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 code that makes the Dir aware that a drive is an
19 autochanger (so the user doesn't need to use the Autochanger = yes
23 - Keep same dcr when switching device ...
24 - Job retention period in a Pool (and hence Volume). The job would
26 - Detect resource deadlock in Migrate when same job wants to read
27 and write the same device.
28 - Make hardlink code at line 240 of find_one.c use binary search.
29 - Queue warning/error messages during restore so that they
30 are reported at the end of the report rather than being
31 hidden in the file listing ...
32 - Fix Maximum Changer Wait (and others) to accept qualifiers.
33 - Look at -D_FORTIFY_SOURCE=2
34 - Add Win32 FileSet definition somewhere
35 - Look at fixing restore status stats in SD.
36 - Make selection of Database used in restore correspond to
38 - Implement a mode that says when a hard read error is
39 encountered, read many times (as it currently does), and if the
40 block cannot be read, skip to the next block, and try again. If
41 that fails, skip to the next file and try again, ...
43 create table LevelType (LevelType binary(1), LevelTypeLong tinyblob);
44 insert into LevelType (LevelType,LevelTypeLong) values
48 - Add ACL to restore only to original location.
49 - Add a recursive mark command (rmark) to restore.
50 - "Minimum Job Interval = nnn" sets minimum interval between Jobs
51 of the same level and does not permit multiple simultaneous
52 running of that Job (i.e. lets any previous invocation finish
53 before doing Interval testing).
54 - Look at simplifying File exclusions.
55 - Fix store_yesno to be store_bitmask.
56 - New directive "Delete purged Volumes"
57 - new pool XXX with ScratchPoolId = MyScratchPool's PoolId and
58 let it fill itself, and RecyclePoolId = XXX's PoolId so I can
59 see if it become stable and I just have to supervise
61 - If I want to remove this pool, I set RecyclePoolId = MyScratchPool's
62 PoolId, and when it is empty remove it.
63 - Figure out how to recycle Scratch volumes back to the Scratch
66 - Allow Check Labels to be used with Bacula labels.
67 - "Resuming" a failed backup (lost line for example) by using the
68 failed backup as a sort of "base" job.
70 - Email to the user when the tape is about to need changing x
71 days before it needs changing.
72 - Command to show next tape that will be used for a job even
73 if the job is not scheduled.
74 --- create_file.c.orig Fri Jul 8 12:13:05 2005
75 +++ create_file.c Fri Jul 8 12:13:07 2005
77 attr->ofname, be.strerror());
80 + } else if(S_ISSOCK(attr->statp.st_mode)) {
81 + Dmsg1(200, "Skipping socket: %s\n", attr->ofname);
83 Dmsg1(200, "Restore node: %s\n", attr->ofname);
84 if (mknod(attr->ofname, attr->statp.st_mode, attr->statp.st_rdev) != 0 && errno != EEXIST) {
85 - From: Arunav Mandal <amandal@trolltech.com>
86 1. When jobs are running and bacula for some reason crashes or if I do a
87 restart it remembers and jobs it was running before it crashed or restarted
88 as of now I loose all jobs if I restart it.
90 2. When spooling and in the midway if client is disconnected for instance a
91 laptop bacula completely discard the spool. It will be nice if it can write
92 that spool to tape so there will be some backups for that client if not all.
94 3. We have around 150 clients machines it will be nice to have a option to
95 upgrade all the client machines bacula version automatically.
97 4. Atleast one connection should be reserved for the bconsole so at heavy load
98 I should connect to the director via bconsole which at sometimes I can't
100 5. Another most important feature that is missing, say at 10am I manually
101 started backup of client abc and it was a full backup since client abc has
102 no backup history and at 10.30am bacula again automatically started backup of
103 client abc as that was in the schedule. So now we have 2 multiple Full
104 backups of the same client and if we again try to start a full backup of
105 client backup abc bacula won't complain. That should be fixed.
107 - Fix bpipe.c so that it does not modify results pointer.
108 ***FIXME*** calling sequence should be changed.
110 #3 Migration (Move, Copy, Archive Jobs)
111 #7 Single Job Writing to Multiple Storage Devices
112 - Reserve blocks other restore jobs when first cannot connect
114 - Add true/false to conf same as yes/no
115 - For Windows disaster recovery see http://unattended.sf.net/
116 - regardless of the retention period, Bacula will not prune the
117 last Full, Diff, or Inc File data until a month after the
118 retention period for the last Full backup that was done.
119 - update volume=xxx --- add status=Full
120 - Remove old spool files on startup.
121 - Exclude SD spool/working directory.
122 - Refuse to prune last valid Full backup. Same goes for Catalog.
124 - Make a callback when Rerun failed levels is called.
125 - Give Python program access to Scheduled jobs.
126 - Add setting Volume State via Python.
127 - Python script to save with Python, not save, save with Bacula.
128 - Python script to do backup.
130 - Change the Priority, Client, Storage, JobStatus (error)
131 at the start of a job.
132 - Why is SpoolDirectory = /home/bacula/spool; not reported
133 as an error when writing a DVD?
134 - Make bootstrap file handle multiple MediaTypes (SD)
135 - Remove all old Device resource code in Dir and code to pass it
136 back in SD -- better, rework it to pass back device statistics.
137 - Check locking of resources -- be sure to lock devices where previously
138 resources were locked.
139 - The last part is left in the spool dir.
142 - In restore don't compare byte count on a raw device -- directory
143 entry does not contain bytes.
144 - To mark files as deleted, run essentially a Verify to disk, and
145 when a file is found missing (MarkId != JobId), then create
146 a new File record with FileIndex == -1. This could be done
147 by the FD at the same time as the backup.
151 MA = (last_MA * 3 + rate) / 4
152 rate = (bytes - last_bytes) / (runtime - last_runtime)
153 - Max Vols limit in Pool off by one?
154 - Implement Files/Bytes,... stats for restore job.
155 - Implement Total Bytes Written, ... for restore job.
156 - Despool attributes simultaneously with data in a separate
157 thread, rejoined at end of data spooling.
158 - 7. Implement new Console commands to allow offlining/reserving drives,
159 and possibly manipulating the autochanger (much asked for).
160 - Add start/end date editing in messages (%t %T, %e?) ...
161 - Add ClientDefs similar to JobDefs.
162 - Print more info when bextract -p accepts a bad block.
163 - Fix FD JobType to be set before RunBeforeJob in FD.
164 - Look at adding full Volume and Pool information to a Volume
165 label so that bscan can get *all* the info.
166 - If the user puts "Purge Oldest Volume = yes" or "Recycle Oldest Volume = yes"
167 and there is only one volume in the pool, refuse to do it -- otherwise
168 he fills the Volume, then immediately starts reusing it.
169 - Implement copies and stripes.
170 - Add history file to console.
171 - Each file on tape creates a JobMedia record. Peter has 4 million
172 files spread over 10000 tape files and four tapes. A restore takes
173 16 hours to build the restore list.
174 - Add and option to check if the file size changed during backup.
175 - Make sure SD deletes spool files on error exit.
176 - Delete old spool files when SD starts.
177 - When labeling tapes, if you enter 000026, Bacula uses
178 the tape index rather than the Volume name 000026.
179 - Add offline tape command to Bacula console.
181 Enter MediaId or Volume name: 32
182 Enter new Volume name: DLT-20Dec04
183 Automatically selected Pool: Default
184 Connecting to Storage daemon DLTDrive at 192.168.68.104:9103 ...
185 Sending relabel command from "DLT-28Jun03" to "DLT-20Dec04" ...
186 block.c:552 Write error at 0:0 on device /dev/nst0. ERR=Bad file descriptor.
187 Error writing final EOF to tape. This tape may not be readable.
188 dev.c:1207 ioctl MTWEOF error on /dev/nst0. ERR=Permission denied.
189 askdir.c:219 NULL Volume name. This shouldn't happen!!!
190 3912 Failed to label Volume: ERR=dev.c:1207 ioctl MTWEOF error on /dev/nst0. ERR=Permission denied.
191 Label command failed for Volume DLT-20Dec04.
192 Do not forget to mount the drive!!!
193 - Bug: if a job is manually scheduled to run later, it does not appear
194 in any status report and cannot be cancelled.
196 ==== Keeping track of deleted files ====
197 My "trick" for keeping track of deletions is the following.
198 Assuming the user turns on this option, after all the files
199 have been backed up, but before the job has terminated, the
200 FD will make a pass through all the files and send their
201 names to the DIR (*exactly* the same as what a Verify job
202 currently does). This will probably be done at the same
203 time the files are being sent to the SD avoiding a second
204 pass. The DIR will then compare that to what is stored in
205 the catalog. Any files in the catalog but not in what the
206 FD sent will receive a catalog File entry that indicates
207 that at that point in time the file was deleted.
209 During a restore, any file initially picked up by some
210 backup (Full, ...) then subsequently having a File entry
211 marked "delete" will be removed from the tree, so will not
212 be restored. If a file with the same name is later OK it
213 will be inserted in the tree -- this already happens. All
214 will be consistent except for possible changes during the
217 Since I'm on the subject, some of you may be wondering what
218 the utility of the in memory tree is if you are going to
219 restore everything (at least it comes up from time to time
220 on the list). Well, it is still *very* useful because it
221 allows only the last item found for a particular filename
222 (full path) to be entered into the tree, and thus if a file
223 is backed up 10 times, only the last copy will be restored.
224 I recently (last Friday) restored a complete directory, and
225 the Full and all the Differential and Incremental backups
226 spanned 3 Volumes. The first Volume was not even mounted
227 because all the files had been updated and hence backed up
228 since the Full backup was made. In this case, the tree
229 saved me a *lot* of time.
231 Make sure this information is stored on the tape too so
232 that it can be restored directly from the tape.
234 Comments from Martin Simmons (I think they are all covered):
235 Ok, that should cover the basics. There are few issues though:
237 - Restore will depend on the catalog. I think it is better to include the
238 extra data in the backup as well, so it can be seen by bscan and bextract.
240 - I'm not sure if it will preserve multiple hard links to the same inode. Or
241 maybe adding or removing links will cause the data to be dumped again?
243 - I'm not sure if it will handle renamed directories. Possibly it will work
244 by dumping the whole tree under a renamed directory?
246 - It remains to be seen how the backup performance of the DIR's will be
247 affected when comparing the catalog for a large filesystem.
252 - Add Pool/Storage override regression test.
253 - Add delete JobId to regression.
254 - Add a regression test for dbcheck.
255 - New test to add bscan to four-concurrent-jobs regression,
256 i.e. after the four-concurrent jobs zap the
257 database as is done in the bscan-test, then use bscan to
258 restore the database, do a restore and compare with the
260 - Add restore of specific JobId to regression (item 3
261 on the restore prompt)
262 - Add IPv6 to regression
263 - Add database test to regression. Test each function like delete,
266 - AntiVir can slow down backups on Win32 systems.
267 - Win32 systems with FAT32 can be much slower than NTFS for
268 more than 1000 files per directory.
272 - A HOLD command to stop all jobs from starting.
273 - A PAUSE command to pause all running jobs ==> release the
275 - Media Type = LTO,LTO-2,LTO-3
276 Media Type Read = LTO,LTO2,LTO3
277 Media Type Write = LTO2, LTO3
279 === From Carsten Menke <bootsy52@gmx.net>
281 Following is a list of what I think in the situations where I'm faced with,
282 could be a usefull enhancement to bacula, which I'm certain other users will
283 benefit from as well.
285 1. NextJob/NextJobs Directive within a Job Resource in the form of
286 NextJobs = job1,job2.
289 I currently solved the problem with running multiple jobs each after each
290 by setting the Max Wait Time for a job to 8 hours, and give
291 the jobs different Priorities. However, there scenarios where
292 1 Job is directly depending on another job, so if the former job fails,
293 the job after it needn't to be run
294 while maybe other jobs should run despite of that
297 A Backup Job and a Verify job, if the backup job fails there is no need to run
298 the verify job, as the backup job already failed. However, one may like
299 to backup the Catalog to disk despite of that the main backup job failed.
302 I see that this is related to the Event Handlers which are on the ToDo
303 list, also it is maybe a good idea to check for the return value and
304 execute different actions based on the return value
307 3. offline capability to bconsole
310 Currently I use a script which I execute within the last Job via the
311 RunAfterJob Directive, to release and eject the tape.
312 So I have to call bconsole "release=Storage-Name" and afterwards
313 mt -f /dev/nst0 eject to get the tape out.
315 If I have multiple Storage Devices, than these may not be /dev/nst0 and
316 I have to modify the script or call it with parameters etc.
317 This would actually not be needed, as everything is already defined
318 in bacula-sd.conf and if I can invoke bconsole with the
319 storage name via $1 in the script than I'm done and information is
322 4. %s for Storage Name added to the chars being substituted in "RunAfterJob"
326 For the reason mentioned in 3. to have the ability to call a
327 script with /scripts/foobar %s and in the script use $1
328 to pass the Storage Name to bconsole
330 5. Setting Volume State within a Job Resource
333 Instead of using "Maximum Volume Jobs" in the Pool Resource,
334 I would have the possibilty to define
335 in a Job Resource that after this certain job is run, the Volume State
336 should be set to "Volume State = Used", this give more flexibility (IMHO).
338 6. Localization of Bacula Messages
341 Unfortunatley many,many people I work with don't speak english very well.
342 So if at least the Reporting messages would be localized then they
343 would understand that they have to change the tape,etc. etc.
345 I volunteer to do the german translations, and if I can convince my wife also
346 french and Morre (western african language).
348 7. OK, this is evil, probably bound to security risks and maybe not possible
349 due to the design of bacula.
351 Implementation of Backtics ( `command` ) for shell comand execution to
352 the "Label Format" Directive.
356 Currently I have defined BACULA_DAY_OF_WEEK="day1|day2..." resulting in
357 Label Format = "HolyBackup-${BACULA_DAY_OF_WEEK[${WeekDay}]}". If I could
358 use backticks than I could use "Label Format = HolyBackup-`date +%A` to have
359 the localized name for the day of the week appended to the
360 format string. Then I have the tape labeled automatically with weekday
361 name in the correct language.
363 - Yes, that is surely the case. I probably should turn those into Warning
364 errors. In addition, you just made me think that it might not be bad to
365 add an option to check the file size after backing up the file and
366 report if it changes. This would be done as an option because it would
369 Kern, good idea. If you do do that, mention in the output: file
370 shrunk, or file expanded, just to make it obvious to the user
371 (without having to the refer to file size), just how the file size
374 Would this option be for all file, or just one file? Or a fileset?
375 - Make output from status use html table tags for nicely
376 presenting in a browser.
377 - Can one write tapes faster with 8192 byte block sizes?
378 - Document security problems with the same password for everyone in
379 rpm and Win32 releases.
380 - Browse generations of files.
381 - I've seen an error when my catalog's File table fills up. I
382 then have to recreate the File table with a larger maximum row
383 size. Relevant information is at
384 http://dev.mysql.com/doc/mysql/en/Full_table.html ; I think the
385 "Installing and Configuring MySQL" chapter should talk a bit
386 about this potential problem, and recommend a solution.
387 - For Solaris must use POSIX awk.
388 - Want speed of writing to tape while despooling.
389 - Supported autochanger:
397 Wangtek 6525ES (SCSI-1 QIC drive, 525MB), under Linux 2.4.something,
398 bacula 1.36.0/1 works with blocksize 16k INSIDE bacula-sd.conf.
399 - Add regex from http://www.pcre.org to Bacula for Win32.
400 - Use only shell tools no make in CDROM package.
401 - Include within include does it work?
402 - Implement a Pool of type Cleaning?
403 - Implement VolReadTime and VolWriteTime in SD
404 - Modify Backing up Your Database to include a bootstrap file.
405 - Think about making certain database errors fatal.
406 - Look at correcting the time jump in the scheduler for daylight
407 savings time changes.
408 - Add a "real" timer to network connections.
409 - Promote to Full = Time period
410 - Check dates entered by user for correctness (month/day/... ranges)
411 - Compress restore Volume listing by date and first file.
412 - Look at patches/bacula_db.b2z postgresql that loops during restore.
414 - Perhaps add read/write programs and/or plugins to FileSets.
415 - How to handle backing up portables ...
416 - Add some sort of guaranteed Interval for upgrading jobs.
417 - Can we write the state file after every job terminates? On Win32
418 the system crashes and the state file is not updated.
421 Documentation to do: (any release a little bit at a time)
422 - Doc to do unmount before removing magazine.
423 - Alternative to static linking "ldd prog" save all binaries listed,
424 restore them and point LD_LIBRARY_PATH to them.
425 - Document add "</dev/null >/dev/null 2>&1" to the bacula-fd command line
426 - Document query file format.
427 - Add more documentation for bsr files.
428 - Document problems with Verify and pruning.
429 - Document how to use multiple databases.
430 - VXA drives have a "cleaning required"
431 indicator, but Exabyte recommends preventive cleaning after every 75
434 In this context, it should be noted that Exabyte has a command-line
435 vxatool utility available for free download. (The current version is
436 vxatool-3.72.) It can get diagnostic info, read, write and erase tapes,
437 test the drive, unload tapes, change drive settings, flash new firmware,
439 Of particular interest in this context is that vxatool <device> -i will
440 report, among other details, the time since last cleaning in tape motion
441 minutes. This information can be retrieved (and settings changed, for
442 that matter) through the generic-SCSI device even when Bacula has the
443 regular tape device locked. (Needless to say, I don't recommend
444 changing tape settings while a job is running.)
445 - Lookup HP cleaning recommendations.
446 - Lookup HP tape replacement recommendations (see trouble shooting autochanger)
447 - Document doing table repair
450 ===================================
451 - Add macro expansions in JobDefs.
452 Run Before Job = "SomeFile %{Level} %{Client}"
453 Write Bootstrap="/some/dir/%{JobName}_%{Client}.bsr"
454 - Use non-blocking network I/O but if no data is available, use
456 - Use gather write() for network I/O.
457 - Autorestart on crash.
458 - Add bandwidth limiting.
459 - Add acks every once and a while from the SD to keep
460 the line from timing out.
461 - When an error in input occurs and conio beeps, you can back
462 up through the prompt.
463 - Detect fixed tape block mode during positioning by looking at
464 block numbers in btape "test". Possibly adjust in Bacula.
465 - Fix list volumes to output volume retention in some other
466 units, perhaps via a directive.
467 - If opening a tape in read/write mode fails attempt to open
468 it in read-only mode, and mark the tape for read only.
469 - Allow Simultaneous Priorities = yes => run up to Max concurrent jobs even
470 with multiple priorities.
471 - If you use restore replace=never, the directory attributes for
472 non-existent directories will not be restored properly.
474 - see lzma401.zip in others directory for new compression
476 - Minimal autochanger handling in Bacula and in btape.
477 - Look into how tar does not save sockets and the possiblity of
478 not saving them in Bacula (Martin Simmons reported this).
479 - Fix restore jobs so that multiple jobs can run if they
480 are not using the same tape(s).
481 - Allow the user to select JobType for manual pruning/purging.
482 - bscan does not put first of two volumes back with all info in
484 - Implement the FreeBSD nodump flag in chflags.
485 - Figure out how to make named console messages go only to that
486 console and to the non-restricted console (new console class?).
487 - Make restricted console prompt for password if *ask* is set or
488 perhaps if password is undefined.
489 - Implement "from ISO-date/time every x hours/days/weeks/months" in
492 ==== from Marc Schoechlin
493 - the help-command should be more verbose
494 (it should explain the paramters of the different
496 -> its time-comsuming to consult the manual anytime
497 you need a special parameter
498 -> maybe its more easy to maintain this, if the
499 descriptions of that commands are outsourced to
501 - the cd-command should allow complete paths
502 i.e. cd /foo/bar/foo/bar
503 -> if a customer mails me the path to a certain file,
504 its faster to enter the specified directory
505 - if the password is not configured in bconsole.conf
506 you should be asked for it.
507 -> sometimes you like to do restore on a customer-machine
508 which shouldnt know the password for bacula.
509 -> adding the password to the file favours admins
510 to forget to remove the password after usage
512 the protection of that file is less important
513 - long-listed-output of commands should be scrollable
514 like the unix more/less-command does
515 -> if someone runs 200 and more machines, the lists could
516 be a little long and complex
517 - command-output should be shown column by column
518 to reduce scrolling and to increase clarity
520 - lsmark should list the selected files with full
522 - wildcards for selecting and file and directories would be nice
523 - any actions should be interuptable with STRG+C
524 - command-expansion would be pretty cool
526 - When the replace Never option is set, new directory permissions
527 are not restored. See bug 213. To fix this requires creating a
528 list of newly restored directories so that those directory
529 permissions *can* be restored.
530 - Compaction of Disk space by "migrating" Volumes that have pruned
531 Jobs (what criteria? size, #jobs, time).
532 - Add prune all command
533 - Document fact that purge can destroy a part of a restore by purging
534 one volume while others remain valid -- perhaps mark Jobs.
535 - Add multiple-media-types.txt
536 - look at mxt-changer.html
537 - Make ? do a help command (no return needed).
538 - Implement restore directory.
539 - Document streams and how to implement them.
540 - Try not to re-backup a file if a new hard link is added.
541 - Add feature to backup hard links only, but not the data.
542 - Fix stream handling to be simpler.
543 - Add Priority and Bootstrap to Run a Job.
544 - Eliminate Restore "Run Restore Job" prompt by allowing new "run command
546 - Remove View FileSet button from Run a Job dialog.
547 - Handle prompt for restore job at end of Restore command.
548 - Add display of total selected files to Restore window.
549 - Add tree pane to left of window.
550 - Add progress meter.
551 - Max wait time or max run time causes seg fault -- see runtime-bug.txt
552 - Document writing to a CD/DVD with Bacula.
553 - Add a "base" package to the window installer for pthreadsVCE.dll
554 which is needed by all packages.
555 - Add message to user to check for fixed block size when the forward
556 space test fails in btape.
557 - When unmarking a directory check if all files below are unmarked and
558 then remove the + flag -- in the restore tree.
559 - Possibly implement: Action = Unmount Device="TapeDrive1" in Admin jobs.
560 - Setup lrrd graphs: (http://www.linpro.no/projects/lrrd/) Mike Acar.
561 - Revisit the question of multiple Volumes (disk) on a single device.
562 - Add a block copy option to bcopy.
563 - Investigate adding Mac Resource Forks.
564 - Finish work on Gnome restore GUI.
565 - Fix "llist jobid=xx" where no fileset or client exists.
566 - For each job type (Admin, Restore, ...) require only the really necessary
567 fields.- Pass Director resource name as an option to the Console.
568 - Add a "batch" mode to the Console (no unsolicited queries, ...).
569 - Add a .list all files in the restore tree (probably also a list all files)
570 Do both a long and short form.
571 - Allow browsing the catalog to see all versions of a file (with
572 stat data on each file).
573 - Restore attributes of directory if replace=never set but directory
575 - Use SHA1 on authentication if possible.
576 - See comtest-xxx.zip for Windows code to talk to USB.
577 - Add John's appended files:
578 Appended = { /files/server/logs/http/*log }
579 and such files would be treated as follows.On a FULL backup, they would
580 be backed up like any other file.On an INCREMENTAL backup, where a
581 previous INCREMENTAL or FULL was already in thecatalogue and the length
582 of the file wasgreater than the length of the last backup, only thedata
583 added since the last backup will be dumped.On an INCREMENTAL backup, if
584 the length of the file is less than thelength of the file with the same
585 name last backed up, the completefile is dumped.On Windows systems, with
586 creation date of files, we can be evensmarter about this and not count
587 entirely upon the length.On a restore, the full and all incrementals
588 since it will beapplied in sequence to restore the file.
589 - Check new HAVE_WIN32 open bits.
590 - Check if the tape has moved before writing.
591 - Handling removable disks -- see below:
592 - Keep track of tape use time, and report when cleaning is necessary.
593 - Add FromClient and ToClient keywords on restore command (or
594 BackupClient RestoreClient).
595 - Implement a JobSet, which groups any number of jobs. If the
596 JobSet is started, all the jobs are started together.
597 Allow Pool, Level, and Schedule overrides.
598 - Enhance cancel to timeout BSOCK packets after a specific delay.
599 - Do scheduling by UTC using gmtime_r() in run_conf, scheduler, and
600 ua_status.!!! Thanks to Alan Brown for this tip.
601 - Look at updating Volume Jobs so that Max Volume Jobs = 1 will work
602 correctly for multiple simultaneous jobs.
603 - Correct code so that FileSet MD5 is calculated for < and | filename
605 - Implement the Media record flag that indicates that the Volume does disk
607 - Implement VolAddr, which is used when Volume is addressed like a disk,
608 and form it from VolFile and VolBlock.
609 - Make multiple restore jobs for multiple media types specifying
610 the proper storage type.
611 - Fix fast block rejection (stored/read_record.c:118). It passes a null
612 pointer (rec) to try_repositioning().
613 - Look at extracting Win data from BackupRead.
614 - Implement RestoreJobRetention? Maybe better "JobRetention" in a Job,
615 which would take precidence over the Catalog "JobRetention".
616 - Implement Label Format in Add and Label console commands.
617 - Possibly up network buffers to 65K. Put on variable.
618 - Put email tape request delays on one or more variables. User wants
619 to cancel the job after a certain time interval. Maximum Mount Wait?
620 - Job, Client, Device, Pool, or Volume?
621 Is it possible to make this a directive which is *optional* in multiple
622 resources, like Level? If so, I think I'd make it an optional directive
623 in Job, Client, and Pool, with precedence such that Job overrides Client
624 which in turn overrides Pool.
626 - New Storage specifications:
627 - Want to write to multiple storage devices simultaneously
628 - Want to write to multiple storage devices sequentially (in one job)
629 - Want to read/write simultaneously
630 - Key is MediaType -- it must match
632 Passed to SD as a sort of BSR record called Storage Specification
636 MediaType -> Next MediaType
638 Device -> Next Device
640 Allow multiple Storage specifications
648 Allow Multiple Pool specifications (note, Pool currently
650 Allow Multiple MediaType specifications in Dir conf
651 Allow Multiple Device specifications in Dir conf
652 Perhaps keep this in a single SSR
653 Tie a Volume to a specific device by using a MediaType that
654 is contained in only one device.
655 In SD allow Device to have Multiple MediaTypes
657 - Ideas from Jerry Scharf:
658 First let's point out some big pluses that bacula has for this
660 more importantly it's active. Thank you so much for that
661 even more important, it's not flaky
662 it has an open access catalog, opening many possibilities
663 it's pushing toward heterogeneous systems capability
665 Macintosh file client
666 macs are an interesting niche, but I fear a server is a rathole
667 working bare iron recovery for windows
668 the option for inc/diff backups not reset on fileset revision
669 a) use both change and inode update time against base time
670 b) do the full catalog check (expensive but accurate)
671 sizing guide (how much system is needed to back up N systems/files)
672 consultants on using bacula in building a disaster recovery system
674 or how to get at fancy things that one could do with bacula
675 logwatch code for bacula logs (or similar)
676 linux distro inclusion of bacula (brings good and bad, but necessary)
677 win2k/XP server capability (icky but you asked)
678 support for Oracle database ??
680 - Look at adding SQL server and Exchange support for Windows.
681 - Make dev->file and dev->block_num signed integers so that -1 can
682 be an invalid value which happens with BSR.
683 - Create VolAddr for disk files in place of VolFile and VolBlock. This
684 is needed to properly specify ranges.
685 - Add progress of files/bytes to SD and FD.
686 - Print warning message if FileId > 4 billion
687 - do a "messages" before the first prompt in Console
688 - Client does not show busy during Estimate command.
689 - Implement Console mtx commands.
690 - Implement a Mount Command and an Unmount Command where
691 the users could specify a system command to be performed
692 to do the mount, after which Bacula could attempt to
693 read the device. This is for Removeable media such as a CDROM.
694 - Most likely, this mount command would be invoked explicitly
695 by the user using the current Console "mount" and "unmount"
696 commands -- the Storage Daemon would do the right thing
697 depending on the exact nature of the device.
698 - As with tape drives, when Bacula wanted a new removable
699 disk mounted, it would unmount the old one, and send a message
700 to the user, who would then use "mount" as described above
701 once he had actually inserted the disk.
702 - Implement dump/print label to UA
703 - Spool to disk only when the tape is full, then when a tape is hung move
705 - bextract is sending everything to the log file ****FIXME****
706 - Allow multiple Storage specifications (or multiple names on
707 a single Storage specification) in the Job record. Thus a job
708 can be backed up to a number of storage devices.
709 - Implement some way for the File daemon to contact the Director
710 to start a job or pass its DHCP obtained IP number.
711 - Implement a query tape prompt/replace feature for a console
712 - Copy console @ code to gnome2-console
713 - Make AES the only encryption algorithm see
714 http://csrc.nist.gov/CryptoToolkit/aes/). It's
715 an officially adopted standard, has survived peer
716 review, and provides keys up to 256 bits.
717 - Take a careful look at SetACL http://setacl.sourceforge.net
718 - Make tree walk routines like cd, ls, ... more user friendly
719 by handling spaces better.
720 - Make sure that Bacula rechecks the tape after the 20 min wait.
721 - Set IO_NOWAIT on Bacula TCP/IP packets.
722 - Try doing a raw partition backup and restore by mounting a
725 Yes, it would allow to highly automatic the request for new tapes. If a
726 tape is empty, bacula reads the barcodes (native or simulated), and if
727 an unused tape is found, it runs the label command with all the
728 necessary parameters.
730 By the way can bacula automatically "move" an empty/purged volume say
731 in the "short" pool to the "long" pool if this pool runs out of volume
733 - What to do about "list files job=xxx".
734 - Get and test MySQL 4.0
735 - Look at how fuser works and /proc/PID/fd that is how Nic found the
736 file descriptor leak in Bacula.
737 - Implement WrapCounters in Counters.
738 - Add heartbeat from FD to SD if hb interval expires.
739 - Can we dynamically change FileSets?
740 - If pool specified to label command and Label Format is specified,
741 automatically generate the Volume name.
742 - Why can't SQL do the filename sort for restore?
743 - Add ExhautiveRestoreSearch
744 - Look at the possibility of loading only the necessary
745 data into the restore tree (i.e. do it one directory at a
746 time as the user walks through the tree).
747 - Possibly use the hash code if the user selects all for a restore command.
748 - Fix "restore all" to bypass building the tree.
749 - Prohibit backing up archive device (findlib/find_one.c:128)
750 - Implement Release Device in the Job resource to unmount a drive.
751 - Implement Acquire Device in the Job resource to mount a drive,
752 be sure this works with admin jobs so that the user can get
753 prompted to insert the correct tape. Possibly some way to say to
754 run the job but don't save the files.
755 - Make things like list where a file is saved case independent for
758 - Use autochanger to handle multiple devices.
759 - On Windows with very long path names, it may be impossible to create
760 a file (and thus restore it) because the total length is too long.
761 We must cd into the directory then create the file without the
763 - Implement a Recycle command
764 - Test a second language e.g. french.
765 - Start working on Base jobs.
766 - Implement UnsavedFiles DB record.
767 - From Phil Stracchino:
768 It would probably be a per-client option, and would be called
769 something like, say, "Automatically purge obsoleted jobs". What it
770 would do is, when you successfully complete a Differential backup of a
771 client, it would automatically purge all Incremental backups for that
772 client that are rendered redundant by that Differential. Likewise,
773 when a Full backup on a client completed, it would automatically purge
774 all Differential and Incremental jobs obsoleted by that Full backup.
775 This would let people minimize the number of tapes they're keeping on
776 hand without having to master the art of retention times.
777 - When doing a Backup send all attributes back to the Director, who
778 would then figure out what files have been deleted.
779 - Currently in mount.c:236 the SD simply creates a Volume. It should have
780 explicit permission to do so. It should also mark the tape in error
781 if there is an error.
782 - Cancel waiting for Client connect in SD if FD goes away.
784 - Implement timeout in response() when it should come quickly.
785 - Implement a Slot priority (loaded/not loaded).
786 - Implement "vacation" Incremental only saves.
787 - Implement create "FileSet"?
788 - Add prefixlinks to where or not where absolute links to FD.
789 - Issue message to mount a new tape before the rewind.
790 - Simplified client job initiation for portables.
791 - If SD cannot open a drive, make it periodically retry.
792 - Add more of the config info to the tape label.
794 - If tape is marked read-only, then try opening it read-only rather than
795 failing, and remember that it cannot be written.
796 - Refine SD waiting output:
797 Device is being positioned
798 > Device is being positioned for append
799 > Device is being positioned to file x
801 - Figure out some way to estimate output size and to avoid splitting
802 a backup across two Volumes -- this could be useful for writing CDROMs
803 where you really prefer not to have it split -- not serious.
804 - Have SD compute MD5 or SHA1 and compare to what FD computes.
805 - Make VolumeToCatalog calculate an MD5 or SHA1 from the
806 actual data on the Volume and compare it.
807 - Implement Bacula plugins -- design API
808 - Make bcopy read through bad tape records.
809 - Program files (i.e. execute a program to read/write files).
810 Pass read date of last backup, size of file last time.
811 - Add Signature type to File DB record.
812 - CD into subdirectory when open()ing files for backup to
813 speed up things. Test with testfind().
814 - Priority job to go to top of list.
815 - Why are save/restore of device different sizes (sparse?) Yup! Fix it.
816 - Implement some way for the Console to dynamically create a job.
817 - Solaris -I on tar for include list
818 - Need a verbose mode in restore, perhaps to bsr.
819 - bscan without -v is too quiet -- perhaps show jobs.
820 - Add code to reject whole blocks if not wanted on restore.
821 - Check if we can increase Bacula FD priorty in Win2000
822 - Make sure the MaxVolFiles is fully implemented in SD
823 - Check if both CatalogFiles and UseCatalog are set to SD.
824 - Possibly add email to Watchdog if drive is unmounted too
825 long and a job is waiting on the drive.
826 - After unmount, if restore job started, ask to mount.
827 - Add UA rc and history files.
828 - put termcap (used by console) in ./configure and
829 allow -with-termcap-dir.
830 - Fix Autoprune for Volumes to respect need for full save.
831 - Compare tape to Client files (attributes, or attributes and data)
832 - Make all database Ids 64 bit.
833 - Allow console commands to detach or run in background.
834 - Fix status delay on storage daemon during rewind.
835 - Add SD message variables to control operator wait time
836 - Maximum Operator Wait
837 - Minimum Message Interval
838 - Maximum Message Interval
839 - Send Operator message when cannot read tape label.
840 - Verify level=Volume (scan only), level=Data (compare of data to file).
841 Verify level=Catalog, level=InitCatalog
843 - Add keyword search to show command in Console.
844 - Events : tape has more than xxx bytes.
845 - Complete code in Bacula Resources -- this will permit
846 reading a new config file at any time.
847 - Handle ctl-c in Console
848 - Implement script driven addition of File daemon to config files.
849 - Think about how to make Bacula work better with File (non-tape) archives.
850 - Write Unix emulator for Windows.
851 - Put memory utilization in Status output of each daemon
852 if full status requested or if some level of debug on.
853 - Make database type selectable by .conf files i.e. at runtime
854 - Set flag for uname -a. Add to Volume label.
855 - Restore files modified after date
856 - SET LD_RUN_PATH=$HOME/mysql/lib/mysql
857 - Remove duplicate fields from jcr (e.g. jcr.level and jcr.jr.Level, ...).
858 - Timout a job or terminate if link goes down, or reopen link and query.
859 - Concept of precious tapes (cannot be reused).
860 - Make bcopy copy with a single tape drive.
861 - Permit changing ownership during restore.
864 > My suggestion: Add a feature on the systray menu-icon menu to request
865 > an immediate backup now. This would be useful for laptop users who may
866 > not be on the network when the regular scheduled backup is run.
868 > My wife's suggestion: Add a setting to the win32 client to allow it to
869 > shut down the machine after backup is complete (after, of course,
870 > displaying a "System will shut down in one minute, click here to cancel"
871 > warning dialog). This would be useful for sites that want user
872 > woorkstations to be shut down overnight to save power.
875 - Autolabel should be specified by DIR instead of SD.
878 - AutoScan (check checksum of tape)
879 - Format command = "format /dev/nst0"
883 - Seek resolution (usually corresponds to buffer size)
884 - EODErrorCode=ENOSPC or code
885 - Partial Read error code
886 - Partial write error code
887 - Nonformatted read error
888 - Nonformatted write error
889 - WriteProtected error
893 - IgnoreCloseErrors=yes
903 - FD sends unsaved file list to Director at end of job (see
905 - File daemon should build list of files skipped, and then
906 at end of save retry and report any errors.
907 - Write a Storage daemon that uses pipes and
908 standard Unix programs to write to the tape.
910 - Need something that monitors the JCR queue and
911 times out jobs by asking the deamons where they are.
912 - Enhance Jmsg code to permit buffering and saving to disk.
913 - device driver = "xxxx" for drives.
915 - Ensure that /dev/null works
916 - Need report class for messages. Perhaps
917 report resource where report=group of messages
918 - enhance scan_attrib and rename scan_jobtype, and
919 fill in code for "since" option
920 - Director needs a time after which the report status is sent
921 anyway -- or better yet, a retry time for the job.
922 - Don't reschedule a job if previous incarnation is still running.
923 - Some way to automatically backup everything is needed????
924 - Need a structure for pending actions:
926 - termination status (part of buffered msgs?)
928 Read, Write, Clean, Delete
929 - Login to Bacula; Bacula users with different permissions:
930 owner, group, user, quotas
931 - Store info on each file system type (probably in the job header on tape.
932 This could be the output of df; or perhaps some sort of /etc/mtab record.
934 ========= ideas ===============
935 From: "Jerry K. Schieffer" <jerry@skylinetechnology.com>
936 To: <kern@sibbald.com>
937 Subject: RE: [Bacula-users] future large programming jobs
938 Date: Thu, 26 Feb 2004 11:34:54 -0600
940 I noticed the subject thread and thought I would offer the following
941 merely as sources of ideas, i.e. something to think about, not even as
942 strong as a request. In my former life (before retiring) I often
943 dealt with backups and storage management issues/products as a
944 developer and as a consultant. I am currently migrating my personal
945 network from amanda to bacula specifically because of the ability to
946 cross media boundaries during storing backups.
947 Are you familiar with the commercial product called ADSM (I think IBM
948 now sells it under the Tivoli label)? It has a couple of interesting
949 ideas that may apply to the following topics.
951 1. Migration: Consider that when you need to restore a system, there
952 may be pressure to hurry. If all the information for a single client
953 can eventually end up on the same media (and in chronological order),
954 the restore is facillitated by not having to search past information
955 from other clients. ADSM has the concept of "client affinity" that
956 may be associated with it's storage pools. It seems to me that this
957 concept (as an optional feature) might fit in your architecture for
960 ADSM also has the concept of defining one or more storage pools as
961 "copy pools" (almost mirrors, but only in the sense of contents).
962 These pools provide the ability to have duplicte data stored both
963 onsite and offsite. The copy process can be scheduled to be handled
964 by their storage manager during periods when there is no backup
965 activity. Again, the migration process might be a place to consider
966 implementing something like this.
969 > It strikes me that it would be very nice to be able to do things
971 > have the Job(s) backing up the machines run, and once they have all
972 > completed, start a migration job to copy the data from disks Volumes
974 > a tape library and then to offsite storage. Maybe this can already
976 > done with some careful scheduling and Job prioritzation; the events
977 > mechanism described below would probably make it very easy.
979 This is the goal. In the first step (before events), you simply
981 the Migration to tape later.
983 2. Base jobs: In ADSM, each copy of each stored file is tracked in
984 the database. Once a file (unique by path and metadata such as dates,
985 size, ownership, etc.) is in a copy pool, no more copies are made. In
986 other words, when you start ADSM, it begins like your concept of a
987 base job. After that it is in the "incremental" mode. You can
988 configure the number of "generations" of files to be retained, plus a
989 retention date after which even old generations are purged. The
990 database tracks the contents of media and projects the percentage of
991 each volume that is valid. When the valid content of a volume drops
992 below a configured percentage, the valid data are migrated to another
993 volume and the old volume is marked as empty. Note, this requires
994 ADSM to have an idea of the contents of a client, i.e. marking the
995 database when an existing file was deleted, but this would solve your
996 issue of restoring a client without restoring deleted files.
998 This is pretty far from what bacula now does, but if you are going to
999 rip things up for Base jobs,.....
1000 Also, the benefits of this are huge for very large shops, especially
1001 with media robots, but are a pain for shops with manual media
1005 > Base jobs sound pretty useful, but I'm not dying for them.
1007 Nobody is dying for them, but when you see what it does, you will die
1010 3. Restoring deleted files: Since I think my comments in (2) above
1011 have low probability of implementation, I'll also suggest that you
1012 could approach the issue of deleted files by a mechanism of having the
1013 fd report to the dir, a list of all files on the client for every
1014 backup job. The dir could note in the database entry for each file
1015 the date that the file was seen. Then if a restore as of date X takes
1016 place, only files that exist from before X until after X would be
1017 restored. Probably the major cost here is the extra date container in
1018 each row of the files table.
1020 Thanks for "listening". I hope some of this helps. If you want to
1021 contact me, please send me an email - I read some but not all of the
1022 mailing list traffic and might miss a reply there.
1024 Please accept my compliments for bacula. It is doing a great job for
1025 me!! I sympathize with you in the need to wrestle with excelence in
1026 execution vs. excelence in feature inclusion.
1031 ==============================
1034 - Design at hierarchial storage for Bacula. Migration and Clone.
1035 - Implement FSM (File System Modules).
1036 - Audit M_ error codes to ensure they are correct and consistent.
1037 - Add variable break characters to lex analyzer.
1038 Either a bit mask or a string of chars so that
1039 the caller can change the break characters.
1040 - Make a single T_BREAK to replace T_COMMA, etc.
1041 - Ensure that File daemon and Storage daemon can
1042 continue a save if the Director goes down (this
1043 is NOT currently the case). Must detect socket error,
1044 buffer messages for later.
1045 - Enhance time/duration input to allow multiple qualifiers e.g. 3d2h
1046 - Add ability to backup to two Storage devices (two SD sessions) at
1047 the same time -- e.g. onsite, offsite.
1048 - Add the ability to consolidate old backup sets (basically do a restore
1049 to tape and appropriately update the catalog). Compress Volume sets.
1050 Might need to spool via file is only one drive is available.
1051 - Compress or consolidate Volumes of old possibly deleted files. Perhaps
1052 someway to do so with every volume that has less than x% valid
1056 Migration: Move a backup from one Volume to another
1057 Clone: Copy a backup -- two Volumes
1059 Bacula Migration is based on Jobs (apparently Networker is file by file).
1061 Migration triggered by:
1065 Highwater mark (keep total size)
1070 ======================================================
1072 It is somewhat like a Full save becomes an incremental since
1073 the Base job (or jobs) plus other non-base files.
1075 - A Base backup is same as Full backup, just different type.
1076 - New BaseFiles table that contains:
1078 BaseJobId - Base JobId referenced for this FileId (needed ???)
1079 JobId - JobId currently running
1080 FileId - File not backed up, exists in Base Job
1081 FileIndex - FileIndex from Base Job.
1082 i.e. for each base file that exists but is not saved because
1083 it has not changed, the File daemon sends the JobId, BaseId,
1084 FileId, FileIndex back to the Director who creates the DB entry.
1085 - To initiate a Base save, the Director sends the FD
1086 the FileId, and full filename for each file in the Base.
1087 - When the FD finds a Base file, he requests the Director to
1088 send him the full File entry (stat packet plus MD5), or
1089 conversely, the FD sends it to the Director and the Director
1090 says yes or no. This can be quite rapid if the FileId is kept
1091 by the FD for each Base Filename.
1092 - It is probably better to have the comparison done by the FD
1093 despite the fact that the File entry must be sent across the
1095 - An alternative would be to send the FD the whole File entry
1096 from the start. The disadvantage is that it requires a lot of
1097 space. The advantage is that it requires less communications
1099 - The Job record must be updated to indicate that one or more
1101 - At end of Job, FD returns:
1102 1. Count of base files/bytes not written to tape (i.e. matches)
1103 2. Count of base file that were saved i.e. had changed.
1104 - No tape record would be written for a Base file that matches, in the
1105 same way that no tape record is written for Incremental jobs where
1106 the file is not saved because it is unchanged.
1107 - On a restore, all the Base file records must explicitly be
1108 found from the BaseFile tape. I.e. for each Full save that is marked
1109 to have one or more Base Jobs, search the BaseFile for all occurrences
1111 - An optimization might be to make the BaseFile have:
1117 This would avoid the need to explicitly fetch each File record for
1118 the Base job. The Base Job record will be fetched to get the
1119 VolSessionId and VolSessionTime.
1120 =========================================================
1123 ==========================================================
1125 For each Incremental job that is run, there may be files that
1126 were found but not saved because they were locked (this applies
1127 only to Windows). Such a system could send back to the Director
1128 a list of Unsaved files.
1130 - New UnSavedFiles table that contains:
1134 - Then in the next Incremental job, the list of Unsaved Files will be
1135 feed to the FD, who will ensure that they are explicitly chosen even
1136 if standard date/time check would not have selected them.
1137 =============================================================
1141 Multiple drive autochanger data: see Alan Brown
1142 mtx -f xxx unloadStorage Element 1 is Already Full(drive 0 was empty)
1143 Unloading Data Transfer Element into Storage Element 1...source Element
1144 Address 480 is Empty
1146 (drive 0 was empty and so was slot 1)
1147 > mtx -f xxx load 15 0
1148 no response, just returns to the command prompt when complete.
1149 > mtx -f xxx status Storage Changer /dev/changer:2 Drives, 60 Slots ( 2 Import/Export )
1150 Data Transfer Element 0:Full (Storage Element 15 Loaded):VolumeTag = HX001
1151 Data Transfer Element 1:Empty
1152 Storage Element 1:Empty
1153 Storage Element 2:Full :VolumeTag=HX002
1154 Storage Element 3:Full :VolumeTag=HX003
1155 Storage Element 4:Full :VolumeTag=HX004
1156 Storage Element 5:Full :VolumeTag=HX005
1157 Storage Element 6:Full :VolumeTag=HX006
1158 Storage Element 7:Full :VolumeTag=HX007
1159 Storage Element 8:Full :VolumeTag=HX008
1160 Storage Element 9:Full :VolumeTag=HX009
1161 Storage Element 10:Full :VolumeTag=HX010
1162 Storage Element 11:Empty
1163 Storage Element 12:Empty
1164 Storage Element 13:Empty
1165 Storage Element 14:Empty
1166 Storage Element 15:Empty
1167 Storage Element 16:Empty....
1168 Storage Element 28:Empty
1169 Storage Element 29:Full :VolumeTag=CLNU01L1
1170 Storage Element 30:Empty....
1171 Storage Element 57:Empty
1172 Storage Element 58:Full :VolumeTag=NEX261L2
1173 Storage Element 59 IMPORT/EXPORT:Empty
1174 Storage Element 60 IMPORT/EXPORT:Empty
1176 Unloading Data Transfer Element into Storage Element 15...done
1178 (just to verify it remembers where it came from, however it can be
1179 overrriden with mtx unload {slotnumber} to go to any storage slot.)
1181 There needs to be a table of drive # to devices somewhere - If there are
1182 multiple changers or drives there may not be a 1:1 correspondance between
1183 changer drive number and system device name - and depending on the way the
1184 drives are hooked up to scsi busses, they may not be linearly numbered
1185 from an offset point either.something like
1187 Autochanger drives = 2
1188 Autochanger drive 0 = /dev/nst1
1189 Autochanger drive 1 = /dev/nst2
1190 IMHO, it would be _safest_ to use explicit mtx unload commands at all
1191 times, not just for multidrive changers. For a 1 drive changer, that's
1197 MTX's manpage (1.2.15):
1198 unload [<slotnum>] [ <drivenum> ]
1199 Unloads media from drive <drivenum> into slot
1200 <slotnum>. If <drivenum> is omitted, defaults to
1201 drive 0 (as do all commands). If <slotnum> is
1202 omitted, defaults to the slot that the drive was
1203 loaded from. Note that there's currently no way
1204 to say 'unload drive 1's media to the slot it
1205 came from', other than to explicitly use that
1206 slot number as the destination.AB
1212 undef# camcontrol devlist
1213 <WANGTEK 51000 SCSI M74H 12B3> at scbus0 target 2 lun 0 (pass0,sa0)
1214 <ARCHIVE 4586XX 28887-XXX 4BGD> at scbus0 target 4 lun 0 (pass1,sa1)
1215 <ARCHIVE 4586XX 28887-XXX 4BGD> at scbus0 target 4 lun 1 (pass2)
1217 tapeinfo -f /dev/sg0 with a bad tape in drive 1:
1218 [kern@rufus mtx-1.2.17kes]$ ./tapeinfo -f /dev/sg0
1219 Product Type: Tape Drive
1221 Product ID: 'C5713A '
1223 Attached Changer: No
1224 TapeAlert[3]: Hard Error: Uncorrectable read/write error.
1225 TapeAlert[20]: Clean Now: The tape drive neads cleaning NOW.
1232 Medium Type: Not Loaded
1235 DataCompEnabled: yes
1236 DataCompCapable: yes
1237 DataDeCompEnabled: yes
1244 Handling removable disks
1246 From: Karl Cunningham <karlc@keckec.com>
1248 My backups are only to hard disk these days, in removable bays. This is my
1249 idea of how a backup to hard disk would work more smoothly. Some of these
1250 things Bacula does already, but I mention them for completeness. If others
1251 have better ways to do this, I'd like to hear about it.
1253 1. Accommodate several disks, rotated similar to how tapes are. Identified
1254 by partition volume ID or perhaps by the name of a subdirectory.
1255 2. Abort & notify the admin if the wrong disk is in the bay.
1256 3. Write backups to different subdirectories for each machine to be backed
1258 4. Volumes (files) get created as needed in the proper subdirectory, one
1260 5. When a disk is recycled, remove or zero all old backup files. This is
1261 important as the disk being recycled may be close to full. This may be
1262 better done manually since the backup files for many machines may be
1263 scattered in many subdirectories.
1268 - Make sure that all do_prompt() calls in Dir check for
1269 -1 (error) and -2 (cancel) returns.
1270 - Fix foreach_jcr() to have free_jcr() inside next().
1271 jcr=jcr_walk_start();
1272 for ( ; jcr; (jcr=jcr_walk_next(jcr)) )
1275 - A Volume taken from Scratch should take on the retention period
1277 - Correct doc for Maximum Changer Wait (and others) accepting only
1279 - Implement status that shows why a job is being held in reserve, or
1280 rather why none of the drives are suitable.
1281 - Implement a way to disable a drive (so you can use the second
1282 drive of an autochanger, and the first one will not be used or
1284 - Make sure Maximum Volumes is respected in Pools when adding
1285 Volumes (e.g. when pulling a Scratch volume).