9 - Document cleaning up the spool files:
10 db, pid, state, bsr, mail, conmsg, spool
11 - Document the multiple-drive-changer.txt script.
12 - Pruning with Admin job.
13 - Does WildFile match against full name? Doc.
14 - %d and %v only valid on Director, not for ClientRunBefore/After.
19 - Fix re-read of last block to check if job has actually written
20 a block, and check if block was written by a different job
21 (i.e. multiple simultaneous jobs writing).
22 - JobStatus and Termination codes.
23 - Some users claim that they must do two prune commands to get a
24 Volume marked as purged.
25 - Print warning message if LANG environment variable does not specify
27 === Migration from David ===
31 Name = "<poolname>-migrate"
35 Migration Selection Type = LowestUtil | OldestVol | PoolOccupancy |
36 Client | PoolResidence | Volume | JobName | SQLquery
37 Migration Selection Pattern = "regexp"
38 Next Pool = <override>
41 There should be no need for a Level (migration is always Full, since you
42 don't calculate differential/incremental differences for migration),
43 Storage should be determined by the volume types in the pool, and Client
44 is really a selection issue. Migration should always occur to the
45 NextPool defined in the pool definition. If no nextpool is defined, the
46 job should end with a reason of "no place to go". If Next Pool statement
47 is present, we override the check in the pool definition and use the
50 Here's how I'd define Migration Selection Types:
52 LowestUtil -- Identify the volume in the pool with the least data on it
53 and empty it. No Migration Selection Pattern required.
55 OldestVol -- Identify the LRU volume with data written, and empty it. No
56 Migration Selection Pattern required.
58 PoolOccupancy -- if pool occupancy exceeds <highmig>, migrate volumes
59 (starting with most full volumes) until pool occupancy drops below
60 <lowmig>. Pool highmig and lowmig values are in pool definition, no
61 Migration Selection Pattern required.
63 Client -- Migrate data from selected client only. Migration Selection
64 Pattern regexp provides pattern to select client names, eg ^FS00* makes
65 all client names starting with FS00 eligible for migration.
67 PoolResidence -- Migrate data sitting in pool for longer than
68 PoolResidence value in pool definition. Migration Selection Pattern
69 optional; if specified, override value in pool definition (value in
72 Volume -- Migrate all data on specified volumes. Migration Selection
73 Pattern regexp provides selection criteria for volumes to be migrated.
74 Volumes must exist in pool to be eligible for migration.
76 Jobname -- Migration all jobs matching name. Migration Selection Pattern
77 regexp provides pattern to select jobnames existing in pool.
79 SQLQuery -- Migrate all jobuids returned by the supplied SQL query.
80 Migration Selection Pattern contains SQL query to execute; should return
81 a list of 1 or more jobuids to migrate.
83 [ possibly a Python event -- kes ]
85 - Document export LDFLAGS=-L/usr/lib64
86 - Don't attempt to restore from "Disabled" Volumes.
87 - Network error on Win32 should set Win32 error code.
88 - What happens when you rename a Disk Volume?
89 - Job retention period in a Pool (and hence Volume). The job would
91 - Detect resource deadlock in Migrate when same job wants to read
92 and write the same device.
93 - Make hardlink code at line 240 of find_one.c use binary search.
94 - Queue warning/error messages during restore so that they
95 are reported at the end of the report rather than being
96 hidden in the file listing ...
97 - Look at -D_FORTIFY_SOURCE=2
98 - Add Win32 FileSet definition somewhere
99 - Look at fixing restore status stats in SD.
100 - Make selection of Database used in restore correspond to
102 - Implement a mode that says when a hard read error is
103 encountered, read many times (as it currently does), and if the
104 block cannot be read, skip to the next block, and try again. If
105 that fails, skip to the next file and try again, ...
107 create table LevelType (LevelType binary(1), LevelTypeLong tinyblob);
108 insert into LevelType (LevelType,LevelTypeLong) values
112 - Add ACL to restore only to original location.
113 - Add a recursive mark command (rmark) to restore.
114 - "Minimum Job Interval = nnn" sets minimum interval between Jobs
115 of the same level and does not permit multiple simultaneous
116 running of that Job (i.e. lets any previous invocation finish
117 before doing Interval testing).
118 - Look at simplifying File exclusions.
119 - New directive "Delete purged Volumes"
120 - new pool XXX with ScratchPoolId = MyScratchPool's PoolId and
121 let it fill itself, and RecyclePoolId = XXX's PoolId so I can
122 see if it become stable and I just have to supervise
124 - If I want to remove this pool, I set RecyclePoolId = MyScratchPool's
125 PoolId, and when it is empty remove it.
126 - Figure out how to recycle Scratch volumes back to the Scratch Pool.
128 - Allow Check Labels to be used with Bacula labels.
129 - "Resuming" a failed backup (lost line for example) by using the
130 failed backup as a sort of "base" job.
132 - Email to the user when the tape is about to need changing x
133 days before it needs changing.
134 - Command to show next tape that will be used for a job even
135 if the job is not scheduled.
136 - From: Arunav Mandal <amandal@trolltech.com>
137 1. When jobs are running and bacula for some reason crashes or if I do a
138 restart it remembers and jobs it was running before it crashed or restarted
139 as of now I loose all jobs if I restart it.
141 2. When spooling and in the midway if client is disconnected for instance a
142 laptop bacula completely discard the spool. It will be nice if it can write
143 that spool to tape so there will be some backups for that client if not all.
145 3. We have around 150 clients machines it will be nice to have a option to
146 upgrade all the client machines bacula version automatically.
148 4. Atleast one connection should be reserved for the bconsole so at heavy load
149 I should connect to the director via bconsole which at sometimes I can't
151 5. Another most important feature that is missing, say at 10am I manually
152 started backup of client abc and it was a full backup since client abc has
153 no backup history and at 10.30am bacula again automatically started backup of
154 client abc as that was in the schedule. So now we have 2 multiple Full
155 backups of the same client and if we again try to start a full backup of
156 client backup abc bacula won't complain. That should be fixed.
158 - Fix bpipe.c so that it does not modify results pointer.
159 ***FIXME*** calling sequence should be changed.
160 - For Windows disaster recovery see http://unattended.sf.net/
161 - regardless of the retention period, Bacula will not prune the
162 last Full, Diff, or Inc File data until a month after the
163 retention period for the last Full backup that was done.
164 - update volume=xxx --- add status=Full
165 - Remove old spool files on startup.
166 - Exclude SD spool/working directory.
167 - Refuse to prune last valid Full backup. Same goes for Catalog.
169 - Make a callback when Rerun failed levels is called.
170 - Give Python program access to Scheduled jobs.
171 - Add setting Volume State via Python.
172 - Python script to save with Python, not save, save with Bacula.
173 - Python script to do backup.
175 - Change the Priority, Client, Storage, JobStatus (error)
176 at the start of a job.
177 - Why is SpoolDirectory = /home/bacula/spool; not reported
178 as an error when writing a DVD?
179 - Make bootstrap file handle multiple MediaTypes (SD)
180 - Remove all old Device resource code in Dir and code to pass it
181 back in SD -- better, rework it to pass back device statistics.
182 - Check locking of resources -- be sure to lock devices where previously
183 resources were locked.
184 - The last part is left in the spool dir.
187 - In restore don't compare byte count on a raw device -- directory
188 entry does not contain bytes.
189 - To mark files as deleted, run essentially a Verify to disk, and
190 when a file is found missing (MarkId != JobId), then create
191 a new File record with FileIndex == -1. This could be done
192 by the FD at the same time as the backup.
196 MA = (last_MA * 3 + rate) / 4
197 rate = (bytes - last_bytes) / (runtime - last_runtime)
198 - Max Vols limit in Pool off by one?
199 - Implement Files/Bytes,... stats for restore job.
200 - Implement Total Bytes Written, ... for restore job.
201 - Despool attributes simultaneously with data in a separate
202 thread, rejoined at end of data spooling.
203 - 7. Implement new Console commands to allow offlining/reserving drives,
204 and possibly manipulating the autochanger (much asked for).
205 - Add start/end date editing in messages (%t %T, %e?) ...
206 - Add ClientDefs similar to JobDefs.
207 - Print more info when bextract -p accepts a bad block.
208 - Fix FD JobType to be set before RunBeforeJob in FD.
209 - Look at adding full Volume and Pool information to a Volume
210 label so that bscan can get *all* the info.
211 - If the user puts "Purge Oldest Volume = yes" or "Recycle Oldest Volume = yes"
212 and there is only one volume in the pool, refuse to do it -- otherwise
213 he fills the Volume, then immediately starts reusing it.
214 - Implement copies and stripes.
215 - Add history file to console.
216 - Each file on tape creates a JobMedia record. Peter has 4 million
217 files spread over 10000 tape files and four tapes. A restore takes
218 16 hours to build the restore list.
219 - Add and option to check if the file size changed during backup.
220 - Make sure SD deletes spool files on error exit.
221 - Delete old spool files when SD starts.
222 - When labeling tapes, if you enter 000026, Bacula uses
223 the tape index rather than the Volume name 000026.
224 - Add offline tape command to Bacula console.
226 Enter MediaId or Volume name: 32
227 Enter new Volume name: DLT-20Dec04
228 Automatically selected Pool: Default
229 Connecting to Storage daemon DLTDrive at 192.168.68.104:9103 ...
230 Sending relabel command from "DLT-28Jun03" to "DLT-20Dec04" ...
231 block.c:552 Write error at 0:0 on device /dev/nst0. ERR=Bad file descriptor.
232 Error writing final EOF to tape. This tape may not be readable.
233 dev.c:1207 ioctl MTWEOF error on /dev/nst0. ERR=Permission denied.
234 askdir.c:219 NULL Volume name. This shouldn't happen!!!
235 3912 Failed to label Volume: ERR=dev.c:1207 ioctl MTWEOF error on /dev/nst0. ERR=Permission denied.
236 Label command failed for Volume DLT-20Dec04.
237 Do not forget to mount the drive!!!
238 - Bug: if a job is manually scheduled to run later, it does not appear
239 in any status report and cannot be cancelled.
241 ==== Keeping track of deleted files ====
242 My "trick" for keeping track of deletions is the following.
243 Assuming the user turns on this option, after all the files
244 have been backed up, but before the job has terminated, the
245 FD will make a pass through all the files and send their
246 names to the DIR (*exactly* the same as what a Verify job
247 currently does). This will probably be done at the same
248 time the files are being sent to the SD avoiding a second
249 pass. The DIR will then compare that to what is stored in
250 the catalog. Any files in the catalog but not in what the
251 FD sent will receive a catalog File entry that indicates
252 that at that point in time the file was deleted.
254 During a restore, any file initially picked up by some
255 backup (Full, ...) then subsequently having a File entry
256 marked "delete" will be removed from the tree, so will not
257 be restored. If a file with the same name is later OK it
258 will be inserted in the tree -- this already happens. All
259 will be consistent except for possible changes during the
262 Since I'm on the subject, some of you may be wondering what
263 the utility of the in memory tree is if you are going to
264 restore everything (at least it comes up from time to time
265 on the list). Well, it is still *very* useful because it
266 allows only the last item found for a particular filename
267 (full path) to be entered into the tree, and thus if a file
268 is backed up 10 times, only the last copy will be restored.
269 I recently (last Friday) restored a complete directory, and
270 the Full and all the Differential and Incremental backups
271 spanned 3 Volumes. The first Volume was not even mounted
272 because all the files had been updated and hence backed up
273 since the Full backup was made. In this case, the tree
274 saved me a *lot* of time.
276 Make sure this information is stored on the tape too so
277 that it can be restored directly from the tape.
279 Comments from Martin Simmons (I think they are all covered):
280 Ok, that should cover the basics. There are few issues though:
282 - Restore will depend on the catalog. I think it is better to include the
283 extra data in the backup as well, so it can be seen by bscan and bextract.
285 - I'm not sure if it will preserve multiple hard links to the same inode. Or
286 maybe adding or removing links will cause the data to be dumped again?
288 - I'm not sure if it will handle renamed directories. Possibly it will work
289 by dumping the whole tree under a renamed directory?
291 - It remains to be seen how the backup performance of the DIR's will be
292 affected when comparing the catalog for a large filesystem.
296 How about introducing a Type = MgmtPolicy job type? That job type would
297 be responsible for scanning the Bacula environment looking for specific
298 conditions, and submitting the appropriate jobs for implementing said
302 Name = "Migration-Policy"
304 Policy Selection Job Type = Migrate
305 Scope = "<keyword> <operator> <regexp>"
306 Threshold = "<keyword> <operator> <regexp>"
307 Job Template = <template-name>
310 Where <keyword> is any legal job keyword, <operator> is a comparison
311 operator (=,<,>,!=, logical operators AND/OR/NOT) and <regexp> is a
312 appropriate regexp. I could see an argument for Scope and Threshold
313 being SQL queries if we want to support full flexibility. The
314 Migration-Policy job would then get scheduled as frequently as a site
315 felt necessary (suggested default: every 15 minutes).
320 Name = "Migration-Policy"
322 Policy Selection Job Type = Migration
324 Threshold = "Migration Selection Type = LowestUtil"
325 Job Template = "MigrationTemplate"
328 would select all pools for examination and generate a job based on
329 MigrationTemplate to automatically select the volume with the lowest
330 usage and migrate it's contents to the nextpool defined for that pool.
332 This policy abstraction would be really handy for adjusting the behavior
333 of Bacula according to site-selectable criteria (one thing that pops
334 into mind is Amanda's ability to automatically adjust backup levels
335 depending on various criteria).
341 - Add Pool/Storage override regression test.
342 - Add delete JobId to regression.
343 - Add a regression test for dbcheck.
344 - New test to add bscan to four-concurrent-jobs regression,
345 i.e. after the four-concurrent jobs zap the
346 database as is done in the bscan-test, then use bscan to
347 restore the database, do a restore and compare with the
349 - Add restore of specific JobId to regression (item 3
350 on the restore prompt)
351 - Add IPv6 to regression
352 - Add database test to regression. Test each function like delete,
355 - AntiVir can slow down backups on Win32 systems.
356 - Win32 systems with FAT32 can be much slower than NTFS for
357 more than 1000 files per directory.
361 - A HOLD command to stop all jobs from starting.
362 - A PAUSE command to pause all running jobs ==> release the
364 - Media Type = LTO,LTO-2,LTO-3
365 Media Type Read = LTO,LTO2,LTO3
366 Media Type Write = LTO2, LTO3
368 === From Carsten Menke <bootsy52@gmx.net>
370 Following is a list of what I think in the situations where I'm faced with,
371 could be a usefull enhancement to bacula, which I'm certain other users will
372 benefit from as well.
374 1. NextJob/NextJobs Directive within a Job Resource in the form of
375 NextJobs = job1,job2.
378 I currently solved the problem with running multiple jobs each after each
379 by setting the Max Wait Time for a job to 8 hours, and give
380 the jobs different Priorities. However, there scenarios where
381 1 Job is directly depending on another job, so if the former job fails,
382 the job after it needn't to be run
383 while maybe other jobs should run despite of that
386 A Backup Job and a Verify job, if the backup job fails there is no need to run
387 the verify job, as the backup job already failed. However, one may like
388 to backup the Catalog to disk despite of that the main backup job failed.
391 I see that this is related to the Event Handlers which are on the ToDo
392 list, also it is maybe a good idea to check for the return value and
393 execute different actions based on the return value
396 3. offline capability to bconsole
399 Currently I use a script which I execute within the last Job via the
400 RunAfterJob Directive, to release and eject the tape.
401 So I have to call bconsole "release=Storage-Name" and afterwards
402 mt -f /dev/nst0 eject to get the tape out.
404 If I have multiple Storage Devices, than these may not be /dev/nst0 and
405 I have to modify the script or call it with parameters etc.
406 This would actually not be needed, as everything is already defined
407 in bacula-sd.conf and if I can invoke bconsole with the
408 storage name via $1 in the script than I'm done and information is
411 4. %s for Storage Name added to the chars being substituted in "RunAfterJob"
415 For the reason mentioned in 3. to have the ability to call a
416 script with /scripts/foobar %s and in the script use $1
417 to pass the Storage Name to bconsole
419 5. Setting Volume State within a Job Resource
422 Instead of using "Maximum Volume Jobs" in the Pool Resource,
423 I would have the possibilty to define
424 in a Job Resource that after this certain job is run, the Volume State
425 should be set to "Volume State = Used", this give more flexibility (IMHO).
427 6. Localization of Bacula Messages
430 Unfortunatley many,many people I work with don't speak english very well.
431 So if at least the Reporting messages would be localized then they
432 would understand that they have to change the tape,etc. etc.
434 I volunteer to do the german translations, and if I can convince my wife also
435 french and Morre (western african language).
437 7. OK, this is evil, probably bound to security risks and maybe not possible
438 due to the design of bacula.
440 Implementation of Backtics ( `command` ) for shell comand execution to
441 the "Label Format" Directive.
445 Currently I have defined BACULA_DAY_OF_WEEK="day1|day2..." resulting in
446 Label Format = "HolyBackup-${BACULA_DAY_OF_WEEK[${WeekDay}]}". If I could
447 use backticks than I could use "Label Format = HolyBackup-`date +%A` to have
448 the localized name for the day of the week appended to the
449 format string. Then I have the tape labeled automatically with weekday
450 name in the correct language.
452 - Yes, that is surely the case. I probably should turn those into Warning
453 errors. In addition, you just made me think that it might not be bad to
454 add an option to check the file size after backing up the file and
455 report if it changes. This would be done as an option because it would
458 Kern, good idea. If you do do that, mention in the output: file
459 shrunk, or file expanded, just to make it obvious to the user
460 (without having to the refer to file size), just how the file size
463 Would this option be for all file, or just one file? Or a fileset?
464 - Make output from status use html table tags for nicely
465 presenting in a browser.
466 - Can one write tapes faster with 8192 byte block sizes?
467 - Document security problems with the same password for everyone in
468 rpm and Win32 releases.
469 - Browse generations of files.
470 - I've seen an error when my catalog's File table fills up. I
471 then have to recreate the File table with a larger maximum row
472 size. Relevant information is at
473 http://dev.mysql.com/doc/mysql/en/Full_table.html ; I think the
474 "Installing and Configuring MySQL" chapter should talk a bit
475 about this potential problem, and recommend a solution.
476 - For Solaris must use POSIX awk.
477 - Want speed of writing to tape while despooling.
478 - Supported autochanger:
486 Wangtek 6525ES (SCSI-1 QIC drive, 525MB), under Linux 2.4.something,
487 bacula 1.36.0/1 works with blocksize 16k INSIDE bacula-sd.conf.
488 - Add regex from http://www.pcre.org to Bacula for Win32.
489 - Use only shell tools no make in CDROM package.
490 - Include within include does it work?
491 - Implement a Pool of type Cleaning?
492 - Implement VolReadTime and VolWriteTime in SD
493 - Modify Backing up Your Database to include a bootstrap file.
494 - Think about making certain database errors fatal.
495 - Look at correcting the time jump in the scheduler for daylight
496 savings time changes.
497 - Add a "real" timer to network connections.
498 - Promote to Full = Time period
499 - Check dates entered by user for correctness (month/day/... ranges)
500 - Compress restore Volume listing by date and first file.
501 - Look at patches/bacula_db.b2z postgresql that loops during restore.
503 - Perhaps add read/write programs and/or plugins to FileSets.
504 - How to handle backing up portables ...
505 - Add some sort of guaranteed Interval for upgrading jobs.
506 - Can we write the state file after every job terminates? On Win32
507 the system crashes and the state file is not updated.
510 Documentation to do: (any release a little bit at a time)
511 - Doc to do unmount before removing magazine.
512 - Alternative to static linking "ldd prog" save all binaries listed,
513 restore them and point LD_LIBRARY_PATH to them.
514 - Document add "</dev/null >/dev/null 2>&1" to the bacula-fd command line
515 - Document query file format.
516 - Add more documentation for bsr files.
517 - Document problems with Verify and pruning.
518 - Document how to use multiple databases.
519 - VXA drives have a "cleaning required"
520 indicator, but Exabyte recommends preventive cleaning after every 75
523 In this context, it should be noted that Exabyte has a command-line
524 vxatool utility available for free download. (The current version is
525 vxatool-3.72.) It can get diagnostic info, read, write and erase tapes,
526 test the drive, unload tapes, change drive settings, flash new firmware,
528 Of particular interest in this context is that vxatool <device> -i will
529 report, among other details, the time since last cleaning in tape motion
530 minutes. This information can be retrieved (and settings changed, for
531 that matter) through the generic-SCSI device even when Bacula has the
532 regular tape device locked. (Needless to say, I don't recommend
533 changing tape settings while a job is running.)
534 - Lookup HP cleaning recommendations.
535 - Lookup HP tape replacement recommendations (see trouble shooting autochanger)
536 - Document doing table repair
539 ===================================
540 - Add macro expansions in JobDefs.
541 Run Before Job = "SomeFile %{Level} %{Client}"
542 Write Bootstrap="/some/dir/%{JobName}_%{Client}.bsr"
543 - Use non-blocking network I/O but if no data is available, use
545 - Use gather write() for network I/O.
546 - Autorestart on crash.
547 - Add bandwidth limiting.
548 - Add acks every once and a while from the SD to keep
549 the line from timing out.
550 - When an error in input occurs and conio beeps, you can back
551 up through the prompt.
552 - Detect fixed tape block mode during positioning by looking at
553 block numbers in btape "test". Possibly adjust in Bacula.
554 - Fix list volumes to output volume retention in some other
555 units, perhaps via a directive.
556 - If opening a tape in read/write mode fails attempt to open
557 it in read-only mode, and mark the tape for read only.
558 - Allow Simultaneous Priorities = yes => run up to Max concurrent jobs even
559 with multiple priorities.
560 - If you use restore replace=never, the directory attributes for
561 non-existent directories will not be restored properly.
563 - see lzma401.zip in others directory for new compression
565 - Minimal autochanger handling in Bacula and in btape.
566 - Look into how tar does not save sockets and the possiblity of
567 not saving them in Bacula (Martin Simmons reported this).
568 - Fix restore jobs so that multiple jobs can run if they
569 are not using the same tape(s).
570 - Allow the user to select JobType for manual pruning/purging.
571 - bscan does not put first of two volumes back with all info in
573 - Implement the FreeBSD nodump flag in chflags.
574 - Figure out how to make named console messages go only to that
575 console and to the non-restricted console (new console class?).
576 - Make restricted console prompt for password if *ask* is set or
577 perhaps if password is undefined.
578 - Implement "from ISO-date/time every x hours/days/weeks/months" in
581 ==== from Marc Schoechlin
582 - the help-command should be more verbose
583 (it should explain the paramters of the different
585 -> its time-comsuming to consult the manual anytime
586 you need a special parameter
587 -> maybe its more easy to maintain this, if the
588 descriptions of that commands are outsourced to
590 - the cd-command should allow complete paths
591 i.e. cd /foo/bar/foo/bar
592 -> if a customer mails me the path to a certain file,
593 its faster to enter the specified directory
594 - if the password is not configured in bconsole.conf
595 you should be asked for it.
596 -> sometimes you like to do restore on a customer-machine
597 which shouldnt know the password for bacula.
598 -> adding the password to the file favours admins
599 to forget to remove the password after usage
601 the protection of that file is less important
602 - long-listed-output of commands should be scrollable
603 like the unix more/less-command does
604 -> if someone runs 200 and more machines, the lists could
605 be a little long and complex
606 - command-output should be shown column by column
607 to reduce scrolling and to increase clarity
609 - lsmark should list the selected files with full
611 - wildcards for selecting and file and directories would be nice
612 - any actions should be interuptable with STRG+C
613 - command-expansion would be pretty cool
615 - When the replace Never option is set, new directory permissions
616 are not restored. See bug 213. To fix this requires creating a
617 list of newly restored directories so that those directory
618 permissions *can* be restored.
619 - Compaction of Disk space by "migrating" Volumes that have pruned
620 Jobs (what criteria? size, #jobs, time).
621 - Add prune all command
622 - Document fact that purge can destroy a part of a restore by purging
623 one volume while others remain valid -- perhaps mark Jobs.
624 - Add multiple-media-types.txt
625 - look at mxt-changer.html
626 - Make ? do a help command (no return needed).
627 - Implement restore directory.
628 - Document streams and how to implement them.
629 - Try not to re-backup a file if a new hard link is added.
630 - Add feature to backup hard links only, but not the data.
631 - Fix stream handling to be simpler.
632 - Add Priority and Bootstrap to Run a Job.
633 - Eliminate Restore "Run Restore Job" prompt by allowing new "run command
635 - Remove View FileSet button from Run a Job dialog.
636 - Handle prompt for restore job at end of Restore command.
637 - Add display of total selected files to Restore window.
638 - Add tree pane to left of window.
639 - Add progress meter.
640 - Max wait time or max run time causes seg fault -- see runtime-bug.txt
641 - Document writing to a CD/DVD with Bacula.
642 - Add a "base" package to the window installer for pthreadsVCE.dll
643 which is needed by all packages.
644 - Add message to user to check for fixed block size when the forward
645 space test fails in btape.
646 - When unmarking a directory check if all files below are unmarked and
647 then remove the + flag -- in the restore tree.
648 - Possibly implement: Action = Unmount Device="TapeDrive1" in Admin jobs.
649 - Setup lrrd graphs: (http://www.linpro.no/projects/lrrd/) Mike Acar.
650 - Revisit the question of multiple Volumes (disk) on a single device.
651 - Add a block copy option to bcopy.
652 - Investigate adding Mac Resource Forks.
653 - Finish work on Gnome restore GUI.
654 - Fix "llist jobid=xx" where no fileset or client exists.
655 - For each job type (Admin, Restore, ...) require only the really necessary
656 fields.- Pass Director resource name as an option to the Console.
657 - Add a "batch" mode to the Console (no unsolicited queries, ...).
658 - Add a .list all files in the restore tree (probably also a list all files)
659 Do both a long and short form.
660 - Allow browsing the catalog to see all versions of a file (with
661 stat data on each file).
662 - Restore attributes of directory if replace=never set but directory
664 - Use SHA1 on authentication if possible.
665 - See comtest-xxx.zip for Windows code to talk to USB.
666 - Add John's appended files:
667 Appended = { /files/server/logs/http/*log }
668 and such files would be treated as follows.On a FULL backup, they would
669 be backed up like any other file.On an INCREMENTAL backup, where a
670 previous INCREMENTAL or FULL was already in thecatalogue and the length
671 of the file wasgreater than the length of the last backup, only thedata
672 added since the last backup will be dumped.On an INCREMENTAL backup, if
673 the length of the file is less than thelength of the file with the same
674 name last backed up, the completefile is dumped.On Windows systems, with
675 creation date of files, we can be evensmarter about this and not count
676 entirely upon the length.On a restore, the full and all incrementals
677 since it will beapplied in sequence to restore the file.
678 - Check new HAVE_WIN32 open bits.
679 - Check if the tape has moved before writing.
680 - Handling removable disks -- see below:
681 - Keep track of tape use time, and report when cleaning is necessary.
682 - Add FromClient and ToClient keywords on restore command (or
683 BackupClient RestoreClient).
684 - Implement a JobSet, which groups any number of jobs. If the
685 JobSet is started, all the jobs are started together.
686 Allow Pool, Level, and Schedule overrides.
687 - Enhance cancel to timeout BSOCK packets after a specific delay.
688 - Do scheduling by UTC using gmtime_r() in run_conf, scheduler, and
689 ua_status.!!! Thanks to Alan Brown for this tip.
690 - Look at updating Volume Jobs so that Max Volume Jobs = 1 will work
691 correctly for multiple simultaneous jobs.
692 - Correct code so that FileSet MD5 is calculated for < and | filename
694 - Implement the Media record flag that indicates that the Volume does disk
696 - Implement VolAddr, which is used when Volume is addressed like a disk,
697 and form it from VolFile and VolBlock.
698 - Make multiple restore jobs for multiple media types specifying
699 the proper storage type.
700 - Fix fast block rejection (stored/read_record.c:118). It passes a null
701 pointer (rec) to try_repositioning().
702 - Look at extracting Win data from BackupRead.
703 - Implement RestoreJobRetention? Maybe better "JobRetention" in a Job,
704 which would take precidence over the Catalog "JobRetention".
705 - Implement Label Format in Add and Label console commands.
706 - Possibly up network buffers to 65K. Put on variable.
707 - Put email tape request delays on one or more variables. User wants
708 to cancel the job after a certain time interval. Maximum Mount Wait?
709 - Job, Client, Device, Pool, or Volume?
710 Is it possible to make this a directive which is *optional* in multiple
711 resources, like Level? If so, I think I'd make it an optional directive
712 in Job, Client, and Pool, with precedence such that Job overrides Client
713 which in turn overrides Pool.
715 - New Storage specifications:
716 - Want to write to multiple storage devices simultaneously
717 - Want to write to multiple storage devices sequentially (in one job)
718 - Want to read/write simultaneously
719 - Key is MediaType -- it must match
721 Passed to SD as a sort of BSR record called Storage Specification
725 MediaType -> Next MediaType
727 Device -> Next Device
729 Allow multiple Storage specifications
737 Allow Multiple Pool specifications (note, Pool currently
739 Allow Multiple MediaType specifications in Dir conf
740 Allow Multiple Device specifications in Dir conf
741 Perhaps keep this in a single SSR
742 Tie a Volume to a specific device by using a MediaType that
743 is contained in only one device.
744 In SD allow Device to have Multiple MediaTypes
746 - Ideas from Jerry Scharf:
747 First let's point out some big pluses that bacula has for this
749 more importantly it's active. Thank you so much for that
750 even more important, it's not flaky
751 it has an open access catalog, opening many possibilities
752 it's pushing toward heterogeneous systems capability
754 Macintosh file client
755 macs are an interesting niche, but I fear a server is a rathole
756 working bare iron recovery for windows
757 the option for inc/diff backups not reset on fileset revision
758 a) use both change and inode update time against base time
759 b) do the full catalog check (expensive but accurate)
760 sizing guide (how much system is needed to back up N systems/files)
761 consultants on using bacula in building a disaster recovery system
763 or how to get at fancy things that one could do with bacula
764 logwatch code for bacula logs (or similar)
765 linux distro inclusion of bacula (brings good and bad, but necessary)
766 win2k/XP server capability (icky but you asked)
767 support for Oracle database ??
769 - Look at adding SQL server and Exchange support for Windows.
770 - Make dev->file and dev->block_num signed integers so that -1 can
771 be an invalid value which happens with BSR.
772 - Create VolAddr for disk files in place of VolFile and VolBlock. This
773 is needed to properly specify ranges.
774 - Add progress of files/bytes to SD and FD.
775 - Print warning message if FileId > 4 billion
776 - do a "messages" before the first prompt in Console
777 - Client does not show busy during Estimate command.
778 - Implement Console mtx commands.
779 - Implement a Mount Command and an Unmount Command where
780 the users could specify a system command to be performed
781 to do the mount, after which Bacula could attempt to
782 read the device. This is for Removeable media such as a CDROM.
783 - Most likely, this mount command would be invoked explicitly
784 by the user using the current Console "mount" and "unmount"
785 commands -- the Storage Daemon would do the right thing
786 depending on the exact nature of the device.
787 - As with tape drives, when Bacula wanted a new removable
788 disk mounted, it would unmount the old one, and send a message
789 to the user, who would then use "mount" as described above
790 once he had actually inserted the disk.
791 - Implement dump/print label to UA
792 - Spool to disk only when the tape is full, then when a tape is hung move
794 - bextract is sending everything to the log file ****FIXME****
795 - Allow multiple Storage specifications (or multiple names on
796 a single Storage specification) in the Job record. Thus a job
797 can be backed up to a number of storage devices.
798 - Implement some way for the File daemon to contact the Director
799 to start a job or pass its DHCP obtained IP number.
800 - Implement a query tape prompt/replace feature for a console
801 - Copy console @ code to gnome2-console
802 - Make AES the only encryption algorithm see
803 http://csrc.nist.gov/CryptoToolkit/aes/). It's
804 an officially adopted standard, has survived peer
805 review, and provides keys up to 256 bits.
806 - Take a careful look at SetACL http://setacl.sourceforge.net
807 - Make tree walk routines like cd, ls, ... more user friendly
808 by handling spaces better.
809 - Make sure that Bacula rechecks the tape after the 20 min wait.
810 - Set IO_NOWAIT on Bacula TCP/IP packets.
811 - Try doing a raw partition backup and restore by mounting a
814 Yes, it would allow to highly automatic the request for new tapes. If a
815 tape is empty, bacula reads the barcodes (native or simulated), and if
816 an unused tape is found, it runs the label command with all the
817 necessary parameters.
819 By the way can bacula automatically "move" an empty/purged volume say
820 in the "short" pool to the "long" pool if this pool runs out of volume
822 - What to do about "list files job=xxx".
823 - Get and test MySQL 4.0
824 - Look at how fuser works and /proc/PID/fd that is how Nic found the
825 file descriptor leak in Bacula.
826 - Implement WrapCounters in Counters.
827 - Add heartbeat from FD to SD if hb interval expires.
828 - Can we dynamically change FileSets?
829 - If pool specified to label command and Label Format is specified,
830 automatically generate the Volume name.
831 - Why can't SQL do the filename sort for restore?
832 - Add ExhautiveRestoreSearch
833 - Look at the possibility of loading only the necessary
834 data into the restore tree (i.e. do it one directory at a
835 time as the user walks through the tree).
836 - Possibly use the hash code if the user selects all for a restore command.
837 - Fix "restore all" to bypass building the tree.
838 - Prohibit backing up archive device (findlib/find_one.c:128)
839 - Implement Release Device in the Job resource to unmount a drive.
840 - Implement Acquire Device in the Job resource to mount a drive,
841 be sure this works with admin jobs so that the user can get
842 prompted to insert the correct tape. Possibly some way to say to
843 run the job but don't save the files.
844 - Make things like list where a file is saved case independent for
847 - Use autochanger to handle multiple devices.
848 - On Windows with very long path names, it may be impossible to create
849 a file (and thus restore it) because the total length is too long.
850 We must cd into the directory then create the file without the
852 - Implement a Recycle command
853 - Test a second language e.g. french.
854 - Start working on Base jobs.
855 - Implement UnsavedFiles DB record.
856 - From Phil Stracchino:
857 It would probably be a per-client option, and would be called
858 something like, say, "Automatically purge obsoleted jobs". What it
859 would do is, when you successfully complete a Differential backup of a
860 client, it would automatically purge all Incremental backups for that
861 client that are rendered redundant by that Differential. Likewise,
862 when a Full backup on a client completed, it would automatically purge
863 all Differential and Incremental jobs obsoleted by that Full backup.
864 This would let people minimize the number of tapes they're keeping on
865 hand without having to master the art of retention times.
866 - When doing a Backup send all attributes back to the Director, who
867 would then figure out what files have been deleted.
868 - Currently in mount.c:236 the SD simply creates a Volume. It should have
869 explicit permission to do so. It should also mark the tape in error
870 if there is an error.
871 - Cancel waiting for Client connect in SD if FD goes away.
873 - Implement timeout in response() when it should come quickly.
874 - Implement a Slot priority (loaded/not loaded).
875 - Implement "vacation" Incremental only saves.
876 - Implement create "FileSet"?
877 - Add prefixlinks to where or not where absolute links to FD.
878 - Issue message to mount a new tape before the rewind.
879 - Simplified client job initiation for portables.
880 - If SD cannot open a drive, make it periodically retry.
881 - Add more of the config info to the tape label.
883 - If tape is marked read-only, then try opening it read-only rather than
884 failing, and remember that it cannot be written.
885 - Refine SD waiting output:
886 Device is being positioned
887 > Device is being positioned for append
888 > Device is being positioned to file x
890 - Figure out some way to estimate output size and to avoid splitting
891 a backup across two Volumes -- this could be useful for writing CDROMs
892 where you really prefer not to have it split -- not serious.
893 - Have SD compute MD5 or SHA1 and compare to what FD computes.
894 - Make VolumeToCatalog calculate an MD5 or SHA1 from the
895 actual data on the Volume and compare it.
896 - Implement Bacula plugins -- design API
897 - Make bcopy read through bad tape records.
898 - Program files (i.e. execute a program to read/write files).
899 Pass read date of last backup, size of file last time.
900 - Add Signature type to File DB record.
901 - CD into subdirectory when open()ing files for backup to
902 speed up things. Test with testfind().
903 - Priority job to go to top of list.
904 - Why are save/restore of device different sizes (sparse?) Yup! Fix it.
905 - Implement some way for the Console to dynamically create a job.
906 - Solaris -I on tar for include list
907 - Need a verbose mode in restore, perhaps to bsr.
908 - bscan without -v is too quiet -- perhaps show jobs.
909 - Add code to reject whole blocks if not wanted on restore.
910 - Check if we can increase Bacula FD priorty in Win2000
911 - Make sure the MaxVolFiles is fully implemented in SD
912 - Check if both CatalogFiles and UseCatalog are set to SD.
913 - Possibly add email to Watchdog if drive is unmounted too
914 long and a job is waiting on the drive.
915 - After unmount, if restore job started, ask to mount.
916 - Add UA rc and history files.
917 - put termcap (used by console) in ./configure and
918 allow -with-termcap-dir.
919 - Fix Autoprune for Volumes to respect need for full save.
920 - Compare tape to Client files (attributes, or attributes and data)
921 - Make all database Ids 64 bit.
922 - Allow console commands to detach or run in background.
923 - Fix status delay on storage daemon during rewind.
924 - Add SD message variables to control operator wait time
925 - Maximum Operator Wait
926 - Minimum Message Interval
927 - Maximum Message Interval
928 - Send Operator message when cannot read tape label.
929 - Verify level=Volume (scan only), level=Data (compare of data to file).
930 Verify level=Catalog, level=InitCatalog
932 - Add keyword search to show command in Console.
933 - Events : tape has more than xxx bytes.
934 - Complete code in Bacula Resources -- this will permit
935 reading a new config file at any time.
936 - Handle ctl-c in Console
937 - Implement script driven addition of File daemon to config files.
938 - Think about how to make Bacula work better with File (non-tape) archives.
939 - Write Unix emulator for Windows.
940 - Put memory utilization in Status output of each daemon
941 if full status requested or if some level of debug on.
942 - Make database type selectable by .conf files i.e. at runtime
943 - Set flag for uname -a. Add to Volume label.
944 - Restore files modified after date
945 - SET LD_RUN_PATH=$HOME/mysql/lib/mysql
946 - Remove duplicate fields from jcr (e.g. jcr.level and jcr.jr.Level, ...).
947 - Timout a job or terminate if link goes down, or reopen link and query.
948 - Concept of precious tapes (cannot be reused).
949 - Make bcopy copy with a single tape drive.
950 - Permit changing ownership during restore.
953 > My suggestion: Add a feature on the systray menu-icon menu to request
954 > an immediate backup now. This would be useful for laptop users who may
955 > not be on the network when the regular scheduled backup is run.
957 > My wife's suggestion: Add a setting to the win32 client to allow it to
958 > shut down the machine after backup is complete (after, of course,
959 > displaying a "System will shut down in one minute, click here to cancel"
960 > warning dialog). This would be useful for sites that want user
961 > woorkstations to be shut down overnight to save power.
964 - Autolabel should be specified by DIR instead of SD.
967 - AutoScan (check checksum of tape)
968 - Format command = "format /dev/nst0"
972 - Seek resolution (usually corresponds to buffer size)
973 - EODErrorCode=ENOSPC or code
974 - Partial Read error code
975 - Partial write error code
976 - Nonformatted read error
977 - Nonformatted write error
978 - WriteProtected error
982 - IgnoreCloseErrors=yes
992 - FD sends unsaved file list to Director at end of job (see
994 - File daemon should build list of files skipped, and then
995 at end of save retry and report any errors.
996 - Write a Storage daemon that uses pipes and
997 standard Unix programs to write to the tape.
999 - Need something that monitors the JCR queue and
1000 times out jobs by asking the deamons where they are.
1001 - Enhance Jmsg code to permit buffering and saving to disk.
1002 - device driver = "xxxx" for drives.
1003 - Verify from Volume
1004 - Ensure that /dev/null works
1005 - Need report class for messages. Perhaps
1006 report resource where report=group of messages
1007 - enhance scan_attrib and rename scan_jobtype, and
1008 fill in code for "since" option
1009 - Director needs a time after which the report status is sent
1010 anyway -- or better yet, a retry time for the job.
1011 - Don't reschedule a job if previous incarnation is still running.
1012 - Some way to automatically backup everything is needed????
1013 - Need a structure for pending actions:
1015 - termination status (part of buffered msgs?)
1017 Read, Write, Clean, Delete
1018 - Login to Bacula; Bacula users with different permissions:
1019 owner, group, user, quotas
1020 - Store info on each file system type (probably in the job header on tape.
1021 This could be the output of df; or perhaps some sort of /etc/mtab record.
1023 ========= ideas ===============
1024 From: "Jerry K. Schieffer" <jerry@skylinetechnology.com>
1025 To: <kern@sibbald.com>
1026 Subject: RE: [Bacula-users] future large programming jobs
1027 Date: Thu, 26 Feb 2004 11:34:54 -0600
1029 I noticed the subject thread and thought I would offer the following
1030 merely as sources of ideas, i.e. something to think about, not even as
1031 strong as a request. In my former life (before retiring) I often
1032 dealt with backups and storage management issues/products as a
1033 developer and as a consultant. I am currently migrating my personal
1034 network from amanda to bacula specifically because of the ability to
1035 cross media boundaries during storing backups.
1036 Are you familiar with the commercial product called ADSM (I think IBM
1037 now sells it under the Tivoli label)? It has a couple of interesting
1038 ideas that may apply to the following topics.
1040 1. Migration: Consider that when you need to restore a system, there
1041 may be pressure to hurry. If all the information for a single client
1042 can eventually end up on the same media (and in chronological order),
1043 the restore is facillitated by not having to search past information
1044 from other clients. ADSM has the concept of "client affinity" that
1045 may be associated with it's storage pools. It seems to me that this
1046 concept (as an optional feature) might fit in your architecture for
1049 ADSM also has the concept of defining one or more storage pools as
1050 "copy pools" (almost mirrors, but only in the sense of contents).
1051 These pools provide the ability to have duplicte data stored both
1052 onsite and offsite. The copy process can be scheduled to be handled
1053 by their storage manager during periods when there is no backup
1054 activity. Again, the migration process might be a place to consider
1055 implementing something like this.
1058 > It strikes me that it would be very nice to be able to do things
1060 > have the Job(s) backing up the machines run, and once they have all
1061 > completed, start a migration job to copy the data from disks Volumes
1063 > a tape library and then to offsite storage. Maybe this can already
1065 > done with some careful scheduling and Job prioritzation; the events
1066 > mechanism described below would probably make it very easy.
1068 This is the goal. In the first step (before events), you simply
1070 the Migration to tape later.
1072 2. Base jobs: In ADSM, each copy of each stored file is tracked in
1073 the database. Once a file (unique by path and metadata such as dates,
1074 size, ownership, etc.) is in a copy pool, no more copies are made. In
1075 other words, when you start ADSM, it begins like your concept of a
1076 base job. After that it is in the "incremental" mode. You can
1077 configure the number of "generations" of files to be retained, plus a
1078 retention date after which even old generations are purged. The
1079 database tracks the contents of media and projects the percentage of
1080 each volume that is valid. When the valid content of a volume drops
1081 below a configured percentage, the valid data are migrated to another
1082 volume and the old volume is marked as empty. Note, this requires
1083 ADSM to have an idea of the contents of a client, i.e. marking the
1084 database when an existing file was deleted, but this would solve your
1085 issue of restoring a client without restoring deleted files.
1087 This is pretty far from what bacula now does, but if you are going to
1088 rip things up for Base jobs,.....
1089 Also, the benefits of this are huge for very large shops, especially
1090 with media robots, but are a pain for shops with manual media
1094 > Base jobs sound pretty useful, but I'm not dying for them.
1096 Nobody is dying for them, but when you see what it does, you will die
1099 3. Restoring deleted files: Since I think my comments in (2) above
1100 have low probability of implementation, I'll also suggest that you
1101 could approach the issue of deleted files by a mechanism of having the
1102 fd report to the dir, a list of all files on the client for every
1103 backup job. The dir could note in the database entry for each file
1104 the date that the file was seen. Then if a restore as of date X takes
1105 place, only files that exist from before X until after X would be
1106 restored. Probably the major cost here is the extra date container in
1107 each row of the files table.
1109 Thanks for "listening". I hope some of this helps. If you want to
1110 contact me, please send me an email - I read some but not all of the
1111 mailing list traffic and might miss a reply there.
1113 Please accept my compliments for bacula. It is doing a great job for
1114 me!! I sympathize with you in the need to wrestle with excelence in
1115 execution vs. excelence in feature inclusion.
1120 ==============================
1123 - Design at hierarchial storage for Bacula. Migration and Clone.
1124 - Implement FSM (File System Modules).
1125 - Audit M_ error codes to ensure they are correct and consistent.
1126 - Add variable break characters to lex analyzer.
1127 Either a bit mask or a string of chars so that
1128 the caller can change the break characters.
1129 - Make a single T_BREAK to replace T_COMMA, etc.
1130 - Ensure that File daemon and Storage daemon can
1131 continue a save if the Director goes down (this
1132 is NOT currently the case). Must detect socket error,
1133 buffer messages for later.
1134 - Enhance time/duration input to allow multiple qualifiers e.g. 3d2h
1135 - Add ability to backup to two Storage devices (two SD sessions) at
1136 the same time -- e.g. onsite, offsite.
1137 - Add the ability to consolidate old backup sets (basically do a restore
1138 to tape and appropriately update the catalog). Compress Volume sets.
1139 Might need to spool via file is only one drive is available.
1140 - Compress or consolidate Volumes of old possibly deleted files. Perhaps
1141 someway to do so with every volume that has less than x% valid
1145 Migration: Move a backup from one Volume to another
1146 Clone: Copy a backup -- two Volumes
1148 Bacula Migration is based on Jobs (apparently Networker is file by file).
1150 Migration triggered by:
1154 Highwater mark (keep total size)
1159 ======================================================
1161 It is somewhat like a Full save becomes an incremental since
1162 the Base job (or jobs) plus other non-base files.
1164 - A Base backup is same as Full backup, just different type.
1165 - New BaseFiles table that contains:
1167 BaseJobId - Base JobId referenced for this FileId (needed ???)
1168 JobId - JobId currently running
1169 FileId - File not backed up, exists in Base Job
1170 FileIndex - FileIndex from Base Job.
1171 i.e. for each base file that exists but is not saved because
1172 it has not changed, the File daemon sends the JobId, BaseId,
1173 FileId, FileIndex back to the Director who creates the DB entry.
1174 - To initiate a Base save, the Director sends the FD
1175 the FileId, and full filename for each file in the Base.
1176 - When the FD finds a Base file, he requests the Director to
1177 send him the full File entry (stat packet plus MD5), or
1178 conversely, the FD sends it to the Director and the Director
1179 says yes or no. This can be quite rapid if the FileId is kept
1180 by the FD for each Base Filename.
1181 - It is probably better to have the comparison done by the FD
1182 despite the fact that the File entry must be sent across the
1184 - An alternative would be to send the FD the whole File entry
1185 from the start. The disadvantage is that it requires a lot of
1186 space. The advantage is that it requires less communications
1188 - The Job record must be updated to indicate that one or more
1190 - At end of Job, FD returns:
1191 1. Count of base files/bytes not written to tape (i.e. matches)
1192 2. Count of base file that were saved i.e. had changed.
1193 - No tape record would be written for a Base file that matches, in the
1194 same way that no tape record is written for Incremental jobs where
1195 the file is not saved because it is unchanged.
1196 - On a restore, all the Base file records must explicitly be
1197 found from the BaseFile tape. I.e. for each Full save that is marked
1198 to have one or more Base Jobs, search the BaseFile for all occurrences
1200 - An optimization might be to make the BaseFile have:
1206 This would avoid the need to explicitly fetch each File record for
1207 the Base job. The Base Job record will be fetched to get the
1208 VolSessionId and VolSessionTime.
1209 =========================================================
1212 ==========================================================
1214 For each Incremental job that is run, there may be files that
1215 were found but not saved because they were locked (this applies
1216 only to Windows). Such a system could send back to the Director
1217 a list of Unsaved files.
1219 - New UnSavedFiles table that contains:
1223 - Then in the next Incremental job, the list of Unsaved Files will be
1224 feed to the FD, who will ensure that they are explicitly chosen even
1225 if standard date/time check would not have selected them.
1226 =============================================================
1230 Multiple drive autochanger data: see Alan Brown
1231 mtx -f xxx unloadStorage Element 1 is Already Full(drive 0 was empty)
1232 Unloading Data Transfer Element into Storage Element 1...source Element
1233 Address 480 is Empty
1235 (drive 0 was empty and so was slot 1)
1236 > mtx -f xxx load 15 0
1237 no response, just returns to the command prompt when complete.
1238 > mtx -f xxx status Storage Changer /dev/changer:2 Drives, 60 Slots ( 2 Import/Export )
1239 Data Transfer Element 0:Full (Storage Element 15 Loaded):VolumeTag = HX001
1240 Data Transfer Element 1:Empty
1241 Storage Element 1:Empty
1242 Storage Element 2:Full :VolumeTag=HX002
1243 Storage Element 3:Full :VolumeTag=HX003
1244 Storage Element 4:Full :VolumeTag=HX004
1245 Storage Element 5:Full :VolumeTag=HX005
1246 Storage Element 6:Full :VolumeTag=HX006
1247 Storage Element 7:Full :VolumeTag=HX007
1248 Storage Element 8:Full :VolumeTag=HX008
1249 Storage Element 9:Full :VolumeTag=HX009
1250 Storage Element 10:Full :VolumeTag=HX010
1251 Storage Element 11:Empty
1252 Storage Element 12:Empty
1253 Storage Element 13:Empty
1254 Storage Element 14:Empty
1255 Storage Element 15:Empty
1256 Storage Element 16:Empty....
1257 Storage Element 28:Empty
1258 Storage Element 29:Full :VolumeTag=CLNU01L1
1259 Storage Element 30:Empty....
1260 Storage Element 57:Empty
1261 Storage Element 58:Full :VolumeTag=NEX261L2
1262 Storage Element 59 IMPORT/EXPORT:Empty
1263 Storage Element 60 IMPORT/EXPORT:Empty
1265 Unloading Data Transfer Element into Storage Element 15...done
1267 (just to verify it remembers where it came from, however it can be
1268 overrriden with mtx unload {slotnumber} to go to any storage slot.)
1270 There needs to be a table of drive # to devices somewhere - If there are
1271 multiple changers or drives there may not be a 1:1 correspondance between
1272 changer drive number and system device name - and depending on the way the
1273 drives are hooked up to scsi busses, they may not be linearly numbered
1274 from an offset point either.something like
1276 Autochanger drives = 2
1277 Autochanger drive 0 = /dev/nst1
1278 Autochanger drive 1 = /dev/nst2
1279 IMHO, it would be _safest_ to use explicit mtx unload commands at all
1280 times, not just for multidrive changers. For a 1 drive changer, that's
1286 MTX's manpage (1.2.15):
1287 unload [<slotnum>] [ <drivenum> ]
1288 Unloads media from drive <drivenum> into slot
1289 <slotnum>. If <drivenum> is omitted, defaults to
1290 drive 0 (as do all commands). If <slotnum> is
1291 omitted, defaults to the slot that the drive was
1292 loaded from. Note that there's currently no way
1293 to say 'unload drive 1's media to the slot it
1294 came from', other than to explicitly use that
1295 slot number as the destination.AB
1301 undef# camcontrol devlist
1302 <WANGTEK 51000 SCSI M74H 12B3> at scbus0 target 2 lun 0 (pass0,sa0)
1303 <ARCHIVE 4586XX 28887-XXX 4BGD> at scbus0 target 4 lun 0 (pass1,sa1)
1304 <ARCHIVE 4586XX 28887-XXX 4BGD> at scbus0 target 4 lun 1 (pass2)
1306 tapeinfo -f /dev/sg0 with a bad tape in drive 1:
1307 [kern@rufus mtx-1.2.17kes]$ ./tapeinfo -f /dev/sg0
1308 Product Type: Tape Drive
1310 Product ID: 'C5713A '
1312 Attached Changer: No
1313 TapeAlert[3]: Hard Error: Uncorrectable read/write error.
1314 TapeAlert[20]: Clean Now: The tape drive neads cleaning NOW.
1321 Medium Type: Not Loaded
1324 DataCompEnabled: yes
1325 DataCompCapable: yes
1326 DataDeCompEnabled: yes
1333 Handling removable disks
1335 From: Karl Cunningham <karlc@keckec.com>
1337 My backups are only to hard disk these days, in removable bays. This is my
1338 idea of how a backup to hard disk would work more smoothly. Some of these
1339 things Bacula does already, but I mention them for completeness. If others
1340 have better ways to do this, I'd like to hear about it.
1342 1. Accommodate several disks, rotated similar to how tapes are. Identified
1343 by partition volume ID or perhaps by the name of a subdirectory.
1344 2. Abort & notify the admin if the wrong disk is in the bay.
1345 3. Write backups to different subdirectories for each machine to be backed
1347 4. Volumes (files) get created as needed in the proper subdirectory, one
1349 5. When a disk is recycled, remove or zero all old backup files. This is
1350 important as the disk being recycled may be close to full. This may be
1351 better done manually since the backup files for many machines may be
1352 scattered in many subdirectories.
1357 - Make sure that all do_prompt() calls in Dir check for
1358 -1 (error) and -2 (cancel) returns.
1359 - Fix foreach_jcr() to have free_jcr() inside next().
1360 jcr=jcr_walk_start();
1361 for ( ; jcr; (jcr=jcr_walk_next(jcr)) )
1364 - A Volume taken from Scratch should take on the retention period
1366 - Correct doc for Maximum Changer Wait (and others) accepting only
1368 - Implement status that shows why a job is being held in reserve, or
1369 rather why none of the drives are suitable.
1370 - Implement a way to disable a drive (so you can use the second
1371 drive of an autochanger, and the first one will not be used or
1373 - Make sure Maximum Volumes is respected in Pools when adding
1374 Volumes (e.g. when pulling a Scratch volume).
1375 - Keep same dcr when switching device ...
1376 - Implement code that makes the Dir aware that a drive is an
1377 autochanger (so the user doesn't need to use the Autochanger = yes
1379 - Make catalog respect ACL.
1380 - Add recycle count to Media record.
1381 - Add initial write date to Media record.
1382 - Fix store_yesno to be store_bitmask.
1383 --- create_file.c.orig Fri Jul 8 12:13:05 2005
1384 +++ create_file.c Fri Jul 8 12:13:07 2005
1386 attr->ofname, be.strerror());
1389 + } else if(S_ISSOCK(attr->statp.st_mode)) {
1390 + Dmsg1(200, "Skipping socket: %s\n", attr->ofname);
1392 Dmsg1(200, "Restore node: %s\n", attr->ofname);
1393 if (mknod(attr->ofname, attr->statp.st_mode, attr->statp.st_rdev) != 0 && errno != EEXIST) {
1394 - Add true/false to conf same as yes/no
1395 - Reserve blocks other restore jobs when first cannot connect to SD.
1396 - Fix Maximum Changer Wait, Maximum Open Wait, Maximum Rewind Wait to
1397 accept time qualifiers.
1398 - Does ClientRunAfterJob fail the job on a bad return code?