9 - Document cleaning up the spool files:
10 db, pid, state, bsr, mail, conmsg, spool
11 - Document the multiple-drive-changer.txt script.
12 - Pruning with Admin job.
13 - Does WildFile match against full name? Doc.
14 - %d and %v only valid on Director, not for ClientRunBefore/After.
15 - During tests with the 260 char fix code, I found one problem:
16 if the system "sees" a long path once, it seems to forget it's
17 working drive (e.g. c:\), which will lead to a problem during
18 the next job (create bootstrap file will fail). Here is the
19 workaround: specify absolute working and pid directory in
20 bacula-fd.conf (e.g. c:\bacula\working instead of
22 - Document techniques for restoring large numbers of files.
23 - Document setting my.cnf to big file usage.
24 - Add example of proper index output to doc. show index from File;
25 - Correct the Include syntax in the m4.xxx files in examples/conf
26 - Document JobStatus and Termination codes.
27 - Fix the error with the "DVI file can't be opened" while
28 building the French PDF.
29 - Document more DVD stuff
37 - Document all the little details of setting up certificates for
38 the Bacula data encryption code.
39 - Document more precisely how to use master keys -- especially
40 for disaster recovery.
44 - Check if gnome-console works with TLS.
45 - Ensure that the SD re-reads the Media record if the JobFiles
46 does not match -- it may have been updated by another job.
47 - Look at moving the Storage directive from the Job to the
48 Pool in the default conf files.
49 - Migration Volume span bug
52 - Test FIFO backup/restore -- make regression
54 - Add encryption regression tests
55 - Test Volume compatibility between machine architectures
56 - Encryption documentation
57 - Wrong jobbytes with query 12 (todo)
58 - bacula-1.38.2-ssl.patch
59 - Bare-metal recovery Windows (todo)
64 - Implement Python event for backing up/restoring a file.
65 - Change dbcheck to tell users to use native tools for fixing
66 broken databases, and to ensure they have the proper indexes.
67 - add udev rules for Bacula devices.
68 - If a job terminates, the DIR connection can close before the
69 Volume info is updated, leaving the File count wrong.
70 - Look at why SIGPIPE during connection can cause seg fault in
71 writing the daemon message, when Dir dropped to bacula:bacula
72 - Look at zlib 32 => 64 problems.
73 - Try turning on disk seek code.
74 - Possibly turn on St. Bernard code.
75 - Fix bextract to restore ACLs, or better yet, use common routines.
76 - Do we migrate appendable Volumes?
77 - Remove queue.c code.
78 - Some users claim that they must do two prune commands to get a
79 Volume marked as purged.
80 - Print warning message if LANG environment variable does not specify
82 - New dot commands from Arno.
83 .show device=xxx lists information from one storage device, including
84 devices (I'm not even sure that information exists in the DIR...)
85 .move eject device=xxx mostly the same as 'unmount xxx' but perhaps with
86 better machine-readable output like "Ok" or "Error busy"
87 .move eject device=xxx toslot=yyy the same as above, but with a new
88 target slot. The catalog should be updated accordingly.
89 .move transfer device=xxx fromslot=yyy toslot=zzz
92 - It appears to me that you have run into some sort of race
93 condition where two threads want to use the same Volume and they
94 were both given access. Normally that is no problem. However,
95 one thread wanted the particular Volume in drive 0, but it was
96 loaded into drive 1 so it decided to unload it from drive 1 and
97 then loaded it into drive 0, while the second thread went on
98 thinking that the Volume could be used in drive 1 not realizing
99 that in between time, it was loaded in drive 0.
100 I'll look at the code to see if there is some way we can avoid
101 this kind of problem. Probably the best solution is to make the
102 first thread simply start using the Volume in drive 1 rather than
103 transferring it to drive 0.
104 - After pruning, check to see if the Volume retention period has
106 - Check to see if jcr->stime is lost during rescheduling of
108 - Fix re-read of last block to check if job has actually written
109 a block, and check if block was written by a different job
110 (i.e. multiple simultaneous jobs writing).
111 - Figure out how to configure query.sql. Suggestion to use m4:
112 == changequote.m4 ===
113 changequote(`[',`]')dnl
114 ==== query.sql.in ===
115 :List next 20 volumes to expire
117 Pool.Name AS PoolName,
122 [ FROM_UNIXTIME(UNIX_TIMESTAMP(Media.LastWritten) Media.VolRetention) AS Expire, ])dnl
124 [ media.lastwritten + interval '1 second' * media.volretention as expire, ])dnl
128 ON Media.PoolId=Pool.PoolId
129 WHERE Media.LastWritten>0
133 Command: m4 -DmySQL changequote.m4 query.sql.in >query.sql
135 The problem is that it requires m4, which is not present on all machines
137 - Get Perl replacement for bregex.c
138 - Given all the problems with FIFOs, I think the solution is to do something a
139 little different, though I will look at the code and see if there is not some
140 simple solution (i.e. some bug that was introduced). What might be a better
141 solution would be to use a FIFO as a sort of "key" to tell Bacula to read and
142 write data to a program rather than the FIFO. For example, suppose you
147 Then, I could imagine if you backup and restore this file with a direct
148 reference as is currently done for fifos, instead, during backup Bacula will
151 /home/kern/my-fifo.backup
153 and read the data that my-fifo.backup writes to stdout. For restore, Bacula
156 /home/kern/my-fifo.restore
158 and send the data backed up to stdout. These programs can either be an
159 executable or a shell script and they need only read/write to stdin/stdout.
161 I think this would give a lot of flexibility to the user without making any
162 significant changes to Bacula.
167 select FilenameId from Filename where Name='';
168 # Get list of all directories referenced in a Backup.
169 select Path.Path from Path,File where File.JobId=nnn and
170 File.FilenameId=(FilenameId-from-above) and File.PathId=Path.PathId
171 order by Path.Path ASC;
173 - Look into using Dart for testing
174 http://public.kitware.com/Dart/HTML/Index.shtml
176 - Look into replacing autotools with cmake
177 http://www.cmake.org/HTML/Index.html
179 === Migration from David ===
180 What I'd like to see:
183 Name = "<poolname>-migrate"
187 Migration Selection Type = LowestUtil | OldestVol | PoolOccupancy |
188 Client | PoolResidence | Volume | JobName | SQLquery
189 Migration Selection Pattern = "regexp"
190 Next Pool = <override>
193 There should be no need for a Level (migration is always Full, since you
194 don't calculate differential/incremental differences for migration),
195 Storage should be determined by the volume types in the pool, and Client
196 is really a selection issue. Migration should always occur to the
197 NextPool defined in the pool definition. If no nextpool is defined, the
198 job should end with a reason of "no place to go". If Next Pool statement
199 is present, we override the check in the pool definition and use the
202 Here's how I'd define Migration Selection Types:
205 Client -- Migrate data from selected client only. Migration Selection
206 Pattern regexp provides pattern to select client names, eg ^FS00* makes
207 all client names starting with FS00 eligible for migration.
209 Jobname -- Migration all jobs matching name. Migration Selection Pattern
210 regexp provides pattern to select jobnames existing in pool.
212 Volume -- Migrate all data on specified volumes. Migration Selection
213 Pattern regexp provides selection criteria for volumes to be migrated.
214 Volumes must exist in pool to be eligible for migration.
218 LowestUtil -- Identify the volume in the pool with the least data on it
219 and empty it. No Migration Selection Pattern required.
221 OldestVol -- Identify the LRU volume with data written, and empty it. No
222 Migration Selection Pattern required.
224 PoolOccupancy -- if pool occupancy exceeds <highmig>, migrate volumes
225 (starting with most full volumes) until pool occupancy drops below
226 <lowmig>. Pool highmig and lowmig values are in pool definition, no
227 Migration Selection Pattern required.
231 SQLQuery -- Migrate all jobuids returned by the supplied SQL query.
232 Migration Selection Pattern contains SQL query to execute; should return
233 a list of 1 or more jobuids to migrate.
235 PoolResidence -- Migrate data sitting in pool for longer than
236 PoolResidence value in pool definition. Migration Selection Pattern
237 optional; if specified, override value in pool definition (value in
241 [ possibly a Python event -- kes ]
243 - Mount on an Autochanger with no tape in the drive causes:
244 Automatically selected Storage: LTO-changer
245 Enter autochanger drive[0]: 0
246 3301 Issuing autochanger "loaded drive 0" command.
247 3302 Autochanger "loaded drive 0", result: nothing loaded.
248 3301 Issuing autochanger "loaded drive 0" command.
249 3302 Autochanger "loaded drive 0", result: nothing loaded.
250 3902 Cannot mount Volume on Storage Device "LTO-Drive1" (/dev/nst0) because:
251 Couldn't rewind device "LTO-Drive1" (/dev/nst0): ERR=dev.c:678 Rewind error on "LTO-Drive1" (/dev/nst0). ERR=No medium found.
252 3905 Device "LTO-Drive1" (/dev/nst0) open but no Bacula volume is mounted.
253 If this is not a blank tape, try unmounting and remounting the Volume.
254 - If Drive 0 is blocked, and drive 1 is set "Autoselect=no", drive 1 will
256 - Autochanger did not change volumes.
257 select * from Storage;
258 +-----------+-------------+-------------+
259 | StorageId | Name | AutoChanger |
260 +-----------+-------------+-------------+
261 | 1 | LTO-changer | 0 |
262 +-----------+-------------+-------------+
263 05-May 03:50 roxie-sd: 3302 Autochanger "loaded drive 0", result is Slot 11.
264 05-May 03:50 roxie-sd: Tibs.2006-05-05_03.05.02 Warning: Director wanted Volume "LT
265 Current Volume "LT0-002" not acceptable because:
266 1997 Volume "LT0-002" not in catalog.
267 05-May 03:50 roxie-sd: Tibs.2006-05-05_03.05.02 Error: Autochanger Volume "LT0-002"
268 Setting InChanger to zero in catalog.
269 05-May 03:50 roxie-dir: Tibs.2006-05-05_03.05.02 Error: Unable to get Media record
271 05-May 03:50 roxie-sd: Tibs.2006-05-05_03.05.02 Fatal error: Error getting Volume i
272 05-May 03:50 roxie-sd: Tibs.2006-05-05_03.05.02 Fatal error: Job 530 canceled.
273 05-May 03:50 roxie-sd: Tibs.2006-05-05_03.05.02 Fatal error: spool.c:249 Fatal appe
274 05-May 03:49 Tibs: Tibs.2006-05-05_03.05.02 Fatal error: c:\cygwin\home\kern\bacula
283 FirstWritten: 2006-05-05 03:11:54
284 LastWritten: 2006-05-05 03:50:23
285 LabelDate: 2005-12-26 16:52:40
296 VolRetention: 31,536,000
308 Note VolStatus is blank!!!!!
315 FirstWritten: 0000-00-00 00:00:00
316 LastWritten: 0000-00-00 00:00:00
317 LabelDate: 2005-12-26 16:52:40
328 VolRetention: 31,536,000
341 Automatically selected Storage: LTO-changer
342 Enter autochanger drive[0]: 0
343 3301 Issuing autochanger "loaded drive 0" command.
344 3302 Autochanger "loaded drive 0", result: nothing loaded.
345 3301 Issuing autochanger "loaded drive 0" command.
346 3302 Autochanger "loaded drive 0", result: nothing loaded.
347 3902 Cannot mount Volume on Storage Device "LTO-Drive1" (/dev/nst0) because:
348 Couldn't rewind device "LTO-Drive1" (/dev/nst0): ERR=dev.c:678 Rewind error on "LTO-Drive1" (/dev/nst0). ERR=No medium found.
350 3905 Device "LTO-Drive1" (/dev/nst0) open but no Bacula volume is mounted.
351 If this is not a blank tape, try unmounting and remounting the Volume.
353 - Add VolumeState (enable, disable, archive)
354 - Add VolumeLock to prevent all but lock holder (SD) from updating
355 the Volume data (with the exception of VolumeState).
356 - The btape fill command does not seem to use the Autochanger
357 - Make Windows installer default to system disk drive.
358 - Look at using ioctl(FIOBMAP, ...) on Linux, and
359 DeviceIoControl(..., FSCTL_QUERY_ALLOCATED_RANGES, ...) on
360 Win32 for sparse files.
361 http://www.flexhex.com/docs/articles/sparse-files.phtml
362 http://www.informatik.uni-frankfurt.de/~loizides/reiserfs/fibmap.html
363 - Directive: at <event> "command"
364 - Command: pycmd "command" generates "command" event. How to
365 attach to a specific job?
366 - Integrate Christopher's St. Bernard code.
367 - run_cmd() returns int should return JobId_t
368 - get_next_jobid_from_list() returns int should return JobId_t
369 - Document export LDFLAGS=-L/usr/lib64
370 - Don't attempt to restore from "Disabled" Volumes.
371 - Network error on Win32 should set Win32 error code.
372 - What happens when you rename a Disk Volume?
373 - Job retention period in a Pool (and hence Volume). The job would
375 - Detect resource deadlock in Migrate when same job wants to read
376 and write the same device.
377 - Queue warning/error messages during restore so that they
378 are reported at the end of the report rather than being
379 hidden in the file listing ...
380 - Look at -D_FORTIFY_SOURCE=2
381 - Add Win32 FileSet definition somewhere
382 - Look at fixing restore status stats in SD.
383 - Make selection of Database used in restore correspond to
385 - Look at using ioctl(FIMAP) and FIGETBSZ for sparse files.
386 http://www.informatik.uni-frankfurt.de/~loizides/reiserfs/fibmap.html
387 - Implement a mode that says when a hard read error is
388 encountered, read many times (as it currently does), and if the
389 block cannot be read, skip to the next block, and try again. If
390 that fails, skip to the next file and try again, ...
392 create table LevelType (LevelType binary(1), LevelTypeLong tinyblob);
393 insert into LevelType (LevelType,LevelTypeLong) values
397 - Add ACL to restore only to original location.
398 - Show files/second in client status output.
399 - Add a recursive mark command (rmark) to restore.
400 - "Minimum Job Interval = nnn" sets minimum interval between Jobs
401 of the same level and does not permit multiple simultaneous
402 running of that Job (i.e. lets any previous invocation finish
403 before doing Interval testing).
404 - Look at simplifying File exclusions.
405 - New directive "Delete purged Volumes"
406 - new pool XXX with ScratchPoolId = MyScratchPool's PoolId and
407 let it fill itself, and RecyclePoolId = XXX's PoolId so I can
408 see if it become stable and I just have to supervise
410 - If I want to remove this pool, I set RecyclePoolId = MyScratchPool's
411 PoolId, and when it is empty remove it.
412 - Figure out how to recycle Scratch volumes back to the Scratch Pool.
414 - Allow Check Labels to be used with Bacula labels.
415 - "Resuming" a failed backup (lost line for example) by using the
416 failed backup as a sort of "base" job.
418 - Email to the user when the tape is about to need changing x
419 days before it needs changing.
420 - Command to show next tape that will be used for a job even
421 if the job is not scheduled.
422 - From: Arunav Mandal <amandal@trolltech.com>
423 1. When jobs are running and bacula for some reason crashes or if I do a
424 restart it remembers and jobs it was running before it crashed or restarted
425 as of now I loose all jobs if I restart it.
427 2. When spooling and in the midway if client is disconnected for instance a
428 laptop bacula completely discard the spool. It will be nice if it can write
429 that spool to tape so there will be some backups for that client if not all.
431 3. We have around 150 clients machines it will be nice to have a option to
432 upgrade all the client machines bacula version automatically.
434 4. Atleast one connection should be reserved for the bconsole so at heavy load
435 I should connect to the director via bconsole which at sometimes I can't
437 5. Another most important feature that is missing, say at 10am I manually
438 started backup of client abc and it was a full backup since client abc has
439 no backup history and at 10.30am bacula again automatically started backup of
440 client abc as that was in the schedule. So now we have 2 multiple Full
441 backups of the same client and if we again try to start a full backup of
442 client backup abc bacula won't complain. That should be fixed.
444 - Fix bpipe.c so that it does not modify results pointer.
445 ***FIXME*** calling sequence should be changed.
446 - For Windows disaster recovery see http://unattended.sf.net/
447 - regardless of the retention period, Bacula will not prune the
448 last Full, Diff, or Inc File data until a month after the
449 retention period for the last Full backup that was done.
450 - update volume=xxx --- add status=Full
451 - Remove old spool files on startup.
452 - Exclude SD spool/working directory.
453 - Refuse to prune last valid Full backup. Same goes for Catalog.
455 - Make a callback when Rerun failed levels is called.
456 - Give Python program access to Scheduled jobs.
457 - Add setting Volume State via Python.
458 - Python script to save with Python, not save, save with Bacula.
459 - Python script to do backup.
461 - Change the Priority, Client, Storage, JobStatus (error)
462 at the start of a job.
463 - Why is SpoolDirectory = /home/bacula/spool; not reported
464 as an error when writing a DVD?
465 - Make bootstrap file handle multiple MediaTypes (SD)
466 - Remove all old Device resource code in Dir and code to pass it
467 back in SD -- better, rework it to pass back device statistics.
468 - Check locking of resources -- be sure to lock devices where previously
469 resources were locked.
470 - The last part is left in the spool dir.
473 - In restore don't compare byte count on a raw device -- directory
474 entry does not contain bytes.
478 MA = (last_MA * 3 + rate) / 4
479 rate = (bytes - last_bytes) / (runtime - last_runtime)
480 - Max Vols limit in Pool off by one?
481 - Implement Files/Bytes,... stats for restore job.
482 - Implement Total Bytes Written, ... for restore job.
483 - Despool attributes simultaneously with data in a separate
484 thread, rejoined at end of data spooling.
485 - 7. Implement new Console commands to allow offlining/reserving drives,
486 and possibly manipulating the autochanger (much asked for).
487 - Add start/end date editing in messages (%t %T, %e?) ...
488 - Add ClientDefs similar to JobDefs.
489 - Print more info when bextract -p accepts a bad block.
490 - Fix FD JobType to be set before RunBeforeJob in FD.
491 - Look at adding full Volume and Pool information to a Volume
492 label so that bscan can get *all* the info.
493 - If the user puts "Purge Oldest Volume = yes" or "Recycle Oldest Volume = yes"
494 and there is only one volume in the pool, refuse to do it -- otherwise
495 he fills the Volume, then immediately starts reusing it.
496 - Implement copies and stripes.
497 - Add history file to console.
498 - Each file on tape creates a JobMedia record. Peter has 4 million
499 files spread over 10000 tape files and four tapes. A restore takes
500 16 hours to build the restore list.
501 - Add and option to check if the file size changed during backup.
502 - Make sure SD deletes spool files on error exit.
503 - Delete old spool files when SD starts.
504 - When labeling tapes, if you enter 000026, Bacula uses
505 the tape index rather than the Volume name 000026.
506 - Add offline tape command to Bacula console.
508 Enter MediaId or Volume name: 32
509 Enter new Volume name: DLT-20Dec04
510 Automatically selected Pool: Default
511 Connecting to Storage daemon DLTDrive at 192.168.68.104:9103 ...
512 Sending relabel command from "DLT-28Jun03" to "DLT-20Dec04" ...
513 block.c:552 Write error at 0:0 on device /dev/nst0. ERR=Bad file descriptor.
514 Error writing final EOF to tape. This tape may not be readable.
515 dev.c:1207 ioctl MTWEOF error on /dev/nst0. ERR=Permission denied.
516 askdir.c:219 NULL Volume name. This shouldn't happen!!!
517 3912 Failed to label Volume: ERR=dev.c:1207 ioctl MTWEOF error on /dev/nst0. ERR=Permission denied.
518 Label command failed for Volume DLT-20Dec04.
519 Do not forget to mount the drive!!!
520 - Bug: if a job is manually scheduled to run later, it does not appear
521 in any status report and cannot be cancelled.
523 ==== Keeping track of deleted/new files ====
524 - To mark files as deleted, run essentially a Verify to disk, and
525 when a file is found missing (MarkId != JobId), then create
526 a new File record with FileIndex == -1. This could be done
527 by the FD at the same time as the backup.
529 My "trick" for keeping track of deletions is the following.
530 Assuming the user turns on this option, after all the files
531 have been backed up, but before the job has terminated, the
532 FD will make a pass through all the files and send their
533 names to the DIR (*exactly* the same as what a Verify job
534 currently does). This will probably be done at the same
535 time the files are being sent to the SD avoiding a second
536 pass. The DIR will then compare that to what is stored in
537 the catalog. Any files in the catalog but not in what the
538 FD sent will receive a catalog File entry that indicates
539 that at that point in time the file was deleted. This
540 either transmitted to the FD or simultaneously computed in
541 the FD, so that the FD can put a record on the tape that
542 indicates that the file has been deleted at this point.
543 A delete file entry could potentially be one with a FileIndex
544 of 0 or perhaps -1 (need to check if FileIndex is used for
545 some other thing as many of the Bacula fields are "overloaded"
548 During a restore, any file initially picked up by some
549 backup (Full, ...) then subsequently having a File entry
550 marked "delete" will be removed from the tree, so will not
551 be restored. If a file with the same name is later OK it
552 will be inserted in the tree -- this already happens. All
553 will be consistent except for possible changes during the
556 Since I'm on the subject, some of you may be wondering what
557 the utility of the in memory tree is if you are going to
558 restore everything (at least it comes up from time to time
559 on the list). Well, it is still *very* useful because it
560 allows only the last item found for a particular filename
561 (full path) to be entered into the tree, and thus if a file
562 is backed up 10 times, only the last copy will be restored.
563 I recently (last Friday) restored a complete directory, and
564 the Full and all the Differential and Incremental backups
565 spanned 3 Volumes. The first Volume was not even mounted
566 because all the files had been updated and hence backed up
567 since the Full backup was made. In this case, the tree
568 saved me a *lot* of time.
570 Make sure this information is stored on the tape too so
571 that it can be restored directly from the tape.
573 All the code (with the exception of formally generating and
574 saving the delete file entries) already exists in the Verify
575 Catalog command. It explicitly recognizes added/deleted files since
576 the last InitCatalog. It is more or less a "simple" matter of
577 taking that code and adapting it slightly to work for backups.
579 Comments from Martin Simmons (I think they are all covered):
580 Ok, that should cover the basics. There are few issues though:
582 - Restore will depend on the catalog. I think it is better to include the
583 extra data in the backup as well, so it can be seen by bscan and bextract.
585 - I'm not sure if it will preserve multiple hard links to the same inode. Or
586 maybe adding or removing links will cause the data to be dumped again?
588 - I'm not sure if it will handle renamed directories. Possibly it will work
589 by dumping the whole tree under a renamed directory?
591 - It remains to be seen how the backup performance of the DIR's will be
592 affected when comparing the catalog for a large filesystem.
596 How about introducing a Type = MgmtPolicy job type? That job type would
597 be responsible for scanning the Bacula environment looking for specific
598 conditions, and submitting the appropriate jobs for implementing said
602 Name = "Migration-Policy"
604 Policy Selection Job Type = Migrate
605 Scope = "<keyword> <operator> <regexp>"
606 Threshold = "<keyword> <operator> <regexp>"
607 Job Template = <template-name>
610 Where <keyword> is any legal job keyword, <operator> is a comparison
611 operator (=,<,>,!=, logical operators AND/OR/NOT) and <regexp> is a
612 appropriate regexp. I could see an argument for Scope and Threshold
613 being SQL queries if we want to support full flexibility. The
614 Migration-Policy job would then get scheduled as frequently as a site
615 felt necessary (suggested default: every 15 minutes).
620 Name = "Migration-Policy"
622 Policy Selection Job Type = Migration
624 Threshold = "Migration Selection Type = LowestUtil"
625 Job Template = "MigrationTemplate"
628 would select all pools for examination and generate a job based on
629 MigrationTemplate to automatically select the volume with the lowest
630 usage and migrate it's contents to the nextpool defined for that pool.
632 This policy abstraction would be really handy for adjusting the behavior
633 of Bacula according to site-selectable criteria (one thing that pops
634 into mind is Amanda's ability to automatically adjust backup levels
635 depending on various criteria).
641 - Add Pool/Storage override regression test.
642 - Add delete JobId to regression.
643 - Add a regression test for dbcheck.
644 - New test to add bscan to four-concurrent-jobs regression,
645 i.e. after the four-concurrent jobs zap the
646 database as is done in the bscan-test, then use bscan to
647 restore the database, do a restore and compare with the
649 - Add restore of specific JobId to regression (item 3
650 on the restore prompt)
651 - Add IPv6 to regression
652 - Add database test to regression. Test each function like delete,
655 - AntiVir can slow down backups on Win32 systems.
656 - Win32 systems with FAT32 can be much slower than NTFS for
657 more than 1000 files per directory.
661 - A HOLD command to stop all jobs from starting.
662 - A PAUSE command to pause all running jobs ==> release the
664 - Media Type = LTO,LTO-2,LTO-3
665 Media Type Read = LTO,LTO2,LTO3
666 Media Type Write = LTO2, LTO3
668 === From Carsten Menke <bootsy52@gmx.net>
670 Following is a list of what I think in the situations where I'm faced with,
671 could be a usefull enhancement to bacula, which I'm certain other users will
672 benefit from as well.
674 1. NextJob/NextJobs Directive within a Job Resource in the form of
675 NextJobs = job1,job2.
678 I currently solved the problem with running multiple jobs each after each
679 by setting the Max Wait Time for a job to 8 hours, and give
680 the jobs different Priorities. However, there scenarios where
681 1 Job is directly depending on another job, so if the former job fails,
682 the job after it needn't to be run
683 while maybe other jobs should run despite of that
686 A Backup Job and a Verify job, if the backup job fails there is no need to run
687 the verify job, as the backup job already failed. However, one may like
688 to backup the Catalog to disk despite of that the main backup job failed.
691 I see that this is related to the Event Handlers which are on the ToDo
692 list, also it is maybe a good idea to check for the return value and
693 execute different actions based on the return value
696 3. offline capability to bconsole
699 Currently I use a script which I execute within the last Job via the
700 RunAfterJob Directive, to release and eject the tape.
701 So I have to call bconsole "release=Storage-Name" and afterwards
702 mt -f /dev/nst0 eject to get the tape out.
704 If I have multiple Storage Devices, than these may not be /dev/nst0 and
705 I have to modify the script or call it with parameters etc.
706 This would actually not be needed, as everything is already defined
707 in bacula-sd.conf and if I can invoke bconsole with the
708 storage name via $1 in the script than I'm done and information is
711 4. %s for Storage Name added to the chars being substituted in "RunAfterJob"
715 For the reason mentioned in 3. to have the ability to call a
716 script with /scripts/foobar %s and in the script use $1
717 to pass the Storage Name to bconsole
719 5. Setting Volume State within a Job Resource
722 Instead of using "Maximum Volume Jobs" in the Pool Resource,
723 I would have the possibilty to define
724 in a Job Resource that after this certain job is run, the Volume State
725 should be set to "Volume State = Used", this give more flexibility (IMHO).
727 6. Localization of Bacula Messages
730 Unfortunatley many,many people I work with don't speak english very well.
731 So if at least the Reporting messages would be localized then they
732 would understand that they have to change the tape,etc. etc.
734 I volunteer to do the german translations, and if I can convince my wife also
735 french and Morre (western african language).
737 7. OK, this is evil, probably bound to security risks and maybe not possible
738 due to the design of bacula.
740 Implementation of Backtics ( `command` ) for shell comand execution to
741 the "Label Format" Directive.
745 Currently I have defined BACULA_DAY_OF_WEEK="day1|day2..." resulting in
746 Label Format = "HolyBackup-${BACULA_DAY_OF_WEEK[${WeekDay}]}". If I could
747 use backticks than I could use "Label Format = HolyBackup-`date +%A` to have
748 the localized name for the day of the week appended to the
749 format string. Then I have the tape labeled automatically with weekday
750 name in the correct language.
752 - Yes, that is surely the case. I probably should turn those into Warning
753 errors. In addition, you just made me think that it might not be bad to
754 add an option to check the file size after backing up the file and
755 report if it changes. This would be done as an option because it would
758 Kern, good idea. If you do do that, mention in the output: file
759 shrunk, or file expanded, just to make it obvious to the user
760 (without having to the refer to file size), just how the file size
763 Would this option be for all file, or just one file? Or a fileset?
764 - Make output from status use html table tags for nicely
765 presenting in a browser.
766 - Can one write tapes faster with 8192 byte block sizes?
767 - Document security problems with the same password for everyone in
768 rpm and Win32 releases.
769 - Browse generations of files.
770 - I've seen an error when my catalog's File table fills up. I
771 then have to recreate the File table with a larger maximum row
772 size. Relevant information is at
773 http://dev.mysql.com/doc/mysql/en/Full_table.html ; I think the
774 "Installing and Configuring MySQL" chapter should talk a bit
775 about this potential problem, and recommend a solution.
776 - For Solaris must use POSIX awk.
777 - Want speed of writing to tape while despooling.
778 - Supported autochanger:
786 Wangtek 6525ES (SCSI-1 QIC drive, 525MB), under Linux 2.4.something,
787 bacula 1.36.0/1 works with blocksize 16k INSIDE bacula-sd.conf.
788 - Add regex from http://www.pcre.org to Bacula for Win32.
789 - Use only shell tools no make in CDROM package.
790 - Include within include does it work?
791 - Implement a Pool of type Cleaning?
792 - Implement VolReadTime and VolWriteTime in SD
793 - Modify Backing up Your Database to include a bootstrap file.
794 - Think about making certain database errors fatal.
795 - Look at correcting the time jump in the scheduler for daylight
796 savings time changes.
797 - Add a "real" timer to network connections.
798 - Promote to Full = Time period
799 - Check dates entered by user for correctness (month/day/... ranges)
800 - Compress restore Volume listing by date and first file.
801 - Look at patches/bacula_db.b2z postgresql that loops during restore.
803 - Perhaps add read/write programs and/or plugins to FileSets.
804 - How to handle backing up portables ...
805 - Add some sort of guaranteed Interval for upgrading jobs.
806 - Can we write the state file after every job terminates? On Win32
807 the system crashes and the state file is not updated.
810 Documentation to do: (any release a little bit at a time)
811 - Doc to do unmount before removing magazine.
812 - Alternative to static linking "ldd prog" save all binaries listed,
813 restore them and point LD_LIBRARY_PATH to them.
814 - Document add "</dev/null >/dev/null 2>&1" to the bacula-fd command line
815 - Document query file format.
816 - Add more documentation for bsr files.
817 - Document problems with Verify and pruning.
818 - Document how to use multiple databases.
819 - VXA drives have a "cleaning required"
820 indicator, but Exabyte recommends preventive cleaning after every 75
823 In this context, it should be noted that Exabyte has a command-line
824 vxatool utility available for free download. (The current version is
825 vxatool-3.72.) It can get diagnostic info, read, write and erase tapes,
826 test the drive, unload tapes, change drive settings, flash new firmware,
828 Of particular interest in this context is that vxatool <device> -i will
829 report, among other details, the time since last cleaning in tape motion
830 minutes. This information can be retrieved (and settings changed, for
831 that matter) through the generic-SCSI device even when Bacula has the
832 regular tape device locked. (Needless to say, I don't recommend
833 changing tape settings while a job is running.)
834 - Lookup HP cleaning recommendations.
835 - Lookup HP tape replacement recommendations (see trouble shooting autochanger)
836 - Document doing table repair
839 ===================================
840 - Add macro expansions in JobDefs.
841 Run Before Job = "SomeFile %{Level} %{Client}"
842 Write Bootstrap="/some/dir/%{JobName}_%{Client}.bsr"
843 - Use non-blocking network I/O but if no data is available, use
845 - Use gather write() for network I/O.
846 - Autorestart on crash.
847 - Add bandwidth limiting.
848 - Add acks every once and a while from the SD to keep
849 the line from timing out.
850 - When an error in input occurs and conio beeps, you can back
851 up through the prompt.
852 - Detect fixed tape block mode during positioning by looking at
853 block numbers in btape "test". Possibly adjust in Bacula.
854 - Fix list volumes to output volume retention in some other
855 units, perhaps via a directive.
856 - If opening a tape in read/write mode fails attempt to open
857 it in read-only mode, and mark the tape for read only.
858 - Allow Simultaneous Priorities = yes => run up to Max concurrent jobs even
859 with multiple priorities.
860 - If you use restore replace=never, the directory attributes for
861 non-existent directories will not be restored properly.
863 - see lzma401.zip in others directory for new compression
865 - Minimal autochanger handling in Bacula and in btape.
866 - Look into how tar does not save sockets and the possiblity of
867 not saving them in Bacula (Martin Simmons reported this).
868 - Fix restore jobs so that multiple jobs can run if they
869 are not using the same tape(s).
870 - Allow the user to select JobType for manual pruning/purging.
871 - bscan does not put first of two volumes back with all info in
873 - Implement the FreeBSD nodump flag in chflags.
874 - Figure out how to make named console messages go only to that
875 console and to the non-restricted console (new console class?).
876 - Make restricted console prompt for password if *ask* is set or
877 perhaps if password is undefined.
878 - Implement "from ISO-date/time every x hours/days/weeks/months" in
881 ==== from Marc Schoechlin
882 - the help-command should be more verbose
883 (it should explain the paramters of the different
885 -> its time-comsuming to consult the manual anytime
886 you need a special parameter
887 -> maybe its more easy to maintain this, if the
888 descriptions of that commands are outsourced to
890 - the cd-command should allow complete paths
891 i.e. cd /foo/bar/foo/bar
892 -> if a customer mails me the path to a certain file,
893 its faster to enter the specified directory
894 - if the password is not configured in bconsole.conf
895 you should be asked for it.
896 -> sometimes you like to do restore on a customer-machine
897 which shouldnt know the password for bacula.
898 -> adding the password to the file favours admins
899 to forget to remove the password after usage
901 the protection of that file is less important
902 - long-listed-output of commands should be scrollable
903 like the unix more/less-command does
904 -> if someone runs 200 and more machines, the lists could
905 be a little long and complex
906 - command-output should be shown column by column
907 to reduce scrolling and to increase clarity
909 - lsmark should list the selected files with full
911 - wildcards for selecting and file and directories would be nice
912 - any actions should be interuptable with STRG+C
913 - command-expansion would be pretty cool
915 - When the replace Never option is set, new directory permissions
916 are not restored. See bug 213. To fix this requires creating a
917 list of newly restored directories so that those directory
918 permissions *can* be restored.
919 - Compaction of Disk space by "migrating" Volumes that have pruned
920 Jobs (what criteria? size, #jobs, time).
921 - Add prune all command
922 - Document fact that purge can destroy a part of a restore by purging
923 one volume while others remain valid -- perhaps mark Jobs.
924 - Add multiple-media-types.txt
925 - look at mxt-changer.html
926 - Make ? do a help command (no return needed).
927 - Implement restore directory.
928 - Document streams and how to implement them.
929 - Try not to re-backup a file if a new hard link is added.
930 - Add feature to backup hard links only, but not the data.
931 - Fix stream handling to be simpler.
932 - Add Priority and Bootstrap to Run a Job.
933 - Eliminate Restore "Run Restore Job" prompt by allowing new "run command
935 - Remove View FileSet button from Run a Job dialog.
936 - Handle prompt for restore job at end of Restore command.
937 - Add display of total selected files to Restore window.
938 - Add tree pane to left of window.
939 - Add progress meter.
940 - Max wait time or max run time causes seg fault -- see runtime-bug.txt
941 - Document writing to a CD/DVD with Bacula.
942 - Add a "base" package to the window installer for pthreadsVCE.dll
943 which is needed by all packages.
944 - Add message to user to check for fixed block size when the forward
945 space test fails in btape.
946 - When unmarking a directory check if all files below are unmarked and
947 then remove the + flag -- in the restore tree.
948 - Possibly implement: Action = Unmount Device="TapeDrive1" in Admin jobs.
949 - Setup lrrd graphs: (http://www.linpro.no/projects/lrrd/) Mike Acar.
950 - Revisit the question of multiple Volumes (disk) on a single device.
951 - Add a block copy option to bcopy.
952 - Investigate adding Mac Resource Forks.
953 - Finish work on Gnome restore GUI.
954 - Fix "llist jobid=xx" where no fileset or client exists.
955 - For each job type (Admin, Restore, ...) require only the really necessary
956 fields.- Pass Director resource name as an option to the Console.
957 - Add a "batch" mode to the Console (no unsolicited queries, ...).
958 - Add a .list all files in the restore tree (probably also a list all files)
959 Do both a long and short form.
960 - Allow browsing the catalog to see all versions of a file (with
961 stat data on each file).
962 - Restore attributes of directory if replace=never set but directory
964 - Use SHA1 on authentication if possible.
965 - See comtest-xxx.zip for Windows code to talk to USB.
966 - Add John's appended files:
967 Appended = { /files/server/logs/http/*log }
968 and such files would be treated as follows.On a FULL backup, they would
969 be backed up like any other file.On an INCREMENTAL backup, where a
970 previous INCREMENTAL or FULL was already in thecatalogue and the length
971 of the file wasgreater than the length of the last backup, only thedata
972 added since the last backup will be dumped.On an INCREMENTAL backup, if
973 the length of the file is less than thelength of the file with the same
974 name last backed up, the completefile is dumped.On Windows systems, with
975 creation date of files, we can be evensmarter about this and not count
976 entirely upon the length.On a restore, the full and all incrementals
977 since it will beapplied in sequence to restore the file.
978 - Check new HAVE_WIN32 open bits.
979 - Check if the tape has moved before writing.
980 - Handling removable disks -- see below:
981 - Keep track of tape use time, and report when cleaning is necessary.
982 - Add FromClient and ToClient keywords on restore command (or
983 BackupClient RestoreClient).
984 - Implement a JobSet, which groups any number of jobs. If the
985 JobSet is started, all the jobs are started together.
986 Allow Pool, Level, and Schedule overrides.
987 - Enhance cancel to timeout BSOCK packets after a specific delay.
988 - Do scheduling by UTC using gmtime_r() in run_conf, scheduler, and
989 ua_status.!!! Thanks to Alan Brown for this tip.
990 - Look at updating Volume Jobs so that Max Volume Jobs = 1 will work
991 correctly for multiple simultaneous jobs.
992 - Correct code so that FileSet MD5 is calculated for < and | filename
994 - Implement the Media record flag that indicates that the Volume does disk
996 - Implement VolAddr, which is used when Volume is addressed like a disk,
997 and form it from VolFile and VolBlock.
998 - Make multiple restore jobs for multiple media types specifying
999 the proper storage type.
1000 - Fix fast block rejection (stored/read_record.c:118). It passes a null
1001 pointer (rec) to try_repositioning().
1002 - Look at extracting Win data from BackupRead.
1003 - Implement RestoreJobRetention? Maybe better "JobRetention" in a Job,
1004 which would take precidence over the Catalog "JobRetention".
1005 - Implement Label Format in Add and Label console commands.
1006 - Possibly up network buffers to 65K. Put on variable.
1007 - Put email tape request delays on one or more variables. User wants
1008 to cancel the job after a certain time interval. Maximum Mount Wait?
1009 - Job, Client, Device, Pool, or Volume?
1010 Is it possible to make this a directive which is *optional* in multiple
1011 resources, like Level? If so, I think I'd make it an optional directive
1012 in Job, Client, and Pool, with precedence such that Job overrides Client
1013 which in turn overrides Pool.
1015 - New Storage specifications:
1016 - Want to write to multiple storage devices simultaneously
1017 - Want to write to multiple storage devices sequentially (in one job)
1018 - Want to read/write simultaneously
1019 - Key is MediaType -- it must match
1021 Passed to SD as a sort of BSR record called Storage Specification
1025 MediaType -> Next MediaType
1027 Device -> Next Device
1029 Allow multiple Storage specifications
1037 Allow Multiple Pool specifications (note, Pool currently
1039 Allow Multiple MediaType specifications in Dir conf
1040 Allow Multiple Device specifications in Dir conf
1041 Perhaps keep this in a single SSR
1042 Tie a Volume to a specific device by using a MediaType that
1043 is contained in only one device.
1044 In SD allow Device to have Multiple MediaTypes
1046 - Ideas from Jerry Scharf:
1047 First let's point out some big pluses that bacula has for this
1049 more importantly it's active. Thank you so much for that
1050 even more important, it's not flaky
1051 it has an open access catalog, opening many possibilities
1052 it's pushing toward heterogeneous systems capability
1054 Macintosh file client
1055 macs are an interesting niche, but I fear a server is a rathole
1056 working bare iron recovery for windows
1057 the option for inc/diff backups not reset on fileset revision
1058 a) use both change and inode update time against base time
1059 b) do the full catalog check (expensive but accurate)
1060 sizing guide (how much system is needed to back up N systems/files)
1061 consultants on using bacula in building a disaster recovery system
1062 an integration guide
1063 or how to get at fancy things that one could do with bacula
1064 logwatch code for bacula logs (or similar)
1065 linux distro inclusion of bacula (brings good and bad, but necessary)
1066 win2k/XP server capability (icky but you asked)
1067 support for Oracle database ??
1069 - Look at adding SQL server and Exchange support for Windows.
1070 - Make dev->file and dev->block_num signed integers so that -1 can
1071 be an invalid value which happens with BSR.
1072 - Create VolAddr for disk files in place of VolFile and VolBlock. This
1073 is needed to properly specify ranges.
1074 - Add progress of files/bytes to SD and FD.
1075 - Print warning message if FileId > 4 billion
1076 - do a "messages" before the first prompt in Console
1077 - Client does not show busy during Estimate command.
1078 - Implement Console mtx commands.
1079 - Implement a Mount Command and an Unmount Command where
1080 the users could specify a system command to be performed
1081 to do the mount, after which Bacula could attempt to
1082 read the device. This is for Removeable media such as a CDROM.
1083 - Most likely, this mount command would be invoked explicitly
1084 by the user using the current Console "mount" and "unmount"
1085 commands -- the Storage Daemon would do the right thing
1086 depending on the exact nature of the device.
1087 - As with tape drives, when Bacula wanted a new removable
1088 disk mounted, it would unmount the old one, and send a message
1089 to the user, who would then use "mount" as described above
1090 once he had actually inserted the disk.
1091 - Implement dump/print label to UA
1092 - Spool to disk only when the tape is full, then when a tape is hung move
1094 - bextract is sending everything to the log file ****FIXME****
1095 - Allow multiple Storage specifications (or multiple names on
1096 a single Storage specification) in the Job record. Thus a job
1097 can be backed up to a number of storage devices.
1098 - Implement some way for the File daemon to contact the Director
1099 to start a job or pass its DHCP obtained IP number.
1100 - Implement a query tape prompt/replace feature for a console
1101 - Copy console @ code to gnome2-console
1102 - Make AES the only encryption algorithm see
1103 http://csrc.nist.gov/CryptoToolkit/aes/). It's
1104 an officially adopted standard, has survived peer
1105 review, and provides keys up to 256 bits.
1106 - Take a careful look at SetACL http://setacl.sourceforge.net
1107 - Make tree walk routines like cd, ls, ... more user friendly
1108 by handling spaces better.
1109 - Make sure that Bacula rechecks the tape after the 20 min wait.
1110 - Set IO_NOWAIT on Bacula TCP/IP packets.
1111 - Try doing a raw partition backup and restore by mounting a
1113 - From Lars Kellers:
1114 Yes, it would allow to highly automatic the request for new tapes. If a
1115 tape is empty, bacula reads the barcodes (native or simulated), and if
1116 an unused tape is found, it runs the label command with all the
1117 necessary parameters.
1119 By the way can bacula automatically "move" an empty/purged volume say
1120 in the "short" pool to the "long" pool if this pool runs out of volume
1122 - What to do about "list files job=xxx".
1123 - Get and test MySQL 4.0
1124 - Look at how fuser works and /proc/PID/fd that is how Nic found the
1125 file descriptor leak in Bacula.
1126 - Implement WrapCounters in Counters.
1127 - Add heartbeat from FD to SD if hb interval expires.
1128 - Can we dynamically change FileSets?
1129 - If pool specified to label command and Label Format is specified,
1130 automatically generate the Volume name.
1131 - Why can't SQL do the filename sort for restore?
1132 - Add ExhautiveRestoreSearch
1133 - Look at the possibility of loading only the necessary
1134 data into the restore tree (i.e. do it one directory at a
1135 time as the user walks through the tree).
1136 - Possibly use the hash code if the user selects all for a restore command.
1137 - Fix "restore all" to bypass building the tree.
1138 - Prohibit backing up archive device (findlib/find_one.c:128)
1139 - Implement Release Device in the Job resource to unmount a drive.
1140 - Implement Acquire Device in the Job resource to mount a drive,
1141 be sure this works with admin jobs so that the user can get
1142 prompted to insert the correct tape. Possibly some way to say to
1143 run the job but don't save the files.
1144 - Make things like list where a file is saved case independent for
1147 - Use autochanger to handle multiple devices.
1148 - On Windows with very long path names, it may be impossible to create
1149 a file (and thus restore it) because the total length is too long.
1150 We must cd into the directory then create the file without the
1152 - Implement a Recycle command
1153 - Test a second language e.g. french.
1154 - Start working on Base jobs.
1155 - Implement UnsavedFiles DB record.
1156 - From Phil Stracchino:
1157 It would probably be a per-client option, and would be called
1158 something like, say, "Automatically purge obsoleted jobs". What it
1159 would do is, when you successfully complete a Differential backup of a
1160 client, it would automatically purge all Incremental backups for that
1161 client that are rendered redundant by that Differential. Likewise,
1162 when a Full backup on a client completed, it would automatically purge
1163 all Differential and Incremental jobs obsoleted by that Full backup.
1164 This would let people minimize the number of tapes they're keeping on
1165 hand without having to master the art of retention times.
1166 - When doing a Backup send all attributes back to the Director, who
1167 would then figure out what files have been deleted.
1168 - Currently in mount.c:236 the SD simply creates a Volume. It should have
1169 explicit permission to do so. It should also mark the tape in error
1170 if there is an error.
1171 - Cancel waiting for Client connect in SD if FD goes away.
1173 - Implement timeout in response() when it should come quickly.
1174 - Implement a Slot priority (loaded/not loaded).
1175 - Implement "vacation" Incremental only saves.
1176 - Implement create "FileSet"?
1177 - Add prefixlinks to where or not where absolute links to FD.
1178 - Issue message to mount a new tape before the rewind.
1179 - Simplified client job initiation for portables.
1180 - If SD cannot open a drive, make it periodically retry.
1181 - Add more of the config info to the tape label.
1183 - If tape is marked read-only, then try opening it read-only rather than
1184 failing, and remember that it cannot be written.
1185 - Refine SD waiting output:
1186 Device is being positioned
1187 > Device is being positioned for append
1188 > Device is being positioned to file x
1190 - Figure out some way to estimate output size and to avoid splitting
1191 a backup across two Volumes -- this could be useful for writing CDROMs
1192 where you really prefer not to have it split -- not serious.
1193 - Have SD compute MD5 or SHA1 and compare to what FD computes.
1194 - Make VolumeToCatalog calculate an MD5 or SHA1 from the
1195 actual data on the Volume and compare it.
1196 - Implement Bacula plugins -- design API
1197 - Make bcopy read through bad tape records.
1198 - Program files (i.e. execute a program to read/write files).
1199 Pass read date of last backup, size of file last time.
1200 - Add Signature type to File DB record.
1201 - CD into subdirectory when open()ing files for backup to
1202 speed up things. Test with testfind().
1203 - Priority job to go to top of list.
1204 - Why are save/restore of device different sizes (sparse?) Yup! Fix it.
1205 - Implement some way for the Console to dynamically create a job.
1206 - Solaris -I on tar for include list
1207 - Need a verbose mode in restore, perhaps to bsr.
1208 - bscan without -v is too quiet -- perhaps show jobs.
1209 - Add code to reject whole blocks if not wanted on restore.
1210 - Check if we can increase Bacula FD priorty in Win2000
1211 - Make sure the MaxVolFiles is fully implemented in SD
1212 - Check if both CatalogFiles and UseCatalog are set to SD.
1213 - Possibly add email to Watchdog if drive is unmounted too
1214 long and a job is waiting on the drive.
1215 - After unmount, if restore job started, ask to mount.
1216 - Add UA rc and history files.
1217 - put termcap (used by console) in ./configure and
1218 allow -with-termcap-dir.
1219 - Fix Autoprune for Volumes to respect need for full save.
1220 - Compare tape to Client files (attributes, or attributes and data)
1221 - Make all database Ids 64 bit.
1222 - Allow console commands to detach or run in background.
1223 - Fix status delay on storage daemon during rewind.
1224 - Add SD message variables to control operator wait time
1225 - Maximum Operator Wait
1226 - Minimum Message Interval
1227 - Maximum Message Interval
1228 - Send Operator message when cannot read tape label.
1229 - Verify level=Volume (scan only), level=Data (compare of data to file).
1230 Verify level=Catalog, level=InitCatalog
1232 - Add keyword search to show command in Console.
1233 - Events : tape has more than xxx bytes.
1234 - Complete code in Bacula Resources -- this will permit
1235 reading a new config file at any time.
1236 - Handle ctl-c in Console
1237 - Implement script driven addition of File daemon to config files.
1238 - Think about how to make Bacula work better with File (non-tape) archives.
1239 - Write Unix emulator for Windows.
1240 - Put memory utilization in Status output of each daemon
1241 if full status requested or if some level of debug on.
1242 - Make database type selectable by .conf files i.e. at runtime
1243 - Set flag for uname -a. Add to Volume label.
1244 - Restore files modified after date
1245 - SET LD_RUN_PATH=$HOME/mysql/lib/mysql
1246 - Remove duplicate fields from jcr (e.g. jcr.level and jcr.jr.Level, ...).
1247 - Timout a job or terminate if link goes down, or reopen link and query.
1248 - Concept of precious tapes (cannot be reused).
1249 - Make bcopy copy with a single tape drive.
1250 - Permit changing ownership during restore.
1253 > My suggestion: Add a feature on the systray menu-icon menu to request
1254 > an immediate backup now. This would be useful for laptop users who may
1255 > not be on the network when the regular scheduled backup is run.
1257 > My wife's suggestion: Add a setting to the win32 client to allow it to
1258 > shut down the machine after backup is complete (after, of course,
1259 > displaying a "System will shut down in one minute, click here to cancel"
1260 > warning dialog). This would be useful for sites that want user
1261 > woorkstations to be shut down overnight to save power.
1264 - Autolabel should be specified by DIR instead of SD.
1266 - Add media capacity
1267 - AutoScan (check checksum of tape)
1268 - Format command = "format /dev/nst0"
1272 - Seek resolution (usually corresponds to buffer size)
1273 - EODErrorCode=ENOSPC or code
1274 - Partial Read error code
1275 - Partial write error code
1276 - Nonformatted read error
1277 - Nonformatted write error
1278 - WriteProtected error
1282 - IgnoreCloseErrors=yes
1292 - FD sends unsaved file list to Director at end of job (see
1294 - File daemon should build list of files skipped, and then
1295 at end of save retry and report any errors.
1296 - Write a Storage daemon that uses pipes and
1297 standard Unix programs to write to the tape.
1299 - Need something that monitors the JCR queue and
1300 times out jobs by asking the deamons where they are.
1301 - Enhance Jmsg code to permit buffering and saving to disk.
1302 - device driver = "xxxx" for drives.
1303 - Verify from Volume
1304 - Ensure that /dev/null works
1305 - Need report class for messages. Perhaps
1306 report resource where report=group of messages
1307 - enhance scan_attrib and rename scan_jobtype, and
1308 fill in code for "since" option
1309 - Director needs a time after which the report status is sent
1310 anyway -- or better yet, a retry time for the job.
1311 - Don't reschedule a job if previous incarnation is still running.
1312 - Some way to automatically backup everything is needed????
1313 - Need a structure for pending actions:
1315 - termination status (part of buffered msgs?)
1317 Read, Write, Clean, Delete
1318 - Login to Bacula; Bacula users with different permissions:
1319 owner, group, user, quotas
1320 - Store info on each file system type (probably in the job header on tape.
1321 This could be the output of df; or perhaps some sort of /etc/mtab record.
1323 ========= ideas ===============
1324 From: "Jerry K. Schieffer" <jerry@skylinetechnology.com>
1325 To: <kern@sibbald.com>
1326 Subject: RE: [Bacula-users] future large programming jobs
1327 Date: Thu, 26 Feb 2004 11:34:54 -0600
1329 I noticed the subject thread and thought I would offer the following
1330 merely as sources of ideas, i.e. something to think about, not even as
1331 strong as a request. In my former life (before retiring) I often
1332 dealt with backups and storage management issues/products as a
1333 developer and as a consultant. I am currently migrating my personal
1334 network from amanda to bacula specifically because of the ability to
1335 cross media boundaries during storing backups.
1336 Are you familiar with the commercial product called ADSM (I think IBM
1337 now sells it under the Tivoli label)? It has a couple of interesting
1338 ideas that may apply to the following topics.
1340 1. Migration: Consider that when you need to restore a system, there
1341 may be pressure to hurry. If all the information for a single client
1342 can eventually end up on the same media (and in chronological order),
1343 the restore is facillitated by not having to search past information
1344 from other clients. ADSM has the concept of "client affinity" that
1345 may be associated with it's storage pools. It seems to me that this
1346 concept (as an optional feature) might fit in your architecture for
1349 ADSM also has the concept of defining one or more storage pools as
1350 "copy pools" (almost mirrors, but only in the sense of contents).
1351 These pools provide the ability to have duplicte data stored both
1352 onsite and offsite. The copy process can be scheduled to be handled
1353 by their storage manager during periods when there is no backup
1354 activity. Again, the migration process might be a place to consider
1355 implementing something like this.
1358 > It strikes me that it would be very nice to be able to do things
1360 > have the Job(s) backing up the machines run, and once they have all
1361 > completed, start a migration job to copy the data from disks Volumes
1363 > a tape library and then to offsite storage. Maybe this can already
1365 > done with some careful scheduling and Job prioritzation; the events
1366 > mechanism described below would probably make it very easy.
1368 This is the goal. In the first step (before events), you simply
1370 the Migration to tape later.
1372 2. Base jobs: In ADSM, each copy of each stored file is tracked in
1373 the database. Once a file (unique by path and metadata such as dates,
1374 size, ownership, etc.) is in a copy pool, no more copies are made. In
1375 other words, when you start ADSM, it begins like your concept of a
1376 base job. After that it is in the "incremental" mode. You can
1377 configure the number of "generations" of files to be retained, plus a
1378 retention date after which even old generations are purged. The
1379 database tracks the contents of media and projects the percentage of
1380 each volume that is valid. When the valid content of a volume drops
1381 below a configured percentage, the valid data are migrated to another
1382 volume and the old volume is marked as empty. Note, this requires
1383 ADSM to have an idea of the contents of a client, i.e. marking the
1384 database when an existing file was deleted, but this would solve your
1385 issue of restoring a client without restoring deleted files.
1387 This is pretty far from what bacula now does, but if you are going to
1388 rip things up for Base jobs,.....
1389 Also, the benefits of this are huge for very large shops, especially
1390 with media robots, but are a pain for shops with manual media
1394 > Base jobs sound pretty useful, but I'm not dying for them.
1396 Nobody is dying for them, but when you see what it does, you will die
1399 3. Restoring deleted files: Since I think my comments in (2) above
1400 have low probability of implementation, I'll also suggest that you
1401 could approach the issue of deleted files by a mechanism of having the
1402 fd report to the dir, a list of all files on the client for every
1403 backup job. The dir could note in the database entry for each file
1404 the date that the file was seen. Then if a restore as of date X takes
1405 place, only files that exist from before X until after X would be
1406 restored. Probably the major cost here is the extra date container in
1407 each row of the files table.
1409 Thanks for "listening". I hope some of this helps. If you want to
1410 contact me, please send me an email - I read some but not all of the
1411 mailing list traffic and might miss a reply there.
1413 Please accept my compliments for bacula. It is doing a great job for
1414 me!! I sympathize with you in the need to wrestle with excelence in
1415 execution vs. excelence in feature inclusion.
1420 ==============================
1423 - Design at hierarchial storage for Bacula. Migration and Clone.
1424 - Implement FSM (File System Modules).
1425 - Audit M_ error codes to ensure they are correct and consistent.
1426 - Add variable break characters to lex analyzer.
1427 Either a bit mask or a string of chars so that
1428 the caller can change the break characters.
1429 - Make a single T_BREAK to replace T_COMMA, etc.
1430 - Ensure that File daemon and Storage daemon can
1431 continue a save if the Director goes down (this
1432 is NOT currently the case). Must detect socket error,
1433 buffer messages for later.
1434 - Enhance time/duration input to allow multiple qualifiers e.g. 3d2h
1435 - Add ability to backup to two Storage devices (two SD sessions) at
1436 the same time -- e.g. onsite, offsite.
1437 - Add the ability to consolidate old backup sets (basically do a restore
1438 to tape and appropriately update the catalog). Compress Volume sets.
1439 Might need to spool via file is only one drive is available.
1440 - Compress or consolidate Volumes of old possibly deleted files. Perhaps
1441 someway to do so with every volume that has less than x% valid
1445 Migration: Move a backup from one Volume to another
1446 Clone: Copy a backup -- two Volumes
1448 Bacula Migration is based on Jobs (apparently Networker is file by file).
1450 Migration triggered by:
1454 Highwater mark (keep total size)
1459 ======================================================
1461 It is somewhat like a Full save becomes an incremental since
1462 the Base job (or jobs) plus other non-base files.
1464 - A Base backup is same as Full backup, just different type.
1465 - New BaseFiles table that contains:
1467 BaseJobId - Base JobId referenced for this FileId (needed ???)
1468 JobId - JobId currently running
1469 FileId - File not backed up, exists in Base Job
1470 FileIndex - FileIndex from Base Job.
1471 i.e. for each base file that exists but is not saved because
1472 it has not changed, the File daemon sends the JobId, BaseId,
1473 FileId, FileIndex back to the Director who creates the DB entry.
1474 - To initiate a Base save, the Director sends the FD
1475 the FileId, and full filename for each file in the Base.
1476 - When the FD finds a Base file, he requests the Director to
1477 send him the full File entry (stat packet plus MD5), or
1478 conversely, the FD sends it to the Director and the Director
1479 says yes or no. This can be quite rapid if the FileId is kept
1480 by the FD for each Base Filename.
1481 - It is probably better to have the comparison done by the FD
1482 despite the fact that the File entry must be sent across the
1484 - An alternative would be to send the FD the whole File entry
1485 from the start. The disadvantage is that it requires a lot of
1486 space. The advantage is that it requires less communications
1488 - The Job record must be updated to indicate that one or more
1490 - At end of Job, FD returns:
1491 1. Count of base files/bytes not written to tape (i.e. matches)
1492 2. Count of base file that were saved i.e. had changed.
1493 - No tape record would be written for a Base file that matches, in the
1494 same way that no tape record is written for Incremental jobs where
1495 the file is not saved because it is unchanged.
1496 - On a restore, all the Base file records must explicitly be
1497 found from the BaseFile tape. I.e. for each Full save that is marked
1498 to have one or more Base Jobs, search the BaseFile for all occurrences
1500 - An optimization might be to make the BaseFile have:
1506 This would avoid the need to explicitly fetch each File record for
1507 the Base job. The Base Job record will be fetched to get the
1508 VolSessionId and VolSessionTime.
1509 =========================================================
1512 ==========================================================
1514 For each Incremental job that is run, there may be files that
1515 were found but not saved because they were locked (this applies
1516 only to Windows). Such a system could send back to the Director
1517 a list of Unsaved files.
1519 - New UnSavedFiles table that contains:
1523 - Then in the next Incremental job, the list of Unsaved Files will be
1524 feed to the FD, who will ensure that they are explicitly chosen even
1525 if standard date/time check would not have selected them.
1526 =============================================================
1530 Multiple drive autochanger data: see Alan Brown
1531 mtx -f xxx unloadStorage Element 1 is Already Full(drive 0 was empty)
1532 Unloading Data Transfer Element into Storage Element 1...source Element
1533 Address 480 is Empty
1535 (drive 0 was empty and so was slot 1)
1536 > mtx -f xxx load 15 0
1537 no response, just returns to the command prompt when complete.
1538 > mtx -f xxx status Storage Changer /dev/changer:2 Drives, 60 Slots ( 2 Import/Export )
1539 Data Transfer Element 0:Full (Storage Element 15 Loaded):VolumeTag = HX001
1540 Data Transfer Element 1:Empty
1541 Storage Element 1:Empty
1542 Storage Element 2:Full :VolumeTag=HX002
1543 Storage Element 3:Full :VolumeTag=HX003
1544 Storage Element 4:Full :VolumeTag=HX004
1545 Storage Element 5:Full :VolumeTag=HX005
1546 Storage Element 6:Full :VolumeTag=HX006
1547 Storage Element 7:Full :VolumeTag=HX007
1548 Storage Element 8:Full :VolumeTag=HX008
1549 Storage Element 9:Full :VolumeTag=HX009
1550 Storage Element 10:Full :VolumeTag=HX010
1551 Storage Element 11:Empty
1552 Storage Element 12:Empty
1553 Storage Element 13:Empty
1554 Storage Element 14:Empty
1555 Storage Element 15:Empty
1556 Storage Element 16:Empty....
1557 Storage Element 28:Empty
1558 Storage Element 29:Full :VolumeTag=CLNU01L1
1559 Storage Element 30:Empty....
1560 Storage Element 57:Empty
1561 Storage Element 58:Full :VolumeTag=NEX261L2
1562 Storage Element 59 IMPORT/EXPORT:Empty
1563 Storage Element 60 IMPORT/EXPORT:Empty
1565 Unloading Data Transfer Element into Storage Element 15...done
1567 (just to verify it remembers where it came from, however it can be
1568 overrriden with mtx unload {slotnumber} to go to any storage slot.)
1570 There needs to be a table of drive # to devices somewhere - If there are
1571 multiple changers or drives there may not be a 1:1 correspondance between
1572 changer drive number and system device name - and depending on the way the
1573 drives are hooked up to scsi busses, they may not be linearly numbered
1574 from an offset point either.something like
1576 Autochanger drives = 2
1577 Autochanger drive 0 = /dev/nst1
1578 Autochanger drive 1 = /dev/nst2
1579 IMHO, it would be _safest_ to use explicit mtx unload commands at all
1580 times, not just for multidrive changers. For a 1 drive changer, that's
1586 MTX's manpage (1.2.15):
1587 unload [<slotnum>] [ <drivenum> ]
1588 Unloads media from drive <drivenum> into slot
1589 <slotnum>. If <drivenum> is omitted, defaults to
1590 drive 0 (as do all commands). If <slotnum> is
1591 omitted, defaults to the slot that the drive was
1592 loaded from. Note that there's currently no way
1593 to say 'unload drive 1's media to the slot it
1594 came from', other than to explicitly use that
1595 slot number as the destination.AB
1601 undef# camcontrol devlist
1602 <WANGTEK 51000 SCSI M74H 12B3> at scbus0 target 2 lun 0 (pass0,sa0)
1603 <ARCHIVE 4586XX 28887-XXX 4BGD> at scbus0 target 4 lun 0 (pass1,sa1)
1604 <ARCHIVE 4586XX 28887-XXX 4BGD> at scbus0 target 4 lun 1 (pass2)
1606 tapeinfo -f /dev/sg0 with a bad tape in drive 1:
1607 [kern@rufus mtx-1.2.17kes]$ ./tapeinfo -f /dev/sg0
1608 Product Type: Tape Drive
1610 Product ID: 'C5713A '
1612 Attached Changer: No
1613 TapeAlert[3]: Hard Error: Uncorrectable read/write error.
1614 TapeAlert[20]: Clean Now: The tape drive neads cleaning NOW.
1621 Medium Type: Not Loaded
1624 DataCompEnabled: yes
1625 DataCompCapable: yes
1626 DataDeCompEnabled: yes
1633 Handling removable disks
1635 From: Karl Cunningham <karlc@keckec.com>
1637 My backups are only to hard disk these days, in removable bays. This is my
1638 idea of how a backup to hard disk would work more smoothly. Some of these
1639 things Bacula does already, but I mention them for completeness. If others
1640 have better ways to do this, I'd like to hear about it.
1642 1. Accommodate several disks, rotated similar to how tapes are. Identified
1643 by partition volume ID or perhaps by the name of a subdirectory.
1644 2. Abort & notify the admin if the wrong disk is in the bay.
1645 3. Write backups to different subdirectories for each machine to be backed
1647 4. Volumes (files) get created as needed in the proper subdirectory, one
1649 5. When a disk is recycled, remove or zero all old backup files. This is
1650 important as the disk being recycled may be close to full. This may be
1651 better done manually since the backup files for many machines may be
1652 scattered in many subdirectories.
1657 - Make sure that all do_prompt() calls in Dir check for
1658 -1 (error) and -2 (cancel) returns.
1659 - Fix foreach_jcr() to have free_jcr() inside next().
1660 jcr=jcr_walk_start();
1661 for ( ; jcr; (jcr=jcr_walk_next(jcr)) )
1664 - A Volume taken from Scratch should take on the retention period
1666 - Correct doc for Maximum Changer Wait (and others) accepting only
1668 - Implement status that shows why a job is being held in reserve, or
1669 rather why none of the drives are suitable.
1670 - Implement a way to disable a drive (so you can use the second
1671 drive of an autochanger, and the first one will not be used or
1673 - Make sure Maximum Volumes is respected in Pools when adding
1674 Volumes (e.g. when pulling a Scratch volume).
1675 - Keep same dcr when switching device ...
1676 - Implement code that makes the Dir aware that a drive is an
1677 autochanger (so the user doesn't need to use the Autochanger = yes
1679 - Make catalog respect ACL.
1680 - Add recycle count to Media record.
1681 - Add initial write date to Media record.
1682 - Fix store_yesno to be store_bitmask.
1683 --- create_file.c.orig Fri Jul 8 12:13:05 2005
1684 +++ create_file.c Fri Jul 8 12:13:07 2005
1686 attr->ofname, be.strerror());
1689 + } else if(S_ISSOCK(attr->statp.st_mode)) {
1690 + Dmsg1(200, "Skipping socket: %s\n", attr->ofname);
1692 Dmsg1(200, "Restore node: %s\n", attr->ofname);
1693 if (mknod(attr->ofname, attr->statp.st_mode, attr->statp.st_rdev) != 0 && errno != EEXIST) {
1694 - Add true/false to conf same as yes/no
1695 - Reserve blocks other restore jobs when first cannot connect to SD.
1696 - Fix Maximum Changer Wait, Maximum Open Wait, Maximum Rewind Wait to
1697 accept time qualifiers.
1698 - Does ClientRunAfterJob fail the job on a bad return code?
1699 - Make hardlink code at line 240 of find_one.c use binary search.
1700 - Add ACL error messages in src/filed/acl.c.
1701 - Make authentication failures single threaded.
1702 - Make Dir and SD authentication errors single threaded.
1703 - Fix catreq.c digestbuf at line 411 in src/dird/catreq.c
1704 - Make base64.c (bin_to_base64) take a buffer length
1705 argument to avoid overruns.
1706 and verify that other buffers cannot overrun.
1707 - Implement VolumeState as discussed with Arno.
1708 - Add LocationId to update volume
1716 - Add Comment to Media record
1717 - Fix auth compatibility with 1.38
1718 - Update dbcheck to include Log table
1719 - Update llist to include new fields.
1720 - Make unmount unload autochanger. Make mount load slot.
1721 - Fix bscan to report the JobType when restoring a job.
1722 - Fix wx-console scanning problem with commas in names.
1723 - Add manpages to the list of directories for make install. Notify
1725 - Add bconsole option to use stdin/out instead of conio.
1726 - Fix ClientRunBefore/AfterJob compatibility.
1727 - Ensure that connection to daemon failure always indicates what
1728 daemon it was trying to connect to.
1729 - Freespace on DVD requested over and over even with no intervening
1731 - .update volume [enabled|disabled|*see below]
1732 > However, I could easily imagine an option to "update slots" that says
1733 > "enable=yes|no" that would automatically enable or disable all the Volumes
1734 > found in the autochanger. This will permit the user to optionally mark all
1735 > the Volumes in the magazine disabled prior to taking them offsite, and mark
1736 > them all enabled when bringing them back on site. Coupled with the options
1737 > to the slots keyword, you can apply the enable/disable to any or all volumes.
1738 - Restricted consoles start in the Default catalog even if it
1740 - When reading through parts on the DVD, the DVD is mounted and
1741 unmounted for each part.
1742 - Make sure that the restore options don't permit "seeing" other
1744 - Restore of a raw drive should not try to check the volume size.
1745 - Lock tape drive door when open()
1746 - Make release unload any autochanger.
1747 - Arno's reservation deadlock.
1749 - Make sure the new level=Full syntax is used in all
1750 example conf files (especially in the manual).
1751 - Fix prog copyright (SD) all other files.
1752 - Document need for UTF-8 format