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.
69 - Implement Diff,Inc Retention Periods
72 - Encrypt sd_auth_key = s with director's key = d
73 k[i] = s[i] + (d[i] & 0xF)) & 0xFF + 'A' skip -
74 Decrypt key = k with director's key
75 x = k[i] - (d[i] & 0xF))
77 x = k[i] - (d[i] & 0xF) + 16
80 - Eliminate: /var is a different filesystem. Will not descend from / into /var
81 - Separate Files and Directories in catalog
82 - Create FileVersions table
83 - Look at rsysnc for incremental updates and dedupping
84 - Add MD5 or SHA1 check in SD for data validation
85 - modify pruning to keep a fixed number of versions of a file,
87 - finish implementation of fdcalled -- see ua_run.c:105
88 - Fix problem in postgresql.c in my_postgresql_query, where the
89 generation of the error message doesn't differentiate result==NULL
90 and a bad status from that result. Not only that, the result is
91 cleared on a bail_out without having generated the error message.
92 - Erabt if min_block_size > max_block_size
94 - Implement wait on multiple objects
98 - Implement SDErrors (must return from SD)
99 - Implement USB keyboard support in rescue CD.
100 - Implement continue spooling while despooling.
101 - Remove all install temp files in Win32 PLUGINSDIR.
102 - Audit retention periods to make sure everything is 64 bit.
103 - Use E'xxx' to escape PostgreSQL strings.
104 - No where in restore causes kaboom.
105 - Performance: multiple spool files for a single job.
106 - Performance: despool attributes when despooling data (problem
107 multiplexing Dir connection).
108 - Make restore use the in-use volume reservation algorithm.
109 - Look at mincore: http://insights.oetiker.ch/linux/fadvise.html
110 - Unicode input http://en.wikipedia.org/wiki/Byte_Order_Mark
111 - Add TLS to bat (should be done).
112 - When Pool specifies Storage command override does not work.
113 - Implement wait_for_sysop() message display in wait_for_device(), which
114 now prints warnings too often.
115 - Ensure that each device in an Autochanger has a different
117 - Add Catalog = to Pool resource so that pools will exist
118 in only one catalog -- currently Pools are "global".
119 - Look at sg_logs -a /dev/sg0 for getting soft errors.
120 - btape "test" command with Offline on Unmount = yes
122 This test is essential to Bacula.
124 I'm going to write one record in file 0,
125 two records in file 1,
126 and three records in file 2
128 02-Feb 11:00 btape: ABORTING due to ERROR in dev.c:715
129 dev.c:714 Bad call to rewind. Device "LTO" (/dev/nst0) not open
130 02-Feb 11:00 btape: Fatal Error because: Bacula interrupted by signal 11: Segmentation violation
131 Kaboom! btape, btape got signal 11. Attempting traceback.
133 - Encryption -- email from Landon
134 > The backup encryption algorithm is currently not configurable, and is
135 > set to AES_128_CBC in src/filed/backup.c. The encryption code
136 > supports a number of different ciphers (as well as adding arbitrary
137 > new ones) -- only a small bit of code would be required to map a
138 > configuration string value to a CRYPTO_CIPHER_* value, if anyone is
139 > interested in implementing this functionality.
141 - Why doesn't @"xxx abc" work in a conf file?
142 - Figure out some way to "automatically" backup conf changes.
143 - Add the OS version back to the Win32 client info.
144 - Restarted jobs have a NULL in the from field.
145 - Modify SD status command to indicate when the SD is writing
146 to a DVD (the device is not open -- see bug #732).
147 - Look at the possibility of adding "SET NAMES UTF8" for MySQL,
148 and possibly changing the blobs into varchar.
149 - Ensure that the SD re-reads the Media record if the JobFiles
150 does not match -- it may have been updated by another job.
151 - Look at moving the Storage directive from the Job to the
152 Pool in the default conf files.
154 - Test Volume compatibility between machine architectures
155 - Encryption documentation
156 - Wrong jobbytes with query 12 (todo)
157 - Bare-metal recovery Windows (todo)
162 - Access Mode = Read-Only, Read-Write, Unavailable, Destroyed, Offsite
164 - Maximum number of scratch volumes
166 - Next Pool (already have)
167 - Reclamation threshold
169 - Reuse delay (after all files purged from volume before it can be used)
170 - Copy Pool = xx, yyy (or multiple lines).
172 - Allow pool selection during restore.
174 - Average tape size from Eric
175 SELECT COALESCE(media_avg_size.volavg,0) * count(Media.MediaId) AS volmax, GROUP BY Media.MediaType, Media.PoolId, media_avg_size.volavg
176 count(Media.MediaId) AS volnum,
177 sum(Media.VolBytes) AS voltotal,
178 Media.PoolId AS PoolId,
179 Media.MediaType AS MediaType
181 LEFT JOIN (SELECT avg(Media.VolBytes) AS volavg,
182 Media.MediaType AS MediaType
184 WHERE Media.VolStatus = 'Full'
185 GROUP BY Media.MediaType
186 ) AS media_avg_size ON (Media.MediaType = media_avg_size.MediaType)
187 GROUP BY Media.MediaType, Media.PoolId, media_avg_size.volavg
191 - Add doc for bweb -- especially Installation
193 http://www.orangecrate.com/modules.php?name=News&file=article&sid=501
195 - Despool attributes in separate thread
198 - Check why restore repeatedly sends Rechdrs between
199 each data chunk -- according to James Harper 9Jan07.
202 - Full at least once a month, ...
203 - Cancel Inc if Diff/Full running
204 - More intelligent re-run
205 - New/deleted file backup
207 - Incremental backup -- rsync, Stow
211 - Try to fix bscan not working with multiple DVD volumes bug #912.
212 - Look at mondo/mindi
213 - Don't restore Solaris Door files:
214 #define S_IFDOOR in st_mode.
215 see: http://docs.sun.com/app/docs/doc/816-5173/6mbb8ae23?a=view#indexterm-360
216 - Make Bacula by default not backup tmpfs, procfs, sysfs, ...
217 - Fix hardlinked immutable files when linking a second file, the
218 immutable flag must be removed prior to trying to link it.
219 - Implement Python event for backing up/restoring a file.
220 - Change dbcheck to tell users to use native tools for fixing
221 broken databases, and to ensure they have the proper indexes.
222 - add udev rules for Bacula devices.
223 - If a job terminates, the DIR connection can close before the
224 Volume info is updated, leaving the File count wrong.
225 - Look at why SIGPIPE during connection can cause seg fault in
226 writing the daemon message, when Dir dropped to bacula:bacula
227 - Look at zlib 32 => 64 problems.
228 - Possibly turn on St. Bernard code.
229 - Fix bextract to restore ACLs, or better yet, use common routines.
230 - Do we migrate appendable Volumes?
231 - Remove queue.c code.
232 - Print warning message if LANG environment variable does not specify
234 - New dot commands from Arno.
235 .show device=xxx lists information from one storage device, including
236 devices (I'm not even sure that information exists in the DIR...)
237 .move eject device=xxx mostly the same as 'unmount xxx' but perhaps with
238 better machine-readable output like "Ok" or "Error busy"
239 .move eject device=xxx toslot=yyy the same as above, but with a new
240 target slot. The catalog should be updated accordingly.
241 .move transfer device=xxx fromslot=yyy toslot=zzz
244 - Article: http://www.heise.de/open/news/meldung/83231
245 - Article: http://www.golem.de/0701/49756.html
246 - Article: http://lwn.net/Articles/209809/
247 - Article: http://www.onlamp.com/pub/a/onlamp/2004/01/09/bacula.html
248 - Article: http://www.linuxdevcenter.com/pub/a/linux/2005/04/07/bacula.html
249 - Article: http://www.osreviews.net/reviews/admin/bacula
250 - Article: http://www.debianhelp.co.uk/baculaweb.htm
252 - Wikis mentioning Bacula
253 http://wiki.finkproject.org/index.php/Admin:Backups
254 http://wiki.linuxquestions.org/wiki/Bacula
255 http://www.openpkg.org/product/packages/?package=bacula
256 http://www.iterating.com/products/Bacula
257 http://net-snmp.sourceforge.net/wiki/index.php/Net-snmp_extensions
258 http://www.section6.net/wiki/index.php/Using_Bacula_for_Tape_Backups
259 http://bacula.darwinports.com/
260 http://wiki.mandriva.com/en/Releases/Corporate/Server_4/Notes#Bacula
261 http://en.wikipedia.org/wiki/Bacula
264 http://www.devco.net/pubwiki/Bacula/
265 http://paramount.ind.wpi.edu/wiki/doku.php
266 http://gentoo-wiki.com/HOWTO_Backup
267 http://www.georglutz.de/wiki/Bacula
268 http://www.clarkconnect.com/wiki/index.php?title=Modules_-_LAN_Backup/Recovery
269 http://linuxwiki.de/Bacula (in German)
271 - Possibly allow SD to spool even if a tape is not mounted.
272 - It appears to me that you have run into some sort of race
273 condition where two threads want to use the same Volume and they
274 were both given access. Normally that is no problem. However,
275 one thread wanted the particular Volume in drive 0, but it was
276 loaded into drive 1 so it decided to unload it from drive 1 and
277 then loaded it into drive 0, while the second thread went on
278 thinking that the Volume could be used in drive 1 not realizing
279 that in between time, it was loaded in drive 0.
280 I'll look at the code to see if there is some way we can avoid
281 this kind of problem. Probably the best solution is to make the
282 first thread simply start using the Volume in drive 1 rather than
283 transferring it to drive 0.
284 - Fix re-read of last block to check if job has actually written
285 a block, and check if block was written by a different job
286 (i.e. multiple simultaneous jobs writing).
287 - Figure out how to configure query.sql. Suggestion to use m4:
288 == changequote.m4 ===
289 changequote(`[',`]')dnl
290 ==== query.sql.in ===
291 :List next 20 volumes to expire
293 Pool.Name AS PoolName,
298 [ FROM_UNIXTIME(UNIX_TIMESTAMP(Media.LastWritten) Media.VolRetention) AS Expire, ])dnl
300 [ media.lastwritten + interval '1 second' * media.volretention as expire, ])dnl
304 ON Media.PoolId=Pool.PoolId
305 WHERE Media.LastWritten>0
309 Command: m4 -DmySQL changequote.m4 query.sql.in >query.sql
311 The problem is that it requires m4, which is not present on all machines
313 - Given all the problems with FIFOs, I think the solution is to do something a
314 little different, though I will look at the code and see if there is not some
315 simple solution (i.e. some bug that was introduced). What might be a better
316 solution would be to use a FIFO as a sort of "key" to tell Bacula to read and
317 write data to a program rather than the FIFO. For example, suppose you
322 Then, I could imagine if you backup and restore this file with a direct
323 reference as is currently done for fifos, instead, during backup Bacula will
326 /home/kern/my-fifo.backup
328 and read the data that my-fifo.backup writes to stdout. For restore, Bacula
331 /home/kern/my-fifo.restore
333 and send the data backed up to stdout. These programs can either be an
334 executable or a shell script and they need only read/write to stdin/stdout.
336 I think this would give a lot of flexibility to the user without making any
337 significant changes to Bacula.
342 select FilenameId from Filename where Name='';
343 # Get list of all directories referenced in a Backup.
344 select Path.Path from Path,File where File.JobId=nnn and
345 File.FilenameId=(FilenameId-from-above) and File.PathId=Path.PathId
346 order by Path.Path ASC;
348 - Look into using Dart for testing
349 http://public.kitware.com/Dart/HTML/Index.shtml
351 - Look into replacing autotools with cmake
352 http://www.cmake.org/HTML/Index.html
354 === Migration from David ===
355 What I'd like to see:
358 Name = "<poolname>-migrate"
362 Migration Selection Type = LowestUtil | OldestVol | PoolOccupancy |
363 Client | PoolResidence | Volume | JobName | SQLquery
364 Migration Selection Pattern = "regexp"
365 Next Pool = <override>
368 There should be no need for a Level (migration is always Full, since you
369 don't calculate differential/incremental differences for migration),
370 Storage should be determined by the volume types in the pool, and Client
371 is really a selection issue. Migration should always occur to the
372 NextPool defined in the pool definition. If no nextpool is defined, the
373 job should end with a reason of "no place to go". If Next Pool statement
374 is present, we override the check in the pool definition and use the
377 Here's how I'd define Migration Selection Types:
380 Client -- Migrate data from selected client only. Migration Selection
381 Pattern regexp provides pattern to select client names, eg ^FS00* makes
382 all client names starting with FS00 eligible for migration.
384 Jobname -- Migration all jobs matching name. Migration Selection Pattern
385 regexp provides pattern to select jobnames existing in pool.
387 Volume -- Migrate all data on specified volumes. Migration Selection
388 Pattern regexp provides selection criteria for volumes to be migrated.
389 Volumes must exist in pool to be eligible for migration.
393 LowestUtil -- Identify the volume in the pool with the least data on it
394 and empty it. No Migration Selection Pattern required.
396 OldestVol -- Identify the LRU volume with data written, and empty it. No
397 Migration Selection Pattern required.
399 PoolOccupancy -- if pool occupancy exceeds <highmig>, migrate volumes
400 (starting with most full volumes) until pool occupancy drops below
401 <lowmig>. Pool highmig and lowmig values are in pool definition, no
402 Migration Selection Pattern required.
406 SQLQuery -- Migrate all jobuids returned by the supplied SQL query.
407 Migration Selection Pattern contains SQL query to execute; should return
408 a list of 1 or more jobuids to migrate.
410 PoolResidence -- Migrate data sitting in pool for longer than
411 PoolResidence value in pool definition. Migration Selection Pattern
412 optional; if specified, override value in pool definition (value in
416 [ possibly a Python event -- kes ]
418 - Mount on an Autochanger with no tape in the drive causes:
419 Automatically selected Storage: LTO-changer
420 Enter autochanger drive[0]: 0
421 3301 Issuing autochanger "loaded drive 0" command.
422 3302 Autochanger "loaded drive 0", result: nothing loaded.
423 3301 Issuing autochanger "loaded drive 0" command.
424 3302 Autochanger "loaded drive 0", result: nothing loaded.
425 3902 Cannot mount Volume on Storage Device "LTO-Drive1" (/dev/nst0) because:
426 Couldn't rewind device "LTO-Drive1" (/dev/nst0): ERR=dev.c:678 Rewind error on "LTO-Drive1" (/dev/nst0). ERR=No medium found.
427 3905 Device "LTO-Drive1" (/dev/nst0) open but no Bacula volume is mounted.
428 If this is not a blank tape, try unmounting and remounting the Volume.
429 - If Drive 0 is blocked, and drive 1 is set "Autoselect=no", drive 1 will
431 - Autochanger did not change volumes.
432 select * from Storage;
433 +-----------+-------------+-------------+
434 | StorageId | Name | AutoChanger |
435 +-----------+-------------+-------------+
436 | 1 | LTO-changer | 0 |
437 +-----------+-------------+-------------+
438 05-May 03:50 roxie-sd: 3302 Autochanger "loaded drive 0", result is Slot 11.
439 05-May 03:50 roxie-sd: Tibs.2006-05-05_03.05.02 Warning: Director wanted Volume "LT
440 Current Volume "LT0-002" not acceptable because:
441 1997 Volume "LT0-002" not in catalog.
442 05-May 03:50 roxie-sd: Tibs.2006-05-05_03.05.02 Error: Autochanger Volume "LT0-002"
443 Setting InChanger to zero in catalog.
444 05-May 03:50 roxie-dir: Tibs.2006-05-05_03.05.02 Error: Unable to get Media record
446 05-May 03:50 roxie-sd: Tibs.2006-05-05_03.05.02 Fatal error: Error getting Volume i
447 05-May 03:50 roxie-sd: Tibs.2006-05-05_03.05.02 Fatal error: Job 530 canceled.
448 05-May 03:50 roxie-sd: Tibs.2006-05-05_03.05.02 Fatal error: spool.c:249 Fatal appe
449 05-May 03:49 Tibs: Tibs.2006-05-05_03.05.02 Fatal error: c:\cygwin\home\kern\bacula
458 FirstWritten: 2006-05-05 03:11:54
459 LastWritten: 2006-05-05 03:50:23
460 LabelDate: 2005-12-26 16:52:40
471 VolRetention: 31,536,000
483 Note VolStatus is blank!!!!!
490 FirstWritten: 0000-00-00 00:00:00
491 LastWritten: 0000-00-00 00:00:00
492 LabelDate: 2005-12-26 16:52:40
503 VolRetention: 31,536,000
516 Automatically selected Storage: LTO-changer
517 Enter autochanger drive[0]: 0
518 3301 Issuing autochanger "loaded drive 0" command.
519 3302 Autochanger "loaded drive 0", result: nothing loaded.
520 3301 Issuing autochanger "loaded drive 0" command.
521 3302 Autochanger "loaded drive 0", result: nothing loaded.
522 3902 Cannot mount Volume on Storage Device "LTO-Drive1" (/dev/nst0) because:
523 Couldn't rewind device "LTO-Drive1" (/dev/nst0): ERR=dev.c:678 Rewind error on "LTO-Drive1" (/dev/nst0). ERR=No medium found.
525 3905 Device "LTO-Drive1" (/dev/nst0) open but no Bacula volume is mounted.
526 If this is not a blank tape, try unmounting and remounting the Volume.
528 - http://www.dwheeler.com/essays/commercial-floss.html
529 - Add VolumeLock to prevent all but lock holder (SD) from updating
530 the Volume data (with the exception of VolumeState).
531 - The btape fill command does not seem to use the Autochanger
532 - Make Windows installer default to system disk drive.
533 - Look at using ioctl(FIOBMAP, ...) on Linux, and
534 DeviceIoControl(..., FSCTL_QUERY_ALLOCATED_RANGES, ...) on
535 Win32 for sparse files.
536 http://www.flexhex.com/docs/articles/sparse-files.phtml
537 http://www.informatik.uni-frankfurt.de/~loizides/reiserfs/fibmap.html
538 - Directive: at <event> "command"
539 - Command: pycmd "command" generates "command" event. How to
540 attach to a specific job?
541 - Integrate Christopher's St. Bernard code.
542 - run_cmd() returns int should return JobId_t
543 - get_next_jobid_from_list() returns int should return JobId_t
544 - Document export LDFLAGS=-L/usr/lib64
545 - Don't attempt to restore from "Disabled" Volumes.
546 - Network error on Win32 should set Win32 error code.
547 - What happens when you rename a Disk Volume?
548 - Job retention period in a Pool (and hence Volume). The job would
550 - Look at -D_FORTIFY_SOURCE=2
551 - Add Win32 FileSet definition somewhere
552 - Look at fixing restore status stats in SD.
553 - Look at using ioctl(FIMAP) and FIGETBSZ for sparse files.
554 http://www.informatik.uni-frankfurt.de/~loizides/reiserfs/fibmap.html
555 - Implement a mode that says when a hard read error is
556 encountered, read many times (as it currently does), and if the
557 block cannot be read, skip to the next block, and try again. If
558 that fails, skip to the next file and try again, ...
560 create table LevelType (LevelType binary(1), LevelTypeLong tinyblob);
561 insert into LevelType (LevelType,LevelTypeLong) values
565 - Show files/second in client status output.
566 - Add a recursive mark command (rmark) to restore.
567 - "Minimum Job Interval = nnn" sets minimum interval between Jobs
568 of the same level and does not permit multiple simultaneous
569 running of that Job (i.e. lets any previous invocation finish
570 before doing Interval testing).
571 - Look at simplifying File exclusions.
572 - New directive "Delete purged Volumes"
573 - new pool XXX with ScratchPoolId = MyScratchPool's PoolId and
574 let it fill itself, and RecyclePoolId = XXX's PoolId so I can
575 see if it become stable and I just have to supervise
577 - If I want to remove this pool, I set RecyclePoolId = MyScratchPool's
578 PoolId, and when it is empty remove it.
579 - Figure out how to recycle Scratch volumes back to the Scratch Pool.
581 - Allow Check Labels to be used with Bacula labels.
582 - "Resuming" a failed backup (lost line for example) by using the
583 failed backup as a sort of "base" job.
585 - Email to the user when the tape is about to need changing x
586 days before it needs changing.
587 - Command to show next tape that will be used for a job even
588 if the job is not scheduled.
589 - From: Arunav Mandal <amandal@trolltech.com>
590 1. When jobs are running and bacula for some reason crashes or if I do a
591 restart it remembers and jobs it was running before it crashed or restarted
592 as of now I loose all jobs if I restart it.
594 2. When spooling and in the midway if client is disconnected for instance a
595 laptop bacula completely discard the spool. It will be nice if it can write
596 that spool to tape so there will be some backups for that client if not all.
598 3. We have around 150 clients machines it will be nice to have a option to
599 upgrade all the client machines bacula version automatically.
601 4. Atleast one connection should be reserved for the bconsole so at heavy load
602 I should connect to the director via bconsole which at sometimes I can't
604 5. Another most important feature that is missing, say at 10am I manually
605 started backup of client abc and it was a full backup since client abc has
606 no backup history and at 10.30am bacula again automatically started backup of
607 client abc as that was in the schedule. So now we have 2 multiple Full
608 backups of the same client and if we again try to start a full backup of
609 client backup abc bacula won't complain. That should be fixed.
611 - Fix bpipe.c so that it does not modify results pointer.
612 ***FIXME*** calling sequence should be changed.
613 - For Windows disaster recovery see http://unattended.sf.net/
614 - regardless of the retention period, Bacula will not prune the
615 last Full, Diff, or Inc File data until a month after the
616 retention period for the last Full backup that was done.
617 - update volume=xxx --- add status=Full
618 - Remove old spool files on startup.
619 - Exclude SD spool/working directory.
620 - Refuse to prune last valid Full backup. Same goes for Catalog.
622 - Make a callback when Rerun failed levels is called.
623 - Give Python program access to Scheduled jobs.
624 - Add setting Volume State via Python.
625 - Python script to save with Python, not save, save with Bacula.
626 - Python script to do backup.
628 - Change the Priority, Client, Storage, JobStatus (error)
629 at the start of a job.
630 - Why is SpoolDirectory = /home/bacula/spool; not reported
631 as an error when writing a DVD?
632 - Make bootstrap file handle multiple MediaTypes (SD)
633 - Remove all old Device resource code in Dir and code to pass it
634 back in SD -- better, rework it to pass back device statistics.
635 - Check locking of resources -- be sure to lock devices where previously
636 resources were locked.
637 - The last part is left in the spool dir.
640 - In restore don't compare byte count on a raw device -- directory
641 entry does not contain bytes.
645 MA = (last_MA * 3 + rate) / 4
646 rate = (bytes - last_bytes) / (runtime - last_runtime)
647 - Max Vols limit in Pool off by one?
648 - Implement Files/Bytes,... stats for restore job.
649 - Implement Total Bytes Written, ... for restore job.
650 - Despool attributes simultaneously with data in a separate
651 thread, rejoined at end of data spooling.
652 - 7. Implement new Console commands to allow offlining/reserving drives,
653 and possibly manipulating the autochanger (much asked for).
654 - Add start/end date editing in messages (%t %T, %e?) ...
655 - Add ClientDefs similar to JobDefs.
656 - Print more info when bextract -p accepts a bad block.
657 - Fix FD JobType to be set before RunBeforeJob in FD.
658 - Look at adding full Volume and Pool information to a Volume
659 label so that bscan can get *all* the info.
660 - If the user puts "Purge Oldest Volume = yes" or "Recycle Oldest Volume = yes"
661 and there is only one volume in the pool, refuse to do it -- otherwise
662 he fills the Volume, then immediately starts reusing it.
663 - Implement copies and stripes.
664 - Add history file to console.
665 - Each file on tape creates a JobMedia record. Peter has 4 million
666 files spread over 10000 tape files and four tapes. A restore takes
667 16 hours to build the restore list.
668 - Add and option to check if the file size changed during backup.
669 - Make sure SD deletes spool files on error exit.
670 - Delete old spool files when SD starts.
671 - When labeling tapes, if you enter 000026, Bacula uses
672 the tape index rather than the Volume name 000026.
673 - Add offline tape command to Bacula console.
675 Enter MediaId or Volume name: 32
676 Enter new Volume name: DLT-20Dec04
677 Automatically selected Pool: Default
678 Connecting to Storage daemon DLTDrive at 192.168.68.104:9103 ...
679 Sending relabel command from "DLT-28Jun03" to "DLT-20Dec04" ...
680 block.c:552 Write error at 0:0 on device /dev/nst0. ERR=Bad file descriptor.
681 Error writing final EOF to tape. This tape may not be readable.
682 dev.c:1207 ioctl MTWEOF error on /dev/nst0. ERR=Permission denied.
683 askdir.c:219 NULL Volume name. This shouldn't happen!!!
684 3912 Failed to label Volume: ERR=dev.c:1207 ioctl MTWEOF error on /dev/nst0. ERR=Permission denied.
685 Label command failed for Volume DLT-20Dec04.
686 Do not forget to mount the drive!!!
687 - Bug: if a job is manually scheduled to run later, it does not appear
688 in any status report and cannot be cancelled.
690 ==== Keeping track of deleted/new files ====
691 - To mark files as deleted, run essentially a Verify to disk, and
692 when a file is found missing (MarkId != JobId), then create
693 a new File record with FileIndex == -1. This could be done
694 by the FD at the same time as the backup.
696 My "trick" for keeping track of deletions is the following.
697 Assuming the user turns on this option, after all the files
698 have been backed up, but before the job has terminated, the
699 FD will make a pass through all the files and send their
700 names to the DIR (*exactly* the same as what a Verify job
701 currently does). This will probably be done at the same
702 time the files are being sent to the SD avoiding a second
703 pass. The DIR will then compare that to what is stored in
704 the catalog. Any files in the catalog but not in what the
705 FD sent will receive a catalog File entry that indicates
706 that at that point in time the file was deleted. This
707 either transmitted to the FD or simultaneously computed in
708 the FD, so that the FD can put a record on the tape that
709 indicates that the file has been deleted at this point.
710 A delete file entry could potentially be one with a FileIndex
711 of 0 or perhaps -1 (need to check if FileIndex is used for
712 some other thing as many of the Bacula fields are "overloaded"
715 During a restore, any file initially picked up by some
716 backup (Full, ...) then subsequently having a File entry
717 marked "delete" will be removed from the tree, so will not
718 be restored. If a file with the same name is later OK it
719 will be inserted in the tree -- this already happens. All
720 will be consistent except for possible changes during the
723 Since I'm on the subject, some of you may be wondering what
724 the utility of the in memory tree is if you are going to
725 restore everything (at least it comes up from time to time
726 on the list). Well, it is still *very* useful because it
727 allows only the last item found for a particular filename
728 (full path) to be entered into the tree, and thus if a file
729 is backed up 10 times, only the last copy will be restored.
730 I recently (last Friday) restored a complete directory, and
731 the Full and all the Differential and Incremental backups
732 spanned 3 Volumes. The first Volume was not even mounted
733 because all the files had been updated and hence backed up
734 since the Full backup was made. In this case, the tree
735 saved me a *lot* of time.
737 Make sure this information is stored on the tape too so
738 that it can be restored directly from the tape.
740 All the code (with the exception of formally generating and
741 saving the delete file entries) already exists in the Verify
742 Catalog command. It explicitly recognizes added/deleted files since
743 the last InitCatalog. It is more or less a "simple" matter of
744 taking that code and adapting it slightly to work for backups.
746 Comments from Martin Simmons (I think they are all covered):
747 Ok, that should cover the basics. There are few issues though:
749 - Restore will depend on the catalog. I think it is better to include the
750 extra data in the backup as well, so it can be seen by bscan and bextract.
752 - I'm not sure if it will preserve multiple hard links to the same inode. Or
753 maybe adding or removing links will cause the data to be dumped again?
755 - I'm not sure if it will handle renamed directories. Possibly it will work
756 by dumping the whole tree under a renamed directory?
758 - It remains to be seen how the backup performance of the DIR's will be
759 affected when comparing the catalog for a large filesystem.
763 How about introducing a Type = MgmtPolicy job type? That job type would
764 be responsible for scanning the Bacula environment looking for specific
765 conditions, and submitting the appropriate jobs for implementing said
769 Name = "Migration-Policy"
771 Policy Selection Job Type = Migrate
772 Scope = "<keyword> <operator> <regexp>"
773 Threshold = "<keyword> <operator> <regexp>"
774 Job Template = <template-name>
777 Where <keyword> is any legal job keyword, <operator> is a comparison
778 operator (=,<,>,!=, logical operators AND/OR/NOT) and <regexp> is a
779 appropriate regexp. I could see an argument for Scope and Threshold
780 being SQL queries if we want to support full flexibility. The
781 Migration-Policy job would then get scheduled as frequently as a site
782 felt necessary (suggested default: every 15 minutes).
787 Name = "Migration-Policy"
789 Policy Selection Job Type = Migration
791 Threshold = "Migration Selection Type = LowestUtil"
792 Job Template = "MigrationTemplate"
795 would select all pools for examination and generate a job based on
796 MigrationTemplate to automatically select the volume with the lowest
797 usage and migrate it's contents to the nextpool defined for that pool.
799 This policy abstraction would be really handy for adjusting the behavior
800 of Bacula according to site-selectable criteria (one thing that pops
801 into mind is Amanda's ability to automatically adjust backup levels
802 depending on various criteria).
808 - Add Pool/Storage override regression test.
809 - Add delete JobId to regression.
810 - Add a regression test for dbcheck.
811 - New test to add bscan to four-concurrent-jobs regression,
812 i.e. after the four-concurrent jobs zap the
813 database as is done in the bscan-test, then use bscan to
814 restore the database, do a restore and compare with the
816 - Add restore of specific JobId to regression (item 3
817 on the restore prompt)
818 - Add IPv6 to regression
819 - Add database test to regression. Test each function like delete,
822 - AntiVir can slow down backups on Win32 systems.
823 - Win32 systems with FAT32 can be much slower than NTFS for
824 more than 1000 files per directory.
828 - A HOLD command to stop all jobs from starting.
829 - A PAUSE command to pause all running jobs ==> release the
831 - Media Type = LTO,LTO-2,LTO-3
832 Media Type Read = LTO,LTO2,LTO3
833 Media Type Write = LTO2, LTO3
835 === From Carsten Menke <bootsy52@gmx.net>
837 Following is a list of what I think in the situations where I'm faced with,
838 could be a usefull enhancement to bacula, which I'm certain other users will
839 benefit from as well.
841 1. NextJob/NextJobs Directive within a Job Resource in the form of
842 NextJobs = job1,job2.
845 I currently solved the problem with running multiple jobs each after each
846 by setting the Max Wait Time for a job to 8 hours, and give
847 the jobs different Priorities. However, there scenarios where
848 1 Job is directly depending on another job, so if the former job fails,
849 the job after it needn't to be run
850 while maybe other jobs should run despite of that
853 A Backup Job and a Verify job, if the backup job fails there is no need to run
854 the verify job, as the backup job already failed. However, one may like
855 to backup the Catalog to disk despite of that the main backup job failed.
858 I see that this is related to the Event Handlers which are on the ToDo
859 list, also it is maybe a good idea to check for the return value and
860 execute different actions based on the return value
863 3. offline capability to bconsole
866 Currently I use a script which I execute within the last Job via the
867 RunAfterJob Directive, to release and eject the tape.
868 So I have to call bconsole "release=Storage-Name" and afterwards
869 mt -f /dev/nst0 eject to get the tape out.
871 If I have multiple Storage Devices, than these may not be /dev/nst0 and
872 I have to modify the script or call it with parameters etc.
873 This would actually not be needed, as everything is already defined
874 in bacula-sd.conf and if I can invoke bconsole with the
875 storage name via $1 in the script than I'm done and information is
878 4. %s for Storage Name added to the chars being substituted in "RunAfterJob"
882 For the reason mentioned in 3. to have the ability to call a
883 script with /scripts/foobar %s and in the script use $1
884 to pass the Storage Name to bconsole
886 5. Setting Volume State within a Job Resource
889 Instead of using "Maximum Volume Jobs" in the Pool Resource,
890 I would have the possibilty to define
891 in a Job Resource that after this certain job is run, the Volume State
892 should be set to "Volume State = Used", this give more flexibility (IMHO).
894 6. Localization of Bacula Messages
897 Unfortunatley many,many people I work with don't speak english very well.
898 So if at least the Reporting messages would be localized then they
899 would understand that they have to change the tape,etc. etc.
901 I volunteer to do the german translations, and if I can convince my wife also
902 french and Morre (western african language).
904 7. OK, this is evil, probably bound to security risks and maybe not possible
905 due to the design of bacula.
907 Implementation of Backtics ( `command` ) for shell comand execution to
908 the "Label Format" Directive.
912 Currently I have defined BACULA_DAY_OF_WEEK="day1|day2..." resulting in
913 Label Format = "HolyBackup-${BACULA_DAY_OF_WEEK[${WeekDay}]}". If I could
914 use backticks than I could use "Label Format = HolyBackup-`date +%A` to have
915 the localized name for the day of the week appended to the
916 format string. Then I have the tape labeled automatically with weekday
917 name in the correct language.
919 - Yes, that is surely the case. I probably should turn those into Warning
920 errors. In addition, you just made me think that it might not be bad to
921 add an option to check the file size after backing up the file and
922 report if it changes. This would be done as an option because it would
925 Kern, good idea. If you do do that, mention in the output: file
926 shrunk, or file expanded, just to make it obvious to the user
927 (without having to the refer to file size), just how the file size
930 Would this option be for all file, or just one file? Or a fileset?
931 - Make output from status use html table tags for nicely
932 presenting in a browser.
933 - Can one write tapes faster with 8192 byte block sizes?
934 - Document security problems with the same password for everyone in
935 rpm and Win32 releases.
936 - Browse generations of files.
937 - I've seen an error when my catalog's File table fills up. I
938 then have to recreate the File table with a larger maximum row
939 size. Relevant information is at
940 http://dev.mysql.com/doc/mysql/en/Full_table.html ; I think the
941 "Installing and Configuring MySQL" chapter should talk a bit
942 about this potential problem, and recommend a solution.
943 - For Solaris must use POSIX awk.
944 - Want speed of writing to tape while despooling.
945 - Supported autochanger:
953 Wangtek 6525ES (SCSI-1 QIC drive, 525MB), under Linux 2.4.something,
954 bacula 1.36.0/1 works with blocksize 16k INSIDE bacula-sd.conf.
955 - Add regex from http://www.pcre.org to Bacula for Win32.
956 - Use only shell tools no make in CDROM package.
957 - Include within include does it work?
958 - Implement a Pool of type Cleaning?
959 - Implement VolReadTime and VolWriteTime in SD
960 - Modify Backing up Your Database to include a bootstrap file.
961 - Think about making certain database errors fatal.
962 - Look at correcting the time jump in the scheduler for daylight
963 savings time changes.
964 - Add a "real" timer to network connections.
965 - Promote to Full = Time period
966 - Check dates entered by user for correctness (month/day/... ranges)
967 - Compress restore Volume listing by date and first file.
968 - Look at patches/bacula_db.b2z postgresql that loops during restore.
970 - Perhaps add read/write programs and/or plugins to FileSets.
971 - How to handle backing up portables ...
972 - Add some sort of guaranteed Interval for upgrading jobs.
973 - Can we write the state file after every job terminates? On Win32
974 the system crashes and the state file is not updated.
977 Documentation to do: (any release a little bit at a time)
978 - Doc to do unmount before removing magazine.
979 - Alternative to static linking "ldd prog" save all binaries listed,
980 restore them and point LD_LIBRARY_PATH to them.
981 - Document add "</dev/null >/dev/null 2>&1" to the bacula-fd command line
982 - Document query file format.
983 - Add more documentation for bsr files.
984 - Document problems with Verify and pruning.
985 - Document how to use multiple databases.
986 - VXA drives have a "cleaning required"
987 indicator, but Exabyte recommends preventive cleaning after every 75
990 In this context, it should be noted that Exabyte has a command-line
991 vxatool utility available for free download. (The current version is
992 vxatool-3.72.) It can get diagnostic info, read, write and erase tapes,
993 test the drive, unload tapes, change drive settings, flash new firmware,
995 Of particular interest in this context is that vxatool <device> -i will
996 report, among other details, the time since last cleaning in tape motion
997 minutes. This information can be retrieved (and settings changed, for
998 that matter) through the generic-SCSI device even when Bacula has the
999 regular tape device locked. (Needless to say, I don't recommend
1000 changing tape settings while a job is running.)
1001 - Lookup HP cleaning recommendations.
1002 - Lookup HP tape replacement recommendations (see trouble shooting autochanger)
1003 - Document doing table repair
1006 ===================================
1007 - Add macro expansions in JobDefs.
1008 Run Before Job = "SomeFile %{Level} %{Client}"
1009 Write Bootstrap="/some/dir/%{JobName}_%{Client}.bsr"
1010 - Use non-blocking network I/O but if no data is available, use
1012 - Use gather write() for network I/O.
1013 - Autorestart on crash.
1014 - Add bandwidth limiting.
1015 - Add acks every once and a while from the SD to keep
1016 the line from timing out.
1017 - When an error in input occurs and conio beeps, you can back
1018 up through the prompt.
1019 - Detect fixed tape block mode during positioning by looking at
1020 block numbers in btape "test". Possibly adjust in Bacula.
1021 - Fix list volumes to output volume retention in some other
1022 units, perhaps via a directive.
1023 - Allow Simultaneous Priorities = yes => run up to Max concurrent jobs even
1024 with multiple priorities.
1025 - If you use restore replace=never, the directory attributes for
1026 non-existent directories will not be restored properly.
1028 - see lzma401.zip in others directory for new compression
1030 - Allow the user to select JobType for manual pruning/purging.
1031 - bscan does not put first of two volumes back with all info in
1033 - Implement the FreeBSD nodump flag in chflags.
1034 - Figure out how to make named console messages go only to that
1035 console and to the non-restricted console (new console class?).
1036 - Make restricted console prompt for password if *ask* is set or
1037 perhaps if password is undefined.
1038 - Implement "from ISO-date/time every x hours/days/weeks/months" in
1041 ==== from Marc Schoechlin
1042 - the help-command should be more verbose
1043 (it should explain the paramters of the different
1045 -> its time-comsuming to consult the manual anytime
1046 you need a special parameter
1047 -> maybe its more easy to maintain this, if the
1048 descriptions of that commands are outsourced to
1050 - the cd-command should allow complete paths
1051 i.e. cd /foo/bar/foo/bar
1052 -> if a customer mails me the path to a certain file,
1053 its faster to enter the specified directory
1054 - if the password is not configured in bconsole.conf
1055 you should be asked for it.
1056 -> sometimes you like to do restore on a customer-machine
1057 which shouldnt know the password for bacula.
1058 -> adding the password to the file favours admins
1059 to forget to remove the password after usage
1061 the protection of that file is less important
1062 - long-listed-output of commands should be scrollable
1063 like the unix more/less-command does
1064 -> if someone runs 200 and more machines, the lists could
1065 be a little long and complex
1066 - command-output should be shown column by column
1067 to reduce scrolling and to increase clarity
1069 - lsmark should list the selected files with full
1071 - wildcards for selecting and file and directories would be nice
1072 - any actions should be interuptable with STRG+C
1073 - command-expansion would be pretty cool
1075 - When the replace Never option is set, new directory permissions
1076 are not restored. See bug 213. To fix this requires creating a
1077 list of newly restored directories so that those directory
1078 permissions *can* be restored.
1079 - Add prune all command
1080 - Document fact that purge can destroy a part of a restore by purging
1081 one volume while others remain valid -- perhaps mark Jobs.
1082 - Add multiple-media-types.txt
1083 - look at mxt-changer.html
1084 - Make ? do a help command (no return needed).
1085 - Implement restore directory.
1086 - Document streams and how to implement them.
1087 - Try not to re-backup a file if a new hard link is added.
1088 - Add feature to backup hard links only, but not the data.
1089 - Fix stream handling to be simpler.
1090 - Add Priority and Bootstrap to Run a Job.
1091 - Eliminate Restore "Run Restore Job" prompt by allowing new "run command
1093 - Remove View FileSet button from Run a Job dialog.
1094 - Handle prompt for restore job at end of Restore command.
1095 - Add display of total selected files to Restore window.
1096 - Add tree pane to left of window.
1097 - Add progress meter.
1098 - Max wait time or max run time causes seg fault -- see runtime-bug.txt
1099 - Add message to user to check for fixed block size when the forward
1100 space test fails in btape.
1101 - When unmarking a directory check if all files below are unmarked and
1102 then remove the + flag -- in the restore tree.
1103 - Possibly implement: Action = Unmount Device="TapeDrive1" in Admin jobs.
1104 - Setup lrrd graphs: (http://www.linpro.no/projects/lrrd/) Mike Acar.
1105 - Revisit the question of multiple Volumes (disk) on a single device.
1106 - Add a block copy option to bcopy.
1107 - Finish work on Gnome restore GUI.
1108 - Fix "llist jobid=xx" where no fileset or client exists.
1109 - For each job type (Admin, Restore, ...) require only the really necessary
1110 fields.- Pass Director resource name as an option to the Console.
1111 - Add a "batch" mode to the Console (no unsolicited queries, ...).
1112 - Add a .list all files in the restore tree (probably also a list all files)
1113 Do both a long and short form.
1114 - Allow browsing the catalog to see all versions of a file (with
1115 stat data on each file).
1116 - Restore attributes of directory if replace=never set but directory
1118 - Use SHA1 on authentication if possible.
1119 - See comtest-xxx.zip for Windows code to talk to USB.
1120 - Add John's appended files:
1121 Appended = { /files/server/logs/http/*log }
1122 and such files would be treated as follows.On a FULL backup, they would
1123 be backed up like any other file.On an INCREMENTAL backup, where a
1124 previous INCREMENTAL or FULL was already in thecatalogue and the length
1125 of the file wasgreater than the length of the last backup, only thedata
1126 added since the last backup will be dumped.On an INCREMENTAL backup, if
1127 the length of the file is less than thelength of the file with the same
1128 name last backed up, the completefile is dumped.On Windows systems, with
1129 creation date of files, we can be evensmarter about this and not count
1130 entirely upon the length.On a restore, the full and all incrementals
1131 since it will beapplied in sequence to restore the file.
1132 - Check new HAVE_WIN32 open bits.
1133 - Check if the tape has moved before writing.
1134 - Handling removable disks -- see below:
1135 - Keep track of tape use time, and report when cleaning is necessary.
1136 - Add FromClient and ToClient keywords on restore command (or
1137 BackupClient RestoreClient).
1138 - Implement a JobSet, which groups any number of jobs. If the
1139 JobSet is started, all the jobs are started together.
1140 Allow Pool, Level, and Schedule overrides.
1141 - Enhance cancel to timeout BSOCK packets after a specific delay.
1142 - Do scheduling by UTC using gmtime_r() in run_conf, scheduler, and
1143 ua_status.!!! Thanks to Alan Brown for this tip.
1144 - Look at updating Volume Jobs so that Max Volume Jobs = 1 will work
1145 correctly for multiple simultaneous jobs.
1146 - Correct code so that FileSet MD5 is calculated for < and | filename
1148 - Implement the Media record flag that indicates that the Volume does disk
1150 - Implement VolAddr, which is used when Volume is addressed like a disk,
1151 and form it from VolFile and VolBlock.
1152 - Make multiple restore jobs for multiple media types specifying
1153 the proper storage type.
1154 - Fix fast block rejection (stored/read_record.c:118). It passes a null
1155 pointer (rec) to try_repositioning().
1156 - Look at extracting Win data from BackupRead.
1157 - Implement RestoreJobRetention? Maybe better "JobRetention" in a Job,
1158 which would take precidence over the Catalog "JobRetention".
1159 - Implement Label Format in Add and Label console commands.
1160 - Possibly up network buffers to 65K. Put on variable.
1161 - Put email tape request delays on one or more variables. User wants
1162 to cancel the job after a certain time interval. Maximum Mount Wait?
1163 - Job, Client, Device, Pool, or Volume?
1164 Is it possible to make this a directive which is *optional* in multiple
1165 resources, like Level? If so, I think I'd make it an optional directive
1166 in Job, Client, and Pool, with precedence such that Job overrides Client
1167 which in turn overrides Pool.
1169 - New Storage specifications:
1170 - Want to write to multiple storage devices simultaneously
1171 - Want to write to multiple storage devices sequentially (in one job)
1172 - Want to read/write simultaneously
1173 - Key is MediaType -- it must match
1175 Passed to SD as a sort of BSR record called Storage Specification
1179 MediaType -> Next MediaType
1181 Device -> Next Device
1183 Allow multiple Storage specifications
1191 Allow Multiple Pool specifications (note, Pool currently
1193 Allow Multiple MediaType specifications in Dir conf
1194 Allow Multiple Device specifications in Dir conf
1195 Perhaps keep this in a single SSR
1196 Tie a Volume to a specific device by using a MediaType that
1197 is contained in only one device.
1198 In SD allow Device to have Multiple MediaTypes
1200 - Ideas from Jerry Scharf:
1201 First let's point out some big pluses that bacula has for this
1203 more importantly it's active. Thank you so much for that
1204 even more important, it's not flaky
1205 it has an open access catalog, opening many possibilities
1206 it's pushing toward heterogeneous systems capability
1208 Macintosh file client
1209 macs are an interesting niche, but I fear a server is a rathole
1210 working bare iron recovery for windows
1211 the option for inc/diff backups not reset on fileset revision
1212 a) use both change and inode update time against base time
1213 b) do the full catalog check (expensive but accurate)
1214 sizing guide (how much system is needed to back up N systems/files)
1215 consultants on using bacula in building a disaster recovery system
1216 an integration guide
1217 or how to get at fancy things that one could do with bacula
1218 logwatch code for bacula logs (or similar)
1219 linux distro inclusion of bacula (brings good and bad, but necessary)
1220 win2k/XP server capability (icky but you asked)
1221 support for Oracle database ??
1223 - Look at adding SQL server and Exchange support for Windows.
1224 - Make dev->file and dev->block_num signed integers so that -1 can
1225 be an invalid value which happens with BSR.
1226 - Create VolAddr for disk files in place of VolFile and VolBlock. This
1227 is needed to properly specify ranges.
1228 - Add progress of files/bytes to SD and FD.
1229 - Print warning message if FileId > 4 billion
1230 - do a "messages" before the first prompt in Console
1231 - Client does not show busy during Estimate command.
1232 - Implement Console mtx commands.
1233 - Implement a Mount Command and an Unmount Command where
1234 the users could specify a system command to be performed
1235 to do the mount, after which Bacula could attempt to
1236 read the device. This is for Removeable media such as a CDROM.
1237 - Most likely, this mount command would be invoked explicitly
1238 by the user using the current Console "mount" and "unmount"
1239 commands -- the Storage Daemon would do the right thing
1240 depending on the exact nature of the device.
1241 - As with tape drives, when Bacula wanted a new removable
1242 disk mounted, it would unmount the old one, and send a message
1243 to the user, who would then use "mount" as described above
1244 once he had actually inserted the disk.
1245 - Implement dump/print label to UA
1246 - Spool to disk only when the tape is full, then when a tape is hung move
1248 - bextract is sending everything to the log file ****FIXME****
1249 - Allow multiple Storage specifications (or multiple names on
1250 a single Storage specification) in the Job record. Thus a job
1251 can be backed up to a number of storage devices.
1252 - Implement some way for the File daemon to contact the Director
1253 to start a job or pass its DHCP obtained IP number.
1254 - Implement a query tape prompt/replace feature for a console
1255 - Copy console @ code to gnome2-console
1256 - Make tree walk routines like cd, ls, ... more user friendly
1257 by handling spaces better.
1258 - Make sure that Bacula rechecks the tape after the 20 min wait.
1259 - Set IO_NOWAIT on Bacula TCP/IP packets.
1260 - Try doing a raw partition backup and restore by mounting a
1262 - From Lars Kellers:
1263 Yes, it would allow to highly automatic the request for new tapes. If a
1264 tape is empty, bacula reads the barcodes (native or simulated), and if
1265 an unused tape is found, it runs the label command with all the
1266 necessary parameters.
1268 By the way can bacula automatically "move" an empty/purged volume say
1269 in the "short" pool to the "long" pool if this pool runs out of volume
1271 - What to do about "list files job=xxx".
1272 - Look at how fuser works and /proc/PID/fd that is how Nic found the
1273 file descriptor leak in Bacula.
1274 - Implement WrapCounters in Counters.
1275 - Add heartbeat from FD to SD if hb interval expires.
1276 - Can we dynamically change FileSets?
1277 - If pool specified to label command and Label Format is specified,
1278 automatically generate the Volume name.
1279 - Why can't SQL do the filename sort for restore?
1280 - Add ExhautiveRestoreSearch
1281 - Look at the possibility of loading only the necessary
1282 data into the restore tree (i.e. do it one directory at a
1283 time as the user walks through the tree).
1284 - Possibly use the hash code if the user selects all for a restore command.
1285 - Fix "restore all" to bypass building the tree.
1286 - Prohibit backing up archive device (findlib/find_one.c:128)
1287 - Implement Release Device in the Job resource to unmount a drive.
1288 - Implement Acquire Device in the Job resource to mount a drive,
1289 be sure this works with admin jobs so that the user can get
1290 prompted to insert the correct tape. Possibly some way to say to
1291 run the job but don't save the files.
1292 - Make things like list where a file is saved case independent for
1294 - Use autochanger to handle multiple devices.
1295 - Implement a Recycle command
1296 - Start working on Base jobs.
1297 - Implement UnsavedFiles DB record.
1298 - From Phil Stracchino:
1299 It would probably be a per-client option, and would be called
1300 something like, say, "Automatically purge obsoleted jobs". What it
1301 would do is, when you successfully complete a Differential backup of a
1302 client, it would automatically purge all Incremental backups for that
1303 client that are rendered redundant by that Differential. Likewise,
1304 when a Full backup on a client completed, it would automatically purge
1305 all Differential and Incremental jobs obsoleted by that Full backup.
1306 This would let people minimize the number of tapes they're keeping on
1307 hand without having to master the art of retention times.
1308 - When doing a Backup send all attributes back to the Director, who
1309 would then figure out what files have been deleted.
1310 - Currently in mount.c:236 the SD simply creates a Volume. It should have
1311 explicit permission to do so. It should also mark the tape in error
1312 if there is an error.
1313 - Cancel waiting for Client connect in SD if FD goes away.
1315 - Implement timeout in response() when it should come quickly.
1316 - Implement a Slot priority (loaded/not loaded).
1317 - Implement "vacation" Incremental only saves.
1318 - Implement create "FileSet"?
1319 - Add prefixlinks to where or not where absolute links to FD.
1320 - Issue message to mount a new tape before the rewind.
1321 - Simplified client job initiation for portables.
1322 - If SD cannot open a drive, make it periodically retry.
1323 - Add more of the config info to the tape label.
1325 - Refine SD waiting output:
1326 Device is being positioned
1327 > Device is being positioned for append
1328 > Device is being positioned to file x
1330 - Figure out some way to estimate output size and to avoid splitting
1331 a backup across two Volumes -- this could be useful for writing CDROMs
1332 where you really prefer not to have it split -- not serious.
1333 - Have SD compute MD5 or SHA1 and compare to what FD computes.
1334 - Make VolumeToCatalog calculate an MD5 or SHA1 from the
1335 actual data on the Volume and compare it.
1336 - Implement Bacula plugins -- design API
1337 - Make bcopy read through bad tape records.
1338 - Program files (i.e. execute a program to read/write files).
1339 Pass read date of last backup, size of file last time.
1340 - Add Signature type to File DB record.
1341 - CD into subdirectory when open()ing files for backup to
1342 speed up things. Test with testfind().
1343 - Priority job to go to top of list.
1344 - Why are save/restore of device different sizes (sparse?) Yup! Fix it.
1345 - Implement some way for the Console to dynamically create a job.
1346 - Solaris -I on tar for include list
1347 - Need a verbose mode in restore, perhaps to bsr.
1348 - bscan without -v is too quiet -- perhaps show jobs.
1349 - Add code to reject whole blocks if not wanted on restore.
1350 - Check if we can increase Bacula FD priorty in Win2000
1351 - Make sure the MaxVolFiles is fully implemented in SD
1352 - Check if both CatalogFiles and UseCatalog are set to SD.
1353 - Possibly add email to Watchdog if drive is unmounted too
1354 long and a job is waiting on the drive.
1355 - After unmount, if restore job started, ask to mount.
1356 - Add UA rc and history files.
1357 - put termcap (used by console) in ./configure and
1358 allow -with-termcap-dir.
1359 - Fix Autoprune for Volumes to respect need for full save.
1360 - Compare tape to Client files (attributes, or attributes and data)
1361 - Make all database Ids 64 bit.
1362 - Allow console commands to detach or run in background.
1363 - Add SD message variables to control operator wait time
1364 - Maximum Operator Wait
1365 - Minimum Message Interval
1366 - Maximum Message Interval
1367 - Send Operator message when cannot read tape label.
1368 - Verify level=Volume (scan only), level=Data (compare of data to file).
1369 Verify level=Catalog, level=InitCatalog
1371 - Add keyword search to show command in Console.
1372 - Events : tape has more than xxx bytes.
1373 - Complete code in Bacula Resources -- this will permit
1374 reading a new config file at any time.
1375 - Handle ctl-c in Console
1376 - Implement script driven addition of File daemon to config files.
1377 - Think about how to make Bacula work better with File (non-tape) archives.
1378 - Write Unix emulator for Windows.
1379 - Put memory utilization in Status output of each daemon
1380 if full status requested or if some level of debug on.
1381 - Make database type selectable by .conf files i.e. at runtime
1382 - Set flag for uname -a. Add to Volume label.
1383 - Restore files modified after date
1384 - SET LD_RUN_PATH=$HOME/mysql/lib/mysql
1385 - Remove duplicate fields from jcr (e.g. jcr.level and jcr.jr.Level, ...).
1386 - Timout a job or terminate if link goes down, or reopen link and query.
1387 - Concept of precious tapes (cannot be reused).
1388 - Make bcopy copy with a single tape drive.
1389 - Permit changing ownership during restore.
1392 > My suggestion: Add a feature on the systray menu-icon menu to request
1393 > an immediate backup now. This would be useful for laptop users who may
1394 > not be on the network when the regular scheduled backup is run.
1396 > My wife's suggestion: Add a setting to the win32 client to allow it to
1397 > shut down the machine after backup is complete (after, of course,
1398 > displaying a "System will shut down in one minute, click here to cancel"
1399 > warning dialog). This would be useful for sites that want user
1400 > woorkstations to be shut down overnight to save power.
1403 - Autolabel should be specified by DIR instead of SD.
1405 - Add media capacity
1406 - AutoScan (check checksum of tape)
1407 - Format command = "format /dev/nst0"
1411 - Seek resolution (usually corresponds to buffer size)
1412 - EODErrorCode=ENOSPC or code
1413 - Partial Read error code
1414 - Partial write error code
1415 - Nonformatted read error
1416 - Nonformatted write error
1417 - WriteProtected error
1421 - IgnoreCloseErrors=yes
1431 - FD sends unsaved file list to Director at end of job (see
1433 - File daemon should build list of files skipped, and then
1434 at end of save retry and report any errors.
1435 - Write a Storage daemon that uses pipes and
1436 standard Unix programs to write to the tape.
1438 - Need something that monitors the JCR queue and
1439 times out jobs by asking the deamons where they are.
1440 - Enhance Jmsg code to permit buffering and saving to disk.
1441 - device driver = "xxxx" for drives.
1442 - Verify from Volume
1443 - Ensure that /dev/null works
1444 - Need report class for messages. Perhaps
1445 report resource where report=group of messages
1446 - enhance scan_attrib and rename scan_jobtype, and
1447 fill in code for "since" option
1448 - Director needs a time after which the report status is sent
1449 anyway -- or better yet, a retry time for the job.
1450 - Don't reschedule a job if previous incarnation is still running.
1451 - Some way to automatically backup everything is needed????
1452 - Need a structure for pending actions:
1454 - termination status (part of buffered msgs?)
1456 Read, Write, Clean, Delete
1457 - Login to Bacula; Bacula users with different permissions:
1458 owner, group, user, quotas
1459 - Store info on each file system type (probably in the job header on tape.
1460 This could be the output of df; or perhaps some sort of /etc/mtab record.
1462 ========= ideas ===============
1463 From: "Jerry K. Schieffer" <jerry@skylinetechnology.com>
1464 To: <kern@sibbald.com>
1465 Subject: RE: [Bacula-users] future large programming jobs
1466 Date: Thu, 26 Feb 2004 11:34:54 -0600
1468 I noticed the subject thread and thought I would offer the following
1469 merely as sources of ideas, i.e. something to think about, not even as
1470 strong as a request. In my former life (before retiring) I often
1471 dealt with backups and storage management issues/products as a
1472 developer and as a consultant. I am currently migrating my personal
1473 network from amanda to bacula specifically because of the ability to
1474 cross media boundaries during storing backups.
1475 Are you familiar with the commercial product called ADSM (I think IBM
1476 now sells it under the Tivoli label)? It has a couple of interesting
1477 ideas that may apply to the following topics.
1479 1. Migration: Consider that when you need to restore a system, there
1480 may be pressure to hurry. If all the information for a single client
1481 can eventually end up on the same media (and in chronological order),
1482 the restore is facillitated by not having to search past information
1483 from other clients. ADSM has the concept of "client affinity" that
1484 may be associated with it's storage pools. It seems to me that this
1485 concept (as an optional feature) might fit in your architecture for
1488 ADSM also has the concept of defining one or more storage pools as
1489 "copy pools" (almost mirrors, but only in the sense of contents).
1490 These pools provide the ability to have duplicte data stored both
1491 onsite and offsite. The copy process can be scheduled to be handled
1492 by their storage manager during periods when there is no backup
1493 activity. Again, the migration process might be a place to consider
1494 implementing something like this.
1497 > It strikes me that it would be very nice to be able to do things
1499 > have the Job(s) backing up the machines run, and once they have all
1500 > completed, start a migration job to copy the data from disks Volumes
1502 > a tape library and then to offsite storage. Maybe this can already
1504 > done with some careful scheduling and Job prioritzation; the events
1505 > mechanism described below would probably make it very easy.
1507 This is the goal. In the first step (before events), you simply
1509 the Migration to tape later.
1511 2. Base jobs: In ADSM, each copy of each stored file is tracked in
1512 the database. Once a file (unique by path and metadata such as dates,
1513 size, ownership, etc.) is in a copy pool, no more copies are made. In
1514 other words, when you start ADSM, it begins like your concept of a
1515 base job. After that it is in the "incremental" mode. You can
1516 configure the number of "generations" of files to be retained, plus a
1517 retention date after which even old generations are purged. The
1518 database tracks the contents of media and projects the percentage of
1519 each volume that is valid. When the valid content of a volume drops
1520 below a configured percentage, the valid data are migrated to another
1521 volume and the old volume is marked as empty. Note, this requires
1522 ADSM to have an idea of the contents of a client, i.e. marking the
1523 database when an existing file was deleted, but this would solve your
1524 issue of restoring a client without restoring deleted files.
1526 This is pretty far from what bacula now does, but if you are going to
1527 rip things up for Base jobs,.....
1528 Also, the benefits of this are huge for very large shops, especially
1529 with media robots, but are a pain for shops with manual media
1533 > Base jobs sound pretty useful, but I'm not dying for them.
1535 Nobody is dying for them, but when you see what it does, you will die
1538 3. Restoring deleted files: Since I think my comments in (2) above
1539 have low probability of implementation, I'll also suggest that you
1540 could approach the issue of deleted files by a mechanism of having the
1541 fd report to the dir, a list of all files on the client for every
1542 backup job. The dir could note in the database entry for each file
1543 the date that the file was seen. Then if a restore as of date X takes
1544 place, only files that exist from before X until after X would be
1545 restored. Probably the major cost here is the extra date container in
1546 each row of the files table.
1548 Thanks for "listening". I hope some of this helps. If you want to
1549 contact me, please send me an email - I read some but not all of the
1550 mailing list traffic and might miss a reply there.
1552 Please accept my compliments for bacula. It is doing a great job for
1553 me!! I sympathize with you in the need to wrestle with excelence in
1554 execution vs. excelence in feature inclusion.
1559 ==============================
1562 - Design at hierarchial storage for Bacula. Migration and Clone.
1563 - Implement FSM (File System Modules).
1564 - Audit M_ error codes to ensure they are correct and consistent.
1565 - Add variable break characters to lex analyzer.
1566 Either a bit mask or a string of chars so that
1567 the caller can change the break characters.
1568 - Make a single T_BREAK to replace T_COMMA, etc.
1569 - Ensure that File daemon and Storage daemon can
1570 continue a save if the Director goes down (this
1571 is NOT currently the case). Must detect socket error,
1572 buffer messages for later.
1573 - Enhance time/duration input to allow multiple qualifiers e.g. 3d2h
1574 - Add ability to backup to two Storage devices (two SD sessions) at
1575 the same time -- e.g. onsite, offsite.
1576 - Add the ability to consolidate old backup sets (basically do a restore
1577 to tape and appropriately update the catalog). Compress Volume sets.
1578 Might need to spool via file is only one drive is available.
1579 - Compress or consolidate Volumes of old possibly deleted files. Perhaps
1580 someway to do so with every volume that has less than x% valid
1584 Migration: Move a backup from one Volume to another
1585 Clone: Copy a backup -- two Volumes
1588 ======================================================
1590 It is somewhat like a Full save becomes an incremental since
1591 the Base job (or jobs) plus other non-base files.
1593 - A Base backup is same as Full backup, just different type.
1594 - New BaseFiles table that contains:
1596 BaseJobId - Base JobId referenced for this FileId (needed ???)
1597 JobId - JobId currently running
1598 FileId - File not backed up, exists in Base Job
1599 FileIndex - FileIndex from Base Job.
1600 i.e. for each base file that exists but is not saved because
1601 it has not changed, the File daemon sends the JobId, BaseId,
1602 FileId, FileIndex back to the Director who creates the DB entry.
1603 - To initiate a Base save, the Director sends the FD
1604 the FileId, and full filename for each file in the Base.
1605 - When the FD finds a Base file, he requests the Director to
1606 send him the full File entry (stat packet plus MD5), or
1607 conversely, the FD sends it to the Director and the Director
1608 says yes or no. This can be quite rapid if the FileId is kept
1609 by the FD for each Base Filename.
1610 - It is probably better to have the comparison done by the FD
1611 despite the fact that the File entry must be sent across the
1613 - An alternative would be to send the FD the whole File entry
1614 from the start. The disadvantage is that it requires a lot of
1615 space. The advantage is that it requires less communications
1617 - The Job record must be updated to indicate that one or more
1619 - At end of Job, FD returns:
1620 1. Count of base files/bytes not written to tape (i.e. matches)
1621 2. Count of base file that were saved i.e. had changed.
1622 - No tape record would be written for a Base file that matches, in the
1623 same way that no tape record is written for Incremental jobs where
1624 the file is not saved because it is unchanged.
1625 - On a restore, all the Base file records must explicitly be
1626 found from the BaseFile tape. I.e. for each Full save that is marked
1627 to have one or more Base Jobs, search the BaseFile for all occurrences
1629 - An optimization might be to make the BaseFile have:
1635 This would avoid the need to explicitly fetch each File record for
1636 the Base job. The Base Job record will be fetched to get the
1637 VolSessionId and VolSessionTime.
1638 =========================================================
1641 ==========================================================
1643 For each Incremental job that is run, there may be files that
1644 were found but not saved because they were locked (this applies
1645 only to Windows). Such a system could send back to the Director
1646 a list of Unsaved files.
1648 - New UnSavedFiles table that contains:
1652 - Then in the next Incremental job, the list of Unsaved Files will be
1653 feed to the FD, who will ensure that they are explicitly chosen even
1654 if standard date/time check would not have selected them.
1655 =============================================================
1659 Multiple drive autochanger data: see Alan Brown
1660 mtx -f xxx unloadStorage Element 1 is Already Full(drive 0 was empty)
1661 Unloading Data Transfer Element into Storage Element 1...source Element
1662 Address 480 is Empty
1664 (drive 0 was empty and so was slot 1)
1665 > mtx -f xxx load 15 0
1666 no response, just returns to the command prompt when complete.
1667 > mtx -f xxx status Storage Changer /dev/changer:2 Drives, 60 Slots ( 2 Import/Export )
1668 Data Transfer Element 0:Full (Storage Element 15 Loaded):VolumeTag = HX001
1669 Data Transfer Element 1:Empty
1670 Storage Element 1:Empty
1671 Storage Element 2:Full :VolumeTag=HX002
1672 Storage Element 3:Full :VolumeTag=HX003
1673 Storage Element 4:Full :VolumeTag=HX004
1674 Storage Element 5:Full :VolumeTag=HX005
1675 Storage Element 6:Full :VolumeTag=HX006
1676 Storage Element 7:Full :VolumeTag=HX007
1677 Storage Element 8:Full :VolumeTag=HX008
1678 Storage Element 9:Full :VolumeTag=HX009
1679 Storage Element 10:Full :VolumeTag=HX010
1680 Storage Element 11:Empty
1681 Storage Element 12:Empty
1682 Storage Element 13:Empty
1683 Storage Element 14:Empty
1684 Storage Element 15:Empty
1685 Storage Element 16:Empty....
1686 Storage Element 28:Empty
1687 Storage Element 29:Full :VolumeTag=CLNU01L1
1688 Storage Element 30:Empty....
1689 Storage Element 57:Empty
1690 Storage Element 58:Full :VolumeTag=NEX261L2
1691 Storage Element 59 IMPORT/EXPORT:Empty
1692 Storage Element 60 IMPORT/EXPORT:Empty
1694 Unloading Data Transfer Element into Storage Element 15...done
1696 (just to verify it remembers where it came from, however it can be
1697 overrriden with mtx unload {slotnumber} to go to any storage slot.)
1699 There needs to be a table of drive # to devices somewhere - If there are
1700 multiple changers or drives there may not be a 1:1 correspondance between
1701 changer drive number and system device name - and depending on the way the
1702 drives are hooked up to scsi busses, they may not be linearly numbered
1703 from an offset point either.something like
1705 Autochanger drives = 2
1706 Autochanger drive 0 = /dev/nst1
1707 Autochanger drive 1 = /dev/nst2
1708 IMHO, it would be _safest_ to use explicit mtx unload commands at all
1709 times, not just for multidrive changers. For a 1 drive changer, that's
1715 MTX's manpage (1.2.15):
1716 unload [<slotnum>] [ <drivenum> ]
1717 Unloads media from drive <drivenum> into slot
1718 <slotnum>. If <drivenum> is omitted, defaults to
1719 drive 0 (as do all commands). If <slotnum> is
1720 omitted, defaults to the slot that the drive was
1721 loaded from. Note that there's currently no way
1722 to say 'unload drive 1's media to the slot it
1723 came from', other than to explicitly use that
1724 slot number as the destination.AB
1730 undef# camcontrol devlist
1731 <WANGTEK 51000 SCSI M74H 12B3> at scbus0 target 2 lun 0 (pass0,sa0)
1732 <ARCHIVE 4586XX 28887-XXX 4BGD> at scbus0 target 4 lun 0 (pass1,sa1)
1733 <ARCHIVE 4586XX 28887-XXX 4BGD> at scbus0 target 4 lun 1 (pass2)
1735 tapeinfo -f /dev/sg0 with a bad tape in drive 1:
1736 [kern@rufus mtx-1.2.17kes]$ ./tapeinfo -f /dev/sg0
1737 Product Type: Tape Drive
1739 Product ID: 'C5713A '
1741 Attached Changer: No
1742 TapeAlert[3]: Hard Error: Uncorrectable read/write error.
1743 TapeAlert[20]: Clean Now: The tape drive neads cleaning NOW.
1750 Medium Type: Not Loaded
1753 DataCompEnabled: yes
1754 DataCompCapable: yes
1755 DataDeCompEnabled: yes
1762 Handling removable disks
1764 From: Karl Cunningham <karlc@keckec.com>
1766 My backups are only to hard disk these days, in removable bays. This is my
1767 idea of how a backup to hard disk would work more smoothly. Some of these
1768 things Bacula does already, but I mention them for completeness. If others
1769 have better ways to do this, I'd like to hear about it.
1771 1. Accommodate several disks, rotated similar to how tapes are. Identified
1772 by partition volume ID or perhaps by the name of a subdirectory.
1773 2. Abort & notify the admin if the wrong disk is in the bay.
1774 3. Write backups to different subdirectories for each machine to be backed
1776 4. Volumes (files) get created as needed in the proper subdirectory, one
1778 5. When a disk is recycled, remove or zero all old backup files. This is
1779 important as the disk being recycled may be close to full. This may be
1780 better done manually since the backup files for many machines may be
1781 scattered in many subdirectories.
1786 - Why the heck doesn't bacula drop root priviledges before connecting to
1788 - Look at using posix_fadvise(2) for backups -- see bug #751.
1789 Possibly add the code at findlib/bfile.c:795
1790 /* TCP socket options */
1791 #define TCP_KEEPIDLE 4 /* Start keeplives after this period */
1792 - Fix bnet_connect() code to set a timer and to use time to
1794 - Implement 4th argument to make_catalog_backup that passes hostname.
1795 - Test FIFO backup/restore -- make regression
1796 - Please mount volume "xxx" on Storage device ... should also list
1797 Pool and MediaType in case user needs to create a new volume.
1798 - On restore add Restore Client, Original Client.
1799 01-Apr 00:42 rufus-dir: Start Backup JobId 55, Job=kernsave.2007-04-01_00.42.48
1800 01-Apr 00:42 rufus-sd: Python SD JobStart: JobId=55 Client=Rufus
1801 01-Apr 00:42 rufus-dir: Created new Volume "Full0001" in catalog.
1802 01-Apr 00:42 rufus-dir: Using Device "File"
1803 01-Apr 00:42 rufus-sd: kernsave.2007-04-01_00.42.48 Warning: Device "File" (/tmp) not configured to autolabel Volumes.
1804 01-Apr 00:42 rufus-sd: kernsave.2007-04-01_00.42.48 Warning: Device "File" (/tmp) not configured to autolabel Volumes.
1805 01-Apr 00:42 rufus-sd: Please mount Volume "Full0001" on Storage Device "File" (/tmp) for Job kernsave.2007-04-01_00.42.48
1806 01-Apr 00:44 rufus-sd: Wrote label to prelabeled Volume "Full0001" on device "File" (/tmp)
1807 - Check if gnome-console works with TLS.
1808 - the director seg faulted when I omitted the pool directive from a
1809 job resource. I was experimenting and thought it redundant that I had
1810 specified Pool, Full Backup Pool. and Differential Backup Pool. but
1811 apparently not. This happened when I removed the pool directive and
1812 started the director.
1813 - Add Where: client:/.... to restore job report.
1814 - Ensure that moving a purged Volume in ua_purge.c to the RecyclePool
1815 does the right thing.
1816 - FD-SD quick disconnect
1817 - Building the in memory restore tree is slow.