6 gftp sshfs kile kate lsssci m4 mtx nfs-common nfs-server
7 patch squashfs-tools strace sg3-utils screen scsiadd
8 system-tools-backend telnet dpkg traceroute urar usbutils
9 whois apt-file autofs busybox chkrootkit clamav dmidecode
10 manpages-dev manpages-posix manpages-posix-dev
14 - package sg3-utils, program sg_map
15 - !!! Cannot restore two jobs a the same time that were
16 written simultaneously unless they were totally spooled.
17 - Document cleaning up the spool files:
18 db, pid, state, bsr, mail, conmsg, spool
19 - Document the multiple-drive-changer.txt script.
20 - Pruning with Admin job.
21 - Does WildFile match against full name? Doc.
22 - %d and %v only valid on Director, not for ClientRunBefore/After.
23 - During tests with the 260 char fix code, I found one problem:
24 if the system "sees" a long path once, it seems to forget it's
25 working drive (e.g. c:\), which will lead to a problem during
26 the next job (create bootstrap file will fail). Here is the
27 workaround: specify absolute working and pid directory in
28 bacula-fd.conf (e.g. c:\bacula\working instead of
30 - Document techniques for restoring large numbers of files.
31 - Document setting my.cnf to big file usage.
32 - Correct the Include syntax in the m4.xxx files in examples/conf
33 - Document all the little details of setting up certificates for
34 the Bacula data encryption code.
35 - Document more precisely how to use master keys -- especially
36 for disaster recovery.
40 - Possibly allow SD to spool even if a tape is not mounted.
41 - How to sync remote offices.
43 - Backup up windows system state
44 - Complete Job restart
45 - Look at rsysnc for incremental updates and dedupping
46 - Implement rwlock() for SD that takes why and can_steal to replace
47 existing block/lock mechanism. rlock() would allow multiple readers
48 wlock would allow only one writer.
49 - For Windows disaster recovery see http://unattended.sf.net/
50 - Add "before=" "olderthan=" to FileSet for doing Base of
52 - Show files/second in client status output.
53 - Don't attempt to restore from "Disabled" Volumes.
54 - Have SD compute MD5 or SHA1 and compare to what FD computes.
55 - Make VolumeToCatalog calculate an MD5 or SHA1 from the
56 actual data on the Volume and compare it.
57 - Remove queue.c code.
58 - Implement multiple jobid specification for the cancel command,
59 similar to what is permitted on the update slots command.
60 - Ensure that the SD re-reads the Media record if the JobFiles
61 does not match -- it may have been updated by another job.
62 - Add MD5 or SHA1 check in SD for data validation
63 - When reserving a device to read, check to see if the Volume
64 is already in use, if so wait. Probably will need to pass the
65 Volume. See bug #1313. Create a regression test to simulate
66 this problem and see if VolumePollInterval fixes it. Possibly turn
71 - Why no error message if restore has no permission on the where
73 - Possibly allow manual "purge" to purge a Volume that has not
74 yet been written (even if FirstWritten time is zero) see ua_purge.c
76 - Add disk block detection bsr code (make it work).
78 - Detect deadlocks in reservations.
80 - Add list during dump
81 - Add in plugin code flag
82 - Add bRC_EndJob -- stops more calls to plugin this job
83 - Add bRC_Term (unload plugin)
84 - remove time_t from Jmsg and use utime_t?
85 - Deadlock detection, watchdog sees if counter advances when jobs are
86 running. With debug on, can do a "status" command.
87 - User options for plugins.
88 - Pool Storage override precedence over command line.
89 - Autolabel only if Volume catalog information indicates tape not
90 written. This will avoid overwriting a tape that gets an I/O
91 error on reading the volume label.
92 - I/O error, SD thinks it is not the right Volume, should check slot
93 then disable volume, but Asks for mount.
94 - Can be posible modify package to create and use configuration files in
99 /etc/bacula/bacula-dir.conf
100 /etc/bacula/conf.d/pools.conf
101 /etc/bacula/conf.d/clients.conf
102 /etc/bacula/conf.d/storages.conf
104 and into bacula-dir.conf file include
106 @/etc/bacula/conf.d/pools.conf
107 @/etc/bacula/conf.d/clients.conf
108 @/etc/bacula/conf.d/storages.conf
109 - Possibly add an Inconsistent state when a Volume is in error
111 - Fix #ifdefing so that smartalloc can be disabled. Check manual
112 -- the default is enabled.
113 - Dangling softlinks are not restored properly. For example, take a
114 soft link such as src/testprogs/install-sh, which points to /usr/share/autoconf...
115 move the directory to another machine where the file /usr/share/autoconf does
116 not exist, back it up, then try a full restore. It fails.
117 - Softlinks that point to non-existent file are not restored in restore all,
118 but are restored if the file is individually selected. BUG!
120 - Prune by Job Level (Full, Differential, Incremental)
121 - modify pruning to keep a fixed number of versions of a file,
123 - the cd-command should allow complete paths
124 i.e. cd /foo/bar/foo/bar
125 -> if a customer mails me the path to a certain file,
126 its faster to enter the specified directory
127 - Make tree walk routines like cd, ls, ... more user friendly
128 by handling spaces better.
129 - When doing a restore, if the user does an "update slots"
130 after the job started in order to add a restore volume, the
131 values prior to the update slots will be put into the catalog.
132 Must retrieve catalog record merge it then write it back at the
133 end of the restore job, if we want to do this right.
137 MA = (last_MA * 3 + rate) / 4
138 rate = (bytes - last_bytes) / (runtime - last_runtime)
140 - Add a recursive mark command (rmark) to restore.
141 - Look at simplifying File exclusions.
143 - Separate Files and Directories in catalog
144 - Create FileVersions table
145 - finish implementation of fdcalled -- see ua_run.c:105
146 - Fix problem in postgresql.c in my_postgresql_query, where the
147 generation of the error message doesn't differentiate result==NULL
148 and a bad status from that result. Not only that, the result is
149 cleared on a bail_out without having generated the error message.
150 - Implement SDErrors (must return from SD)
151 - Implement continue spooling while despooling.
152 - Remove all install temp files in Win32 PLUGINSDIR.
153 - No where in restore causes kaboom.
154 - Performance: multiple spool files for a single job.
155 - Performance: despool attributes when despooling data (problem
156 multiplexing Dir connection).
157 - Implement wait_for_sysop() message display in wait_for_device(), which
158 now prints warnings too often.
159 - Ensure that each device in an Autochanger has a different
161 - Look at sg_logs -a /dev/sg0 for getting soft errors.
162 - btape "test" command with Offline on Unmount = yes
164 This test is essential to Bacula.
166 I'm going to write one record in file 0,
167 two records in file 1,
168 and three records in file 2
170 02-Feb 11:00 btape: ABORTING due to ERROR in dev.c:715
171 dev.c:714 Bad call to rewind. Device "LTO" (/dev/nst0) not open
172 02-Feb 11:00 btape: Fatal Error because: Bacula interrupted by signal 11: Segmentation violation
173 Kaboom! btape, btape got signal 11. Attempting traceback.
175 - Encryption -- email from Landon
176 > The backup encryption algorithm is currently not configurable, and is
177 > set to AES_128_CBC in src/filed/backup.c. The encryption code
178 > supports a number of different ciphers (as well as adding arbitrary
179 > new ones) -- only a small bit of code would be required to map a
180 > configuration string value to a CRYPTO_CIPHER_* value, if anyone is
181 > interested in implementing this functionality.
183 - Add the OS version back to the Win32 client info.
184 - Restarted jobs have a NULL in the from field.
185 - Modify SD status command to indicate when the SD is writing
186 to a DVD (the device is not open -- see bug #732).
187 - Look at the possibility of adding "SET NAMES UTF8" for MySQL,
188 and possibly changing the blobs into varchar.
189 - Test Volume compatibility between machine architectures
190 - Encryption documentation
193 - Migration from other vendors
197 - Backup conf/exe (all daemons)
198 - Detect state change of system (verify)
200 - Novell NSS backup http://www.novell.com/coolsolutions/tools/18952.html
201 - Compliance norms that compare restored code hash code.
204 Extract capability (#25)
205 Threshold triggered migration jobs (not currently in list, but will be
207 Client triggered backups
208 Complete rework of the scheduling system (not in list)
209 Performance and usage instrumentation (not in list)
210 See email of 21Aug2007 for details.
211 - Look at: http://tech.groups.yahoo.com/group/cfg2html
212 and http://www.openeyet.nl/scc/ for managing customer changes
216 - Access Mode = Read-Only, Read-Write, Unavailable, Destroyed, Offsite
218 - Maximum number of scratch volumes
220 - Next Pool (already have)
221 - Reclamation threshold
223 - Reuse delay (after all files purged from volume before it can be used)
224 - Copy Pool = xx, yyy (or multiple lines).
226 - Allow pool selection during restore.
228 - Average tape size from Eric
229 SELECT COALESCE(media_avg_size.volavg,0) * count(Media.MediaId) AS volmax, GROUP BY Media.MediaType, Media.PoolId, media_avg_size.volavg
230 count(Media.MediaId) AS volnum,
231 sum(Media.VolBytes) AS voltotal,
232 Media.PoolId AS PoolId,
233 Media.MediaType AS MediaType
235 LEFT JOIN (SELECT avg(Media.VolBytes) AS volavg,
236 Media.MediaType AS MediaType
238 WHERE Media.VolStatus = 'Full'
239 GROUP BY Media.MediaType
240 ) AS media_avg_size ON (Media.MediaType = media_avg_size.MediaType)
241 GROUP BY Media.MediaType, Media.PoolId, media_avg_size.volavg
243 - Despool attributes in separate thread
246 - Check why restore repeatedly sends Rechdrs between
247 each data chunk -- according to James Harper 9Jan07.
250 - More intelligent re-run
251 - Incremental backup -- rsync, Stow
253 - Make Bacula by default not backup tmpfs, procfs, sysfs, ...
254 - Fix hardlinked immutable files when linking a second file, the
255 immutable flag must be removed prior to trying to link it.
256 - Change dbcheck to tell users to use native tools for fixing
257 broken databases, and to ensure they have the proper indexes.
258 - add udev rules for Bacula devices.
259 - If a job terminates, the DIR connection can close before the
260 Volume info is updated, leaving the File count wrong.
261 - Look at why SIGPIPE during connection can cause seg fault in
262 writing the daemon message, when Dir dropped to bacula:bacula
263 - Look at zlib 32 => 64 problems.
264 - Fix bextract to restore ACLs, or better yet, use common routines.
265 - New dot commands from Arno.
266 .show device=xxx lists information from one storage device, including
267 devices (I'm not even sure that information exists in the DIR...)
268 .move eject device=xxx mostly the same as 'unmount xxx' but perhaps with
269 better machine-readable output like "Ok" or "Error busy"
270 .move eject device=xxx toslot=yyy the same as above, but with a new
271 target slot. The catalog should be updated accordingly.
272 .move transfer device=xxx fromslot=yyy toslot=zzz
275 - Article: http://www.heise.de/open/news/meldung/83231
276 - Article: http://www.golem.de/0701/49756.html
277 - Article: http://lwn.net/Articles/209809/
278 - Article: http://www.onlamp.com/pub/a/onlamp/2004/01/09/bacula.html
279 - Article: http://www.linuxdevcenter.com/pub/a/linux/2005/04/07/bacula.html
280 - Article: http://www.osreviews.net/reviews/admin/bacula
281 - Article: http://www.debianhelp.co.uk/baculaweb.htm
283 - Wikis mentioning Bacula
284 http://wiki.finkproject.org/index.php/Admin:Backups
285 http://wiki.linuxquestions.org/wiki/Bacula
286 http://www.openpkg.org/product/packages/?package=bacula
287 http://www.iterating.com/products/Bacula
288 http://net-snmp.sourceforge.net/wiki/index.php/Net-snmp_extensions
289 http://www.section6.net/wiki/index.php/Using_Bacula_for_Tape_Backups
290 http://bacula.darwinports.com/
291 http://wiki.mandriva.com/en/Releases/Corporate/Server_4/Notes#Bacula
292 http://en.wikipedia.org/wiki/Bacula
295 http://www.devco.net/pubwiki/Bacula/
296 http://paramount.ind.wpi.edu/wiki/doku.php
297 http://gentoo-wiki.com/HOWTO_Backup
298 http://www.georglutz.de/wiki/Bacula
299 http://www.clarkconnect.com/wiki/index.php?title=Modules_-_LAN_Backup/Recovery
300 http://linuxwiki.de/Bacula (in German)
302 - Figure out how to configure query.sql. Suggestion to use m4:
303 == changequote.m4 ===
304 changequote(`[',`]')dnl
305 ==== query.sql.in ===
306 :List next 20 volumes to expire
308 Pool.Name AS PoolName,
313 [ FROM_UNIXTIME(UNIX_TIMESTAMP(Media.LastWritten) Media.VolRetention) AS Expire, ])dnl
315 [ media.lastwritten + interval '1 second' * media.volretention as expire, ])dnl
319 ON Media.PoolId=Pool.PoolId
320 WHERE Media.LastWritten>0
324 Command: m4 -DmySQL changequote.m4 query.sql.in >query.sql
326 The problem is that it requires m4, which is not present on all machines
331 select FilenameId from Filename where Name='';
332 # Get list of all directories referenced in a Backup.
333 select Path.Path from Path,File where File.JobId=nnn and
334 File.FilenameId=(FilenameId-from-above) and File.PathId=Path.PathId
335 order by Path.Path ASC;
337 - Mount on an Autochanger with no tape in the drive causes:
338 Automatically selected Storage: LTO-changer
339 Enter autochanger drive[0]: 0
340 3301 Issuing autochanger "loaded drive 0" command.
341 3302 Autochanger "loaded drive 0", result: nothing loaded.
342 3301 Issuing autochanger "loaded drive 0" command.
343 3302 Autochanger "loaded drive 0", result: nothing loaded.
344 3902 Cannot mount Volume on Storage Device "LTO-Drive1" (/dev/nst0) because:
345 Couldn't rewind device "LTO-Drive1" (/dev/nst0): ERR=dev.c:678 Rewind error on "LTO-Drive1" (/dev/nst0). ERR=No medium found.
346 3905 Device "LTO-Drive1" (/dev/nst0) open but no Bacula volume is mounted.
347 If this is not a blank tape, try unmounting and remounting the Volume.
348 - If Drive 0 is blocked, and drive 1 is set "Autoselect=no", drive 1 will
350 - Autochanger did not change volumes.
351 select * from Storage;
352 +-----------+-------------+-------------+
353 | StorageId | Name | AutoChanger |
354 +-----------+-------------+-------------+
355 | 1 | LTO-changer | 0 |
356 +-----------+-------------+-------------+
357 05-May 03:50 roxie-sd: 3302 Autochanger "loaded drive 0", result is Slot 11.
358 05-May 03:50 roxie-sd: Tibs.2006-05-05_03.05.02 Warning: Director wanted Volume "LT
359 Current Volume "LT0-002" not acceptable because:
360 1997 Volume "LT0-002" not in catalog.
361 05-May 03:50 roxie-sd: Tibs.2006-05-05_03.05.02 Error: Autochanger Volume "LT0-002"
362 Setting InChanger to zero in catalog.
363 05-May 03:50 roxie-dir: Tibs.2006-05-05_03.05.02 Error: Unable to get Media record
365 05-May 03:50 roxie-sd: Tibs.2006-05-05_03.05.02 Fatal error: Error getting Volume i
366 05-May 03:50 roxie-sd: Tibs.2006-05-05_03.05.02 Fatal error: Job 530 canceled.
367 05-May 03:50 roxie-sd: Tibs.2006-05-05_03.05.02 Fatal error: spool.c:249 Fatal appe
368 05-May 03:49 Tibs: Tibs.2006-05-05_03.05.02 Fatal error: c:\cygwin\home\kern\bacula
377 FirstWritten: 2006-05-05 03:11:54
378 LastWritten: 2006-05-05 03:50:23
379 LabelDate: 2005-12-26 16:52:40
390 VolRetention: 31,536,000
402 Note VolStatus is blank!!!!!
409 FirstWritten: 0000-00-00 00:00:00
410 LastWritten: 0000-00-00 00:00:00
411 LabelDate: 2005-12-26 16:52:40
422 VolRetention: 31,536,000
435 Automatically selected Storage: LTO-changer
436 Enter autochanger drive[0]: 0
437 3301 Issuing autochanger "loaded drive 0" command.
438 3302 Autochanger "loaded drive 0", result: nothing loaded.
439 3301 Issuing autochanger "loaded drive 0" command.
440 3302 Autochanger "loaded drive 0", result: nothing loaded.
441 3902 Cannot mount Volume on Storage Device "LTO-Drive1" (/dev/nst0) because:
442 Couldn't rewind device "LTO-Drive1" (/dev/nst0): ERR=dev.c:678 Rewind error on "LTO-Drive1" (/dev/nst0). ERR=No medium found.
444 3905 Device "LTO-Drive1" (/dev/nst0) open but no Bacula volume is mounted.
445 If this is not a blank tape, try unmounting and remounting the Volume.
447 - http://www.dwheeler.com/essays/commercial-floss.html
448 - Add VolumeLock to prevent all but lock holder (SD) from updating
449 the Volume data (with the exception of VolumeState).
450 - The btape fill command does not seem to use the Autochanger
451 - Make Windows installer default to system disk drive.
452 - Look at using ioctl(FIOBMAP, ...) on Linux, and
453 DeviceIoControl(..., FSCTL_QUERY_ALLOCATED_RANGES, ...) on
454 Win32 for sparse files.
455 http://www.flexhex.com/docs/articles/sparse-files.phtml
456 http://www.informatik.uni-frankfurt.de/~loizides/reiserfs/fibmap.html
457 - Directive: at <event> "command"
458 - Command: pycmd "command" generates "command" event. How to
459 attach to a specific job?
460 - run_cmd() returns int should return JobId_t
461 - get_next_jobid_from_list() returns int should return JobId_t
462 - Document export LDFLAGS=-L/usr/lib64
463 - Network error on Win32 should set Win32 error code.
464 - What happens when you rename a Disk Volume?
465 - Job retention period in a Pool (and hence Volume). The job would
467 - Add Win32 FileSet definition somewhere
468 - Look at fixing restore status stats in SD.
469 - Look at using ioctl(FIMAP) and FIGETBSZ for sparse files.
470 http://www.informatik.uni-frankfurt.de/~loizides/reiserfs/fibmap.html
471 - Implement a mode that says when a hard read error is
472 encountered, read many times (as it currently does), and if the
473 block cannot be read, skip to the next block, and try again. If
474 that fails, skip to the next file and try again, ...
476 create table LevelType (LevelType binary(1), LevelTypeLong tinyblob);
477 insert into LevelType (LevelType,LevelTypeLong) values
481 - new pool XXX with ScratchPoolId = MyScratchPool's PoolId and
482 let it fill itself, and RecyclePoolId = XXX's PoolId so I can
483 see if it become stable and I just have to supervise
485 - If I want to remove this pool, I set RecyclePoolId = MyScratchPool's
486 PoolId, and when it is empty remove it.
488 - Allow Check Labels to be used with Bacula labels.
489 - "Resuming" a failed backup (lost line for example) by using the
490 failed backup as a sort of "base" job.
492 - Email to the user when the tape is about to need changing x
493 days before it needs changing.
494 - Command to show next tape that will be used for a job even
495 if the job is not scheduled.
496 - From: Arunav Mandal <amandal@trolltech.com>
497 1. When jobs are running and bacula for some reason crashes or if I do a
498 restart it remembers and jobs it was running before it crashed or restarted
499 as of now I loose all jobs if I restart it.
501 2. When spooling and in the midway if client is disconnected for instance a
502 laptop bacula completely discard the spool. It will be nice if it can write
503 that spool to tape so there will be some backups for that client if not all.
505 3. We have around 150 clients machines it will be nice to have a option to
506 upgrade all the client machines bacula version automatically.
508 4. Atleast one connection should be reserved for the bconsole so at heavy load
509 I should connect to the director via bconsole which at sometimes I can't
511 5. Another most important feature that is missing, say at 10am I manually
512 started backup of client abc and it was a full backup since client abc has
513 no backup history and at 10.30am bacula again automatically started backup of
514 client abc as that was in the schedule. So now we have 2 multiple Full
515 backups of the same client and if we again try to start a full backup of
516 client backup abc bacula won't complain. That should be fixed.
518 - regardless of the retention period, Bacula will not prune the
519 last Full, Diff, or Inc File data until a month after the
520 retention period for the last Full backup that was done.
521 - update volume=xxx --- add status=Full
522 - Remove old spool files on startup.
523 - Exclude SD spool/working directory.
524 - Refuse to prune last valid Full backup. Same goes for Catalog.
526 - Make a callback when Rerun failed levels is called.
527 - Give Python program access to Scheduled jobs.
528 - Add setting Volume State via Python.
529 - Python script to save with Python, not save, save with Bacula.
530 - Python script to do backup.
532 - Change the Priority, Client, Storage, JobStatus (error)
533 at the start of a job.
534 - Why is SpoolDirectory = /home/bacula/spool; not reported
535 as an error when writing a DVD?
536 - Make bootstrap file handle multiple MediaTypes (SD)
537 - Remove all old Device resource code in Dir and code to pass it
538 back in SD -- better, rework it to pass back device statistics.
539 - Check locking of resources -- be sure to lock devices where previously
540 resources were locked.
541 - The last part is left in the spool dir.
544 - In restore don't compare byte count on a raw device -- directory
545 entry does not contain bytes.
548 - Max Vols limit in Pool off by one?
549 - Implement Files/Bytes,... stats for restore job.
550 - Implement Total Bytes Written, ... for restore job.
551 - Despool attributes simultaneously with data in a separate
552 thread, rejoined at end of data spooling.
553 - 7. Implement new Console commands to allow offlining/reserving drives,
554 and possibly manipulating the autochanger (much asked for).
555 - Add start/end date editing in messages (%t %T, %e?) ...
556 - Add ClientDefs similar to JobDefs.
557 - Print more info when bextract -p accepts a bad block.
558 - Fix FD JobType to be set before RunBeforeJob in FD.
559 - Look at adding full Volume and Pool information to a Volume
560 label so that bscan can get *all* the info.
561 - If the user puts "Purge Oldest Volume = yes" or "Recycle Oldest Volume = yes"
562 and there is only one volume in the pool, refuse to do it -- otherwise
563 he fills the Volume, then immediately starts reusing it.
564 - Implement copies and stripes.
565 - Add history file to console.
566 - Each file on tape creates a JobMedia record. Peter has 4 million
567 files spread over 10000 tape files and four tapes. A restore takes
568 16 hours to build the restore list.
569 - Add and option to check if the file size changed during backup.
570 - Make sure SD deletes spool files on error exit.
571 - Delete old spool files when SD starts.
572 - When labeling tapes, if you enter 000026, Bacula uses
573 the tape index rather than the Volume name 000026.
574 - Add offline tape command to Bacula console.
576 Enter MediaId or Volume name: 32
577 Enter new Volume name: DLT-20Dec04
578 Automatically selected Pool: Default
579 Connecting to Storage daemon DLTDrive at 192.168.68.104:9103 ...
580 Sending relabel command from "DLT-28Jun03" to "DLT-20Dec04" ...
581 block.c:552 Write error at 0:0 on device /dev/nst0. ERR=Bad file descriptor.
582 Error writing final EOF to tape. This tape may not be readable.
583 dev.c:1207 ioctl MTWEOF error on /dev/nst0. ERR=Permission denied.
584 askdir.c:219 NULL Volume name. This shouldn't happen!!!
585 3912 Failed to label Volume: ERR=dev.c:1207 ioctl MTWEOF error on /dev/nst0. ERR=Permission denied.
586 Label command failed for Volume DLT-20Dec04.
587 Do not forget to mount the drive!!!
588 - Bug: if a job is manually scheduled to run later, it does not appear
589 in any status report and cannot be cancelled.
594 How about introducing a Type = MgmtPolicy job type? That job type would
595 be responsible for scanning the Bacula environment looking for specific
596 conditions, and submitting the appropriate jobs for implementing said
600 Name = "Migration-Policy"
602 Policy Selection Job Type = Migrate
603 Scope = "<keyword> <operator> <regexp>"
604 Threshold = "<keyword> <operator> <regexp>"
605 Job Template = <template-name>
608 Where <keyword> is any legal job keyword, <operator> is a comparison
609 operator (=,<,>,!=, logical operators AND/OR/NOT) and <regexp> is a
610 appropriate regexp. I could see an argument for Scope and Threshold
611 being SQL queries if we want to support full flexibility. The
612 Migration-Policy job would then get scheduled as frequently as a site
613 felt necessary (suggested default: every 15 minutes).
618 Name = "Migration-Policy"
620 Policy Selection Job Type = Migration
622 Threshold = "Migration Selection Type = LowestUtil"
623 Job Template = "MigrationTemplate"
626 would select all pools for examination and generate a job based on
627 MigrationTemplate to automatically select the volume with the lowest
628 usage and migrate it's contents to the nextpool defined for that pool.
630 This policy abstraction would be really handy for adjusting the behavior
631 of Bacula according to site-selectable criteria (one thing that pops
632 into mind is Amanda's ability to automatically adjust backup levels
633 depending on various criteria).
639 - Add Pool/Storage override regression test.
640 - Add delete JobId to regression.
641 - Add a regression test for dbcheck.
642 - New test to add bscan to four-concurrent-jobs regression,
643 i.e. after the four-concurrent jobs zap the
644 database as is done in the bscan-test, then use bscan to
645 restore the database, do a restore and compare with the
647 - Add restore of specific JobId to regression (item 3
648 on the restore prompt)
649 - Add IPv6 to regression
650 - Add database test to regression. Test each function like delete,
653 - AntiVir can slow down backups on Win32 systems.
654 - Win32 systems with FAT32 can be much slower than NTFS for
655 more than 1000 files per directory.
659 - A HOLD command to stop all jobs from starting.
660 - A PAUSE command to pause all running jobs ==> release the
662 - Media Type = LTO,LTO-2,LTO-3
663 Media Type Read = LTO,LTO2,LTO3
664 Media Type Write = LTO2, LTO3
666 === From Carsten Menke <bootsy52@gmx.net>
668 Following is a list of what I think in the situations where I'm faced with,
669 could be a usefull enhancement to bacula, which I'm certain other users will
670 benefit from as well.
672 1. NextJob/NextJobs Directive within a Job Resource in the form of
673 NextJobs = job1,job2.
676 I currently solved the problem with running multiple jobs each after each
677 by setting the Max Wait Time for a job to 8 hours, and give
678 the jobs different Priorities. However, there scenarios where
679 1 Job is directly depending on another job, so if the former job fails,
680 the job after it needn't to be run
681 while maybe other jobs should run despite of that
684 A Backup Job and a Verify job, if the backup job fails there is no need to run
685 the verify job, as the backup job already failed. However, one may like
686 to backup the Catalog to disk despite of that the main backup job failed.
689 I see that this is related to the Event Handlers which are on the ToDo
690 list, also it is maybe a good idea to check for the return value and
691 execute different actions based on the return value
694 3. offline capability to bconsole
697 Currently I use a script which I execute within the last Job via the
698 RunAfterJob Directive, to release and eject the tape.
699 So I have to call bconsole "release=Storage-Name" and afterwards
700 mt -f /dev/nst0 eject to get the tape out.
702 If I have multiple Storage Devices, than these may not be /dev/nst0 and
703 I have to modify the script or call it with parameters etc.
704 This would actually not be needed, as everything is already defined
705 in bacula-sd.conf and if I can invoke bconsole with the
706 storage name via $1 in the script than I'm done and information is
709 4. %s for Storage Name added to the chars being substituted in "RunAfterJob"
713 For the reason mentioned in 3. to have the ability to call a
714 script with /scripts/foobar %s and in the script use $1
715 to pass the Storage Name to bconsole
717 5. Setting Volume State within a Job Resource
720 Instead of using "Maximum Volume Jobs" in the Pool Resource,
721 I would have the possibilty to define
722 in a Job Resource that after this certain job is run, the Volume State
723 should be set to "Volume State = Used", this give more flexibility (IMHO).
725 7. OK, this is evil, probably bound to security risks and maybe not possible
726 due to the design of bacula.
728 Implementation of Backtics ( `command` ) for shell comand execution to
729 the "Label Format" Directive.
733 Currently I have defined BACULA_DAY_OF_WEEK="day1|day2..." resulting in
734 Label Format = "HolyBackup-${BACULA_DAY_OF_WEEK[${WeekDay}]}". If I could
735 use backticks than I could use "Label Format = HolyBackup-`date +%A` to have
736 the localized name for the day of the week appended to the
737 format string. Then I have the tape labeled automatically with weekday
738 name in the correct language.
740 - Make output from status use html table tags for nicely
741 presenting in a browser.
742 - Browse generations of files.
743 - I've seen an error when my catalog's File table fills up. I
744 then have to recreate the File table with a larger maximum row
745 size. Relevant information is at
746 http://dev.mysql.com/doc/mysql/en/Full_table.html ; I think the
747 "Installing and Configuring MySQL" chapter should talk a bit
748 about this potential problem, and recommend a solution.
749 - Want speed of writing to tape while despooling.
750 - Supported autochanger:
758 Wangtek 6525ES (SCSI-1 QIC drive, 525MB), under Linux 2.4.something,
759 bacula 1.36.0/1 works with blocksize 16k INSIDE bacula-sd.conf.
760 - Add regex from http://www.pcre.org to Bacula for Win32.
761 - Use only shell tools no make in CDROM package.
762 - Include within include does it work?
763 - Implement a Pool of type Cleaning?
764 - Think about making certain database errors fatal.
765 - Look at correcting the time jump in the scheduler for daylight
766 savings time changes.
767 - Check dates entered by user for correctness (month/day/... ranges)
768 - Compress restore Volume listing by date and first file.
769 - Look at patches/bacula_db.b2z postgresql that loops during restore.
771 - Perhaps add read/write programs and/or plugins to FileSets.
772 - How to handle backing up portables ...
775 Documentation to do: (any release a little bit at a time)
776 - Doc to do unmount before removing magazine.
777 - Alternative to static linking "ldd prog" save all binaries listed,
778 restore them and point LD_LIBRARY_PATH to them.
779 - Document add "</dev/null >/dev/null 2>&1" to the bacula-fd command line
780 - Document query file format.
781 - Add more documentation for bsr files.
782 - Document problems with Verify and pruning.
783 - Document how to use multiple databases.
784 - VXA drives have a "cleaning required"
785 indicator, but Exabyte recommends preventive cleaning after every 75
788 In this context, it should be noted that Exabyte has a command-line
789 vxatool utility available for free download. (The current version is
790 vxatool-3.72.) It can get diagnostic info, read, write and erase tapes,
791 test the drive, unload tapes, change drive settings, flash new firmware,
793 Of particular interest in this context is that vxatool <device> -i will
794 report, among other details, the time since last cleaning in tape motion
795 minutes. This information can be retrieved (and settings changed, for
796 that matter) through the generic-SCSI device even when Bacula has the
797 regular tape device locked. (Needless to say, I don't recommend
798 changing tape settings while a job is running.)
799 - Lookup HP cleaning recommendations.
800 - Lookup HP tape replacement recommendations (see trouble shooting autochanger)
801 - Document doing table repair
804 ===================================
805 - Add macro expansions in JobDefs.
806 Run Before Job = "SomeFile %{Level} %{Client}"
807 Write Bootstrap="/some/dir/%{JobName}_%{Client}.bsr"
808 - Use non-blocking network I/O but if no data is available, use
810 - Use gather write() for network I/O.
811 - Autorestart on crash.
812 - Add bandwidth limiting.
813 - When an error in input occurs and conio beeps, you can back
814 up through the prompt.
815 - Detect fixed tape block mode during positioning by looking at
816 block numbers in btape "test". Possibly adjust in Bacula.
817 - Fix list volumes to output volume retention in some other
818 units, perhaps via a directive.
819 - If you use restore replace=never, the directory attributes for
820 non-existent directories will not be restored properly.
822 - see lzma401.zip in others directory for new compression
824 - Allow the user to select JobType for manual pruning/purging.
825 - bscan does not put first of two volumes back with all info in
827 - Figure out how to make named console messages go only to that
828 console and to the non-restricted console (new console class?).
829 - Make restricted console prompt for password if *ask* is set or
830 perhaps if password is undefined.
831 - Implement "from ISO-date/time every x hours/days/weeks/months" in
834 ==== from Marc Schoechlin
835 - the help-command should be more verbose
836 (it should explain the paramters of the different
838 -> its time-comsuming to consult the manual anytime
839 you need a special parameter
840 -> maybe its more easy to maintain this, if the
841 descriptions of that commands are outsourced to
843 - if the password is not configured in bconsole.conf
844 you should be asked for it.
845 -> sometimes you like to do restore on a customer-machine
846 which shouldnt know the password for bacula.
847 -> adding the password to the file favours admins
848 to forget to remove the password after usage
850 the protection of that file is less important
851 - long-listed-output of commands should be scrollable
852 like the unix more/less-command does
853 -> if someone runs 200 and more machines, the lists could
854 be a little long and complex
855 - command-output should be shown column by column
856 to reduce scrolling and to increase clarity
858 - lsmark should list the selected files with full
860 - wildcards for selecting and file and directories would be nice
861 - any actions should be interuptable with STRG+C
862 - command-expansion would be pretty cool
864 - When the replace Never option is set, new directory permissions
865 are not restored. See bug 213. To fix this requires creating a
866 list of newly restored directories so that those directory
867 permissions *can* be restored.
868 - Add prune all command
869 - Document fact that purge can destroy a part of a restore by purging
870 one volume while others remain valid -- perhaps mark Jobs.
871 - Add multiple-media-types.txt
872 - look at mxt-changer.html
873 - Make ? do a help command (no return needed).
874 - Implement restore directory.
875 - Document streams and how to implement them.
876 - Try not to re-backup a file if a new hard link is added.
877 - Add feature to backup hard links only, but not the data.
878 - Fix stream handling to be simpler.
879 - Add Priority and Bootstrap to Run a Job.
880 - Eliminate Restore "Run Restore Job" prompt by allowing new "run command
882 - Remove View FileSet button from Run a Job dialog.
883 - Handle prompt for restore job at end of Restore command.
884 - Add display of total selected files to Restore window.
885 - Add tree pane to left of window.
886 - Add progress meter.
887 - Max wait time or max run time causes seg fault -- see runtime-bug.txt
888 - Add message to user to check for fixed block size when the forward
889 space test fails in btape.
890 - When unmarking a directory check if all files below are unmarked and
891 then remove the + flag -- in the restore tree.
892 - Possibly implement: Action = Unmount Device="TapeDrive1" in Admin jobs.
893 - Setup lrrd graphs: (http://www.linpro.no/projects/lrrd/) Mike Acar.
894 - Revisit the question of multiple Volumes (disk) on a single device.
895 - Add a block copy option to bcopy.
896 - Fix "llist jobid=xx" where no fileset or client exists.
897 - For each job type (Admin, Restore, ...) require only the really necessary
898 fields.- Pass Director resource name as an option to the Console.
899 - Add a "batch" mode to the Console (no unsolicited queries, ...).
900 - Allow browsing the catalog to see all versions of a file (with
901 stat data on each file).
902 - Restore attributes of directory if replace=never set but directory
904 - Use SHA1 on authentication if possible.
905 - See comtest-xxx.zip for Windows code to talk to USB.
906 - Add John's appended files:
907 Appended = { /files/server/logs/http/*log }
908 and such files would be treated as follows.On a FULL backup, they would
909 be backed up like any other file.On an INCREMENTAL backup, where a
910 previous INCREMENTAL or FULL was already in thecatalogue and the length
911 of the file wasgreater than the length of the last backup, only thedata
912 added since the last backup will be dumped.On an INCREMENTAL backup, if
913 the length of the file is less than thelength of the file with the same
914 name last backed up, the completefile is dumped.On Windows systems, with
915 creation date of files, we can be evensmarter about this and not count
916 entirely upon the length.On a restore, the full and all incrementals
917 since it will beapplied in sequence to restore the file.
918 - Check new HAVE_WIN32 open bits.
919 - Check if the tape has moved before writing.
920 - Handling removable disks -- see below:
921 - Add FromClient and ToClient keywords on restore command (or
922 BackupClient RestoreClient).
923 - Implement a JobSet, which groups any number of jobs. If the
924 JobSet is started, all the jobs are started together.
925 Allow Pool, Level, and Schedule overrides.
926 - Look at updating Volume Jobs so that Max Volume Jobs = 1 will work
927 correctly for multiple simultaneous jobs.
928 - Implement the Media record flag that indicates that the Volume does disk
930 - Fix fast block rejection (stored/read_record.c:118). It passes a null
931 pointer (rec) to try_repositioning().
932 - Implement RestoreJobRetention? Maybe better "JobRetention" in a Job,
933 which would take precidence over the Catalog "JobRetention".
934 - Implement Label Format in Add and Label console commands.
935 - Put email tape request delays on one or more variables. User wants
936 to cancel the job after a certain time interval. Maximum Mount Wait?
937 - Job, Client, Device, Pool, or Volume?
938 Is it possible to make this a directive which is *optional* in multiple
939 resources, like Level? If so, I think I'd make it an optional directive
940 in Job, Client, and Pool, with precedence such that Job overrides Client
941 which in turn overrides Pool.
943 - New Storage specifications:
944 - Want to write to multiple storage devices simultaneously
945 - Want to write to multiple storage devices sequentially (in one job)
946 - Want to read/write simultaneously
947 - Key is MediaType -- it must match
949 Passed to SD as a sort of BSR record called Storage Specification
953 MediaType -> Next MediaType
955 Device -> Next Device
957 Allow multiple Storage specifications
965 Allow Multiple Pool specifications (note, Pool currently
967 Allow Multiple MediaType specifications in Dir conf
968 Allow Multiple Device specifications in Dir conf
969 Perhaps keep this in a single SSR
970 Tie a Volume to a specific device by using a MediaType that
971 is contained in only one device.
972 In SD allow Device to have Multiple MediaTypes
974 - Ideas from Jerry Scharf:
975 First let's point out some big pluses that bacula has for this
977 more importantly it's active. Thank you so much for that
978 even more important, it's not flaky
979 it has an open access catalog, opening many possibilities
980 it's pushing toward heterogeneous systems capability
982 Macintosh file client
983 working bare iron recovery for windows
984 the option for inc/diff backups not reset on fileset revision
985 a) use both change and inode update time against base time
986 b) do the full catalog check (expensive but accurate)
987 sizing guide (how much system is needed to back up N systems/files)
988 consultants on using bacula in building a disaster recovery system
990 or how to get at fancy things that one could do with bacula
991 logwatch code for bacula logs (or similar)
992 support for Oracle database ??
994 - Look at adding SQL server and Exchange support for Windows.
995 - Add progress of files/bytes to SD and FD.
996 - do a "messages" before the first prompt in Console
997 - Client does not show busy during Estimate command.
998 - Implement Console mtx commands.
999 - Implement a Mount Command and an Unmount Command where
1000 the users could specify a system command to be performed
1001 to do the mount, after which Bacula could attempt to
1002 read the device. This is for Removeable media such as a CDROM.
1003 - Most likely, this mount command would be invoked explicitly
1004 by the user using the current Console "mount" and "unmount"
1005 commands -- the Storage Daemon would do the right thing
1006 depending on the exact nature of the device.
1007 - As with tape drives, when Bacula wanted a new removable
1008 disk mounted, it would unmount the old one, and send a message
1009 to the user, who would then use "mount" as described above
1010 once he had actually inserted the disk.
1011 - Implement dump/print label to UA
1012 - Spool to disk only when the tape is full, then when a tape is hung move
1014 - bextract is sending everything to the log file ****FIXME****
1015 - Allow multiple Storage specifications (or multiple names on
1016 a single Storage specification) in the Job record. Thus a job
1017 can be backed up to a number of storage devices.
1018 - Implement some way for the File daemon to contact the Director
1019 to start a job or pass its DHCP obtained IP number.
1020 - Implement a query tape prompt/replace feature for a console
1021 - Make sure that Bacula rechecks the tape after the 20 min wait.
1022 - Set IO_NOWAIT on Bacula TCP/IP packets.
1023 - Try doing a raw partition backup and restore by mounting a
1025 - From Lars Kellers:
1026 Yes, it would allow to highly automatic the request for new tapes. If a
1027 tape is empty, bacula reads the barcodes (native or simulated), and if
1028 an unused tape is found, it runs the label command with all the
1029 necessary parameters.
1031 By the way can bacula automatically "move" an empty/purged volume say
1032 in the "short" pool to the "long" pool if this pool runs out of volume
1034 - What to do about "list files job=xxx".
1035 - Look at how fuser works and /proc/PID/fd that is how Nic found the
1036 file descriptor leak in Bacula.
1037 - Can we dynamically change FileSets?
1038 - If pool specified to label command and Label Format is specified,
1039 automatically generate the Volume name.
1040 - Add ExhautiveRestoreSearch
1041 - Look at the possibility of loading only the necessary
1042 data into the restore tree (i.e. do it one directory at a
1043 time as the user walks through the tree).
1044 - Possibly use the hash code if the user selects all for a restore command.
1045 - Fix "restore all" to bypass building the tree.
1046 - Prohibit backing up archive device (findlib/find_one.c:128)
1047 - Implement Release Device in the Job resource to unmount a drive.
1048 - Implement Acquire Device in the Job resource to mount a drive,
1049 be sure this works with admin jobs so that the user can get
1050 prompted to insert the correct tape. Possibly some way to say to
1051 run the job but don't save the files.
1052 - Make things like list where a file is saved case independent for
1054 - Implement a Recycle command
1055 - From Phil Stracchino:
1056 It would probably be a per-client option, and would be called
1057 something like, say, "Automatically purge obsoleted jobs". What it
1058 would do is, when you successfully complete a Differential backup of a
1059 client, it would automatically purge all Incremental backups for that
1060 client that are rendered redundant by that Differential. Likewise,
1061 when a Full backup on a client completed, it would automatically purge
1062 all Differential and Incremental jobs obsoleted by that Full backup.
1063 This would let people minimize the number of tapes they're keeping on
1064 hand without having to master the art of retention times.
1065 - When doing a Backup send all attributes back to the Director, who
1066 would then figure out what files have been deleted.
1067 - Currently in mount.c:236 the SD simply creates a Volume. It should have
1068 explicit permission to do so. It should also mark the tape in error
1069 if there is an error.
1070 - Cancel waiting for Client connect in SD if FD goes away.
1072 - Implement timeout in response() when it should come quickly.
1073 - Implement a Slot priority (loaded/not loaded).
1074 - Implement "vacation" Incremental only saves.
1075 - Implement create "FileSet"?
1076 - Add prefixlinks to where or not where absolute links to FD.
1077 - Issue message to mount a new tape before the rewind.
1078 - Simplified client job initiation for portables.
1079 - If SD cannot open a drive, make it periodically retry.
1080 - Add more of the config info to the tape label.
1082 - Refine SD waiting output:
1083 Device is being positioned
1084 > Device is being positioned for append
1085 > Device is being positioned to file x
1087 - Figure out some way to estimate output size and to avoid splitting
1088 a backup across two Volumes -- this could be useful for writing CDROMs
1089 where you really prefer not to have it split -- not serious.
1090 - Make bcopy read through bad tape records.
1091 - Program files (i.e. execute a program to read/write files).
1092 Pass read date of last backup, size of file last time.
1093 - Add Signature type to File DB record.
1094 - CD into subdirectory when open()ing files for backup to
1095 speed up things. Test with testfind().
1096 - Priority job to go to top of list.
1097 - Why are save/restore of device different sizes (sparse?) Yup! Fix it.
1098 - Implement some way for the Console to dynamically create a job.
1099 - Solaris -I on tar for include list
1100 - Need a verbose mode in restore, perhaps to bsr.
1101 - bscan without -v is too quiet -- perhaps show jobs.
1102 - Add code to reject whole blocks if not wanted on restore.
1103 - Check if we can increase Bacula FD priorty in Win2000
1104 - Check if both CatalogFiles and UseCatalog are set to SD.
1105 - Possibly add email to Watchdog if drive is unmounted too
1106 long and a job is waiting on the drive.
1107 - After unmount, if restore job started, ask to mount.
1108 - Add UA rc and history files.
1109 - put termcap (used by console) in ./configure and
1110 allow -with-termcap-dir.
1111 - Fix Autoprune for Volumes to respect need for full save.
1112 - Compare tape to Client files (attributes, or attributes and data)
1113 - Make all database Ids 64 bit.
1114 - Allow console commands to detach or run in background.
1115 - Add SD message variables to control operator wait time
1116 - Maximum Operator Wait
1117 - Minimum Message Interval
1118 - Maximum Message Interval
1119 - Send Operator message when cannot read tape label.
1120 - Verify level=Volume (scan only), level=Data (compare of data to file).
1121 Verify level=Catalog, level=InitCatalog
1123 - Add keyword search to show command in Console.
1124 - Events : tape has more than xxx bytes.
1125 - Complete code in Bacula Resources -- this will permit
1126 reading a new config file at any time.
1127 - Handle ctl-c in Console
1128 - Implement script driven addition of File daemon to config files.
1129 - Think about how to make Bacula work better with File (non-tape) archives.
1130 - Write Unix emulator for Windows.
1131 - Make database type selectable by .conf files i.e. at runtime
1132 - Set flag for uname -a. Add to Volume label.
1133 - Restore files modified after date
1134 - SET LD_RUN_PATH=$HOME/mysql/lib/mysql
1135 - Remove duplicate fields from jcr (e.g. jcr.level and jcr.jr.Level, ...).
1136 - Timout a job or terminate if link goes down, or reopen link and query.
1137 - Concept of precious tapes (cannot be reused).
1138 - Make bcopy copy with a single tape drive.
1139 - Permit changing ownership during restore.
1142 > My suggestion: Add a feature on the systray menu-icon menu to request
1143 > an immediate backup now. This would be useful for laptop users who may
1144 > not be on the network when the regular scheduled backup is run.
1146 > My wife's suggestion: Add a setting to the win32 client to allow it to
1147 > shut down the machine after backup is complete (after, of course,
1148 > displaying a "System will shut down in one minute, click here to cancel"
1149 > warning dialog). This would be useful for sites that want user
1150 > woorkstations to be shut down overnight to save power.
1153 - Autolabel should be specified by DIR instead of SD.
1155 - Add media capacity
1156 - AutoScan (check checksum of tape)
1157 - Format command = "format /dev/nst0"
1161 - Seek resolution (usually corresponds to buffer size)
1162 - EODErrorCode=ENOSPC or code
1163 - Partial Read error code
1164 - Partial write error code
1165 - Nonformatted read error
1166 - Nonformatted write error
1167 - WriteProtected error
1171 - IgnoreCloseErrors=yes
1181 - Write a Storage daemon that uses pipes and
1182 standard Unix programs to write to the tape.
1184 - Need something that monitors the JCR queue and
1185 times out jobs by asking the deamons where they are.
1186 - Verify from Volume
1187 - Need report class for messages. Perhaps
1188 report resource where report=group of messages
1189 - enhance scan_attrib and rename scan_jobtype, and
1190 fill in code for "since" option
1191 - Director needs a time after which the report status is sent
1192 anyway -- or better yet, a retry time for the job.
1193 - Don't reschedule a job if previous incarnation is still running.
1194 - Some way to automatically backup everything is needed????
1195 - Need a structure for pending actions:
1197 - termination status (part of buffered msgs?)
1199 Read, Write, Clean, Delete
1200 - Login to Bacula; Bacula users with different permissions:
1201 owner, group, user, quotas
1202 - Store info on each file system type (probably in the job header on tape.
1203 This could be the output of df; or perhaps some sort of /etc/mtab record.
1205 ========= ideas ===============
1206 From: "Jerry K. Schieffer" <jerry@skylinetechnology.com>
1207 To: <kern@sibbald.com>
1208 Subject: RE: [Bacula-users] future large programming jobs
1209 Date: Thu, 26 Feb 2004 11:34:54 -0600
1211 I noticed the subject thread and thought I would offer the following
1212 merely as sources of ideas, i.e. something to think about, not even as
1213 strong as a request. In my former life (before retiring) I often
1214 dealt with backups and storage management issues/products as a
1215 developer and as a consultant. I am currently migrating my personal
1216 network from amanda to bacula specifically because of the ability to
1217 cross media boundaries during storing backups.
1218 Are you familiar with the commercial product called ADSM (I think IBM
1219 now sells it under the Tivoli label)? It has a couple of interesting
1220 ideas that may apply to the following topics.
1222 1. Migration: Consider that when you need to restore a system, there
1223 may be pressure to hurry. If all the information for a single client
1224 can eventually end up on the same media (and in chronological order),
1225 the restore is facillitated by not having to search past information
1226 from other clients. ADSM has the concept of "client affinity" that
1227 may be associated with it's storage pools. It seems to me that this
1228 concept (as an optional feature) might fit in your architecture for
1231 ADSM also has the concept of defining one or more storage pools as
1232 "copy pools" (almost mirrors, but only in the sense of contents).
1233 These pools provide the ability to have duplicte data stored both
1234 onsite and offsite. The copy process can be scheduled to be handled
1235 by their storage manager during periods when there is no backup
1236 activity. Again, the migration process might be a place to consider
1237 implementing something like this.
1240 > It strikes me that it would be very nice to be able to do things
1242 > have the Job(s) backing up the machines run, and once they have all
1243 > completed, start a migration job to copy the data from disks Volumes
1245 > a tape library and then to offsite storage. Maybe this can already
1247 > done with some careful scheduling and Job prioritzation; the events
1248 > mechanism described below would probably make it very easy.
1250 This is the goal. In the first step (before events), you simply
1252 the Migration to tape later.
1254 2. Base jobs: In ADSM, each copy of each stored file is tracked in
1255 the database. Once a file (unique by path and metadata such as dates,
1256 size, ownership, etc.) is in a copy pool, no more copies are made. In
1257 other words, when you start ADSM, it begins like your concept of a
1258 base job. After that it is in the "incremental" mode. You can
1259 configure the number of "generations" of files to be retained, plus a
1260 retention date after which even old generations are purged. The
1261 database tracks the contents of media and projects the percentage of
1262 each volume that is valid. When the valid content of a volume drops
1263 below a configured percentage, the valid data are migrated to another
1264 volume and the old volume is marked as empty. Note, this requires
1265 ADSM to have an idea of the contents of a client, i.e. marking the
1266 database when an existing file was deleted, but this would solve your
1267 issue of restoring a client without restoring deleted files.
1269 This is pretty far from what bacula now does, but if you are going to
1270 rip things up for Base jobs,.....
1271 Also, the benefits of this are huge for very large shops, especially
1272 with media robots, but are a pain for shops with manual media
1278 ==============================
1281 - Audit M_ error codes to ensure they are correct and consistent.
1282 - Add variable break characters to lex analyzer.
1283 Either a bit mask or a string of chars so that
1284 the caller can change the break characters.
1285 - Make a single T_BREAK to replace T_COMMA, etc.
1286 - Ensure that File daemon and Storage daemon can
1287 continue a save if the Director goes down (this
1288 is NOT currently the case). Must detect socket error,
1289 buffer messages for later.
1290 - Add ability to backup to two Storage devices (two SD sessions) at
1291 the same time -- e.g. onsite, offsite.
1293 ======================================================
1296 Handling removable disks
1298 From: Karl Cunningham <karlc@keckec.com>
1300 My backups are only to hard disk these days, in removable bays. This is my
1301 idea of how a backup to hard disk would work more smoothly. Some of these
1302 things Bacula does already, but I mention them for completeness. If others
1303 have better ways to do this, I'd like to hear about it.
1305 1. Accommodate several disks, rotated similar to how tapes are. Identified
1306 by partition volume ID or perhaps by the name of a subdirectory.
1307 2. Abort & notify the admin if the wrong disk is in the bay.
1308 3. Write backups to different subdirectories for each machine to be backed
1310 4. Volumes (files) get created as needed in the proper subdirectory, one
1312 5. When a disk is recycled, remove or zero all old backup files. This is
1313 important as the disk being recycled may be close to full. This may be
1314 better done manually since the backup files for many machines may be
1315 scattered in many subdirectories.
1323 It is somewhat like a Full save becomes an incremental since
1324 the Base job (or jobs) plus other non-base files.
1326 - A Base backup is same as Full backup, just different type.
1327 - New BaseFiles table that contains:
1329 BaseJobId - Base JobId referenced for this FileId (needed ???)
1330 JobId - JobId currently running
1331 FileId - File not backed up, exists in Base Job
1332 FileIndex - FileIndex from Base Job.
1333 i.e. for each base file that exists but is not saved because
1334 it has not changed, the File daemon sends the JobId, BaseId,
1335 FileId, FileIndex back to the Director who creates the DB entry.
1336 - To initiate a Base save, the Director sends the FD
1337 the FileId, and full filename for each file in the Base.
1338 - When the FD finds a Base file, he requests the Director to
1339 send him the full File entry (stat packet plus MD5), or
1340 conversely, the FD sends it to the Director and the Director
1341 says yes or no. This can be quite rapid if the FileId is kept
1342 by the FD for each Base Filename.
1343 - It is probably better to have the comparison done by the FD
1344 despite the fact that the File entry must be sent across the
1346 - An alternative would be to send the FD the whole File entry
1347 from the start. The disadvantage is that it requires a lot of
1348 space. The advantage is that it requires less communications
1350 - The Job record must be updated to indicate that one or more
1352 - At end of Job, FD returns:
1353 1. Count of base files/bytes not written to tape (i.e. matches)
1354 2. Count of base file that were saved i.e. had changed.
1355 - No tape record would be written for a Base file that matches, in the
1356 same way that no tape record is written for Incremental jobs where
1357 the file is not saved because it is unchanged.
1358 - On a restore, all the Base file records must explicitly be
1359 found from the BaseFile tape. I.e. for each Full save that is marked
1360 to have one or more Base Jobs, search the BaseFile for all occurrences
1362 - An optimization might be to make the BaseFile have:
1368 This would avoid the need to explicitly fetch each File record for
1369 the Base job. The Base Job record will be fetched to get the
1370 VolSessionId and VolSessionTime.
1371 - Fix bpipe.c so that it does not modify results pointer.
1372 ***FIXME*** calling sequence should be changed.
1373 - Fix restore of acls and extended attributes to count ERROR
1374 messages and make errors non-fatal.
1375 - Put save/restore various platform acl/xattrs on a pointer to simplify
1377 - Add blast attributes to DIR to SD.
1378 - Implement unmount of USB volumes.
1379 - Look into using Dart for testing
1380 http://public.kitware.com/Dart/HTML/Index.shtml