9 - Does ClientRunAfterJob fail the job on a bad return code?
10 - Document cleaning up the spool files:
11 db, pid, state, bsr, mail, conmsg, spool
12 - Document the multiple-drive-changer.txt script.
13 - Pruning with Admin job.
14 - Does WildFile match against full name? Doc.
15 - %d and %v only valid on Director, not for ClientRunBefore/After.
18 - Implement a way to disable a drive (so you can use the second
19 drive of an autochanger, and the first one will not be used or
21 - Implement code that makes the Dir aware that a drive is an
22 autochanger (so the user doesn't need to use the Autochanger = yes
26 - Make hardlink code at line 240 of find_one.c use binary search.
27 - Queue warning/error messages during restore so that they
28 are reported at the end of the report rather than being
29 hidden in the file listing ...
30 - Fix Maximum Changer Wait (and others) to accept qualifiers.
31 - Look at -D_FORTIFY_SOURCE=2
32 - Add Win32 FileSet definition somewhere
33 - Look at fixing restore status stats in SD.
34 - Make selection of Database used in restore correspond to
36 - Implement a mode that says when a hard read error is
37 encountered, read many times (as it currently does), and if the
38 block cannot be read, skip to the next block, and try again. If
39 that fails, skip to the next file and try again, ...
41 create table LevelType (LevelType binary(1), LevelTypeLong tinyblob);
42 insert into LevelType (LevelType,LevelTypeLong) values
46 - Add ACL to restore only to original location.
47 - Add a recursive mark command (rmark) to restore.
48 - "Minimum Job Interval = nnn" sets minimum interval between Jobs
49 of the same level and does not permit multiple simultaneous
50 running of that Job (i.e. lets any previous invocation finish
51 before doing Interval testing).
52 - Look at simplifying File exclusions.
53 - Fix store_yesno to be store_bitmask.
54 - New directive "Delete purged Volumes"
55 - new pool XXX with ScratchPoolId = MyScratchPool's PoolId and
56 let it fill itself, and RecyclePoolId = XXX's PoolId so I can
57 see if it become stable and I just have to supervise
59 - If I want to remove this pool, I set RecyclePoolId = MyScratchPool's
60 PoolId, and when it is empty remove it.
61 - Figure out how to recycle Scratch volumes back to the Scratch
64 - Allow Check Labels to be used with Bacula labels.
65 - "Resuming" a failed backup (lost line for example) by using the
66 failed backup as a sort of "base" job.
68 - Email to the user when the tape is about to need changing x
69 days before it needs changing.
70 - Command to show next tape that will be used for a job even
71 if the job is not scheduled.
72 --- create_file.c.orig Fri Jul 8 12:13:05 2005
73 +++ create_file.c Fri Jul 8 12:13:07 2005
75 attr->ofname, be.strerror());
78 + } else if(S_ISSOCK(attr->statp.st_mode)) {
79 + Dmsg1(200, "Skipping socket: %s\n", attr->ofname);
81 Dmsg1(200, "Restore node: %s\n", attr->ofname);
82 if (mknod(attr->ofname, attr->statp.st_mode, attr->statp.st_rdev) != 0 && errno != EEXIST) {
83 - From: Arunav Mandal <amandal@trolltech.com>
84 1. When jobs are running and bacula for some reason crashes or if I do a
85 restart it remembers and jobs it was running before it crashed or restarted
86 as of now I loose all jobs if I restart it.
88 2. When spooling and in the midway if client is disconnected for instance a
89 laptop bacula completely discard the spool. It will be nice if it can write
90 that spool to tape so there will be some backups for that client if not all.
92 3. We have around 150 clients machines it will be nice to have a option to
93 upgrade all the client machines bacula version automatically.
95 4. Atleast one connection should be reserved for the bconsole so at heavy load
96 I should connect to the director via bconsole which at sometimes I can't
98 5. Another most important feature that is missing, say at 10am I manually
99 started backup of client abc and it was a full backup since client abc has
100 no backup history and at 10.30am bacula again automatically started backup of
101 client abc as that was in the schedule. So now we have 2 multiple Full
102 backups of the same client and if we again try to start a full backup of
103 client backup abc bacula won't complain. That should be fixed.
105 - Fix bpipe.c so that it does not modify results pointer.
106 ***FIXME*** calling sequence should be changed.
108 #3 Migration (Move, Copy, Archive Jobs)
109 #7 Single Job Writing to Multiple Storage Devices
110 - Reserve blocks other restore jobs when first cannot connect
112 - Add true/false to conf same as yes/no
113 - For Windows disaster recovery see http://unattended.sf.net/
114 - regardless of the retention period, Bacula will not prune the
115 last Full, Diff, or Inc File data until a month after the
116 retention period for the last Full backup that was done.
117 - update volume=xxx --- add status=Full
118 - Remove old spool files on startup.
119 - Exclude SD spool/working directory.
120 - Refuse to prune last valid Full backup. Same goes for Catalog.
122 - Make a callback when Rerun failed levels is called.
123 - Give Python program access to Scheduled jobs.
124 - Add setting Volume State via Python.
125 - Python script to save with Python, not save, save with Bacula.
126 - Python script to do backup.
128 - Change the Priority, Client, Storage, JobStatus (error)
129 at the start of a job.
130 - Why is SpoolDirectory = /home/bacula/spool; not reported
131 as an error when writing a DVD?
132 - Make bootstrap file handle multiple MediaTypes (SD)
133 - Remove all old Device resource code in Dir and code to pass it
134 back in SD -- better, rework it to pass back device statistics.
135 - Check locking of resources -- be sure to lock devices where previously
136 resources were locked.
137 - The last part is left in the spool dir.
140 - In restore don't compare byte count on a raw device -- directory
141 entry does not contain bytes.
142 - To mark files as deleted, run essentially a Verify to disk, and
143 when a file is found missing (MarkId != JobId), then create
144 a new File record with FileIndex == -1. This could be done
145 by the FD at the same time as the backup.
149 MA = (last_MA * 3 + rate) / 4
150 rate = (bytes - last_bytes) / (runtime - last_runtime)
151 - Max Vols limit in Pool off by one?
152 - Implement Files/Bytes,... stats for restore job.
153 - Implement Total Bytes Written, ... for restore job.
154 - Despool attributes simultaneously with data in a separate
155 thread, rejoined at end of data spooling.
156 - 7. Implement new Console commands to allow offlining/reserving drives,
157 and possibly manipulating the autochanger (much asked for).
158 - Add start/end date editing in messages (%t %T, %e?) ...
159 - Add ClientDefs similar to JobDefs.
160 - Print more info when bextract -p accepts a bad block.
161 - Fix FD JobType to be set before RunBeforeJob in FD.
162 - Look at adding full Volume and Pool information to a Volume
163 label so that bscan can get *all* the info.
164 - If the user puts "Purge Oldest Volume = yes" or "Recycle Oldest Volume = yes"
165 and there is only one volume in the pool, refuse to do it -- otherwise
166 he fills the Volume, then immediately starts reusing it.
167 - Implement copies and stripes.
168 - Add history file to console.
169 - Each file on tape creates a JobMedia record. Peter has 4 million
170 files spread over 10000 tape files and four tapes. A restore takes
171 16 hours to build the restore list.
172 - Add and option to check if the file size changed during backup.
173 - Make sure SD deletes spool files on error exit.
174 - Delete old spool files when SD starts.
175 - When labeling tapes, if you enter 000026, Bacula uses
176 the tape index rather than the Volume name 000026.
177 - Add offline tape command to Bacula console.
179 Enter MediaId or Volume name: 32
180 Enter new Volume name: DLT-20Dec04
181 Automatically selected Pool: Default
182 Connecting to Storage daemon DLTDrive at 192.168.68.104:9103 ...
183 Sending relabel command from "DLT-28Jun03" to "DLT-20Dec04" ...
184 block.c:552 Write error at 0:0 on device /dev/nst0. ERR=Bad file descriptor.
185 Error writing final EOF to tape. This tape may not be readable.
186 dev.c:1207 ioctl MTWEOF error on /dev/nst0. ERR=Permission denied.
187 askdir.c:219 NULL Volume name. This shouldn't happen!!!
188 3912 Failed to label Volume: ERR=dev.c:1207 ioctl MTWEOF error on /dev/nst0. ERR=Permission denied.
189 Label command failed for Volume DLT-20Dec04.
190 Do not forget to mount the drive!!!
191 - Bug: if a job is manually scheduled to run later, it does not appear
192 in any status report and cannot be cancelled.
194 ==== Keeping track of deleted files ====
195 My "trick" for keeping track of deletions is the following.
196 Assuming the user turns on this option, after all the files
197 have been backed up, but before the job has terminated, the
198 FD will make a pass through all the files and send their
199 names to the DIR (*exactly* the same as what a Verify job
200 currently does). This will probably be done at the same
201 time the files are being sent to the SD avoiding a second
202 pass. The DIR will then compare that to what is stored in
203 the catalog. Any files in the catalog but not in what the
204 FD sent will receive a catalog File entry that indicates
205 that at that point in time the file was deleted.
207 During a restore, any file initially picked up by some
208 backup (Full, ...) then subsequently having a File entry
209 marked "delete" will be removed from the tree, so will not
210 be restored. If a file with the same name is later OK it
211 will be inserted in the tree -- this already happens. All
212 will be consistent except for possible changes during the
215 Since I'm on the subject, some of you may be wondering what
216 the utility of the in memory tree is if you are going to
217 restore everything (at least it comes up from time to time
218 on the list). Well, it is still *very* useful because it
219 allows only the last item found for a particular filename
220 (full path) to be entered into the tree, and thus if a file
221 is backed up 10 times, only the last copy will be restored.
222 I recently (last Friday) restored a complete directory, and
223 the Full and all the Differential and Incremental backups
224 spanned 3 Volumes. The first Volume was not even mounted
225 because all the files had been updated and hence backed up
226 since the Full backup was made. In this case, the tree
227 saved me a *lot* of time.
229 Make sure this information is stored on the tape too so
230 that it can be restored directly from the tape.
232 Comments from Martin Simmons (I think they are all covered):
233 Ok, that should cover the basics. There are few issues though:
235 - Restore will depend on the catalog. I think it is better to include the
236 extra data in the backup as well, so it can be seen by bscan and bextract.
238 - I'm not sure if it will preserve multiple hard links to the same inode. Or
239 maybe adding or removing links will cause the data to be dumped again?
241 - I'm not sure if it will handle renamed directories. Possibly it will work
242 by dumping the whole tree under a renamed directory?
244 - It remains to be seen how the backup performance of the DIR's will be
245 affected when comparing the catalog for a large filesystem.
250 - Add Pool/Storage override regression test.
251 - Add delete JobId to regression.
252 - Add a regression test for dbcheck.
253 - New test to add bscan to four-concurrent-jobs regression,
254 i.e. after the four-concurrent jobs zap the
255 database as is done in the bscan-test, then use bscan to
256 restore the database, do a restore and compare with the
258 - Add restore of specific JobId to regression (item 3
259 on the restore prompt)
260 - Add IPv6 to regression
261 - Add database test to regression. Test each function like delete,
264 - AntiVir can slow down backups on Win32 systems.
265 - Win32 systems with FAT32 can be much slower than NTFS for
266 more than 1000 files per directory.
270 - A HOLD command to stop all jobs from starting.
271 - A PAUSE command to pause all running jobs ==> release the
273 - Media Type = LTO,LTO-2,LTO-3
274 Media Type Read = LTO,LTO2,LTO3
275 Media Type Write = LTO2, LTO3
277 === From Carsten Menke <bootsy52@gmx.net>
279 Following is a list of what I think in the situations where I'm faced with,
280 could be a usefull enhancement to bacula, which I'm certain other users will
281 benefit from as well.
283 1. NextJob/NextJobs Directive within a Job Resource in the form of
284 NextJobs = job1,job2.
287 I currently solved the problem with running multiple jobs each after each
288 by setting the Max Wait Time for a job to 8 hours, and give
289 the jobs different Priorities. However, there scenarios where
290 1 Job is directly depending on another job, so if the former job fails,
291 the job after it needn't to be run
292 while maybe other jobs should run despite of that
295 A Backup Job and a Verify job, if the backup job fails there is no need to run
296 the verify job, as the backup job already failed. However, one may like
297 to backup the Catalog to disk despite of that the main backup job failed.
300 I see that this is related to the Event Handlers which are on the ToDo
301 list, also it is maybe a good idea to check for the return value and
302 execute different actions based on the return value
305 3. offline capability to bconsole
308 Currently I use a script which I execute within the last Job via the
309 RunAfterJob Directive, to release and eject the tape.
310 So I have to call bconsole "release=Storage-Name" and afterwards
311 mt -f /dev/nst0 eject to get the tape out.
313 If I have multiple Storage Devices, than these may not be /dev/nst0 and
314 I have to modify the script or call it with parameters etc.
315 This would actually not be needed, as everything is already defined
316 in bacula-sd.conf and if I can invoke bconsole with the
317 storage name via $1 in the script than I'm done and information is
320 4. %s for Storage Name added to the chars being substituted in "RunAfterJob"
324 For the reason mentioned in 3. to have the ability to call a
325 script with /scripts/foobar %s and in the script use $1
326 to pass the Storage Name to bconsole
328 5. Setting Volume State within a Job Resource
331 Instead of using "Maximum Volume Jobs" in the Pool Resource,
332 I would have the possibilty to define
333 in a Job Resource that after this certain job is run, the Volume State
334 should be set to "Volume State = Used", this give more flexibility (IMHO).
336 6. Localization of Bacula Messages
339 Unfortunatley many,many people I work with don't speak english very well.
340 So if at least the Reporting messages would be localized then they
341 would understand that they have to change the tape,etc. etc.
343 I volunteer to do the german translations, and if I can convince my wife also
344 french and Morre (western african language).
346 7. OK, this is evil, probably bound to security risks and maybe not possible
347 due to the design of bacula.
349 Implementation of Backtics ( `command` ) for shell comand execution to
350 the "Label Format" Directive.
354 Currently I have defined BACULA_DAY_OF_WEEK="day1|day2..." resulting in
355 Label Format = "HolyBackup-${BACULA_DAY_OF_WEEK[${WeekDay}]}". If I could
356 use backticks than I could use "Label Format = HolyBackup-`date +%A` to have
357 the localized name for the day of the week appended to the
358 format string. Then I have the tape labeled automatically with weekday
359 name in the correct language.
361 - Yes, that is surely the case. I probably should turn those into Warning
362 errors. In addition, you just made me think that it might not be bad to
363 add an option to check the file size after backing up the file and
364 report if it changes. This would be done as an option because it would
367 Kern, good idea. If you do do that, mention in the output: file
368 shrunk, or file expanded, just to make it obvious to the user
369 (without having to the refer to file size), just how the file size
372 Would this option be for all file, or just one file? Or a fileset?
373 - Make output from status use html table tags for nicely
374 presenting in a browser.
375 - Can one write tapes faster with 8192 byte block sizes?
376 - Document security problems with the same password for everyone in
377 rpm and Win32 releases.
378 - Browse generations of files.
379 - I've seen an error when my catalog's File table fills up. I
380 then have to recreate the File table with a larger maximum row
381 size. Relevant information is at
382 http://dev.mysql.com/doc/mysql/en/Full_table.html ; I think the
383 "Installing and Configuring MySQL" chapter should talk a bit
384 about this potential problem, and recommend a solution.
385 - For Solaris must use POSIX awk.
386 - Want speed of writing to tape while despooling.
387 - Supported autochanger:
395 Wangtek 6525ES (SCSI-1 QIC drive, 525MB), under Linux 2.4.something,
396 bacula 1.36.0/1 works with blocksize 16k INSIDE bacula-sd.conf.
397 - Add regex from http://www.pcre.org to Bacula for Win32.
398 - Use only shell tools no make in CDROM package.
399 - Include within include does it work?
400 - Implement a Pool of type Cleaning?
401 - Implement VolReadTime and VolWriteTime in SD
402 - Modify Backing up Your Database to include a bootstrap file.
403 - Think about making certain database errors fatal.
404 - Look at correcting the time jump in the scheduler for daylight
405 savings time changes.
406 - Add a "real" timer to network connections.
407 - Promote to Full = Time period
408 - Check dates entered by user for correctness (month/day/... ranges)
409 - Compress restore Volume listing by date and first file.
410 - Look at patches/bacula_db.b2z postgresql that loops during restore.
412 - Perhaps add read/write programs and/or plugins to FileSets.
413 - How to handle backing up portables ...
414 - Add some sort of guaranteed Interval for upgrading jobs.
415 - Can we write the state file after every job terminates? On Win32
416 the system crashes and the state file is not updated.
419 Documentation to do: (any release a little bit at a time)
420 - Doc to do unmount before removing magazine.
421 - Alternative to static linking "ldd prog" save all binaries listed,
422 restore them and point LD_LIBRARY_PATH to them.
423 - Document add "</dev/null >/dev/null 2>&1" to the bacula-fd command line
424 - Document query file format.
425 - Add more documentation for bsr files.
426 - Document problems with Verify and pruning.
427 - Document how to use multiple databases.
428 - VXA drives have a "cleaning required"
429 indicator, but Exabyte recommends preventive cleaning after every 75
432 In this context, it should be noted that Exabyte has a command-line
433 vxatool utility available for free download. (The current version is
434 vxatool-3.72.) It can get diagnostic info, read, write and erase tapes,
435 test the drive, unload tapes, change drive settings, flash new firmware,
437 Of particular interest in this context is that vxatool <device> -i will
438 report, among other details, the time since last cleaning in tape motion
439 minutes. This information can be retrieved (and settings changed, for
440 that matter) through the generic-SCSI device even when Bacula has the
441 regular tape device locked. (Needless to say, I don't recommend
442 changing tape settings while a job is running.)
443 - Lookup HP cleaning recommendations.
444 - Lookup HP tape replacement recommendations (see trouble shooting autochanger)
445 - Document doing table repair
448 ===================================
449 - Add macro expansions in JobDefs.
450 Run Before Job = "SomeFile %{Level} %{Client}"
451 Write Bootstrap="/some/dir/%{JobName}_%{Client}.bsr"
452 - Use non-blocking network I/O but if no data is available, use
454 - Use gather write() for network I/O.
455 - Autorestart on crash.
456 - Add bandwidth limiting.
457 - Add acks every once and a while from the SD to keep
458 the line from timing out.
459 - When an error in input occurs and conio beeps, you can back
460 up through the prompt.
461 - Detect fixed tape block mode during positioning by looking at
462 block numbers in btape "test". Possibly adjust in Bacula.
463 - Fix list volumes to output volume retention in some other
464 units, perhaps via a directive.
465 - If opening a tape in read/write mode fails attempt to open
466 it in read-only mode, and mark the tape for read only.
467 - Allow Simultaneous Priorities = yes => run up to Max concurrent jobs even
468 with multiple priorities.
469 - If you use restore replace=never, the directory attributes for
470 non-existent directories will not be restored properly.
472 - see lzma401.zip in others directory for new compression
474 - Minimal autochanger handling in Bacula and in btape.
475 - Look into how tar does not save sockets and the possiblity of
476 not saving them in Bacula (Martin Simmons reported this).
477 - Fix restore jobs so that multiple jobs can run if they
478 are not using the same tape(s).
479 - Allow the user to select JobType for manual pruning/purging.
480 - bscan does not put first of two volumes back with all info in
482 - Implement the FreeBSD nodump flag in chflags.
483 - Figure out how to make named console messages go only to that
484 console and to the non-restricted console (new console class?).
485 - Make restricted console prompt for password if *ask* is set or
486 perhaps if password is undefined.
487 - Implement "from ISO-date/time every x hours/days/weeks/months" in
490 ==== from Marc Schoechlin
491 - the help-command should be more verbose
492 (it should explain the paramters of the different
494 -> its time-comsuming to consult the manual anytime
495 you need a special parameter
496 -> maybe its more easy to maintain this, if the
497 descriptions of that commands are outsourced to
499 - the cd-command should allow complete paths
500 i.e. cd /foo/bar/foo/bar
501 -> if a customer mails me the path to a certain file,
502 its faster to enter the specified directory
503 - if the password is not configured in bconsole.conf
504 you should be asked for it.
505 -> sometimes you like to do restore on a customer-machine
506 which shouldnt know the password for bacula.
507 -> adding the password to the file favours admins
508 to forget to remove the password after usage
510 the protection of that file is less important
511 - long-listed-output of commands should be scrollable
512 like the unix more/less-command does
513 -> if someone runs 200 and more machines, the lists could
514 be a little long and complex
515 - command-output should be shown column by column
516 to reduce scrolling and to increase clarity
518 - lsmark should list the selected files with full
520 - wildcards for selecting and file and directories would be nice
521 - any actions should be interuptable with STRG+C
522 - command-expansion would be pretty cool
524 - When the replace Never option is set, new directory permissions
525 are not restored. See bug 213. To fix this requires creating a
526 list of newly restored directories so that those directory
527 permissions *can* be restored.
528 - Compaction of Disk space by "migrating" Volumes that have pruned
529 Jobs (what criteria? size, #jobs, time).
530 - Add prune all command
531 - Document fact that purge can destroy a part of a restore by purging
532 one volume while others remain valid -- perhaps mark Jobs.
533 - Add multiple-media-types.txt
534 - look at mxt-changer.html
535 - Make ? do a help command (no return needed).
536 - Implement restore directory.
537 - Document streams and how to implement them.
538 - Try not to re-backup a file if a new hard link is added.
539 - Add feature to backup hard links only, but not the data.
540 - Fix stream handling to be simpler.
541 - Add Priority and Bootstrap to Run a Job.
542 - Eliminate Restore "Run Restore Job" prompt by allowing new "run command
544 - Remove View FileSet button from Run a Job dialog.
545 - Handle prompt for restore job at end of Restore command.
546 - Add display of total selected files to Restore window.
547 - Add tree pane to left of window.
548 - Add progress meter.
549 - Max wait time or max run time causes seg fault -- see runtime-bug.txt
550 - Document writing to a CD/DVD with Bacula.
551 - Add a "base" package to the window installer for pthreadsVCE.dll
552 which is needed by all packages.
553 - Add message to user to check for fixed block size when the forward
554 space test fails in btape.
555 - When unmarking a directory check if all files below are unmarked and
556 then remove the + flag -- in the restore tree.
557 - Possibly implement: Action = Unmount Device="TapeDrive1" in Admin jobs.
558 - Setup lrrd graphs: (http://www.linpro.no/projects/lrrd/) Mike Acar.
559 - Revisit the question of multiple Volumes (disk) on a single device.
560 - Add a block copy option to bcopy.
561 - Investigate adding Mac Resource Forks.
562 - Finish work on Gnome restore GUI.
563 - Fix "llist jobid=xx" where no fileset or client exists.
564 - For each job type (Admin, Restore, ...) require only the really necessary
565 fields.- Pass Director resource name as an option to the Console.
566 - Add a "batch" mode to the Console (no unsolicited queries, ...).
567 - Add a .list all files in the restore tree (probably also a list all files)
568 Do both a long and short form.
569 - Allow browsing the catalog to see all versions of a file (with
570 stat data on each file).
571 - Restore attributes of directory if replace=never set but directory
573 - Use SHA1 on authentication if possible.
574 - See comtest-xxx.zip for Windows code to talk to USB.
575 - Add John's appended files:
576 Appended = { /files/server/logs/http/*log }
577 and such files would be treated as follows.On a FULL backup, they would
578 be backed up like any other file.On an INCREMENTAL backup, where a
579 previous INCREMENTAL or FULL was already in thecatalogue and the length
580 of the file wasgreater than the length of the last backup, only thedata
581 added since the last backup will be dumped.On an INCREMENTAL backup, if
582 the length of the file is less than thelength of the file with the same
583 name last backed up, the completefile is dumped.On Windows systems, with
584 creation date of files, we can be evensmarter about this and not count
585 entirely upon the length.On a restore, the full and all incrementals
586 since it will beapplied in sequence to restore the file.
587 - Check new HAVE_WIN32 open bits.
588 - Check if the tape has moved before writing.
589 - Handling removable disks -- see below:
590 - Keep track of tape use time, and report when cleaning is necessary.
591 - Add FromClient and ToClient keywords on restore command (or
592 BackupClient RestoreClient).
593 - Implement a JobSet, which groups any number of jobs. If the
594 JobSet is started, all the jobs are started together.
595 Allow Pool, Level, and Schedule overrides.
596 - Enhance cancel to timeout BSOCK packets after a specific delay.
597 - Do scheduling by UTC using gmtime_r() in run_conf, scheduler, and
598 ua_status.!!! Thanks to Alan Brown for this tip.
599 - Look at updating Volume Jobs so that Max Volume Jobs = 1 will work
600 correctly for multiple simultaneous jobs.
601 - Correct code so that FileSet MD5 is calculated for < and | filename
603 - Implement the Media record flag that indicates that the Volume does disk
605 - Implement VolAddr, which is used when Volume is addressed like a disk,
606 and form it from VolFile and VolBlock.
607 - Make multiple restore jobs for multiple media types specifying
608 the proper storage type.
609 - Fix fast block rejection (stored/read_record.c:118). It passes a null
610 pointer (rec) to try_repositioning().
611 - Look at extracting Win data from BackupRead.
612 - Implement RestoreJobRetention? Maybe better "JobRetention" in a Job,
613 which would take precidence over the Catalog "JobRetention".
614 - Implement Label Format in Add and Label console commands.
615 - Possibly up network buffers to 65K. Put on variable.
616 - Put email tape request delays on one or more variables. User wants
617 to cancel the job after a certain time interval. Maximum Mount Wait?
618 - Job, Client, Device, Pool, or Volume?
619 Is it possible to make this a directive which is *optional* in multiple
620 resources, like Level? If so, I think I'd make it an optional directive
621 in Job, Client, and Pool, with precedence such that Job overrides Client
622 which in turn overrides Pool.
624 - New Storage specifications:
625 - Want to write to multiple storage devices simultaneously
626 - Want to write to multiple storage devices sequentially (in one job)
627 - Want to read/write simultaneously
628 - Key is MediaType -- it must match
630 Passed to SD as a sort of BSR record called Storage Specification
634 MediaType -> Next MediaType
636 Device -> Next Device
638 Allow multiple Storage specifications
646 Allow Multiple Pool specifications (note, Pool currently
648 Allow Multiple MediaType specifications in Dir conf
649 Allow Multiple Device specifications in Dir conf
650 Perhaps keep this in a single SSR
651 Tie a Volume to a specific device by using a MediaType that
652 is contained in only one device.
653 In SD allow Device to have Multiple MediaTypes
655 - Ideas from Jerry Scharf:
656 First let's point out some big pluses that bacula has for this
658 more importantly it's active. Thank you so much for that
659 even more important, it's not flaky
660 it has an open access catalog, opening many possibilities
661 it's pushing toward heterogeneous systems capability
663 Macintosh file client
664 macs are an interesting niche, but I fear a server is a rathole
665 working bare iron recovery for windows
666 the option for inc/diff backups not reset on fileset revision
667 a) use both change and inode update time against base time
668 b) do the full catalog check (expensive but accurate)
669 sizing guide (how much system is needed to back up N systems/files)
670 consultants on using bacula in building a disaster recovery system
672 or how to get at fancy things that one could do with bacula
673 logwatch code for bacula logs (or similar)
674 linux distro inclusion of bacula (brings good and bad, but necessary)
675 win2k/XP server capability (icky but you asked)
676 support for Oracle database ??
678 - Look at adding SQL server and Exchange support for Windows.
679 - Make dev->file and dev->block_num signed integers so that -1 can
680 be an invalid value which happens with BSR.
681 - Create VolAddr for disk files in place of VolFile and VolBlock. This
682 is needed to properly specify ranges.
683 - Add progress of files/bytes to SD and FD.
684 - Print warning message if FileId > 4 billion
685 - do a "messages" before the first prompt in Console
686 - Client does not show busy during Estimate command.
687 - Implement Console mtx commands.
688 - Implement a Mount Command and an Unmount Command where
689 the users could specify a system command to be performed
690 to do the mount, after which Bacula could attempt to
691 read the device. This is for Removeable media such as a CDROM.
692 - Most likely, this mount command would be invoked explicitly
693 by the user using the current Console "mount" and "unmount"
694 commands -- the Storage Daemon would do the right thing
695 depending on the exact nature of the device.
696 - As with tape drives, when Bacula wanted a new removable
697 disk mounted, it would unmount the old one, and send a message
698 to the user, who would then use "mount" as described above
699 once he had actually inserted the disk.
700 - Implement dump/print label to UA
701 - Spool to disk only when the tape is full, then when a tape is hung move
703 - bextract is sending everything to the log file ****FIXME****
704 - Allow multiple Storage specifications (or multiple names on
705 a single Storage specification) in the Job record. Thus a job
706 can be backed up to a number of storage devices.
707 - Implement some way for the File daemon to contact the Director
708 to start a job or pass its DHCP obtained IP number.
709 - Implement a query tape prompt/replace feature for a console
710 - Copy console @ code to gnome2-console
711 - Make AES the only encryption algorithm see
712 http://csrc.nist.gov/CryptoToolkit/aes/). It's
713 an officially adopted standard, has survived peer
714 review, and provides keys up to 256 bits.
715 - Take a careful look at SetACL http://setacl.sourceforge.net
716 - Make tree walk routines like cd, ls, ... more user friendly
717 by handling spaces better.
718 - Make sure that Bacula rechecks the tape after the 20 min wait.
719 - Set IO_NOWAIT on Bacula TCP/IP packets.
720 - Try doing a raw partition backup and restore by mounting a
723 Yes, it would allow to highly automatic the request for new tapes. If a
724 tape is empty, bacula reads the barcodes (native or simulated), and if
725 an unused tape is found, it runs the label command with all the
726 necessary parameters.
728 By the way can bacula automatically "move" an empty/purged volume say
729 in the "short" pool to the "long" pool if this pool runs out of volume
731 - What to do about "list files job=xxx".
732 - Get and test MySQL 4.0
733 - Look at how fuser works and /proc/PID/fd that is how Nic found the
734 file descriptor leak in Bacula.
735 - Implement WrapCounters in Counters.
736 - Add heartbeat from FD to SD if hb interval expires.
737 - Can we dynamically change FileSets?
738 - If pool specified to label command and Label Format is specified,
739 automatically generate the Volume name.
740 - Why can't SQL do the filename sort for restore?
741 - Add ExhautiveRestoreSearch
742 - Look at the possibility of loading only the necessary
743 data into the restore tree (i.e. do it one directory at a
744 time as the user walks through the tree).
745 - Possibly use the hash code if the user selects all for a restore command.
746 - Fix "restore all" to bypass building the tree.
747 - Prohibit backing up archive device (findlib/find_one.c:128)
748 - Implement Release Device in the Job resource to unmount a drive.
749 - Implement Acquire Device in the Job resource to mount a drive,
750 be sure this works with admin jobs so that the user can get
751 prompted to insert the correct tape. Possibly some way to say to
752 run the job but don't save the files.
753 - Make things like list where a file is saved case independent for
756 - Use autochanger to handle multiple devices.
757 - On Windows with very long path names, it may be impossible to create
758 a file (and thus restore it) because the total length is too long.
759 We must cd into the directory then create the file without the
761 - Implement a Recycle command
762 - Test a second language e.g. french.
763 - Start working on Base jobs.
764 - Implement UnsavedFiles DB record.
765 - From Phil Stracchino:
766 It would probably be a per-client option, and would be called
767 something like, say, "Automatically purge obsoleted jobs". What it
768 would do is, when you successfully complete a Differential backup of a
769 client, it would automatically purge all Incremental backups for that
770 client that are rendered redundant by that Differential. Likewise,
771 when a Full backup on a client completed, it would automatically purge
772 all Differential and Incremental jobs obsoleted by that Full backup.
773 This would let people minimize the number of tapes they're keeping on
774 hand without having to master the art of retention times.
775 - When doing a Backup send all attributes back to the Director, who
776 would then figure out what files have been deleted.
777 - Currently in mount.c:236 the SD simply creates a Volume. It should have
778 explicit permission to do so. It should also mark the tape in error
779 if there is an error.
780 - Cancel waiting for Client connect in SD if FD goes away.
782 - Implement timeout in response() when it should come quickly.
783 - Implement a Slot priority (loaded/not loaded).
784 - Implement "vacation" Incremental only saves.
785 - Implement create "FileSet"?
786 - Add prefixlinks to where or not where absolute links to FD.
787 - Issue message to mount a new tape before the rewind.
788 - Simplified client job initiation for portables.
789 - If SD cannot open a drive, make it periodically retry.
790 - Add more of the config info to the tape label.
792 - If tape is marked read-only, then try opening it read-only rather than
793 failing, and remember that it cannot be written.
794 - Refine SD waiting output:
795 Device is being positioned
796 > Device is being positioned for append
797 > Device is being positioned to file x
799 - Figure out some way to estimate output size and to avoid splitting
800 a backup across two Volumes -- this could be useful for writing CDROMs
801 where you really prefer not to have it split -- not serious.
802 - Have SD compute MD5 or SHA1 and compare to what FD computes.
803 - Make VolumeToCatalog calculate an MD5 or SHA1 from the
804 actual data on the Volume and compare it.
805 - Implement Bacula plugins -- design API
806 - Make bcopy read through bad tape records.
807 - Program files (i.e. execute a program to read/write files).
808 Pass read date of last backup, size of file last time.
809 - Add Signature type to File DB record.
810 - CD into subdirectory when open()ing files for backup to
811 speed up things. Test with testfind().
812 - Priority job to go to top of list.
813 - Why are save/restore of device different sizes (sparse?) Yup! Fix it.
814 - Implement some way for the Console to dynamically create a job.
815 - Solaris -I on tar for include list
816 - Need a verbose mode in restore, perhaps to bsr.
817 - bscan without -v is too quiet -- perhaps show jobs.
818 - Add code to reject whole blocks if not wanted on restore.
819 - Check if we can increase Bacula FD priorty in Win2000
820 - Make sure the MaxVolFiles is fully implemented in SD
821 - Check if both CatalogFiles and UseCatalog are set to SD.
822 - Possibly add email to Watchdog if drive is unmounted too
823 long and a job is waiting on the drive.
824 - After unmount, if restore job started, ask to mount.
825 - Add UA rc and history files.
826 - put termcap (used by console) in ./configure and
827 allow -with-termcap-dir.
828 - Fix Autoprune for Volumes to respect need for full save.
829 - Compare tape to Client files (attributes, or attributes and data)
830 - Make all database Ids 64 bit.
831 - Allow console commands to detach or run in background.
832 - Fix status delay on storage daemon during rewind.
833 - Add SD message variables to control operator wait time
834 - Maximum Operator Wait
835 - Minimum Message Interval
836 - Maximum Message Interval
837 - Send Operator message when cannot read tape label.
838 - Verify level=Volume (scan only), level=Data (compare of data to file).
839 Verify level=Catalog, level=InitCatalog
841 - Add keyword search to show command in Console.
842 - Events : tape has more than xxx bytes.
843 - Complete code in Bacula Resources -- this will permit
844 reading a new config file at any time.
845 - Handle ctl-c in Console
846 - Implement script driven addition of File daemon to config files.
847 - Think about how to make Bacula work better with File (non-tape) archives.
848 - Write Unix emulator for Windows.
849 - Put memory utilization in Status output of each daemon
850 if full status requested or if some level of debug on.
851 - Make database type selectable by .conf files i.e. at runtime
852 - Set flag for uname -a. Add to Volume label.
853 - Restore files modified after date
854 - SET LD_RUN_PATH=$HOME/mysql/lib/mysql
855 - Remove duplicate fields from jcr (e.g. jcr.level and jcr.jr.Level, ...).
856 - Timout a job or terminate if link goes down, or reopen link and query.
857 - Concept of precious tapes (cannot be reused).
858 - Make bcopy copy with a single tape drive.
859 - Permit changing ownership during restore.
862 > My suggestion: Add a feature on the systray menu-icon menu to request
863 > an immediate backup now. This would be useful for laptop users who may
864 > not be on the network when the regular scheduled backup is run.
866 > My wife's suggestion: Add a setting to the win32 client to allow it to
867 > shut down the machine after backup is complete (after, of course,
868 > displaying a "System will shut down in one minute, click here to cancel"
869 > warning dialog). This would be useful for sites that want user
870 > woorkstations to be shut down overnight to save power.
873 - Autolabel should be specified by DIR instead of SD.
876 - AutoScan (check checksum of tape)
877 - Format command = "format /dev/nst0"
881 - Seek resolution (usually corresponds to buffer size)
882 - EODErrorCode=ENOSPC or code
883 - Partial Read error code
884 - Partial write error code
885 - Nonformatted read error
886 - Nonformatted write error
887 - WriteProtected error
891 - IgnoreCloseErrors=yes
901 - FD sends unsaved file list to Director at end of job (see
903 - File daemon should build list of files skipped, and then
904 at end of save retry and report any errors.
905 - Write a Storage daemon that uses pipes and
906 standard Unix programs to write to the tape.
908 - Need something that monitors the JCR queue and
909 times out jobs by asking the deamons where they are.
910 - Enhance Jmsg code to permit buffering and saving to disk.
911 - device driver = "xxxx" for drives.
913 - Ensure that /dev/null works
914 - Need report class for messages. Perhaps
915 report resource where report=group of messages
916 - enhance scan_attrib and rename scan_jobtype, and
917 fill in code for "since" option
918 - Director needs a time after which the report status is sent
919 anyway -- or better yet, a retry time for the job.
920 - Don't reschedule a job if previous incarnation is still running.
921 - Some way to automatically backup everything is needed????
922 - Need a structure for pending actions:
924 - termination status (part of buffered msgs?)
926 Read, Write, Clean, Delete
927 - Login to Bacula; Bacula users with different permissions:
928 owner, group, user, quotas
929 - Store info on each file system type (probably in the job header on tape.
930 This could be the output of df; or perhaps some sort of /etc/mtab record.
932 ========= ideas ===============
933 From: "Jerry K. Schieffer" <jerry@skylinetechnology.com>
934 To: <kern@sibbald.com>
935 Subject: RE: [Bacula-users] future large programming jobs
936 Date: Thu, 26 Feb 2004 11:34:54 -0600
938 I noticed the subject thread and thought I would offer the following
939 merely as sources of ideas, i.e. something to think about, not even as
940 strong as a request. In my former life (before retiring) I often
941 dealt with backups and storage management issues/products as a
942 developer and as a consultant. I am currently migrating my personal
943 network from amanda to bacula specifically because of the ability to
944 cross media boundaries during storing backups.
945 Are you familiar with the commercial product called ADSM (I think IBM
946 now sells it under the Tivoli label)? It has a couple of interesting
947 ideas that may apply to the following topics.
949 1. Migration: Consider that when you need to restore a system, there
950 may be pressure to hurry. If all the information for a single client
951 can eventually end up on the same media (and in chronological order),
952 the restore is facillitated by not having to search past information
953 from other clients. ADSM has the concept of "client affinity" that
954 may be associated with it's storage pools. It seems to me that this
955 concept (as an optional feature) might fit in your architecture for
958 ADSM also has the concept of defining one or more storage pools as
959 "copy pools" (almost mirrors, but only in the sense of contents).
960 These pools provide the ability to have duplicte data stored both
961 onsite and offsite. The copy process can be scheduled to be handled
962 by their storage manager during periods when there is no backup
963 activity. Again, the migration process might be a place to consider
964 implementing something like this.
967 > It strikes me that it would be very nice to be able to do things
969 > have the Job(s) backing up the machines run, and once they have all
970 > completed, start a migration job to copy the data from disks Volumes
972 > a tape library and then to offsite storage. Maybe this can already
974 > done with some careful scheduling and Job prioritzation; the events
975 > mechanism described below would probably make it very easy.
977 This is the goal. In the first step (before events), you simply
979 the Migration to tape later.
981 2. Base jobs: In ADSM, each copy of each stored file is tracked in
982 the database. Once a file (unique by path and metadata such as dates,
983 size, ownership, etc.) is in a copy pool, no more copies are made. In
984 other words, when you start ADSM, it begins like your concept of a
985 base job. After that it is in the "incremental" mode. You can
986 configure the number of "generations" of files to be retained, plus a
987 retention date after which even old generations are purged. The
988 database tracks the contents of media and projects the percentage of
989 each volume that is valid. When the valid content of a volume drops
990 below a configured percentage, the valid data are migrated to another
991 volume and the old volume is marked as empty. Note, this requires
992 ADSM to have an idea of the contents of a client, i.e. marking the
993 database when an existing file was deleted, but this would solve your
994 issue of restoring a client without restoring deleted files.
996 This is pretty far from what bacula now does, but if you are going to
997 rip things up for Base jobs,.....
998 Also, the benefits of this are huge for very large shops, especially
999 with media robots, but are a pain for shops with manual media
1003 > Base jobs sound pretty useful, but I'm not dying for them.
1005 Nobody is dying for them, but when you see what it does, you will die
1008 3. Restoring deleted files: Since I think my comments in (2) above
1009 have low probability of implementation, I'll also suggest that you
1010 could approach the issue of deleted files by a mechanism of having the
1011 fd report to the dir, a list of all files on the client for every
1012 backup job. The dir could note in the database entry for each file
1013 the date that the file was seen. Then if a restore as of date X takes
1014 place, only files that exist from before X until after X would be
1015 restored. Probably the major cost here is the extra date container in
1016 each row of the files table.
1018 Thanks for "listening". I hope some of this helps. If you want to
1019 contact me, please send me an email - I read some but not all of the
1020 mailing list traffic and might miss a reply there.
1022 Please accept my compliments for bacula. It is doing a great job for
1023 me!! I sympathize with you in the need to wrestle with excelence in
1024 execution vs. excelence in feature inclusion.
1029 ==============================
1032 - Design at hierarchial storage for Bacula. Migration and Clone.
1033 - Implement FSM (File System Modules).
1034 - Audit M_ error codes to ensure they are correct and consistent.
1035 - Add variable break characters to lex analyzer.
1036 Either a bit mask or a string of chars so that
1037 the caller can change the break characters.
1038 - Make a single T_BREAK to replace T_COMMA, etc.
1039 - Ensure that File daemon and Storage daemon can
1040 continue a save if the Director goes down (this
1041 is NOT currently the case). Must detect socket error,
1042 buffer messages for later.
1043 - Enhance time/duration input to allow multiple qualifiers e.g. 3d2h
1044 - Add ability to backup to two Storage devices (two SD sessions) at
1045 the same time -- e.g. onsite, offsite.
1046 - Add the ability to consolidate old backup sets (basically do a restore
1047 to tape and appropriately update the catalog). Compress Volume sets.
1048 Might need to spool via file is only one drive is available.
1049 - Compress or consolidate Volumes of old possibly deleted files. Perhaps
1050 someway to do so with every volume that has less than x% valid
1054 Migration: Move a backup from one Volume to another
1055 Clone: Copy a backup -- two Volumes
1057 Bacula Migration is based on Jobs (apparently Networker is file by file).
1059 Migration triggered by:
1063 Highwater mark (keep total size)
1068 ======================================================
1070 It is somewhat like a Full save becomes an incremental since
1071 the Base job (or jobs) plus other non-base files.
1073 - A Base backup is same as Full backup, just different type.
1074 - New BaseFiles table that contains:
1076 BaseJobId - Base JobId referenced for this FileId (needed ???)
1077 JobId - JobId currently running
1078 FileId - File not backed up, exists in Base Job
1079 FileIndex - FileIndex from Base Job.
1080 i.e. for each base file that exists but is not saved because
1081 it has not changed, the File daemon sends the JobId, BaseId,
1082 FileId, FileIndex back to the Director who creates the DB entry.
1083 - To initiate a Base save, the Director sends the FD
1084 the FileId, and full filename for each file in the Base.
1085 - When the FD finds a Base file, he requests the Director to
1086 send him the full File entry (stat packet plus MD5), or
1087 conversely, the FD sends it to the Director and the Director
1088 says yes or no. This can be quite rapid if the FileId is kept
1089 by the FD for each Base Filename.
1090 - It is probably better to have the comparison done by the FD
1091 despite the fact that the File entry must be sent across the
1093 - An alternative would be to send the FD the whole File entry
1094 from the start. The disadvantage is that it requires a lot of
1095 space. The advantage is that it requires less communications
1097 - The Job record must be updated to indicate that one or more
1099 - At end of Job, FD returns:
1100 1. Count of base files/bytes not written to tape (i.e. matches)
1101 2. Count of base file that were saved i.e. had changed.
1102 - No tape record would be written for a Base file that matches, in the
1103 same way that no tape record is written for Incremental jobs where
1104 the file is not saved because it is unchanged.
1105 - On a restore, all the Base file records must explicitly be
1106 found from the BaseFile tape. I.e. for each Full save that is marked
1107 to have one or more Base Jobs, search the BaseFile for all occurrences
1109 - An optimization might be to make the BaseFile have:
1115 This would avoid the need to explicitly fetch each File record for
1116 the Base job. The Base Job record will be fetched to get the
1117 VolSessionId and VolSessionTime.
1118 =========================================================
1121 ==========================================================
1123 For each Incremental job that is run, there may be files that
1124 were found but not saved because they were locked (this applies
1125 only to Windows). Such a system could send back to the Director
1126 a list of Unsaved files.
1128 - New UnSavedFiles table that contains:
1132 - Then in the next Incremental job, the list of Unsaved Files will be
1133 feed to the FD, who will ensure that they are explicitly chosen even
1134 if standard date/time check would not have selected them.
1135 =============================================================
1139 Multiple drive autochanger data: see Alan Brown
1140 mtx -f xxx unloadStorage Element 1 is Already Full(drive 0 was empty)
1141 Unloading Data Transfer Element into Storage Element 1...source Element
1142 Address 480 is Empty
1144 (drive 0 was empty and so was slot 1)
1145 > mtx -f xxx load 15 0
1146 no response, just returns to the command prompt when complete.
1147 > mtx -f xxx status Storage Changer /dev/changer:2 Drives, 60 Slots ( 2 Import/Export )
1148 Data Transfer Element 0:Full (Storage Element 15 Loaded):VolumeTag = HX001
1149 Data Transfer Element 1:Empty
1150 Storage Element 1:Empty
1151 Storage Element 2:Full :VolumeTag=HX002
1152 Storage Element 3:Full :VolumeTag=HX003
1153 Storage Element 4:Full :VolumeTag=HX004
1154 Storage Element 5:Full :VolumeTag=HX005
1155 Storage Element 6:Full :VolumeTag=HX006
1156 Storage Element 7:Full :VolumeTag=HX007
1157 Storage Element 8:Full :VolumeTag=HX008
1158 Storage Element 9:Full :VolumeTag=HX009
1159 Storage Element 10:Full :VolumeTag=HX010
1160 Storage Element 11:Empty
1161 Storage Element 12:Empty
1162 Storage Element 13:Empty
1163 Storage Element 14:Empty
1164 Storage Element 15:Empty
1165 Storage Element 16:Empty....
1166 Storage Element 28:Empty
1167 Storage Element 29:Full :VolumeTag=CLNU01L1
1168 Storage Element 30:Empty....
1169 Storage Element 57:Empty
1170 Storage Element 58:Full :VolumeTag=NEX261L2
1171 Storage Element 59 IMPORT/EXPORT:Empty
1172 Storage Element 60 IMPORT/EXPORT:Empty
1174 Unloading Data Transfer Element into Storage Element 15...done
1176 (just to verify it remembers where it came from, however it can be
1177 overrriden with mtx unload {slotnumber} to go to any storage slot.)
1179 There needs to be a table of drive # to devices somewhere - If there are
1180 multiple changers or drives there may not be a 1:1 correspondance between
1181 changer drive number and system device name - and depending on the way the
1182 drives are hooked up to scsi busses, they may not be linearly numbered
1183 from an offset point either.something like
1185 Autochanger drives = 2
1186 Autochanger drive 0 = /dev/nst1
1187 Autochanger drive 1 = /dev/nst2
1188 IMHO, it would be _safest_ to use explicit mtx unload commands at all
1189 times, not just for multidrive changers. For a 1 drive changer, that's
1195 MTX's manpage (1.2.15):
1196 unload [<slotnum>] [ <drivenum> ]
1197 Unloads media from drive <drivenum> into slot
1198 <slotnum>. If <drivenum> is omitted, defaults to
1199 drive 0 (as do all commands). If <slotnum> is
1200 omitted, defaults to the slot that the drive was
1201 loaded from. Note that there's currently no way
1202 to say 'unload drive 1's media to the slot it
1203 came from', other than to explicitly use that
1204 slot number as the destination.AB
1210 undef# camcontrol devlist
1211 <WANGTEK 51000 SCSI M74H 12B3> at scbus0 target 2 lun 0 (pass0,sa0)
1212 <ARCHIVE 4586XX 28887-XXX 4BGD> at scbus0 target 4 lun 0 (pass1,sa1)
1213 <ARCHIVE 4586XX 28887-XXX 4BGD> at scbus0 target 4 lun 1 (pass2)
1215 tapeinfo -f /dev/sg0 with a bad tape in drive 1:
1216 [kern@rufus mtx-1.2.17kes]$ ./tapeinfo -f /dev/sg0
1217 Product Type: Tape Drive
1219 Product ID: 'C5713A '
1221 Attached Changer: No
1222 TapeAlert[3]: Hard Error: Uncorrectable read/write error.
1223 TapeAlert[20]: Clean Now: The tape drive neads cleaning NOW.
1230 Medium Type: Not Loaded
1233 DataCompEnabled: yes
1234 DataCompCapable: yes
1235 DataDeCompEnabled: yes
1242 Handling removable disks
1244 From: Karl Cunningham <karlc@keckec.com>
1246 My backups are only to hard disk these days, in removable bays. This is my
1247 idea of how a backup to hard disk would work more smoothly. Some of these
1248 things Bacula does already, but I mention them for completeness. If others
1249 have better ways to do this, I'd like to hear about it.
1251 1. Accommodate several disks, rotated similar to how tapes are. Identified
1252 by partition volume ID or perhaps by the name of a subdirectory.
1253 2. Abort & notify the admin if the wrong disk is in the bay.
1254 3. Write backups to different subdirectories for each machine to be backed
1256 4. Volumes (files) get created as needed in the proper subdirectory, one
1258 5. When a disk is recycled, remove or zero all old backup files. This is
1259 important as the disk being recycled may be close to full. This may be
1260 better done manually since the backup files for many machines may be
1261 scattered in many subdirectories.
1266 - Make sure that all do_prompt() calls in Dir check for
1267 -1 (error) and -2 (cancel) returns.
1268 - Fix foreach_jcr() to have free_jcr() inside next().
1269 jcr=jcr_walk_start();
1270 for ( ; jcr; (jcr=jcr_walk_next(jcr)) )
1273 - A Volume taken from Scratch should take on the retention period
1275 - Correct doc for Maximum Changer Wait (and others) accepting only
1277 - Implement status that shows why a job is being held in reserve, or
1278 rather why none of the drives are suitable.