7 Version 1.37 Kern (see below)
8 ========================================================
10 Final items for 1.37 before release:
12 - --without-openssl breaks at least on Solaris.
13 3. Document all the new features (about half done).
14 - VSS. Shall I write "Include Writer 'WMI Writer", "MSDEWriter"
15 Let me explain this: An windows application can (no must)
16 register as VSS writer. This means that the applications opts to
17 be notified if a backup (or restore) occurs. If it then gets
18 this message, it will immediately store a consistent state to
19 disk. Examples for these writers are "MSDE" (Microsoft database
20 engine), "Event Log Writer", "Registry Writer" plus 3rd
21 party-writers. If you have a non-vss aware application (e.g.
22 SQL Anywhere or probably MySQL), a shadow copy is still generated
23 and the open files can be backed up, but there is no guarantee
24 that the file is consistent.
26 At least the Microsoft example makes a significant effort to
27 determine which writers may be involved when a drive or file is
28 to be shadow copied. Of course, every single writer may fail....
29 So they offer a user interface to explicitly include or exclude a
30 writer when creating a VSS shadow copy. I personally would not
31 like to bother the user with this - at least not right now
32 (efforts for exchanging lists between fd and director + efforts
35 But I personally would like to have an information message about
36 the individual writers involved in the backup-process ("vssadmin
37 list writers" produced 4 entries on my xp-client and 7 on my w2k3
38 server, please try this on your machine to understand the system
41 - Multiple drive autochanger support
42 - Support for ANSI/IBM labels.
43 - Seven new options keywords in a FileSet resource:
44 ignorecase, fstype, hfsplussupport, wilddir, wildfile, regexdir,
45 and regexfile thanks to Pruben Guldberg). See below for details.
46 - Restore of all files for a Job or set of jobs even if the file
47 records have been removed from the catalog.
48 - Restore of a directory (non-recursive, i.e. only one level).
49 - Support for TLS (ssl) between all the daemon connections thanks
51 - Any Volume in the Pool named Scratch may be reassigned to any
52 other Pool when a new Volume is needed.
53 - Unicode filename support for Win32 (thanks to Thorsten Engel)
54 - Volume Shadow Copy support for Win32 thus the capability to
55 backup exclusively opened files (thanks to Thorsten Engel).
56 A VSS enabled Win32 FD is available. You must explicitly
57 turn on VSS with "Enable VSS = yes" in your FileSet resource.
58 - SQLite3 support, but it seems to run at 1/2 to 1/4 the speed of
60 - New Job directive "Prefer Mounted Volumes = yes|no" causes the
61 SD to select either an Autochanger or a drive with a valid
62 Volume already mounted in preference. If none is available,
63 it will select the first available drive.
64 - New Run directive in Job resource of DIR. It permits
65 cloning of jobs. To clone a copy of the current job, use
66 Run = "job-name level=%l since=\"%s\""
67 Note, job-name is normally the same name as the job that
68 is running but there is no restriction on what you put. If you
69 want to start the job by hand and use job overrides such as
70 storage=xxx, realize that the job will be started with the
71 default storage values not the overrides. The level=%l guarantees
72 that the chosen level of the job is the same, and the since=...
73 ensures that the job uses *exactly* the same time/date for incremental
74 and differential jobs. The since=... is ignored when level=Full.
75 A cloned job will not start additional clones, so it is not possible
77 - New Options keywords in a FileSet directive:
79 Will do a wild card match against directories (files will not
82 Will do a wild card match against files (directories will not
85 Will do a regular expression match against directories (files
88 Will do a regular expression match against files( directories
90 - IgnoreCase = yes | no
91 Will ignore case in wild card and regular expression matches.
92 This is handy for Windows where filename case is not significant.
94 where string is a filesystem type: ext2, jfs, ntfs, proc,
95 reiserfs, xfs, usbdevfs, sysfs, smbfs, iso9660. For ext3
96 systems, use ext2. You may have multiple fstype directives
97 and thus permit multiple filesystem types. If the type
98 specified on the fstype directive does not match the
99 filesystem for a particular directive, that directory will
100 not be backed up. This directive can be used to prevent
101 backing up non-local filesystems.
102 - HFS Plus Support = yes | no
103 If set, Mac OS X resource forks will be saved and restored.
104 - Label Type = ANSI | IBM | Bacula
105 Implemented in Director Pool resource and in SD Device resource.
106 If it is specified in the SD Device resource, it will take
107 precedence over the value passed from the Director to the SD.
108 - Check Labels = yes | no
109 Implemented in the SD Device resource. If you intend to read
110 ANSI or IBM labels, this *must* be set. Even if the volume
111 is not ANSI labeled, you can set this to yes, and Bacula will
112 check the label type.
113 - Scripts Directory = <directory> name. Defines the directory from
114 which Bacula scripts will be called for events. In fact, Bacula
115 appends this name to the standard Python list of search directories,
116 so the script could also be in any of the Python system directories.
117 - In FileSet, you can exclude backing up of hardlinks (if you have
118 a lot, it can be very expensive), by using:
120 in the Options section. Patch supplied by David R Bosso. Thanks.
121 - MaximumPartSize = bytes (SD, Device resource)
122 Defines the maximum part size.
123 - Requires Mount = Yes/No (SD, Device resource)
124 Defines if the device require to be mounted to be read, and if it
125 must be written in a special way. If it set, the following directives
126 must be defined in the same Device resource:
127 + Mount Point = directory
128 Directory where the device must be mounted.
129 + Mount Command = name-string
130 Command that must be executed to mount the device. Before the command
131 is executed, %a is replaced with the Archive Device, and %m with the
133 + Unmount Command = name-string
134 Command that must be executed to unmount the device. Before the
135 command is executed, %a is replaced with the Archive Device, and
136 %m with the Mount Point.
137 + Write Part Command = name-string
138 Command that must be executed to write a part to the device. Before
139 the command is executed, %a is replaced with the Archive Device, %m
140 with the Mount Point, %n with the current part number (0-based),
141 and %v with the current part filename.
142 + Free Space Command = name-string
143 Command that must be executed to check how much free space is left
144 on the device. Before the command is executed, %a is replaced with
145 the Archive Device, %m with the Mount Point, %n with the current part
146 number (0-based), and %v with the current part filename.
147 - Write Part After Job = Yes/No (DIR, Job Resource, and Schedule Resource)
148 If this directive is set to yes (default no), a new part file will be
149 created after the job is finished.
150 - A pile of new Directives to support TLS. Please see the TLS chapter
153 - "python restart" restarts the Python interpreter. Rather brutal, make
154 sure no Python scripts are running. This permits you to change
155 a Python script and ge
156 - With Python 2.3, there are a few compiler warnings.
157 - You must add --with-openssl to the configure command line if
158 you want TLS communications encryption support.
159 7. Write a bacula-web document
160 9. Run the regression scripts on Solaris and FreeBSD
161 - Figure out how to package gui, and rescue programs.
165 - Document cleaning up the spool files:
166 db, pid, state, bsr, mail, conmsg, spool
167 - Document the multiple-drive-changer.txt script.
168 - Pruning with Admin job.
172 - Fix bpipe.c so that it does not modify results pointer.
173 ***FIXME*** calling sequence should be changed.
175 #3 Migration (Move, Copy, Archive Jobs)
176 #7 Single Job Writing to Multiple Storage Devices
177 - Reserve blocks other restore jobs when first cannot connect
179 - Add true/false to conf same as yes/no
180 - For Windows disaster recovery see http://unattended.sf.net/
181 - regardless of the retention period, Bacula will not prune the
182 last Full, Diff, or Inc File data until a month after the
183 retention period for the last Full backup that was done.
184 - update volume=xxx --- add status=Full
185 - Remove old spool files on startup.
186 - Exclude SD spool/working directory.
187 - Refuse to prune last valid Full backup. Same goes for Catalog.
189 - Make a callback when Rerun failed levels is called.
190 - Give Python program access to Scheduled jobs.
191 - Add setting Volume State via Python.
192 - Python script to save with Python, not save, save with Bacula.
193 - Python script to do backup.
195 - Change the Priority, Client, Storage, JobStatus (error)
196 at the start of a job.
197 - Why is SpoolDirectory = /home/bacula/spool; not reported
198 as an error when writing a DVD?
199 - Make bootstrap file handle multiple MediaTypes (SD)
200 - Remove all old Device resource code in Dir and code to pass it
201 back in SD -- better, rework it to pass back device statistics.
202 - Check locking of resources -- be sure to lock devices where previously
203 resources were locked.
204 - The last part is left in the spool dir.
207 - In restore don't compare byte count on a raw device -- directory
208 entry does not contain bytes.
209 - To mark files as deleted, run essentially a Verify to disk, and
210 when a file is found missing (MarkId != JobId), then create
211 a new File record with FileIndex == -1. This could be done
212 by the FD at the same time as the backup.
216 MA = (last_MA * 3 + rate) / 4
217 rate = (bytes - last_bytes) / (runtime - last_runtime)
218 - Max Vols limit in Pool off by one?
219 - Implement Files/Bytes,... stats for restore job.
220 - Implement Total Bytes Written, ... for restore job.
221 - Despool attributes simultaneously with data in a separate
222 thread, rejoined at end of data spooling.
223 - 7. Implement new Console commands to allow offlining/reserving drives,
224 and possibly manipulating the autochanger (much asked for).
225 - Add start/end date editing in messages (%t %T, %e?) ...
226 - Add ClientDefs similar to JobDefs.
227 - Print more info when bextract -p accepts a bad block.
228 - Fix FD JobType to be set before RunBeforeJob in FD.
229 - Look at adding full Volume and Pool information to a Volume
230 label so that bscan can get *all* the info.
231 - If the user puts "Purge Oldest Volume = yes" or "Recycle Oldest Volume = yes"
232 and there is only one volume in the pool, refuse to do it -- otherwise
233 he fills the Volume, then immediately starts reusing it.
234 - Implement copies and stripes.
235 - Add history file to console.
236 - Each file on tape creates a JobMedia record. Peter has 4 million
237 files spread over 10000 tape files and four tapes. A restore takes
238 16 hours to build the restore list.
239 - Add and option to check if the file size changed during backup.
240 - Make sure SD deletes spool files on error exit.
241 - Delete old spool files when SD starts.
242 - When labeling tapes, if you enter 000026, Bacula uses
243 the tape index rather than the Volume name 000026.
244 - Add offline tape command to Bacula console.
246 Enter MediaId or Volume name: 32
247 Enter new Volume name: DLT-20Dec04
248 Automatically selected Pool: Default
249 Connecting to Storage daemon DLTDrive at 192.168.68.104:9103 ...
250 Sending relabel command from "DLT-28Jun03" to "DLT-20Dec04" ...
251 block.c:552 Write error at 0:0 on device /dev/nst0. ERR=Bad file descriptor.
252 Error writing final EOF to tape. This tape may not be readable.
253 dev.c:1207 ioctl MTWEOF error on /dev/nst0. ERR=Permission denied.
254 askdir.c:219 NULL Volume name. This shouldn't happen!!!
255 3912 Failed to label Volume: ERR=dev.c:1207 ioctl MTWEOF error on /dev/nst0. ERR=Permission denied.
256 Label command failed for Volume DLT-20Dec04.
257 Do not forget to mount the drive!!!
258 - Bug: if a job is manually scheduled to run later, it does not appear
259 in any status report and cannot be cancelled.
261 ==== Keeping track of deleted files ====
262 My "trick" for keeping track of deletions is the following.
263 Assuming the user turns on this option, after all the files
264 have been backed up, but before the job has terminated, the
265 FD will make a pass through all the files and send their
266 names to the DIR (*exactly* the same as what a Verify job
267 currently does). This will probably be done at the same
268 time the files are being sent to the SD avoiding a second
269 pass. The DIR will then compare that to what is stored in
270 the catalog. Any files in the catalog but not in what the
271 FD sent will receive a catalog File entry that indicates
272 that at that point in time the file was deleted.
274 During a restore, any file initially picked up by some
275 backup (Full, ...) then subsequently having a File entry
276 marked "delete" will be removed from the tree, so will not
277 be restored. If a file with the same name is later OK it
278 will be inserted in the tree -- this already happens. All
279 will be consistent except for possible changes during the
282 Since I'm on the subject, some of you may be wondering what
283 the utility of the in memory tree is if you are going to
284 restore everything (at least it comes up from time to time
285 on the list). Well, it is still *very* useful because it
286 allows only the last item found for a particular filename
287 (full path) to be entered into the tree, and thus if a file
288 is backed up 10 times, only the last copy will be restored.
289 I recently (last Friday) restored a complete directory, and
290 the Full and all the Differential and Incremental backups
291 spanned 3 Volumes. The first Volume was not even mounted
292 because all the files had been updated and hence backed up
293 since the Full backup was made. In this case, the tree
294 saved me a *lot* of time.
296 Make sure this information is stored on the tape too so
297 that it can be restored directly from the tape.
299 Comments from Martin Simmons (I think they are all covered):
300 Ok, that should cover the basics. There are few issues though:
302 - Restore will depend on the catalog. I think it is better to include the
303 extra data in the backup as well, so it can be seen by bscan and bextract.
305 - I'm not sure if it will preserve multiple hard links to the same inode. Or
306 maybe adding or removing links will cause the data to be dumped again?
308 - I'm not sure if it will handle renamed directories. Possibly it will work
309 by dumping the whole tree under a renamed directory?
311 - It remains to be seen how the backup performance of the DIR's will be
312 affected when comparing the catalog for a large filesystem.
317 - Add Pool/Storage override regression test.
318 - Add delete JobId to regression.
319 - Add a regression test for dbcheck.
320 - New test to add bscan to four-concurrent-jobs regression,
321 i.e. after the four-concurrent jobs zap the
322 database as is done in the bscan-test, then use bscan to
323 restore the database, do a restore and compare with the
325 - Add restore of specific JobId to regression (item 3
326 on the restore prompt)
327 - Add IPv6 to regression
328 - Add database test to regression. Test each function like delete,
331 - AntiVir can slow down backups on Win32 systems.
332 - Win32 systems with FAT32 can be much slower than NTFS for
333 more than 1000 files per directory.
337 - A HOLD command to stop all jobs from starting.
338 - A PAUSE command to pause all running jobs ==> release the
340 - Media Type = LTO,LTO-2,LTO-3
341 Media Type Read = LTO,LTO2,LTO3
342 Media Type Write = LTO2, LTO3
344 === From Carsten Menke <bootsy52@gmx.net>
346 Following is a list of what I think in the situations where I'm faced with,
347 could be a usefull enhancement to bacula, which I'm certain other users will
348 benefit from as well.
350 1. NextJob/NextJobs Directive within a Job Resource in the form of
351 NextJobs = job1,job2.
354 I currently solved the problem with running multiple jobs each after each
355 by setting the Max Wait Time for a job to 8 hours, and give
356 the jobs different Priorities. However, there scenarios where
357 1 Job is directly depending on another job, so if the former job fails,
358 the job after it needn't to be run
359 while maybe other jobs should run despite of that
362 A Backup Job and a Verify job, if the backup job fails there is no need to run
363 the verify job, as the backup job already failed. However, one may like
364 to backup the Catalog to disk despite of that the main backup job failed.
367 I see that this is related to the Event Handlers which are on the ToDo
368 list, also it is maybe a good idea to check for the return value and
369 execute different actions based on the return value
372 3. offline capability to bconsole
375 Currently I use a script which I execute within the last Job via the
376 RunAfterJob Directive, to release and eject the tape.
377 So I have to call bconsole "release=Storage-Name" and afterwards
378 mt -f /dev/nst0 eject to get the tape out.
380 If I have multiple Storage Devices, than these may not be /dev/nst0 and
381 I have to modify the script or call it with parameters etc.
382 This would actually not be needed, as everything is already defined
383 in bacula-sd.conf and if I can invoke bconsole with the
384 storage name via $1 in the script than I'm done and information is
387 4. %s for Storage Name added to the chars being substituted in "RunAfterJob"
391 For the reason mentioned in 3. to have the ability to call a
392 script with /scripts/foobar %s and in the script use $1
393 to pass the Storage Name to bconsole
395 5. Setting Volume State within a Job Resource
398 Instead of using "Maximum Volume Jobs" in the Pool Resource,
399 I would have the possibilty to define
400 in a Job Resource that after this certain job is run, the Volume State
401 should be set to "Volume State = Used", this give more flexibility (IMHO).
403 6. Localization of Bacula Messages
406 Unfortunatley many,many people I work with don't speak english very well.
407 So if at least the Reporting messages would be localized then they
408 would understand that they have to change the tape,etc. etc.
410 I volunteer to do the german translations, and if I can convince my wife also
411 french and Morre (western african language).
413 7. OK, this is evil, probably bound to security risks and maybe not possible
414 due to the design of bacula.
416 Implementation of Backtics ( `command` ) for shell comand execution to
417 the "Label Format" Directive.
421 Currently I have defined BACULA_DAY_OF_WEEK="day1|day2..." resulting in
422 Label Format = "HolyBackup-${BACULA_DAY_OF_WEEK[${WeekDay}]}". If I could
423 use backticks than I could use "Label Format = HolyBackup-`date +%A` to have
424 the localized name for the day of the week appended to the
425 format string. Then I have the tape labeled automatically with weekday
426 name in the correct language.
428 - Yes, that is surely the case. I probably should turn those into Warning
429 errors. In addition, you just made me think that it might not be bad to
430 add an option to check the file size after backing up the file and
431 report if it changes. This would be done as an option because it would
434 Kern, good idea. If you do do that, mention in the output: file
435 shrunk, or file expanded, just to make it obvious to the user
436 (without having to the refer to file size), just how the file size
439 Would this option be for all file, or just one file? Or a fileset?
440 - Make output from status use html table tags for nicely
441 presenting in a browser.
442 - Can one write tapes faster with 8192 byte block sizes?
443 - Document security problems with the same password for everyone in
444 rpm and Win32 releases.
445 - Browse generations of files.
446 - I've seen an error when my catalog's File table fills up. I
447 then have to recreate the File table with a larger maximum row
448 size. Relevant information is at
449 http://dev.mysql.com/doc/mysql/en/Full_table.html ; I think the
450 "Installing and Configuring MySQL" chapter should talk a bit
451 about this potential problem, and recommend a solution.
452 - For Solaris must use POSIX awk.
453 - Want speed of writing to tape while despooling.
454 - Supported autochanger:
462 Wangtek 6525ES (SCSI-1 QIC drive, 525MB), under Linux 2.4.something,
463 bacula 1.36.0/1 works with blocksize 16k INSIDE bacula-sd.conf.
464 - Add regex from http://www.pcre.org to Bacula for Win32.
465 - Use only shell tools no make in CDROM package.
466 - Include within include does it work?
467 - Implement a Pool of type Cleaning?
468 - Implement VolReadTime and VolWriteTime in SD
469 - Modify Backing up Your Database to include a bootstrap file.
470 - Think about making certain database errors fatal.
471 - Look at correcting the time jump in the scheduler for daylight
472 savings time changes.
473 - Add a "real" timer to network connections.
474 - Promote to Full = Time period
475 - Check dates entered by user for correctness (month/day/... ranges)
476 - Compress restore Volume listing by date and first file.
477 - Look at patches/bacula_db.b2z postgresql that loops during restore.
479 - Perhaps add read/write programs and/or plugins to FileSets.
480 - How to handle backing up portables ...
481 - Add some sort of guaranteed Interval for upgrading jobs.
482 - Can we write the state file after every job terminates? On Win32
483 the system crashes and the state file is not updated.
486 Documentation to do: (any release a little bit at a time)
487 - Doc to do unmount before removing magazine.
488 - Alternative to static linking "ldd prog" save all binaries listed,
489 restore them and point LD_LIBRARY_PATH to them.
490 - Document add "</dev/null >/dev/null 2>&1" to the bacula-fd command line
491 - Document query file format.
492 - Add more documentation for bsr files.
493 - Document problems with Verify and pruning.
494 - Document how to use multiple databases.
495 - VXA drives have a "cleaning required"
496 indicator, but Exabyte recommends preventive cleaning after every 75
499 In this context, it should be noted that Exabyte has a command-line
500 vxatool utility available for free download. (The current version is
501 vxatool-3.72.) It can get diagnostic info, read, write and erase tapes,
502 test the drive, unload tapes, change drive settings, flash new firmware,
504 Of particular interest in this context is that vxatool <device> -i will
505 report, among other details, the time since last cleaning in tape motion
506 minutes. This information can be retrieved (and settings changed, for
507 that matter) through the generic-SCSI device even when Bacula has the
508 regular tape device locked. (Needless to say, I don't recommend
509 changing tape settings while a job is running.)
510 - Lookup HP cleaning recommendations.
511 - Lookup HP tape replacement recommendations (see trouble shooting autochanger)
512 - Document doing table repair
515 ===================================
516 - Add macro expansions in JobDefs.
517 Run Before Job = "SomeFile %{Level} %{Client}"
518 Write Bootstrap="/some/dir/%{JobName}_%{Client}.bsr"
519 - Use non-blocking network I/O but if no data is available, use
521 - Use gather write() for network I/O.
522 - Autorestart on crash.
523 - Add bandwidth limiting.
524 - Add acks every once and a while from the SD to keep
525 the line from timing out.
526 - When an error in input occurs and conio beeps, you can back
527 up through the prompt.
528 - Detect fixed tape block mode during positioning by looking at
529 block numbers in btape "test". Possibly adjust in Bacula.
530 - Fix list volumes to output volume retention in some other
531 units, perhaps via a directive.
532 - If opening a tape in read/write mode fails attempt to open
533 it in read-only mode, and mark the tape for read only.
534 - Allow Simultaneous Priorities = yes => run up to Max concurrent jobs even
535 with multiple priorities.
536 - If you use restore replace=never, the directory attributes for
537 non-existent directories will not be restored properly.
539 - see lzma401.zip in others directory for new compression
541 - Minimal autochanger handling in Bacula and in btape.
542 - Look into how tar does not save sockets and the possiblity of
543 not saving them in Bacula (Martin Simmons reported this).
544 - Fix restore jobs so that multiple jobs can run if they
545 are not using the same tape(s).
546 - Allow the user to select JobType for manual pruning/purging.
547 - bscan does not put first of two volumes back with all info in
549 - Implement the FreeBSD nodump flag in chflags.
550 - Figure out how to make named console messages go only to that
551 console and to the non-restricted console (new console class?).
552 - Make restricted console prompt for password if *ask* is set or
553 perhaps if password is undefined.
554 - Implement "from ISO-date/time every x hours/days/weeks/months" in
557 ==== from Marc Schoechlin
558 - the help-command should be more verbose
559 (it should explain the paramters of the different
561 -> its time-comsuming to consult the manual anytime
562 you need a special parameter
563 -> maybe its more easy to maintain this, if the
564 descriptions of that commands are outsourced to
566 - the cd-command should allow complete paths
567 i.e. cd /foo/bar/foo/bar
568 -> if a customer mails me the path to a certain file,
569 its faster to enter the specified directory
570 - if the password is not configured in bconsole.conf
571 you should be asked for it.
572 -> sometimes you like to do restore on a customer-machine
573 which shouldnt know the password for bacula.
574 -> adding the password to the file favours admins
575 to forget to remove the password after usage
577 the protection of that file is less important
578 - long-listed-output of commands should be scrollable
579 like the unix more/less-command does
580 -> if someone runs 200 and more machines, the lists could
581 be a little long and complex
582 - command-output should be shown column by column
583 to reduce scrolling and to increase clarity
585 - lsmark should list the selected files with full
587 - wildcards for selecting and file and directories would be nice
588 - any actions should be interuptable with STRG+C
589 - command-expansion would be pretty cool
591 - When the replace Never option is set, new directory permissions
592 are not restored. See bug 213. To fix this requires creating a
593 list of newly restored directories so that those directory
594 permissions *can* be restored.
595 - Compaction of Disk space by "migrating" Volumes that have pruned
596 Jobs (what criteria? size, #jobs, time).
597 - Add prune all command
598 - Document fact that purge can destroy a part of a restore by purging
599 one volume while others remain valid -- perhaps mark Jobs.
600 - Add multiple-media-types.txt
601 - look at mxt-changer.html
602 - Make ? do a help command (no return needed).
603 - Implement restore directory.
604 - Document streams and how to implement them.
605 - Try not to re-backup a file if a new hard link is added.
606 - Add feature to backup hard links only, but not the data.
607 - Fix stream handling to be simpler.
608 - Add Priority and Bootstrap to Run a Job.
609 - Eliminate Restore "Run Restore Job" prompt by allowing new "run command
611 - Remove View FileSet button from Run a Job dialog.
612 - Handle prompt for restore job at end of Restore command.
613 - Add display of total selected files to Restore window.
614 - Add tree pane to left of window.
615 - Add progress meter.
616 - Max wait time or max run time causes seg fault -- see runtime-bug.txt
617 - Document writing to a CD/DVD with Bacula.
618 - Add a "base" package to the window installer for pthreadsVCE.dll
619 which is needed by all packages.
620 - Add message to user to check for fixed block size when the forward
621 space test fails in btape.
622 - When unmarking a directory check if all files below are unmarked and
623 then remove the + flag -- in the restore tree.
624 - Possibly implement: Action = Unmount Device="TapeDrive1" in Admin jobs.
625 - Setup lrrd graphs: (http://www.linpro.no/projects/lrrd/) Mike Acar.
626 - Revisit the question of multiple Volumes (disk) on a single device.
627 - Add a block copy option to bcopy.
628 - Investigate adding Mac Resource Forks.
629 - Finish work on Gnome restore GUI.
630 - Fix "llist jobid=xx" where no fileset or client exists.
631 - For each job type (Admin, Restore, ...) require only the really necessary
632 fields.- Pass Director resource name as an option to the Console.
633 - Add a "batch" mode to the Console (no unsolicited queries, ...).
634 - Add a .list all files in the restore tree (probably also a list all files)
635 Do both a long and short form.
636 - Allow browsing the catalog to see all versions of a file (with
637 stat data on each file).
638 - Restore attributes of directory if replace=never set but directory
640 - Use SHA1 on authentication if possible.
641 - See comtest-xxx.zip for Windows code to talk to USB.
642 - Add John's appended files:
643 Appended = { /files/server/logs/http/*log }
644 and such files would be treated as follows.On a FULL backup, they would
645 be backed up like any other file.On an INCREMENTAL backup, where a
646 previous INCREMENTAL or FULL was already in thecatalogue and the length
647 of the file wasgreater than the length of the last backup, only thedata
648 added since the last backup will be dumped.On an INCREMENTAL backup, if
649 the length of the file is less than thelength of the file with the same
650 name last backed up, the completefile is dumped.On Windows systems, with
651 creation date of files, we can be evensmarter about this and not count
652 entirely upon the length.On a restore, the full and all incrementals
653 since it will beapplied in sequence to restore the file.
654 - Check new HAVE_WIN32 open bits.
655 - Check if the tape has moved before writing.
656 - Handling removable disks -- see below:
657 - Keep track of tape use time, and report when cleaning is necessary.
658 - Add FromClient and ToClient keywords on restore command (or
659 BackupClient RestoreClient).
660 - Implement a JobSet, which groups any number of jobs. If the
661 JobSet is started, all the jobs are started together.
662 Allow Pool, Level, and Schedule overrides.
663 - Enhance cancel to timeout BSOCK packets after a specific delay.
664 - Do scheduling by UTC using gmtime_r() in run_conf, scheduler, and
665 ua_status.!!! Thanks to Alan Brown for this tip.
666 - Look at updating Volume Jobs so that Max Volume Jobs = 1 will work
667 correctly for multiple simultaneous jobs.
668 - Correct code so that FileSet MD5 is calculated for < and | filename
670 - Implement the Media record flag that indicates that the Volume does disk
672 - Implement VolAddr, which is used when Volume is addressed like a disk,
673 and form it from VolFile and VolBlock.
674 - Make multiple restore jobs for multiple media types specifying
675 the proper storage type.
676 - Fix fast block rejection (stored/read_record.c:118). It passes a null
677 pointer (rec) to try_repositioning().
678 - Look at extracting Win data from BackupRead.
679 - Implement RestoreJobRetention? Maybe better "JobRetention" in a Job,
680 which would take precidence over the Catalog "JobRetention".
681 - Implement Label Format in Add and Label console commands.
682 - Possibly up network buffers to 65K. Put on variable.
683 - Put email tape request delays on one or more variables. User wants
684 to cancel the job after a certain time interval. Maximum Mount Wait?
685 - Job, Client, Device, Pool, or Volume?
686 Is it possible to make this a directive which is *optional* in multiple
687 resources, like Level? If so, I think I'd make it an optional directive
688 in Job, Client, and Pool, with precedence such that Job overrides Client
689 which in turn overrides Pool.
691 - New Storage specifications:
692 - Want to write to multiple storage devices simultaneously
693 - Want to write to multiple storage devices sequentially (in one job)
694 - Want to read/write simultaneously
695 - Key is MediaType -- it must match
697 Passed to SD as a sort of BSR record called Storage Specification
701 MediaType -> Next MediaType
703 Device -> Next Device
705 Allow multiple Storage specifications
713 Allow Multiple Pool specifications (note, Pool currently
715 Allow Multiple MediaType specifications in Dir conf
716 Allow Multiple Device specifications in Dir conf
717 Perhaps keep this in a single SSR
718 Tie a Volume to a specific device by using a MediaType that
719 is contained in only one device.
720 In SD allow Device to have Multiple MediaTypes
722 - Ideas from Jerry Scharf:
723 First let's point out some big pluses that bacula has for this
725 more importantly it's active. Thank you so much for that
726 even more important, it's not flaky
727 it has an open access catalog, opening many possibilities
728 it's pushing toward heterogeneous systems capability
730 Macintosh file client
731 macs are an interesting niche, but I fear a server is a rathole
732 working bare iron recovery for windows
733 the option for inc/diff backups not reset on fileset revision
734 a) use both change and inode update time against base time
735 b) do the full catalog check (expensive but accurate)
736 sizing guide (how much system is needed to back up N systems/files)
737 consultants on using bacula in building a disaster recovery system
739 or how to get at fancy things that one could do with bacula
740 logwatch code for bacula logs (or similar)
741 linux distro inclusion of bacula (brings good and bad, but necessary)
742 win2k/XP server capability (icky but you asked)
743 support for Oracle database ??
745 - Look at adding SQL server and Exchange support for Windows.
746 - Make dev->file and dev->block_num signed integers so that -1 can
747 be an invalid value which happens with BSR.
748 - Create VolAddr for disk files in place of VolFile and VolBlock. This
749 is needed to properly specify ranges.
750 - Add progress of files/bytes to SD and FD.
751 - Print warning message if FileId > 4 billion
752 - do a "messages" before the first prompt in Console
753 - Client does not show busy during Estimate command.
754 - Implement Console mtx commands.
755 - Implement a Mount Command and an Unmount Command where
756 the users could specify a system command to be performed
757 to do the mount, after which Bacula could attempt to
758 read the device. This is for Removeable media such as a CDROM.
759 - Most likely, this mount command would be invoked explicitly
760 by the user using the current Console "mount" and "unmount"
761 commands -- the Storage Daemon would do the right thing
762 depending on the exact nature of the device.
763 - As with tape drives, when Bacula wanted a new removable
764 disk mounted, it would unmount the old one, and send a message
765 to the user, who would then use "mount" as described above
766 once he had actually inserted the disk.
767 - Implement dump/print label to UA
768 - Spool to disk only when the tape is full, then when a tape is hung move
770 - bextract is sending everything to the log file ****FIXME****
771 - Allow multiple Storage specifications (or multiple names on
772 a single Storage specification) in the Job record. Thus a job
773 can be backed up to a number of storage devices.
774 - Implement some way for the File daemon to contact the Director
775 to start a job or pass its DHCP obtained IP number.
776 - Implement a query tape prompt/replace feature for a console
777 - Copy console @ code to gnome2-console
778 - Make AES the only encryption algorithm see
779 http://csrc.nist.gov/CryptoToolkit/aes/). It's
780 an officially adopted standard, has survived peer
781 review, and provides keys up to 256 bits.
782 - Take a careful look at SetACL http://setacl.sourceforge.net
783 - Make tree walk routines like cd, ls, ... more user friendly
784 by handling spaces better.
785 - Make sure that Bacula rechecks the tape after the 20 min wait.
786 - Set IO_NOWAIT on Bacula TCP/IP packets.
787 - Try doing a raw partition backup and restore by mounting a
790 Yes, it would allow to highly automatic the request for new tapes. If a
791 tape is empty, bacula reads the barcodes (native or simulated), and if
792 an unused tape is found, it runs the label command with all the
793 necessary parameters.
795 By the way can bacula automatically "move" an empty/purged volume say
796 in the "short" pool to the "long" pool if this pool runs out of volume
798 - What to do about "list files job=xxx".
799 - Get and test MySQL 4.0
800 - Look at how fuser works and /proc/PID/fd that is how Nic found the
801 file descriptor leak in Bacula.
802 - Implement WrapCounters in Counters.
803 - Add heartbeat from FD to SD if hb interval expires.
804 - Can we dynamically change FileSets?
805 - If pool specified to label command and Label Format is specified,
806 automatically generate the Volume name.
807 - Why can't SQL do the filename sort for restore?
808 - Add ExhautiveRestoreSearch
809 - Look at the possibility of loading only the necessary
810 data into the restore tree (i.e. do it one directory at a
811 time as the user walks through the tree).
812 - Possibly use the hash code if the user selects all for a restore command.
813 - Fix "restore all" to bypass building the tree.
814 - Prohibit backing up archive device (findlib/find_one.c:128)
815 - Implement Release Device in the Job resource to unmount a drive.
816 - Implement Acquire Device in the Job resource to mount a drive,
817 be sure this works with admin jobs so that the user can get
818 prompted to insert the correct tape. Possibly some way to say to
819 run the job but don't save the files.
820 - Make things like list where a file is saved case independent for
823 - Use autochanger to handle multiple devices.
824 - On Windows with very long path names, it may be impossible to create
825 a file (and thus restore it) because the total length is too long.
826 We must cd into the directory then create the file without the
828 - Implement a Recycle command
829 - Test a second language e.g. french.
830 - Start working on Base jobs.
831 - Implement UnsavedFiles DB record.
832 - From Phil Stracchino:
833 It would probably be a per-client option, and would be called
834 something like, say, "Automatically purge obsoleted jobs". What it
835 would do is, when you successfully complete a Differential backup of a
836 client, it would automatically purge all Incremental backups for that
837 client that are rendered redundant by that Differential. Likewise,
838 when a Full backup on a client completed, it would automatically purge
839 all Differential and Incremental jobs obsoleted by that Full backup.
840 This would let people minimize the number of tapes they're keeping on
841 hand without having to master the art of retention times.
842 - When doing a Backup send all attributes back to the Director, who
843 would then figure out what files have been deleted.
844 - Currently in mount.c:236 the SD simply creates a Volume. It should have
845 explicit permission to do so. It should also mark the tape in error
846 if there is an error.
847 - Cancel waiting for Client connect in SD if FD goes away.
849 - Implement timeout in response() when it should come quickly.
850 - Implement a Slot priority (loaded/not loaded).
851 - Implement "vacation" Incremental only saves.
852 - Implement create "FileSet"?
853 - Add prefixlinks to where or not where absolute links to FD.
854 - Issue message to mount a new tape before the rewind.
855 - Simplified client job initiation for portables.
856 - If SD cannot open a drive, make it periodically retry.
857 - Add more of the config info to the tape label.
859 - If tape is marked read-only, then try opening it read-only rather than
860 failing, and remember that it cannot be written.
861 - Refine SD waiting output:
862 Device is being positioned
863 > Device is being positioned for append
864 > Device is being positioned to file x
866 - Figure out some way to estimate output size and to avoid splitting
867 a backup across two Volumes -- this could be useful for writing CDROMs
868 where you really prefer not to have it split -- not serious.
869 - Have SD compute MD5 or SHA1 and compare to what FD computes.
870 - Make VolumeToCatalog calculate an MD5 or SHA1 from the
871 actual data on the Volume and compare it.
872 - Implement Bacula plugins -- design API
873 - Make bcopy read through bad tape records.
874 - Program files (i.e. execute a program to read/write files).
875 Pass read date of last backup, size of file last time.
876 - Add Signature type to File DB record.
877 - CD into subdirectory when open()ing files for backup to
878 speed up things. Test with testfind().
879 - Priority job to go to top of list.
880 - Why are save/restore of device different sizes (sparse?) Yup! Fix it.
881 - Implement some way for the Console to dynamically create a job.
882 - Solaris -I on tar for include list
883 - Need a verbose mode in restore, perhaps to bsr.
884 - bscan without -v is too quiet -- perhaps show jobs.
885 - Add code to reject whole blocks if not wanted on restore.
886 - Check if we can increase Bacula FD priorty in Win2000
887 - Make sure the MaxVolFiles is fully implemented in SD
888 - Check if both CatalogFiles and UseCatalog are set to SD.
889 - Possibly add email to Watchdog if drive is unmounted too
890 long and a job is waiting on the drive.
891 - After unmount, if restore job started, ask to mount.
892 - Add UA rc and history files.
893 - put termcap (used by console) in ./configure and
894 allow -with-termcap-dir.
895 - Fix Autoprune for Volumes to respect need for full save.
896 - Compare tape to Client files (attributes, or attributes and data)
897 - Make all database Ids 64 bit.
898 - Allow console commands to detach or run in background.
899 - Fix status delay on storage daemon during rewind.
900 - Add SD message variables to control operator wait time
901 - Maximum Operator Wait
902 - Minimum Message Interval
903 - Maximum Message Interval
904 - Send Operator message when cannot read tape label.
905 - Verify level=Volume (scan only), level=Data (compare of data to file).
906 Verify level=Catalog, level=InitCatalog
908 - Add keyword search to show command in Console.
909 - Events : tape has more than xxx bytes.
910 - Complete code in Bacula Resources -- this will permit
911 reading a new config file at any time.
912 - Handle ctl-c in Console
913 - Implement script driven addition of File daemon to config files.
914 - Think about how to make Bacula work better with File (non-tape) archives.
915 - Write Unix emulator for Windows.
916 - Put memory utilization in Status output of each daemon
917 if full status requested or if some level of debug on.
918 - Make database type selectable by .conf files i.e. at runtime
919 - Set flag for uname -a. Add to Volume label.
920 - Restore files modified after date
921 - SET LD_RUN_PATH=$HOME/mysql/lib/mysql
922 - Remove duplicate fields from jcr (e.g. jcr.level and jcr.jr.Level, ...).
923 - Timout a job or terminate if link goes down, or reopen link and query.
924 - Concept of precious tapes (cannot be reused).
925 - Make bcopy copy with a single tape drive.
926 - Permit changing ownership during restore.
929 > My suggestion: Add a feature on the systray menu-icon menu to request
930 > an immediate backup now. This would be useful for laptop users who may
931 > not be on the network when the regular scheduled backup is run.
933 > My wife's suggestion: Add a setting to the win32 client to allow it to
934 > shut down the machine after backup is complete (after, of course,
935 > displaying a "System will shut down in one minute, click here to cancel"
936 > warning dialog). This would be useful for sites that want user
937 > woorkstations to be shut down overnight to save power.
940 - Autolabel should be specified by DIR instead of SD.
943 - AutoScan (check checksum of tape)
944 - Format command = "format /dev/nst0"
948 - Seek resolution (usually corresponds to buffer size)
949 - EODErrorCode=ENOSPC or code
950 - Partial Read error code
951 - Partial write error code
952 - Nonformatted read error
953 - Nonformatted write error
954 - WriteProtected error
958 - IgnoreCloseErrors=yes
968 - FD sends unsaved file list to Director at end of job (see
970 - File daemon should build list of files skipped, and then
971 at end of save retry and report any errors.
972 - Write a Storage daemon that uses pipes and
973 standard Unix programs to write to the tape.
975 - Need something that monitors the JCR queue and
976 times out jobs by asking the deamons where they are.
977 - Enhance Jmsg code to permit buffering and saving to disk.
978 - device driver = "xxxx" for drives.
980 - Ensure that /dev/null works
981 - Need report class for messages. Perhaps
982 report resource where report=group of messages
983 - enhance scan_attrib and rename scan_jobtype, and
984 fill in code for "since" option
985 - Director needs a time after which the report status is sent
986 anyway -- or better yet, a retry time for the job.
987 - Don't reschedule a job if previous incarnation is still running.
988 - Some way to automatically backup everything is needed????
989 - Need a structure for pending actions:
991 - termination status (part of buffered msgs?)
993 Read, Write, Clean, Delete
994 - Login to Bacula; Bacula users with different permissions:
995 owner, group, user, quotas
996 - Store info on each file system type (probably in the job header on tape.
997 This could be the output of df; or perhaps some sort of /etc/mtab record.
999 ========= ideas ===============
1000 From: "Jerry K. Schieffer" <jerry@skylinetechnology.com>
1001 To: <kern@sibbald.com>
1002 Subject: RE: [Bacula-users] future large programming jobs
1003 Date: Thu, 26 Feb 2004 11:34:54 -0600
1005 I noticed the subject thread and thought I would offer the following
1006 merely as sources of ideas, i.e. something to think about, not even as
1007 strong as a request. In my former life (before retiring) I often
1008 dealt with backups and storage management issues/products as a
1009 developer and as a consultant. I am currently migrating my personal
1010 network from amanda to bacula specifically because of the ability to
1011 cross media boundaries during storing backups.
1012 Are you familiar with the commercial product called ADSM (I think IBM
1013 now sells it under the Tivoli label)? It has a couple of interesting
1014 ideas that may apply to the following topics.
1016 1. Migration: Consider that when you need to restore a system, there
1017 may be pressure to hurry. If all the information for a single client
1018 can eventually end up on the same media (and in chronological order),
1019 the restore is facillitated by not having to search past information
1020 from other clients. ADSM has the concept of "client affinity" that
1021 may be associated with it's storage pools. It seems to me that this
1022 concept (as an optional feature) might fit in your architecture for
1025 ADSM also has the concept of defining one or more storage pools as
1026 "copy pools" (almost mirrors, but only in the sense of contents).
1027 These pools provide the ability to have duplicte data stored both
1028 onsite and offsite. The copy process can be scheduled to be handled
1029 by their storage manager during periods when there is no backup
1030 activity. Again, the migration process might be a place to consider
1031 implementing something like this.
1034 > It strikes me that it would be very nice to be able to do things
1036 > have the Job(s) backing up the machines run, and once they have all
1037 > completed, start a migration job to copy the data from disks Volumes
1039 > a tape library and then to offsite storage. Maybe this can already
1041 > done with some careful scheduling and Job prioritzation; the events
1042 > mechanism described below would probably make it very easy.
1044 This is the goal. In the first step (before events), you simply
1046 the Migration to tape later.
1048 2. Base jobs: In ADSM, each copy of each stored file is tracked in
1049 the database. Once a file (unique by path and metadata such as dates,
1050 size, ownership, etc.) is in a copy pool, no more copies are made. In
1051 other words, when you start ADSM, it begins like your concept of a
1052 base job. After that it is in the "incremental" mode. You can
1053 configure the number of "generations" of files to be retained, plus a
1054 retention date after which even old generations are purged. The
1055 database tracks the contents of media and projects the percentage of
1056 each volume that is valid. When the valid content of a volume drops
1057 below a configured percentage, the valid data are migrated to another
1058 volume and the old volume is marked as empty. Note, this requires
1059 ADSM to have an idea of the contents of a client, i.e. marking the
1060 database when an existing file was deleted, but this would solve your
1061 issue of restoring a client without restoring deleted files.
1063 This is pretty far from what bacula now does, but if you are going to
1064 rip things up for Base jobs,.....
1065 Also, the benefits of this are huge for very large shops, especially
1066 with media robots, but are a pain for shops with manual media
1070 > Base jobs sound pretty useful, but I'm not dying for them.
1072 Nobody is dying for them, but when you see what it does, you will die
1075 3. Restoring deleted files: Since I think my comments in (2) above
1076 have low probability of implementation, I'll also suggest that you
1077 could approach the issue of deleted files by a mechanism of having the
1078 fd report to the dir, a list of all files on the client for every
1079 backup job. The dir could note in the database entry for each file
1080 the date that the file was seen. Then if a restore as of date X takes
1081 place, only files that exist from before X until after X would be
1082 restored. Probably the major cost here is the extra date container in
1083 each row of the files table.
1085 Thanks for "listening". I hope some of this helps. If you want to
1086 contact me, please send me an email - I read some but not all of the
1087 mailing list traffic and might miss a reply there.
1089 Please accept my compliments for bacula. It is doing a great job for
1090 me!! I sympathize with you in the need to wrestle with excelence in
1091 execution vs. excelence in feature inclusion.
1096 ==============================
1099 - Design at hierarchial storage for Bacula. Migration and Clone.
1100 - Implement FSM (File System Modules).
1101 - Audit M_ error codes to ensure they are correct and consistent.
1102 - Add variable break characters to lex analyzer.
1103 Either a bit mask or a string of chars so that
1104 the caller can change the break characters.
1105 - Make a single T_BREAK to replace T_COMMA, etc.
1106 - Ensure that File daemon and Storage daemon can
1107 continue a save if the Director goes down (this
1108 is NOT currently the case). Must detect socket error,
1109 buffer messages for later.
1110 - Enhance time/duration input to allow multiple qualifiers e.g. 3d2h
1111 - Add ability to backup to two Storage devices (two SD sessions) at
1112 the same time -- e.g. onsite, offsite.
1113 - Add the ability to consolidate old backup sets (basically do a restore
1114 to tape and appropriately update the catalog). Compress Volume sets.
1115 Might need to spool via file is only one drive is available.
1116 - Compress or consolidate Volumes of old possibly deleted files. Perhaps
1117 someway to do so with every volume that has less than x% valid
1121 Migration: Move a backup from one Volume to another
1122 Clone: Copy a backup -- two Volumes
1124 Bacula Migration is based on Jobs (apparently Networker is file by file).
1126 Migration triggered by:
1130 Highwater mark (keep total size)
1135 ======================================================
1137 It is somewhat like a Full save becomes an incremental since
1138 the Base job (or jobs) plus other non-base files.
1140 - A Base backup is same as Full backup, just different type.
1141 - New BaseFiles table that contains:
1143 BaseJobId - Base JobId referenced for this FileId (needed ???)
1144 JobId - JobId currently running
1145 FileId - File not backed up, exists in Base Job
1146 FileIndex - FileIndex from Base Job.
1147 i.e. for each base file that exists but is not saved because
1148 it has not changed, the File daemon sends the JobId, BaseId,
1149 FileId, FileIndex back to the Director who creates the DB entry.
1150 - To initiate a Base save, the Director sends the FD
1151 the FileId, and full filename for each file in the Base.
1152 - When the FD finds a Base file, he requests the Director to
1153 send him the full File entry (stat packet plus MD5), or
1154 conversely, the FD sends it to the Director and the Director
1155 says yes or no. This can be quite rapid if the FileId is kept
1156 by the FD for each Base Filename.
1157 - It is probably better to have the comparison done by the FD
1158 despite the fact that the File entry must be sent across the
1160 - An alternative would be to send the FD the whole File entry
1161 from the start. The disadvantage is that it requires a lot of
1162 space. The advantage is that it requires less communications
1164 - The Job record must be updated to indicate that one or more
1166 - At end of Job, FD returns:
1167 1. Count of base files/bytes not written to tape (i.e. matches)
1168 2. Count of base file that were saved i.e. had changed.
1169 - No tape record would be written for a Base file that matches, in the
1170 same way that no tape record is written for Incremental jobs where
1171 the file is not saved because it is unchanged.
1172 - On a restore, all the Base file records must explicitly be
1173 found from the BaseFile tape. I.e. for each Full save that is marked
1174 to have one or more Base Jobs, search the BaseFile for all occurrences
1176 - An optimization might be to make the BaseFile have:
1182 This would avoid the need to explicitly fetch each File record for
1183 the Base job. The Base Job record will be fetched to get the
1184 VolSessionId and VolSessionTime.
1185 =========================================================
1188 ==========================================================
1190 For each Incremental job that is run, there may be files that
1191 were found but not saved because they were locked (this applies
1192 only to Windows). Such a system could send back to the Director
1193 a list of Unsaved files.
1195 - New UnSavedFiles table that contains:
1199 - Then in the next Incremental job, the list of Unsaved Files will be
1200 feed to the FD, who will ensure that they are explicitly chosen even
1201 if standard date/time check would not have selected them.
1202 =============================================================
1206 Multiple drive autochanger data: see Alan Brown
1207 mtx -f xxx unloadStorage Element 1 is Already Full(drive 0 was empty)
1208 Unloading Data Transfer Element into Storage Element 1...source Element
1209 Address 480 is Empty
1211 (drive 0 was empty and so was slot 1)
1212 > mtx -f xxx load 15 0
1213 no response, just returns to the command prompt when complete.
1214 > mtx -f xxx status Storage Changer /dev/changer:2 Drives, 60 Slots ( 2 Import/Export )
1215 Data Transfer Element 0:Full (Storage Element 15 Loaded):VolumeTag = HX001
1216 Data Transfer Element 1:Empty
1217 Storage Element 1:Empty
1218 Storage Element 2:Full :VolumeTag=HX002
1219 Storage Element 3:Full :VolumeTag=HX003
1220 Storage Element 4:Full :VolumeTag=HX004
1221 Storage Element 5:Full :VolumeTag=HX005
1222 Storage Element 6:Full :VolumeTag=HX006
1223 Storage Element 7:Full :VolumeTag=HX007
1224 Storage Element 8:Full :VolumeTag=HX008
1225 Storage Element 9:Full :VolumeTag=HX009
1226 Storage Element 10:Full :VolumeTag=HX010
1227 Storage Element 11:Empty
1228 Storage Element 12:Empty
1229 Storage Element 13:Empty
1230 Storage Element 14:Empty
1231 Storage Element 15:Empty
1232 Storage Element 16:Empty....
1233 Storage Element 28:Empty
1234 Storage Element 29:Full :VolumeTag=CLNU01L1
1235 Storage Element 30:Empty....
1236 Storage Element 57:Empty
1237 Storage Element 58:Full :VolumeTag=NEX261L2
1238 Storage Element 59 IMPORT/EXPORT:Empty
1239 Storage Element 60 IMPORT/EXPORT:Empty
1241 Unloading Data Transfer Element into Storage Element 15...done
1243 (just to verify it remembers where it came from, however it can be
1244 overrriden with mtx unload {slotnumber} to go to any storage slot.)
1246 There needs to be a table of drive # to devices somewhere - If there are
1247 multiple changers or drives there may not be a 1:1 correspondance between
1248 changer drive number and system device name - and depending on the way the
1249 drives are hooked up to scsi busses, they may not be linearly numbered
1250 from an offset point either.something like
1252 Autochanger drives = 2
1253 Autochanger drive 0 = /dev/nst1
1254 Autochanger drive 1 = /dev/nst2
1255 IMHO, it would be _safest_ to use explicit mtx unload commands at all
1256 times, not just for multidrive changers. For a 1 drive changer, that's
1262 MTX's manpage (1.2.15):
1263 unload [<slotnum>] [ <drivenum> ]
1264 Unloads media from drive <drivenum> into slot
1265 <slotnum>. If <drivenum> is omitted, defaults to
1266 drive 0 (as do all commands). If <slotnum> is
1267 omitted, defaults to the slot that the drive was
1268 loaded from. Note that there's currently no way
1269 to say 'unload drive 1's media to the slot it
1270 came from', other than to explicitly use that
1271 slot number as the destination.AB
1277 undef# camcontrol devlist
1278 <WANGTEK 51000 SCSI M74H 12B3> at scbus0 target 2 lun 0 (pass0,sa0)
1279 <ARCHIVE 4586XX 28887-XXX 4BGD> at scbus0 target 4 lun 0 (pass1,sa1)
1280 <ARCHIVE 4586XX 28887-XXX 4BGD> at scbus0 target 4 lun 1 (pass2)
1282 tapeinfo -f /dev/sg0 with a bad tape in drive 1:
1283 [kern@rufus mtx-1.2.17kes]$ ./tapeinfo -f /dev/sg0
1284 Product Type: Tape Drive
1286 Product ID: 'C5713A '
1288 Attached Changer: No
1289 TapeAlert[3]: Hard Error: Uncorrectable read/write error.
1290 TapeAlert[20]: Clean Now: The tape drive neads cleaning NOW.
1297 Medium Type: Not Loaded
1300 DataCompEnabled: yes
1301 DataCompCapable: yes
1302 DataDeCompEnabled: yes
1309 Handling removable disks
1311 From: Karl Cunningham <karlc@keckec.com>
1313 My backups are only to hard disk these days, in removable bays. This is my
1314 idea of how a backup to hard disk would work more smoothly. Some of these
1315 things Bacula does already, but I mention them for completeness. If others
1316 have better ways to do this, I'd like to hear about it.
1318 1. Accommodate several disks, rotated similar to how tapes are. Identified
1319 by partition volume ID or perhaps by the name of a subdirectory.
1320 2. Abort & notify the admin if the wrong disk is in the bay.
1321 3. Write backups to different subdirectories for each machine to be backed
1323 4. Volumes (files) get created as needed in the proper subdirectory, one
1325 5. When a disk is recycled, remove or zero all old backup files. This is
1326 important as the disk being recycled may be close to full. This may be
1327 better done manually since the backup files for many machines may be
1328 scattered in many subdirectories.
1333 - Save mount point for directories not traversed with onefs=yes.
1334 - Add seconds to start and end times in the Job report output.
1335 - if 2 concurrent backups are attempted on the same tape
1336 drive (autoloader) into different tape pools, one of them will exit
1337 fatally instead of halting until the drive is idle
1338 - Update StartTime if job held in Job Queue.
1339 - Look at www.nu2.nu/pebuilder as a helper for full windows
1340 bare metal restore. (done by Scott)
1341 - Fix orphanned buffers:
1342 Orphaned buffer: 24 bytes allocated at line 808 of rufus-dir job.c
1343 Orphaned buffer: 40 bytes allocated at line 45 of rufus-dir alist.c
1344 - Implement Preben's suggestion to add
1345 File System Types = ext2, ext3
1346 to FileSets, thus simplifying backup of *all* local partitions.
1347 - Try to open a device on each Job if it was not opened
1348 when the SD started.
1349 - Add dump of VolSessionId/Time and FileIndex with bls.
1350 - If Bacula does not find the right tape in the Autochanger,
1351 then mark the tape in error and move on rather than asking
1352 for operator intervention.
1353 - Cancel command should include JobId in list of Jobs.
1354 - Add performance testing hooks
1355 - Bootstrap from JobMedia records.
1356 - Implement WildFile and WildDir to solve problem of
1357 saving only *.doc files.
1359 Please use the "label" command to create a new Volume for:
1360 Storage: DDS-4-changer
1364 The defined Storage resources are:
1365 - Copy Changer Device and Changer Command from Autochanger
1366 to Device resource in SD if none given in Device resource.
1367 - 1. Automatic use of more than one drive in an autochanger (done)
1368 - 2. Automatic selection of the correct drive for each Job (i.e.
1369 selects a drive with an appropriate Volume for the Job) (done)
1370 - 6. Allow multiple simultaneous Jobs referencing the same pool write
1371 to several tapes (some new directive(s) are are probably needed for
1374 - Key on Storage rather than Pool (done)
1375 - Allow multiple drives to use same Pool (change jobq.c DIR) (done).
1376 - Synchronize multiple drives so that not more
1377 than one loads a tape and any time (done)
1378 - 4. Use Changer Device and Changer Command specified in the
1379 Autochanger resource, if none is found in the Device resource.
1380 You can continue to specify them in the Device resource if you want
1381 or need them to be different for each device.
1382 - 5. Implement a new Device directive (perhaps "Autoselect = yes/no")
1383 that can allow a Device be part of an Autochanger, and hence the changer
1384 script protected, but if set to no, will prevent the Device from being
1385 automatically selected from the changer. This allows the device to
1386 be directly accessed through its Device name, but not through the
1388 #6 Select one from among Multiple Storage Devices for Job
1389 #5 Events that call a Python program
1390 (Implemented in Dir/SD)
1391 - Make sure the Device name is in the Query packet returned.
1392 - Don't start a second file job if one is already running.
1393 - Implement EOF/EOV labels for ANSI labels
1394 - Implement IBM labels.
1395 - When Python creates a new label, the tape is immediately
1396 recycled and no label created. This happens when using
1397 autolabeling -- even when Python doesn't generate the name.
1398 - Scratch Pool where the volumes can be re-assigned to any Pool.
1399 - 28-Mar 23:19 rufus-sd: acquire.c:379 Device "DDS-4" (/dev/nst0)
1400 is busy reading. Job 6 canceled.
1401 - Remove separate thread for opening devices in SD. On the other
1402 hand, don't block waiting for open() for devices.
1403 - Fix code to either handle updating NumVol or to calculate it in
1405 - Ensure that you cannot exclude a directory or a file explicitly
1407 #4 Embedded Python Scripting
1408 (Implemented in Dir/SD/FD)
1409 - Add Python writable variable for changing the Priority,
1410 Client, Storage, JobStatus (error), ...
1413 - Add disk seeking on restore; turn off seek on tapes.
1415 - Look at dird_conf.c:1000: warning: `int size'
1416 might be used uninitialized in this function
1417 - Indicate when a Job is purged/pruned during restore.
1418 - Implement some way to turn off automatic pruning in Jobs.
1419 - Implement a way an Admin Job can prune, possibly multiple
1420 clients -- Python script?
1421 - Look at Preben's acl.c error handling code.
1422 - SD crashes after a tape restore then doing a backup.
1423 - If drive is opened read/write, close it and re-open
1424 read-only if doing a restore, and vice-versa.
1426 data-fd: RestoreFiles.2004-12-07_15.56.42 Error:
1427 > ..\findlib\../../findlib/create_file.c:275 Could not open e:/: ERR=Der
1428 > Prozess kann nicht auf die Datei zugreifen, da sie von einem anderen
1429 > Prozess verwendet wird.
1430 Restore restores all files, but then fails at the end trying
1431 to set the attributes of e:
1432 from failed jobs.- Resolve the problem between Device name and Archive name,
1433 and fix SD messages.
1434 - Tell the "restore" user when browsing is no longer possible.
1435 - Add a restore directory-x
1436 - Write non-optimized bsrs from the JobMedia and Media records,
1437 even after Files are pruned.
1438 - Delete Stripe and Copy from VolParams to save space.
1439 - Fix option 2 of restore -- list where file is backed up -- require Client,
1440 then list last 20 backups.
1441 - Finish implementation of passing all Storage and Device needs to
1443 - Move test for max wait time exceeded in job.c up -- Peter's idea.
1444 ## Consider moving docs to their own project.
1445 ## Move rescue to its own project.
1446 - Add client version to the Client name line that prints in
1448 - Fix the Rescue CDROM.
1449 - By the way: on page http://www.bacula.org/?page=tapedrives , at the
1450 bottom, the link to "Tape Testing Chapter" is broken. It goes to
1451 /html-manual/... while the others point to /rel-manual/...
1452 - Device resource needs the "name" of the SD.
1453 - Specify a single directory to restore.
1454 - Implement MediaType keyword in bsr?
1455 - Add a date and time stamp at the beginning of every line in the
1456 Job report (Volker Sauer).
1457 - Add level to estimate command.
1458 - Add "limit=n" for "list jobs"
1459 - Make bootstrap filename unique.
1460 - Make Dmsg look at global before calling subroutine.
1462 it seems to be complaining about 12:00pm which should be a valid 12
1463 hour time. I changed the time to 11:59am and everything works fine.
1464 Also 12:00am works fine. 0:00pm also works (which I don't think
1465 should). None of the values 12:00pm - 12:59pm work for that matter.
1466 - Require restore via the restore command or make a restore Job
1467 get the bootstrap file.
1468 - Implement Maximum Job Spool Size
1469 - Fix 3993 error in SD. It forgets to look at autochanger
1470 resource for device command, ...
1471 - 3. Prevent two drives requesting the same Volume in any given
1472 autochanger, by checking if a Volume is mounted on another drive
1474 - Upgrade to MySQL 4.1.12 See:
1475 http://dev.mysql.com/doc/mysql/en/Server_SQL_mode.html
1476 - Add # Job Level date to bsr file
1477 - Implement "PreferMountedVolumes = yes|no" in Job resource.
1478 ## Integrate web-bacula into a new Bacula project with
1480 - Cleaning tapes should have Status "Cleaning" rather than append.
1481 - Make sure that Python has access to Client address/port so that
1482 it can check if Clients are alive.
1483 - Review all items in "restore".
1484 - Fix PostgreSQL GROUP BY problems in restore.
1485 - Fix PostgreSQL sql problems in bugs.
1487 04-Jul 13:01 MainSD: Rufus.2005-07-04_01.05.02 Warning: Director wanted Volume
1489 Current Volume "DLT-04Jul05" not acceptable because:
1490 1997 Volume "DLT-13Feb04" not in catalog.
1491 04-Jul 13:01 MainSD: Please mount Volume "DLT-04Jul05" on Storage Device
1492 "HP DLT 80" (/dev/nst0) for Job Rufus.2005-07-04_01.05.02
1493 ## Create a new GUI chapter explaining all the GUI programs.
1494 - Make "update slots" when pointing to Autochanger, remove
1495 all Volumes from other drives. "update slots all-drives"?
1496 No, this is done by modifying mtx-changer to list what is
1498 - Finish TLS implementation.
1499 - Port limiting -m in iptables to prevent DoS attacks
1500 could cause broken pipes on Bacula.
1501 6. Build and test the Volume Shadow Copy (VSS) for Win32.
1502 - Allow cancel of unknown Job
1503 - State not saved when closing Win32 FD by icon
1504 - bsr-opt-test fails. bsr deleted. Fix.
1505 - Move Python daemon variables from Job to Bacula object.
1506 WorkingDir, ConfigFile
1507 - Document that Bootstrap files can be written with cataloging
1509 - Document details of ANSI/IBM labels
1511 1) ADIC, DLT, FastStor 4000, 7*20GB
1512 - Linux Sony LIB-D81, AIT-3 library works.
1514 to activate, check or disable the hardware compression feature on my
1515 exb-8900 i use the exabyte "MammothTool" you can get it here:
1516 http://www.exabyte.com/support/online/downloads/index.cfm
1517 There is a solaris version of this tool. With option -C 0 or 1 you can
1518 disable or activate compression. Start this tool without any options for
1520 - Document Heartbeat Interval in the dealing with firewalls section.
1521 - Document new CDROM directory.
1522 - On Win32 working directory must have drive letter ????
1523 - On Win32 working directory must be writable by SYSTEM to
1525 - Document that ChangerDevice is used for Alert command.
1526 - Add better documentation on how restores can be done
1527 8. Take one more try at making DVD writing work (no go)