New BaseFiles table that contains:
1567 BaseJobId - Base JobId referenced for this FileId (needed ???)
1568 JobId - JobId currently running
1569 FileId - File not backed up, exists in Base Job
1570 FileIndex - FileIndex from Base Job.
1571 i.e. for each base file that exists but is not saved because
1572 it has not changed, the File daemon sends the JobId, BaseId,
1573 FileId, FileIndex back to the Director who creates the DB entry.
1574 - To initiate a Base save, the Director sends the FD
1575 the FileId, and full filename for each file in the Base.
1576 - When the FD finds a Base file, he requests the Director to
1577 send him the full File entry (stat packet plus MD5), or
1578 conversely, the FD sends it to the Director and the Director
1579 says yes or no. This can be quite rapid if the FileId is kept
1580 by the FD for each Base Filename.
1581 - It is probably better to have the comparison done by the FD
1582 despite the fact that the File entry must be sent across the
1584 - An alternative would be to send the FD the whole File entry
1585 from the start. The disadvantage is that it requires a lot of
1586 space. The advantage is that it requires less communications
1588 - The Job record must be updated to indicate that one or more
1590 - At end of Job, FD returns:
1591 1. Count of base files/bytes not written to tape (i.e. matches)
1592 2. Count of base file that were saved i.e. had changed.
1593 - No tape record would be written for a Base file that matches, in the
1594 same way that no tape record is written for Incremental jobs where
1595 the file is not saved because it is unchanged.
1596 - On a restore, all the Base file records must explicitly be
1597 found from the BaseFile tape. I.e. for each Full save that is marked
1598 to have one or more Base Jobs, search the BaseFile for all occurrences
1600 - An optimization might be to make the BaseFile have:
1606 This would avoid the need to explicitly fetch each File record for
1607 the Base job. The Base Job record will be fetched to get the
1608 VolSessionId and VolSessionTime.
1609 =========================================================
1612 ==========================================================
1614 For each Incremental job that is run, there may be files that
1615 were found but not saved because they were locked (this applies
1616 only to Windows). Such a system could send back to the Director
1617 a list of Unsaved files.
1619 - New UnSavedFiles table that contains:
1623 - Then in the next Incremental job, the list of Unsaved Files will be
1624 feed to the FD, who will ensure that they are explicitly chosen even
1625 if standard date/time check would not have selected them.
1626 =============================================================
1630 Multiple drive autochanger data: see Alan Brown
1631 mtx -f xxx unloadStorage Element 1 is Already Full(drive 0 was empty)
1632 Unloading Data Transfer Element into Storage Element 1...source Element
1633 Address 480 is Empty
1635 (drive 0 was empty and so was slot 1)
1636 > mtx -f xxx load 15 0
1637 no response, just returns to the command prompt when complete.
1638 > mtx -f xxx status Storage Changer /dev/changer:2 Drives, 60 Slots ( 2 Import/Export )
1639 Data Transfer Element 0:Full (Storage Element 15 Loaded):VolumeTag = HX001
1640 Data Transfer Element 1:Empty
1641 Storage Element 1:Empty
1642 Storage Element 2:Full :VolumeTag=HX002
1643 Storage Element 3:Full :VolumeTag=HX003
1644 Storage Element 4:Full :VolumeTag=HX004
1645 Storage Element 5:Full :VolumeTag=HX005
1646 Storage Element 6:Full :VolumeTag=HX006
1647 Storage Element 7:Full :VolumeTag=HX007
1648 Storage Element 8:Full :VolumeTag=HX008
1649 Storage Element 9:Full :VolumeTag=HX009
1650 Storage Element 10:Full :VolumeTag=HX010
1651 Storage Element 11:Empty
1652 Storage Element 12:Empty
1653 Storage Element 13:Empty
1654 Storage Element 14:Empty
1655 Storage Element 15:Empty
1656 Storage Element 16:Empty....
1657 Storage Element 28:Empty
1658 Storage Element 29:Full :VolumeTag=CLNU01L1
1659 Storage Element 30:Empty....
1660 Storage Element 57:Empty
1661 Storage Element 58:Full :VolumeTag=NEX261L2
1662 Storage Element 59 IMPORT/EXPORT:Empty
1663 Storage Element 60 IMPORT/EXPORT:Empty
1665 Unloading Data Transfer Element into Storage Element 15...done
1667 (just to verify it remembers where it came from, however it can be
1668 overrriden with mtx unload {slotnumber} to go to any storage slot.)
1670 There needs to be a table of drive # to devices somewhere - If there are
1671 multiple changers or drives there may not be a 1:1 correspondance between
1672 changer drive number and system device name - and depending on the way the
1673 drives are hooked up to scsi busses, they may not be linearly numbered
1674 from an offset point either.something like
1676 Autochanger drives = 2
1677 Autochanger drive 0 = /dev/nst1
1678 Autochanger drive 1 = /dev/nst2
1679 IMHO, it would be _safest_ to use explicit mtx unload commands at all
1680 times, not just for multidrive changers. For a 1 drive changer, that's
1686 MTX's manpage (1.2.15):
1687 unload [<slotnum>] [ <drivenum> ]
1688 Unloads media from drive <drivenum> into slot
1689 <slotnum>. If <drivenum> is omitted, defaults to
1690 drive 0 (as do all commands). If <slotnum> is
1691 omitted, defaults to the slot that the drive was
1692 loaded from. Note that there's currently no way
1693 to say 'unload drive 1's media to the slot it
1694 came from', other than to explicitly use that
1695 slot number as the destination.AB
1701 undef# camcontrol devlist
1702 <WANGTEK 51000 SCSI M74H 12B3> at scbus0 target 2 lun 0 (pass0,sa0)
1703 <ARCHIVE 4586XX 28887-XXX 4BGD> at scbus0 target 4 lun 0 (pass1,sa1)
1704 <ARCHIVE 4586XX 28887-XXX 4BGD> at scbus0 target 4 lun 1 (pass2)
1706 tapeinfo -f /dev/sg0 with a bad tape in drive 1:
1707 [kern@rufus mtx-1.2.17kes]$ ./tapeinfo -f /dev/sg0
1708 Product Type: Tape Drive
1710 Product ID: 'C5713A '
1712 Attached Changer: No
1713 TapeAlert[3]: Hard Error: Uncorrectable read/write error.
1714 TapeAlert[20]: Clean Now: The tape drive neads cleaning NOW.
1721 Medium Type: Not Loaded
1724 DataCompEnabled: yes
1725 DataCompCapable: yes
1726 DataDeCompEnabled: yes
1733 Handling removable disks
1735 From: Karl Cunningham <karlc@keckec.com>
1737 My backups are only to hard disk these days, in removable bays. This is my
1738 idea of how a backup to hard disk would work more smoothly. Some of these
1739 things Bacula does already, but I mention them for completeness. If others
1740 have better ways to do this, I'd like to hear about it.
1742 1. Accommodate several disks, rotated similar to how tapes are. Identified
1743 by partition volume ID or perhaps by the name of a subdirectory.
1744 2. Abort & notify the admin if the wrong disk is in the bay.
1745 3. Write backups to different subdirectories for each machine to be backed
1747 4. Volumes (files) get created as needed in the proper subdirectory, one
1749 5. When a disk is recycled, remove or zero all old backup files. This is
1750 important as the disk being recycled may be close to full. This may be
1751 better done manually since the backup files for many machines may be
1752 scattered in many subdirectories.
1757 - Why the heck doesn't bacula drop root priviledges before connecting to
1759 - Look at using posix_fadvise(2) for backups -- see bug #751.
1760 Possibly add the code at findlib/bfile.c:795
1761 /* TCP socket options */
1762 #define TCP_KEEPIDLE 4 /* Start keeplives after this period */
1763 - Fix bnet_connect() code to set a timer and to use time to
1765 - Implement 4th argument to make_catalog_backup that passes hostname.
1766 - Test FIFO backup/restore -- make regression
1767 - Please mount volume "xxx" on Storage device ... should also list
1768 Pool and MediaType in case user needs to create a new volume.
1769 - On restore add Restore Client, Original Client.
1770 01-Apr 00:42 rufus-dir: Start Backup JobId 55, Job=kernsave.2007-04-01_00.42.48
1771 01-Apr 00:42 rufus-sd: Python SD JobStart: JobId=55 Client=Rufus
1772 01-Apr 00:42 rufus-dir: Created new Volume "Full0001" in catalog.
1773 01-Apr 00:42 rufus-dir: Using Device "File"
1774 01-Apr 00:42 rufus-sd: kernsave.2007-04-01_00.42.48 Warning: Device "File" (/tmp) not configured to autolabel Volumes.
1775 01-Apr 00:42 rufus-sd: kernsave.2007-04-01_00.42.48 Warning: Device "File" (/tmp) not configured to autolabel Volumes.
1776 01-Apr 00:42 rufus-sd: Please mount Volume "Full0001" on Storage Device "File" (/tmp) for Job kernsave.2007-04-01_00.42.48
1777 01-Apr 00:44 rufus-sd: Wrote label to prelabeled Volume "Full0001" on device "File" (/tmp)
1778 - Check if gnome-console works with TLS.
1779 - the director seg faulted when I omitted the pool directive from a
1780 job resource. I was experimenting and thought it redundant that I had
1781 specified Pool, Full Backup Pool. and Differential Backup Pool. but
1782 apparently not. This happened when I removed the pool directive and
1783 started the director.
1784 - Add Where: client:/.... to restore job report.
1785 - Ensure that moving a purged Volume in ua_purge.c to the RecyclePool
1786 does the right thing.
1787 - FD-SD quick disconnect
1788 - Building the in memory restore tree is slow.