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 - For Windows disaster recovery see http://unattended.sf.net/
41 - Add "before=" "olderthan=" to FileSet for doing Base of
43 - Show files/second in client status output.
44 - Don't attempt to restore from "Disabled" Volumes.
45 - Have SD compute MD5 or SHA1 and compare to what FD computes.
46 - Make VolumeToCatalog calculate an MD5 or SHA1 from the
47 actual data on the Volume and compare it.
48 - Remove queue.c code.
49 - Implement multiple jobid specification for the cancel command,
50 similar to what is permitted on the update slots command.
51 - Ensure that the SD re-reads the Media record if the JobFiles
52 does not match -- it may have been updated by another job.
53 - Look at rsysnc for incremental updates and dedupping
54 - Add MD5 or SHA1 check in SD for data validation
55 - When reserving a device to read, check to see if the Volume
56 is already in use, if so wait. Probably will need to pass the
57 Volume. See bug #1313. Create a regression test to simulate
58 this problem and see if VolumePollInterval fixes it. Possibly turn
63 - Why no error message if restore has no permission on the where
65 - Possibly allow manual "purge" to purge a Volume that has not
66 yet been written (even if FirstWritten time is zero) see ua_purge.c
68 - Add disk block detection bsr code (make it work).
70 - Detect deadlocks in reservations.
72 - Add list during dump
73 - Add in plugin code flag
74 - Add bRC_EndJob -- stops more calls to plugin this job
75 - Add bRC_Term (unload plugin)
76 - remove time_t from Jmsg and use utime_t?
77 - Deadlock detection, watchdog sees if counter advances when jobs are
78 running. With debug on, can do a "status" command.
79 - User options for plugins.
80 - Pool Storage override precedence over command line.
81 - Autolabel only if Volume catalog information indicates tape not
82 written. This will avoid overwriting a tape that gets an I/O
83 error on reading the volume label.
84 - I/O error, SD thinks it is not the right Volume, should check slot
85 then disable volume, but Asks for mount.
86 - Can be posible modify package to create and use configuration files in
91 /etc/bacula/bacula-dir.conf
92 /etc/bacula/conf.d/pools.conf
93 /etc/bacula/conf.d/clients.conf
94 /etc/bacula/conf.d/storages.conf
96 and into bacula-dir.conf file include
98 @/etc/bacula/conf.d/pools.conf
99 @/etc/bacula/conf.d/clients.conf
100 @/etc/bacula/conf.d/storages.conf
101 - Possibly add an Inconsistent state when a Volume is in error
103 - Fix #ifdefing so that smartalloc can be disabled. Check manual
104 -- the default is enabled.
105 - Change calling sequence to delete_job_id_range() in ua_cmds.c
106 the preceding strtok() is done inside the subroutine only once.
107 - Dangling softlinks are not restored properly. For example, take a
108 soft link such as src/testprogs/install-sh, which points to /usr/share/autoconf...
109 move the directory to another machine where the file /usr/share/autoconf does
110 not exist, back it up, then try a full restore. It fails.
111 - Softlinks that point to non-existent file are not restored in restore all,
112 but are restored if the file is individually selected. BUG!
113 - New directive "Delete purged Volumes"
115 - Prune by Job Level (Full, Differential, Incremental)
116 - Strict automatic pruning
117 - Use "./config no-idea no-mdc2 no-rc5" on building OpenSSL for
118 Win32 to avoid patent problems.
119 - modify pruning to keep a fixed number of versions of a file,
121 - the cd-command should allow complete paths
122 i.e. cd /foo/bar/foo/bar
123 -> if a customer mails me the path to a certain file,
124 its faster to enter the specified directory
125 - Make tree walk routines like cd, ls, ... more user friendly
126 by handling spaces better.
127 - When doing a restore, if the user does an "update slots"
128 after the job started in order to add a restore volume, the
129 values prior to the update slots will be put into the catalog.
130 Must retrieve catalog record merge it then write it back at the
131 end of the restore job, if we want to do this right.
135 MA = (last_MA * 3 + rate) / 4
136 rate = (bytes - last_bytes) / (runtime - last_runtime)
138 - Add a recursive mark command (rmark) to restore.
139 - "Minimum Job Interval = nnn" sets minimum interval between Jobs
140 of the same level and does not permit multiple simultaneous
141 running of that Job (i.e. lets any previous invocation finish
142 before doing Interval testing).
143 - Look at simplifying File exclusions.
145 - Separate Files and Directories in catalog
146 - Create FileVersions table
147 - finish implementation of fdcalled -- see ua_run.c:105
148 - Fix problem in postgresql.c in my_postgresql_query, where the
149 generation of the error message doesn't differentiate result==NULL
150 and a bad status from that result. Not only that, the result is
151 cleared on a bail_out without having generated the error message.
152 - Implement SDErrors (must return from SD)
153 - Implement continue spooling while despooling.
154 - Remove all install temp files in Win32 PLUGINSDIR.
155 - Audit retention periods to make sure everything is 64 bit.
156 - No where in restore causes kaboom.
157 - Performance: multiple spool files for a single job.
158 - Performance: despool attributes when despooling data (problem
159 multiplexing Dir connection).
160 - Make restore use the in-use volume reservation algorithm.
161 - When Pool specifies Storage command override does not work.
162 - Implement wait_for_sysop() message display in wait_for_device(), which
163 now prints warnings too often.
164 - Ensure that each device in an Autochanger has a different
166 - Look at sg_logs -a /dev/sg0 for getting soft errors.
167 - btape "test" command with Offline on Unmount = yes
169 This test is essential to Bacula.
171 I'm going to write one record in file 0,
172 two records in file 1,
173 and three records in file 2
175 02-Feb 11:00 btape: ABORTING due to ERROR in dev.c:715
176 dev.c:714 Bad call to rewind. Device "LTO" (/dev/nst0) not open
177 02-Feb 11:00 btape: Fatal Error because: Bacula interrupted by signal 11: Segmentation violation
178 Kaboom! btape, btape got signal 11. Attempting traceback.
180 - Encryption -- email from Landon
181 > The backup encryption algorithm is currently not configurable, and is
182 > set to AES_128_CBC in src/filed/backup.c. The encryption code
183 > supports a number of different ciphers (as well as adding arbitrary
184 > new ones) -- only a small bit of code would be required to map a
185 > configuration string value to a CRYPTO_CIPHER_* value, if anyone is
186 > interested in implementing this functionality.
188 - Add the OS version back to the Win32 client info.
189 - Restarted jobs have a NULL in the from field.
190 - Modify SD status command to indicate when the SD is writing
191 to a DVD (the device is not open -- see bug #732).
192 - Look at the possibility of adding "SET NAMES UTF8" for MySQL,
193 and possibly changing the blobs into varchar.
194 - Test Volume compatibility between machine architectures
195 - Encryption documentation
196 - Bare-metal recovery Windows (todo)
199 - Migration from other vendors
203 - Backup conf/exe (all daemons)
204 - Backup up system state
205 - Detect state change of system (verify)
206 - Synthetic Full, Diff, Inc (Virtual, Reconstructed)
208 - Novell NSS backup http://www.novell.com/coolsolutions/tools/18952.html
209 - Compliance norms that compare restored code hash code.
210 - When glibc crash, get address with
211 info symbol 0x809780c
212 - How to sync remote offices.
215 Extract capability (#25)
216 Continued enhancement of bweb
217 Threshold triggered migration jobs (not currently in list, but will be
219 Client triggered backups
220 Complete rework of the scheduling system (not in list)
221 Performance and usage instrumentation (not in list)
222 See email of 21Aug2007 for details.
223 - Look at: http://tech.groups.yahoo.com/group/cfg2html
224 and http://www.openeyet.nl/scc/ for managing customer changes
228 - Access Mode = Read-Only, Read-Write, Unavailable, Destroyed, Offsite
230 - Maximum number of scratch volumes
232 - Next Pool (already have)
233 - Reclamation threshold
235 - Reuse delay (after all files purged from volume before it can be used)
236 - Copy Pool = xx, yyy (or multiple lines).
238 - Allow pool selection during restore.
240 - Average tape size from Eric
241 SELECT COALESCE(media_avg_size.volavg,0) * count(Media.MediaId) AS volmax, GROUP BY Media.MediaType, Media.PoolId, media_avg_size.volavg
242 count(Media.MediaId) AS volnum,
243 sum(Media.VolBytes) AS voltotal,
244 Media.PoolId AS PoolId,
245 Media.MediaType AS MediaType
247 LEFT JOIN (SELECT avg(Media.VolBytes) AS volavg,
248 Media.MediaType AS MediaType
250 WHERE Media.VolStatus = 'Full'
251 GROUP BY Media.MediaType
252 ) AS media_avg_size ON (Media.MediaType = media_avg_size.MediaType)
253 GROUP BY Media.MediaType, Media.PoolId, media_avg_size.volavg
255 - Despool attributes in separate thread
258 - Check why restore repeatedly sends Rechdrs between
259 each data chunk -- according to James Harper 9Jan07.
262 - More intelligent re-run
263 - Incremental backup -- rsync, Stow
265 - Make Bacula by default not backup tmpfs, procfs, sysfs, ...
266 - Fix hardlinked immutable files when linking a second file, the
267 immutable flag must be removed prior to trying to link it.
268 - Change dbcheck to tell users to use native tools for fixing
269 broken databases, and to ensure they have the proper indexes.
270 - add udev rules for Bacula devices.
271 - If a job terminates, the DIR connection can close before the
272 Volume info is updated, leaving the File count wrong.
273 - Look at why SIGPIPE during connection can cause seg fault in
274 writing the daemon message, when Dir dropped to bacula:bacula
275 - Look at zlib 32 => 64 problems.
276 - Fix bextract to restore ACLs, or better yet, use common routines.
277 - Do we migrate appendable Volumes?
278 - Print warning message if LANG environment variable does not specify
280 - New dot commands from Arno.
281 .show device=xxx lists information from one storage device, including
282 devices (I'm not even sure that information exists in the DIR...)
283 .move eject device=xxx mostly the same as 'unmount xxx' but perhaps with
284 better machine-readable output like "Ok" or "Error busy"
285 .move eject device=xxx toslot=yyy the same as above, but with a new
286 target slot. The catalog should be updated accordingly.
287 .move transfer device=xxx fromslot=yyy toslot=zzz
290 - Article: http://www.heise.de/open/news/meldung/83231
291 - Article: http://www.golem.de/0701/49756.html
292 - Article: http://lwn.net/Articles/209809/
293 - Article: http://www.onlamp.com/pub/a/onlamp/2004/01/09/bacula.html
294 - Article: http://www.linuxdevcenter.com/pub/a/linux/2005/04/07/bacula.html
295 - Article: http://www.osreviews.net/reviews/admin/bacula
296 - Article: http://www.debianhelp.co.uk/baculaweb.htm
298 - Wikis mentioning Bacula
299 http://wiki.finkproject.org/index.php/Admin:Backups
300 http://wiki.linuxquestions.org/wiki/Bacula
301 http://www.openpkg.org/product/packages/?package=bacula
302 http://www.iterating.com/products/Bacula
303 http://net-snmp.sourceforge.net/wiki/index.php/Net-snmp_extensions
304 http://www.section6.net/wiki/index.php/Using_Bacula_for_Tape_Backups
305 http://bacula.darwinports.com/
306 http://wiki.mandriva.com/en/Releases/Corporate/Server_4/Notes#Bacula
307 http://en.wikipedia.org/wiki/Bacula
310 http://www.devco.net/pubwiki/Bacula/
311 http://paramount.ind.wpi.edu/wiki/doku.php
312 http://gentoo-wiki.com/HOWTO_Backup
313 http://www.georglutz.de/wiki/Bacula
314 http://www.clarkconnect.com/wiki/index.php?title=Modules_-_LAN_Backup/Recovery
315 http://linuxwiki.de/Bacula (in German)
317 - Possibly allow SD to spool even if a tape is not mounted.
318 - Figure out how to configure query.sql. Suggestion to use m4:
319 == changequote.m4 ===
320 changequote(`[',`]')dnl
321 ==== query.sql.in ===
322 :List next 20 volumes to expire
324 Pool.Name AS PoolName,
329 [ FROM_UNIXTIME(UNIX_TIMESTAMP(Media.LastWritten) Media.VolRetention) AS Expire, ])dnl
331 [ media.lastwritten + interval '1 second' * media.volretention as expire, ])dnl
335 ON Media.PoolId=Pool.PoolId
336 WHERE Media.LastWritten>0
340 Command: m4 -DmySQL changequote.m4 query.sql.in >query.sql
342 The problem is that it requires m4, which is not present on all machines
347 select FilenameId from Filename where Name='';
348 # Get list of all directories referenced in a Backup.
349 select Path.Path from Path,File where File.JobId=nnn and
350 File.FilenameId=(FilenameId-from-above) and File.PathId=Path.PathId
351 order by Path.Path ASC;
353 - Mount on an Autochanger with no tape in the drive causes:
354 Automatically selected Storage: LTO-changer
355 Enter autochanger drive[0]: 0
356 3301 Issuing autochanger "loaded drive 0" command.
357 3302 Autochanger "loaded drive 0", result: nothing loaded.
358 3301 Issuing autochanger "loaded drive 0" command.
359 3302 Autochanger "loaded drive 0", result: nothing loaded.
360 3902 Cannot mount Volume on Storage Device "LTO-Drive1" (/dev/nst0) because:
361 Couldn't rewind device "LTO-Drive1" (/dev/nst0): ERR=dev.c:678 Rewind error on "LTO-Drive1" (/dev/nst0). ERR=No medium found.
362 3905 Device "LTO-Drive1" (/dev/nst0) open but no Bacula volume is mounted.
363 If this is not a blank tape, try unmounting and remounting the Volume.
364 - If Drive 0 is blocked, and drive 1 is set "Autoselect=no", drive 1 will
366 - Autochanger did not change volumes.
367 select * from Storage;
368 +-----------+-------------+-------------+
369 | StorageId | Name | AutoChanger |
370 +-----------+-------------+-------------+
371 | 1 | LTO-changer | 0 |
372 +-----------+-------------+-------------+
373 05-May 03:50 roxie-sd: 3302 Autochanger "loaded drive 0", result is Slot 11.
374 05-May 03:50 roxie-sd: Tibs.2006-05-05_03.05.02 Warning: Director wanted Volume "LT
375 Current Volume "LT0-002" not acceptable because:
376 1997 Volume "LT0-002" not in catalog.
377 05-May 03:50 roxie-sd: Tibs.2006-05-05_03.05.02 Error: Autochanger Volume "LT0-002"
378 Setting InChanger to zero in catalog.
379 05-May 03:50 roxie-dir: Tibs.2006-05-05_03.05.02 Error: Unable to get Media record
381 05-May 03:50 roxie-sd: Tibs.2006-05-05_03.05.02 Fatal error: Error getting Volume i
382 05-May 03:50 roxie-sd: Tibs.2006-05-05_03.05.02 Fatal error: Job 530 canceled.
383 05-May 03:50 roxie-sd: Tibs.2006-05-05_03.05.02 Fatal error: spool.c:249 Fatal appe
384 05-May 03:49 Tibs: Tibs.2006-05-05_03.05.02 Fatal error: c:\cygwin\home\kern\bacula
393 FirstWritten: 2006-05-05 03:11:54
394 LastWritten: 2006-05-05 03:50:23
395 LabelDate: 2005-12-26 16:52:40
406 VolRetention: 31,536,000
418 Note VolStatus is blank!!!!!
425 FirstWritten: 0000-00-00 00:00:00
426 LastWritten: 0000-00-00 00:00:00
427 LabelDate: 2005-12-26 16:52:40
438 VolRetention: 31,536,000
451 Automatically selected Storage: LTO-changer
452 Enter autochanger drive[0]: 0
453 3301 Issuing autochanger "loaded drive 0" command.
454 3302 Autochanger "loaded drive 0", result: nothing loaded.
455 3301 Issuing autochanger "loaded drive 0" command.
456 3302 Autochanger "loaded drive 0", result: nothing loaded.
457 3902 Cannot mount Volume on Storage Device "LTO-Drive1" (/dev/nst0) because:
458 Couldn't rewind device "LTO-Drive1" (/dev/nst0): ERR=dev.c:678 Rewind error on "LTO-Drive1" (/dev/nst0). ERR=No medium found.
460 3905 Device "LTO-Drive1" (/dev/nst0) open but no Bacula volume is mounted.
461 If this is not a blank tape, try unmounting and remounting the Volume.
463 - http://www.dwheeler.com/essays/commercial-floss.html
464 - Add VolumeLock to prevent all but lock holder (SD) from updating
465 the Volume data (with the exception of VolumeState).
466 - The btape fill command does not seem to use the Autochanger
467 - Make Windows installer default to system disk drive.
468 - Look at using ioctl(FIOBMAP, ...) on Linux, and
469 DeviceIoControl(..., FSCTL_QUERY_ALLOCATED_RANGES, ...) on
470 Win32 for sparse files.
471 http://www.flexhex.com/docs/articles/sparse-files.phtml
472 http://www.informatik.uni-frankfurt.de/~loizides/reiserfs/fibmap.html
473 - Directive: at <event> "command"
474 - Command: pycmd "command" generates "command" event. How to
475 attach to a specific job?
476 - run_cmd() returns int should return JobId_t
477 - get_next_jobid_from_list() returns int should return JobId_t
478 - Document export LDFLAGS=-L/usr/lib64
479 - Network error on Win32 should set Win32 error code.
480 - What happens when you rename a Disk Volume?
481 - Job retention period in a Pool (and hence Volume). The job would
483 - Add Win32 FileSet definition somewhere
484 - Look at fixing restore status stats in SD.
485 - Look at using ioctl(FIMAP) and FIGETBSZ for sparse files.
486 http://www.informatik.uni-frankfurt.de/~loizides/reiserfs/fibmap.html
487 - Implement a mode that says when a hard read error is
488 encountered, read many times (as it currently does), and if the
489 block cannot be read, skip to the next block, and try again. If
490 that fails, skip to the next file and try again, ...
492 create table LevelType (LevelType binary(1), LevelTypeLong tinyblob);
493 insert into LevelType (LevelType,LevelTypeLong) values
497 - new pool XXX with ScratchPoolId = MyScratchPool's PoolId and
498 let it fill itself, and RecyclePoolId = XXX's PoolId so I can
499 see if it become stable and I just have to supervise
501 - If I want to remove this pool, I set RecyclePoolId = MyScratchPool's
502 PoolId, and when it is empty remove it.
504 - Allow Check Labels to be used with Bacula labels.
505 - "Resuming" a failed backup (lost line for example) by using the
506 failed backup as a sort of "base" job.
508 - Email to the user when the tape is about to need changing x
509 days before it needs changing.
510 - Command to show next tape that will be used for a job even
511 if the job is not scheduled.
512 - From: Arunav Mandal <amandal@trolltech.com>
513 1. When jobs are running and bacula for some reason crashes or if I do a
514 restart it remembers and jobs it was running before it crashed or restarted
515 as of now I loose all jobs if I restart it.
517 2. When spooling and in the midway if client is disconnected for instance a
518 laptop bacula completely discard the spool. It will be nice if it can write
519 that spool to tape so there will be some backups for that client if not all.
521 3. We have around 150 clients machines it will be nice to have a option to
522 upgrade all the client machines bacula version automatically.
524 4. Atleast one connection should be reserved for the bconsole so at heavy load
525 I should connect to the director via bconsole which at sometimes I can't
527 5. Another most important feature that is missing, say at 10am I manually
528 started backup of client abc and it was a full backup since client abc has
529 no backup history and at 10.30am bacula again automatically started backup of
530 client abc as that was in the schedule. So now we have 2 multiple Full
531 backups of the same client and if we again try to start a full backup of
532 client backup abc bacula won't complain. That should be fixed.
534 - regardless of the retention period, Bacula will not prune the
535 last Full, Diff, or Inc File data until a month after the
536 retention period for the last Full backup that was done.
537 - update volume=xxx --- add status=Full
538 - Remove old spool files on startup.
539 - Exclude SD spool/working directory.
540 - Refuse to prune last valid Full backup. Same goes for Catalog.
542 - Make a callback when Rerun failed levels is called.
543 - Give Python program access to Scheduled jobs.
544 - Add setting Volume State via Python.
545 - Python script to save with Python, not save, save with Bacula.
546 - Python script to do backup.
548 - Change the Priority, Client, Storage, JobStatus (error)
549 at the start of a job.
550 - Why is SpoolDirectory = /home/bacula/spool; not reported
551 as an error when writing a DVD?
552 - Make bootstrap file handle multiple MediaTypes (SD)
553 - Remove all old Device resource code in Dir and code to pass it
554 back in SD -- better, rework it to pass back device statistics.
555 - Check locking of resources -- be sure to lock devices where previously
556 resources were locked.
557 - The last part is left in the spool dir.
560 - In restore don't compare byte count on a raw device -- directory
561 entry does not contain bytes.
564 - Max Vols limit in Pool off by one?
565 - Implement Files/Bytes,... stats for restore job.
566 - Implement Total Bytes Written, ... for restore job.
567 - Despool attributes simultaneously with data in a separate
568 thread, rejoined at end of data spooling.
569 - 7. Implement new Console commands to allow offlining/reserving drives,
570 and possibly manipulating the autochanger (much asked for).
571 - Add start/end date editing in messages (%t %T, %e?) ...
572 - Add ClientDefs similar to JobDefs.
573 - Print more info when bextract -p accepts a bad block.
574 - Fix FD JobType to be set before RunBeforeJob in FD.
575 - Look at adding full Volume and Pool information to a Volume
576 label so that bscan can get *all* the info.
577 - If the user puts "Purge Oldest Volume = yes" or "Recycle Oldest Volume = yes"
578 and there is only one volume in the pool, refuse to do it -- otherwise
579 he fills the Volume, then immediately starts reusing it.
580 - Implement copies and stripes.
581 - Add history file to console.
582 - Each file on tape creates a JobMedia record. Peter has 4 million
583 files spread over 10000 tape files and four tapes. A restore takes
584 16 hours to build the restore list.
585 - Add and option to check if the file size changed during backup.
586 - Make sure SD deletes spool files on error exit.
587 - Delete old spool files when SD starts.
588 - When labeling tapes, if you enter 000026, Bacula uses
589 the tape index rather than the Volume name 000026.
590 - Add offline tape command to Bacula console.
592 Enter MediaId or Volume name: 32
593 Enter new Volume name: DLT-20Dec04
594 Automatically selected Pool: Default
595 Connecting to Storage daemon DLTDrive at 192.168.68.104:9103 ...
596 Sending relabel command from "DLT-28Jun03" to "DLT-20Dec04" ...
597 block.c:552 Write error at 0:0 on device /dev/nst0. ERR=Bad file descriptor.
598 Error writing final EOF to tape. This tape may not be readable.
599 dev.c:1207 ioctl MTWEOF error on /dev/nst0. ERR=Permission denied.
600 askdir.c:219 NULL Volume name. This shouldn't happen!!!
601 3912 Failed to label Volume: ERR=dev.c:1207 ioctl MTWEOF error on /dev/nst0. ERR=Permission denied.
602 Label command failed for Volume DLT-20Dec04.
603 Do not forget to mount the drive!!!
604 - Bug: if a job is manually scheduled to run later, it does not appear
605 in any status report and cannot be cancelled.
610 How about introducing a Type = MgmtPolicy job type? That job type would
611 be responsible for scanning the Bacula environment looking for specific
612 conditions, and submitting the appropriate jobs for implementing said
616 Name = "Migration-Policy"
618 Policy Selection Job Type = Migrate
619 Scope = "<keyword> <operator> <regexp>"
620 Threshold = "<keyword> <operator> <regexp>"
621 Job Template = <template-name>
624 Where <keyword> is any legal job keyword, <operator> is a comparison
625 operator (=,<,>,!=, logical operators AND/OR/NOT) and <regexp> is a
626 appropriate regexp. I could see an argument for Scope and Threshold
627 being SQL queries if we want to support full flexibility. The
628 Migration-Policy job would then get scheduled as frequently as a site
629 felt necessary (suggested default: every 15 minutes).
634 Name = "Migration-Policy"
636 Policy Selection Job Type = Migration
638 Threshold = "Migration Selection Type = LowestUtil"
639 Job Template = "MigrationTemplate"
642 would select all pools for examination and generate a job based on
643 MigrationTemplate to automatically select the volume with the lowest
644 usage and migrate it's contents to the nextpool defined for that pool.
646 This policy abstraction would be really handy for adjusting the behavior
647 of Bacula according to site-selectable criteria (one thing that pops
648 into mind is Amanda's ability to automatically adjust backup levels
649 depending on various criteria).
655 - Add Pool/Storage override regression test.
656 - Add delete JobId to regression.
657 - Add a regression test for dbcheck.
658 - New test to add bscan to four-concurrent-jobs regression,
659 i.e. after the four-concurrent jobs zap the
660 database as is done in the bscan-test, then use bscan to
661 restore the database, do a restore and compare with the
663 - Add restore of specific JobId to regression (item 3
664 on the restore prompt)
665 - Add IPv6 to regression
666 - Add database test to regression. Test each function like delete,
669 - AntiVir can slow down backups on Win32 systems.
670 - Win32 systems with FAT32 can be much slower than NTFS for
671 more than 1000 files per directory.
675 - A HOLD command to stop all jobs from starting.
676 - A PAUSE command to pause all running jobs ==> release the
678 - Media Type = LTO,LTO-2,LTO-3
679 Media Type Read = LTO,LTO2,LTO3
680 Media Type Write = LTO2, LTO3
682 === From Carsten Menke <bootsy52@gmx.net>
684 Following is a list of what I think in the situations where I'm faced with,
685 could be a usefull enhancement to bacula, which I'm certain other users will
686 benefit from as well.
688 1. NextJob/NextJobs Directive within a Job Resource in the form of
689 NextJobs = job1,job2.
692 I currently solved the problem with running multiple jobs each after each
693 by setting the Max Wait Time for a job to 8 hours, and give
694 the jobs different Priorities. However, there scenarios where
695 1 Job is directly depending on another job, so if the former job fails,
696 the job after it needn't to be run
697 while maybe other jobs should run despite of that
700 A Backup Job and a Verify job, if the backup job fails there is no need to run
701 the verify job, as the backup job already failed. However, one may like
702 to backup the Catalog to disk despite of that the main backup job failed.
705 I see that this is related to the Event Handlers which are on the ToDo
706 list, also it is maybe a good idea to check for the return value and
707 execute different actions based on the return value
710 3. offline capability to bconsole
713 Currently I use a script which I execute within the last Job via the
714 RunAfterJob Directive, to release and eject the tape.
715 So I have to call bconsole "release=Storage-Name" and afterwards
716 mt -f /dev/nst0 eject to get the tape out.
718 If I have multiple Storage Devices, than these may not be /dev/nst0 and
719 I have to modify the script or call it with parameters etc.
720 This would actually not be needed, as everything is already defined
721 in bacula-sd.conf and if I can invoke bconsole with the
722 storage name via $1 in the script than I'm done and information is
725 4. %s for Storage Name added to the chars being substituted in "RunAfterJob"
729 For the reason mentioned in 3. to have the ability to call a
730 script with /scripts/foobar %s and in the script use $1
731 to pass the Storage Name to bconsole
733 5. Setting Volume State within a Job Resource
736 Instead of using "Maximum Volume Jobs" in the Pool Resource,
737 I would have the possibilty to define
738 in a Job Resource that after this certain job is run, the Volume State
739 should be set to "Volume State = Used", this give more flexibility (IMHO).
741 6. Localization of Bacula Messages
744 Unfortunatley many,many people I work with don't speak english very well.
745 So if at least the Reporting messages would be localized then they
746 would understand that they have to change the tape,etc. etc.
748 I volunteer to do the german translations, and if I can convince my wife also
749 french and Morre (western african language).
751 7. OK, this is evil, probably bound to security risks and maybe not possible
752 due to the design of bacula.
754 Implementation of Backtics ( `command` ) for shell comand execution to
755 the "Label Format" Directive.
759 Currently I have defined BACULA_DAY_OF_WEEK="day1|day2..." resulting in
760 Label Format = "HolyBackup-${BACULA_DAY_OF_WEEK[${WeekDay}]}". If I could
761 use backticks than I could use "Label Format = HolyBackup-`date +%A` to have
762 the localized name for the day of the week appended to the
763 format string. Then I have the tape labeled automatically with weekday
764 name in the correct language.
766 - Make output from status use html table tags for nicely
767 presenting in a browser.
768 - Browse generations of files.
769 - I've seen an error when my catalog's File table fills up. I
770 then have to recreate the File table with a larger maximum row
771 size. Relevant information is at
772 http://dev.mysql.com/doc/mysql/en/Full_table.html ; I think the
773 "Installing and Configuring MySQL" chapter should talk a bit
774 about this potential problem, and recommend a solution.
775 - For Solaris must use POSIX awk.
776 - Want speed of writing to tape while despooling.
777 - Supported autochanger:
785 Wangtek 6525ES (SCSI-1 QIC drive, 525MB), under Linux 2.4.something,
786 bacula 1.36.0/1 works with blocksize 16k INSIDE bacula-sd.conf.
787 - Add regex from http://www.pcre.org to Bacula for Win32.
788 - Use only shell tools no make in CDROM package.
789 - Include within include does it work?
790 - Implement a Pool of type Cleaning?
791 - Implement VolReadTime and VolWriteTime in SD
792 - Modify Backing up Your Database to include a bootstrap file.
793 - Think about making certain database errors fatal.
794 - Look at correcting the time jump in the scheduler for daylight
795 savings time changes.
796 - Add a "real" timer to network connections.
797 - Promote to Full = Time period
798 - Check dates entered by user for correctness (month/day/... ranges)
799 - Compress restore Volume listing by date and first file.
800 - Look at patches/bacula_db.b2z postgresql that loops during restore.
802 - Perhaps add read/write programs and/or plugins to FileSets.
803 - How to handle backing up portables ...
804 - Add some sort of guaranteed Interval for upgrading jobs.
805 - Can we write the state file after every job terminates? On Win32
806 the system crashes and the state file is not updated.
809 Documentation to do: (any release a little bit at a time)
810 - Doc to do unmount before removing magazine.
811 - Alternative to static linking "ldd prog" save all binaries listed,
812 restore them and point LD_LIBRARY_PATH to them.
813 - Document add "</dev/null >/dev/null 2>&1" to the bacula-fd command line
814 - Document query file format.
815 - Add more documentation for bsr files.
816 - Document problems with Verify and pruning.
817 - Document how to use multiple databases.
818 - VXA drives have a "cleaning required"
819 indicator, but Exabyte recommends preventive cleaning after every 75
822 In this context, it should be noted that Exabyte has a command-line
823 vxatool utility available for free download. (The current version is
824 vxatool-3.72.) It can get diagnostic info, read, write and erase tapes,
825 test the drive, unload tapes, change drive settings, flash new firmware,
827 Of particular interest in this context is that vxatool <device> -i will
828 report, among other details, the time since last cleaning in tape motion
829 minutes. This information can be retrieved (and settings changed, for
830 that matter) through the generic-SCSI device even when Bacula has the
831 regular tape device locked. (Needless to say, I don't recommend
832 changing tape settings while a job is running.)
833 - Lookup HP cleaning recommendations.
834 - Lookup HP tape replacement recommendations (see trouble shooting autochanger)
835 - Document doing table repair
838 ===================================
839 - Add macro expansions in JobDefs.
840 Run Before Job = "SomeFile %{Level} %{Client}"
841 Write Bootstrap="/some/dir/%{JobName}_%{Client}.bsr"
842 - Use non-blocking network I/O but if no data is available, use
844 - Use gather write() for network I/O.
845 - Autorestart on crash.
846 - Add bandwidth limiting.
847 - When an error in input occurs and conio beeps, you can back
848 up through the prompt.
849 - Detect fixed tape block mode during positioning by looking at
850 block numbers in btape "test". Possibly adjust in Bacula.
851 - Fix list volumes to output volume retention in some other
852 units, perhaps via a directive.
853 - Allow Simultaneous Priorities = yes => run up to Max concurrent jobs even
854 with multiple priorities.
855 - If you use restore replace=never, the directory attributes for
856 non-existent directories will not be restored properly.
858 - see lzma401.zip in others directory for new compression
860 - Allow the user to select JobType for manual pruning/purging.
861 - bscan does not put first of two volumes back with all info in
863 - Figure out how to make named console messages go only to that
864 console and to the non-restricted console (new console class?).
865 - Make restricted console prompt for password if *ask* is set or
866 perhaps if password is undefined.
867 - Implement "from ISO-date/time every x hours/days/weeks/months" in
870 ==== from Marc Schoechlin
871 - the help-command should be more verbose
872 (it should explain the paramters of the different
874 -> its time-comsuming to consult the manual anytime
875 you need a special parameter
876 -> maybe its more easy to maintain this, if the
877 descriptions of that commands are outsourced to
879 - if the password is not configured in bconsole.conf
880 you should be asked for it.
881 -> sometimes you like to do restore on a customer-machine
882 which shouldnt know the password for bacula.
883 -> adding the password to the file favours admins
884 to forget to remove the password after usage
886 the protection of that file is less important
887 - long-listed-output of commands should be scrollable
888 like the unix more/less-command does
889 -> if someone runs 200 and more machines, the lists could
890 be a little long and complex
891 - command-output should be shown column by column
892 to reduce scrolling and to increase clarity
894 - lsmark should list the selected files with full
896 - wildcards for selecting and file and directories would be nice
897 - any actions should be interuptable with STRG+C
898 - command-expansion would be pretty cool
900 - When the replace Never option is set, new directory permissions
901 are not restored. See bug 213. To fix this requires creating a
902 list of newly restored directories so that those directory
903 permissions *can* be restored.
904 - Add prune all command
905 - Document fact that purge can destroy a part of a restore by purging
906 one volume while others remain valid -- perhaps mark Jobs.
907 - Add multiple-media-types.txt
908 - look at mxt-changer.html
909 - Make ? do a help command (no return needed).
910 - Implement restore directory.
911 - Document streams and how to implement them.
912 - Try not to re-backup a file if a new hard link is added.
913 - Add feature to backup hard links only, but not the data.
914 - Fix stream handling to be simpler.
915 - Add Priority and Bootstrap to Run a Job.
916 - Eliminate Restore "Run Restore Job" prompt by allowing new "run command
918 - Remove View FileSet button from Run a Job dialog.
919 - Handle prompt for restore job at end of Restore command.
920 - Add display of total selected files to Restore window.
921 - Add tree pane to left of window.
922 - Add progress meter.
923 - Max wait time or max run time causes seg fault -- see runtime-bug.txt
924 - Add message to user to check for fixed block size when the forward
925 space test fails in btape.
926 - When unmarking a directory check if all files below are unmarked and
927 then remove the + flag -- in the restore tree.
928 - Possibly implement: Action = Unmount Device="TapeDrive1" in Admin jobs.
929 - Setup lrrd graphs: (http://www.linpro.no/projects/lrrd/) Mike Acar.
930 - Revisit the question of multiple Volumes (disk) on a single device.
931 - Add a block copy option to bcopy.
932 - Fix "llist jobid=xx" where no fileset or client exists.
933 - For each job type (Admin, Restore, ...) require only the really necessary
934 fields.- Pass Director resource name as an option to the Console.
935 - Add a "batch" mode to the Console (no unsolicited queries, ...).
936 - Allow browsing the catalog to see all versions of a file (with
937 stat data on each file).
938 - Restore attributes of directory if replace=never set but directory
940 - Use SHA1 on authentication if possible.
941 - See comtest-xxx.zip for Windows code to talk to USB.
942 - Add John's appended files:
943 Appended = { /files/server/logs/http/*log }
944 and such files would be treated as follows.On a FULL backup, they would
945 be backed up like any other file.On an INCREMENTAL backup, where a
946 previous INCREMENTAL or FULL was already in thecatalogue and the length
947 of the file wasgreater than the length of the last backup, only thedata
948 added since the last backup will be dumped.On an INCREMENTAL backup, if
949 the length of the file is less than thelength of the file with the same
950 name last backed up, the completefile is dumped.On Windows systems, with
951 creation date of files, we can be evensmarter about this and not count
952 entirely upon the length.On a restore, the full and all incrementals
953 since it will beapplied in sequence to restore the file.
954 - Check new HAVE_WIN32 open bits.
955 - Check if the tape has moved before writing.
956 - Handling removable disks -- see below:
957 - Add FromClient and ToClient keywords on restore command (or
958 BackupClient RestoreClient).
959 - Implement a JobSet, which groups any number of jobs. If the
960 JobSet is started, all the jobs are started together.
961 Allow Pool, Level, and Schedule overrides.
962 - Look at updating Volume Jobs so that Max Volume Jobs = 1 will work
963 correctly for multiple simultaneous jobs.
964 - Implement the Media record flag that indicates that the Volume does disk
966 - Implement VolAddr, which is used when Volume is addressed like a disk,
967 and form it from VolFile and VolBlock.
968 - Fix fast block rejection (stored/read_record.c:118). It passes a null
969 pointer (rec) to try_repositioning().
970 - Implement RestoreJobRetention? Maybe better "JobRetention" in a Job,
971 which would take precidence over the Catalog "JobRetention".
972 - Implement Label Format in Add and Label console commands.
973 - Put email tape request delays on one or more variables. User wants
974 to cancel the job after a certain time interval. Maximum Mount Wait?
975 - Job, Client, Device, Pool, or Volume?
976 Is it possible to make this a directive which is *optional* in multiple
977 resources, like Level? If so, I think I'd make it an optional directive
978 in Job, Client, and Pool, with precedence such that Job overrides Client
979 which in turn overrides Pool.
981 - New Storage specifications:
982 - Want to write to multiple storage devices simultaneously
983 - Want to write to multiple storage devices sequentially (in one job)
984 - Want to read/write simultaneously
985 - Key is MediaType -- it must match
987 Passed to SD as a sort of BSR record called Storage Specification
991 MediaType -> Next MediaType
993 Device -> Next Device
995 Allow multiple Storage specifications
1003 Allow Multiple Pool specifications (note, Pool currently
1005 Allow Multiple MediaType specifications in Dir conf
1006 Allow Multiple Device specifications in Dir conf
1007 Perhaps keep this in a single SSR
1008 Tie a Volume to a specific device by using a MediaType that
1009 is contained in only one device.
1010 In SD allow Device to have Multiple MediaTypes
1012 - Ideas from Jerry Scharf:
1013 First let's point out some big pluses that bacula has for this
1015 more importantly it's active. Thank you so much for that
1016 even more important, it's not flaky
1017 it has an open access catalog, opening many possibilities
1018 it's pushing toward heterogeneous systems capability
1020 Macintosh file client
1021 macs are an interesting niche, but I fear a server is a rathole
1022 working bare iron recovery for windows
1023 the option for inc/diff backups not reset on fileset revision
1024 a) use both change and inode update time against base time
1025 b) do the full catalog check (expensive but accurate)
1026 sizing guide (how much system is needed to back up N systems/files)
1027 consultants on using bacula in building a disaster recovery system
1028 an integration guide
1029 or how to get at fancy things that one could do with bacula
1030 logwatch code for bacula logs (or similar)
1031 linux distro inclusion of bacula (brings good and bad, but necessary)
1032 win2k/XP server capability (icky but you asked)
1033 support for Oracle database ??
1035 - Look at adding SQL server and Exchange support for Windows.
1036 - Create VolAddr for disk files in place of VolFile and VolBlock. This
1037 is needed to properly specify ranges.
1038 - Add progress of files/bytes to SD and FD.
1039 - Print warning message if FileId > 4 billion
1040 - do a "messages" before the first prompt in Console
1041 - Client does not show busy during Estimate command.
1042 - Implement Console mtx commands.
1043 - Implement a Mount Command and an Unmount Command where
1044 the users could specify a system command to be performed
1045 to do the mount, after which Bacula could attempt to
1046 read the device. This is for Removeable media such as a CDROM.
1047 - Most likely, this mount command would be invoked explicitly
1048 by the user using the current Console "mount" and "unmount"
1049 commands -- the Storage Daemon would do the right thing
1050 depending on the exact nature of the device.
1051 - As with tape drives, when Bacula wanted a new removable
1052 disk mounted, it would unmount the old one, and send a message
1053 to the user, who would then use "mount" as described above
1054 once he had actually inserted the disk.
1055 - Implement dump/print label to UA
1056 - Spool to disk only when the tape is full, then when a tape is hung move
1058 - bextract is sending everything to the log file ****FIXME****
1059 - Allow multiple Storage specifications (or multiple names on
1060 a single Storage specification) in the Job record. Thus a job
1061 can be backed up to a number of storage devices.
1062 - Implement some way for the File daemon to contact the Director
1063 to start a job or pass its DHCP obtained IP number.
1064 - Implement a query tape prompt/replace feature for a console
1065 - Make sure that Bacula rechecks the tape after the 20 min wait.
1066 - Set IO_NOWAIT on Bacula TCP/IP packets.
1067 - Try doing a raw partition backup and restore by mounting a
1069 - From Lars Kellers:
1070 Yes, it would allow to highly automatic the request for new tapes. If a
1071 tape is empty, bacula reads the barcodes (native or simulated), and if
1072 an unused tape is found, it runs the label command with all the
1073 necessary parameters.
1075 By the way can bacula automatically "move" an empty/purged volume say
1076 in the "short" pool to the "long" pool if this pool runs out of volume
1078 - What to do about "list files job=xxx".
1079 - Look at how fuser works and /proc/PID/fd that is how Nic found the
1080 file descriptor leak in Bacula.
1081 - Can we dynamically change FileSets?
1082 - If pool specified to label command and Label Format is specified,
1083 automatically generate the Volume name.
1084 - Add ExhautiveRestoreSearch
1085 - Look at the possibility of loading only the necessary
1086 data into the restore tree (i.e. do it one directory at a
1087 time as the user walks through the tree).
1088 - Possibly use the hash code if the user selects all for a restore command.
1089 - Fix "restore all" to bypass building the tree.
1090 - Prohibit backing up archive device (findlib/find_one.c:128)
1091 - Implement Release Device in the Job resource to unmount a drive.
1092 - Implement Acquire Device in the Job resource to mount a drive,
1093 be sure this works with admin jobs so that the user can get
1094 prompted to insert the correct tape. Possibly some way to say to
1095 run the job but don't save the files.
1096 - Make things like list where a file is saved case independent for
1098 - Implement a Recycle command
1099 - Start working on Base jobs.
1100 - From Phil Stracchino:
1101 It would probably be a per-client option, and would be called
1102 something like, say, "Automatically purge obsoleted jobs". What it
1103 would do is, when you successfully complete a Differential backup of a
1104 client, it would automatically purge all Incremental backups for that
1105 client that are rendered redundant by that Differential. Likewise,
1106 when a Full backup on a client completed, it would automatically purge
1107 all Differential and Incremental jobs obsoleted by that Full backup.
1108 This would let people minimize the number of tapes they're keeping on
1109 hand without having to master the art of retention times.
1110 - When doing a Backup send all attributes back to the Director, who
1111 would then figure out what files have been deleted.
1112 - Currently in mount.c:236 the SD simply creates a Volume. It should have
1113 explicit permission to do so. It should also mark the tape in error
1114 if there is an error.
1115 - Cancel waiting for Client connect in SD if FD goes away.
1117 - Implement timeout in response() when it should come quickly.
1118 - Implement a Slot priority (loaded/not loaded).
1119 - Implement "vacation" Incremental only saves.
1120 - Implement create "FileSet"?
1121 - Add prefixlinks to where or not where absolute links to FD.
1122 - Issue message to mount a new tape before the rewind.
1123 - Simplified client job initiation for portables.
1124 - If SD cannot open a drive, make it periodically retry.
1125 - Add more of the config info to the tape label.
1127 - Refine SD waiting output:
1128 Device is being positioned
1129 > Device is being positioned for append
1130 > Device is being positioned to file x
1132 - Figure out some way to estimate output size and to avoid splitting
1133 a backup across two Volumes -- this could be useful for writing CDROMs
1134 where you really prefer not to have it split -- not serious.
1135 - Make bcopy read through bad tape records.
1136 - Program files (i.e. execute a program to read/write files).
1137 Pass read date of last backup, size of file last time.
1138 - Add Signature type to File DB record.
1139 - CD into subdirectory when open()ing files for backup to
1140 speed up things. Test with testfind().
1141 - Priority job to go to top of list.
1142 - Why are save/restore of device different sizes (sparse?) Yup! Fix it.
1143 - Implement some way for the Console to dynamically create a job.
1144 - Solaris -I on tar for include list
1145 - Need a verbose mode in restore, perhaps to bsr.
1146 - bscan without -v is too quiet -- perhaps show jobs.
1147 - Add code to reject whole blocks if not wanted on restore.
1148 - Check if we can increase Bacula FD priorty in Win2000
1149 - Check if both CatalogFiles and UseCatalog are set to SD.
1150 - Possibly add email to Watchdog if drive is unmounted too
1151 long and a job is waiting on the drive.
1152 - After unmount, if restore job started, ask to mount.
1153 - Add UA rc and history files.
1154 - put termcap (used by console) in ./configure and
1155 allow -with-termcap-dir.
1156 - Fix Autoprune for Volumes to respect need for full save.
1157 - Compare tape to Client files (attributes, or attributes and data)
1158 - Make all database Ids 64 bit.
1159 - Allow console commands to detach or run in background.
1160 - Add SD message variables to control operator wait time
1161 - Maximum Operator Wait
1162 - Minimum Message Interval
1163 - Maximum Message Interval
1164 - Send Operator message when cannot read tape label.
1165 - Verify level=Volume (scan only), level=Data (compare of data to file).
1166 Verify level=Catalog, level=InitCatalog
1168 - Add keyword search to show command in Console.
1169 - Events : tape has more than xxx bytes.
1170 - Complete code in Bacula Resources -- this will permit
1171 reading a new config file at any time.
1172 - Handle ctl-c in Console
1173 - Implement script driven addition of File daemon to config files.
1174 - Think about how to make Bacula work better with File (non-tape) archives.
1175 - Write Unix emulator for Windows.
1176 - Make database type selectable by .conf files i.e. at runtime
1177 - Set flag for uname -a. Add to Volume label.
1178 - Restore files modified after date
1179 - SET LD_RUN_PATH=$HOME/mysql/lib/mysql
1180 - Remove duplicate fields from jcr (e.g. jcr.level and jcr.jr.Level, ...).
1181 - Timout a job or terminate if link goes down, or reopen link and query.
1182 - Concept of precious tapes (cannot be reused).
1183 - Make bcopy copy with a single tape drive.
1184 - Permit changing ownership during restore.
1187 > My suggestion: Add a feature on the systray menu-icon menu to request
1188 > an immediate backup now. This would be useful for laptop users who may
1189 > not be on the network when the regular scheduled backup is run.
1191 > My wife's suggestion: Add a setting to the win32 client to allow it to
1192 > shut down the machine after backup is complete (after, of course,
1193 > displaying a "System will shut down in one minute, click here to cancel"
1194 > warning dialog). This would be useful for sites that want user
1195 > woorkstations to be shut down overnight to save power.
1198 - Autolabel should be specified by DIR instead of SD.
1200 - Add media capacity
1201 - AutoScan (check checksum of tape)
1202 - Format command = "format /dev/nst0"
1206 - Seek resolution (usually corresponds to buffer size)
1207 - EODErrorCode=ENOSPC or code
1208 - Partial Read error code
1209 - Partial write error code
1210 - Nonformatted read error
1211 - Nonformatted write error
1212 - WriteProtected error
1216 - IgnoreCloseErrors=yes
1226 - Write a Storage daemon that uses pipes and
1227 standard Unix programs to write to the tape.
1229 - Need something that monitors the JCR queue and
1230 times out jobs by asking the deamons where they are.
1231 - Enhance Jmsg code to permit buffering and saving to disk.
1232 - Verify from Volume
1233 - Need report class for messages. Perhaps
1234 report resource where report=group of messages
1235 - enhance scan_attrib and rename scan_jobtype, and
1236 fill in code for "since" option
1237 - Director needs a time after which the report status is sent
1238 anyway -- or better yet, a retry time for the job.
1239 - Don't reschedule a job if previous incarnation is still running.
1240 - Some way to automatically backup everything is needed????
1241 - Need a structure for pending actions:
1243 - termination status (part of buffered msgs?)
1245 Read, Write, Clean, Delete
1246 - Login to Bacula; Bacula users with different permissions:
1247 owner, group, user, quotas
1248 - Store info on each file system type (probably in the job header on tape.
1249 This could be the output of df; or perhaps some sort of /etc/mtab record.
1251 ========= ideas ===============
1252 From: "Jerry K. Schieffer" <jerry@skylinetechnology.com>
1253 To: <kern@sibbald.com>
1254 Subject: RE: [Bacula-users] future large programming jobs
1255 Date: Thu, 26 Feb 2004 11:34:54 -0600
1257 I noticed the subject thread and thought I would offer the following
1258 merely as sources of ideas, i.e. something to think about, not even as
1259 strong as a request. In my former life (before retiring) I often
1260 dealt with backups and storage management issues/products as a
1261 developer and as a consultant. I am currently migrating my personal
1262 network from amanda to bacula specifically because of the ability to
1263 cross media boundaries during storing backups.
1264 Are you familiar with the commercial product called ADSM (I think IBM
1265 now sells it under the Tivoli label)? It has a couple of interesting
1266 ideas that may apply to the following topics.
1268 1. Migration: Consider that when you need to restore a system, there
1269 may be pressure to hurry. If all the information for a single client
1270 can eventually end up on the same media (and in chronological order),
1271 the restore is facillitated by not having to search past information
1272 from other clients. ADSM has the concept of "client affinity" that
1273 may be associated with it's storage pools. It seems to me that this
1274 concept (as an optional feature) might fit in your architecture for
1277 ADSM also has the concept of defining one or more storage pools as
1278 "copy pools" (almost mirrors, but only in the sense of contents).
1279 These pools provide the ability to have duplicte data stored both
1280 onsite and offsite. The copy process can be scheduled to be handled
1281 by their storage manager during periods when there is no backup
1282 activity. Again, the migration process might be a place to consider
1283 implementing something like this.
1286 > It strikes me that it would be very nice to be able to do things
1288 > have the Job(s) backing up the machines run, and once they have all
1289 > completed, start a migration job to copy the data from disks Volumes
1291 > a tape library and then to offsite storage. Maybe this can already
1293 > done with some careful scheduling and Job prioritzation; the events
1294 > mechanism described below would probably make it very easy.
1296 This is the goal. In the first step (before events), you simply
1298 the Migration to tape later.
1300 2. Base jobs: In ADSM, each copy of each stored file is tracked in
1301 the database. Once a file (unique by path and metadata such as dates,
1302 size, ownership, etc.) is in a copy pool, no more copies are made. In
1303 other words, when you start ADSM, it begins like your concept of a
1304 base job. After that it is in the "incremental" mode. You can
1305 configure the number of "generations" of files to be retained, plus a
1306 retention date after which even old generations are purged. The
1307 database tracks the contents of media and projects the percentage of
1308 each volume that is valid. When the valid content of a volume drops
1309 below a configured percentage, the valid data are migrated to another
1310 volume and the old volume is marked as empty. Note, this requires
1311 ADSM to have an idea of the contents of a client, i.e. marking the
1312 database when an existing file was deleted, but this would solve your
1313 issue of restoring a client without restoring deleted files.
1315 This is pretty far from what bacula now does, but if you are going to
1316 rip things up for Base jobs,.....
1317 Also, the benefits of this are huge for very large shops, especially
1318 with media robots, but are a pain for shops with manual media
1322 > Base jobs sound pretty useful, but I'm not dying for them.
1324 Nobody is dying for them, but when you see what it does, you will die
1330 ==============================
1333 - Implement FSM (File System Modules).
1334 - Audit M_ error codes to ensure they are correct and consistent.
1335 - Add variable break characters to lex analyzer.
1336 Either a bit mask or a string of chars so that
1337 the caller can change the break characters.
1338 - Make a single T_BREAK to replace T_COMMA, etc.
1339 - Ensure that File daemon and Storage daemon can
1340 continue a save if the Director goes down (this
1341 is NOT currently the case). Must detect socket error,
1342 buffer messages for later.
1343 - Add ability to backup to two Storage devices (two SD sessions) at
1344 the same time -- e.g. onsite, offsite.
1346 ======================================================
1348 It is somewhat like a Full save becomes an incremental since
1349 the Base job (or jobs) plus other non-base files.
1351 - A Base backup is same as Full backup, just different type.
1352 - New BaseFiles table that contains:
1354 BaseJobId - Base JobId referenced for this FileId (needed ???)
1355 JobId - JobId currently running
1356 FileId - File not backed up, exists in Base Job
1357 FileIndex - FileIndex from Base Job.
1358 i.e. for each base file that exists but is not saved because
1359 it has not changed, the File daemon sends the JobId, BaseId,
1360 FileId, FileIndex back to the Director who creates the DB entry.
1361 - To initiate a Base save, the Director sends the FD
1362 the FileId, and full filename for each file in the Base.
1363 - When the FD finds a Base file, he requests the Director to
1364 send him the full File entry (stat packet plus MD5), or
1365 conversely, the FD sends it to the Director and the Director
1366 says yes or no. This can be quite rapid if the FileId is kept
1367 by the FD for each Base Filename.
1368 - It is probably better to have the comparison done by the FD
1369 despite the fact that the File entry must be sent across the
1371 - An alternative would be to send the FD the whole File entry
1372 from the start. The disadvantage is that it requires a lot of
1373 space. The advantage is that it requires less communications
1375 - The Job record must be updated to indicate that one or more
1377 - At end of Job, FD returns:
1378 1. Count of base files/bytes not written to tape (i.e. matches)
1379 2. Count of base file that were saved i.e. had changed.
1380 - No tape record would be written for a Base file that matches, in the
1381 same way that no tape record is written for Incremental jobs where
1382 the file is not saved because it is unchanged.
1383 - On a restore, all the Base file records must explicitly be
1384 found from the BaseFile tape. I.e. for each Full save that is marked
1385 to have one or more Base Jobs, search the BaseFile for all occurrences
1387 - An optimization might be to make the BaseFile have:
1393 This would avoid the need to explicitly fetch each File record for
1394 the Base job. The Base Job record will be fetched to get the
1395 VolSessionId and VolSessionTime.
1396 =========================================================
1399 Handling removable disks
1401 From: Karl Cunningham <karlc@keckec.com>
1403 My backups are only to hard disk these days, in removable bays. This is my
1404 idea of how a backup to hard disk would work more smoothly. Some of these
1405 things Bacula does already, but I mention them for completeness. If others
1406 have better ways to do this, I'd like to hear about it.
1408 1. Accommodate several disks, rotated similar to how tapes are. Identified
1409 by partition volume ID or perhaps by the name of a subdirectory.
1410 2. Abort & notify the admin if the wrong disk is in the bay.
1411 3. Write backups to different subdirectories for each machine to be backed
1413 4. Volumes (files) get created as needed in the proper subdirectory, one
1415 5. When a disk is recycled, remove or zero all old backup files. This is
1416 important as the disk being recycled may be close to full. This may be
1417 better done manually since the backup files for many machines may be
1418 scattered in many subdirectories.
1425 - Fix bpipe.c so that it does not modify results pointer.
1426 ***FIXME*** calling sequence should be changed.
1427 - Fix restore of acls and extended attributes to count ERROR
1428 messages and make errors non-fatal.
1429 - Put save/restore various platform acl/xattrs on a pointer to simplify
1431 - Add blast attributes to DIR to SD.
1432 - Implement unmount of USB volumes.
1433 - Look into using Dart for testing
1434 http://public.kitware.com/Dart/HTML/Index.shtml