7 Version 1.37 Kern (see below)
8 ========================================================
10 Final items for 1.37 before release:
12 - Tape xxx in drive 0, requested in drive 1
13 - The mount command does not work with drives other than 0.
14 - A mount should cause the SD to re-examine what Slot is
16 - The SD locks on to the first available drive then
17 wants a Volume that is released but in another drive --
19 - Look at fixing restore status stats in SD.
20 - My database is growing
21 - Call GetLastError() in the berrno constructor rather
22 than delaying until strerror.
24 - --without-openssl breaks at least on Solaris.
25 9. Run the regression scripts on Solaris and FreeBSD
26 - Figure out how to package gui, and rescue programs.
28 - Arno had to do -- to get update slots=x to work
29 UPDATE Media SET InChanger=0,Slot=0 WHERE InChanger>0 AND Slot>0; (MySQL)
32 - Document cleaning up the spool files:
33 db, pid, state, bsr, mail, conmsg, spool
34 - Document the multiple-drive-changer.txt script.
35 - Pruning with Admin job.
36 - Restore of all files for a Job or set of jobs even if the file
37 records have been removed from the catalog.
40 - "Resuming" a failed backup (lost line for example) by using the
41 failed backup as a sort of "base" job.
43 - Email to the user when the tape is about to need changing x
44 days before it needs changing.
45 - Command to show next tape that will be used for a job even
46 if the job is not scheduled.
47 --- create_file.c.orig Fri Jul 8 12:13:05 2005
48 +++ create_file.c Fri Jul 8 12:13:07 2005
50 attr->ofname, be.strerror());
53 + } else if(S_ISSOCK(attr->statp.st_mode)) {
54 + Dmsg1(200, "Skipping socket: %s\n", attr->ofname);
56 Dmsg1(200, "Restore node: %s\n", attr->ofname);
57 if (mknod(attr->ofname, attr->statp.st_mode, attr->statp.st_rdev) != 0 && errno != EEXIST) {
59 - Fix bpipe.c so that it does not modify results pointer.
60 ***FIXME*** calling sequence should be changed.
62 #3 Migration (Move, Copy, Archive Jobs)
63 #7 Single Job Writing to Multiple Storage Devices
64 - Reserve blocks other restore jobs when first cannot connect
66 - Add true/false to conf same as yes/no
67 - For Windows disaster recovery see http://unattended.sf.net/
68 - regardless of the retention period, Bacula will not prune the
69 last Full, Diff, or Inc File data until a month after the
70 retention period for the last Full backup that was done.
71 - update volume=xxx --- add status=Full
72 - Remove old spool files on startup.
73 - Exclude SD spool/working directory.
74 - Refuse to prune last valid Full backup. Same goes for Catalog.
76 - Make a callback when Rerun failed levels is called.
77 - Give Python program access to Scheduled jobs.
78 - Add setting Volume State via Python.
79 - Python script to save with Python, not save, save with Bacula.
80 - Python script to do backup.
82 - Change the Priority, Client, Storage, JobStatus (error)
83 at the start of a job.
84 - Why is SpoolDirectory = /home/bacula/spool; not reported
85 as an error when writing a DVD?
86 - Make bootstrap file handle multiple MediaTypes (SD)
87 - Remove all old Device resource code in Dir and code to pass it
88 back in SD -- better, rework it to pass back device statistics.
89 - Check locking of resources -- be sure to lock devices where previously
90 resources were locked.
91 - The last part is left in the spool dir.
94 - In restore don't compare byte count on a raw device -- directory
95 entry does not contain bytes.
96 - To mark files as deleted, run essentially a Verify to disk, and
97 when a file is found missing (MarkId != JobId), then create
98 a new File record with FileIndex == -1. This could be done
99 by the FD at the same time as the backup.
103 MA = (last_MA * 3 + rate) / 4
104 rate = (bytes - last_bytes) / (runtime - last_runtime)
105 - Max Vols limit in Pool off by one?
106 - Implement Files/Bytes,... stats for restore job.
107 - Implement Total Bytes Written, ... for restore job.
108 - Despool attributes simultaneously with data in a separate
109 thread, rejoined at end of data spooling.
110 - 7. Implement new Console commands to allow offlining/reserving drives,
111 and possibly manipulating the autochanger (much asked for).
112 - Add start/end date editing in messages (%t %T, %e?) ...
113 - Add ClientDefs similar to JobDefs.
114 - Print more info when bextract -p accepts a bad block.
115 - Fix FD JobType to be set before RunBeforeJob in FD.
116 - Look at adding full Volume and Pool information to a Volume
117 label so that bscan can get *all* the info.
118 - If the user puts "Purge Oldest Volume = yes" or "Recycle Oldest Volume = yes"
119 and there is only one volume in the pool, refuse to do it -- otherwise
120 he fills the Volume, then immediately starts reusing it.
121 - Implement copies and stripes.
122 - Add history file to console.
123 - Each file on tape creates a JobMedia record. Peter has 4 million
124 files spread over 10000 tape files and four tapes. A restore takes
125 16 hours to build the restore list.
126 - Add and option to check if the file size changed during backup.
127 - Make sure SD deletes spool files on error exit.
128 - Delete old spool files when SD starts.
129 - When labeling tapes, if you enter 000026, Bacula uses
130 the tape index rather than the Volume name 000026.
131 - Add offline tape command to Bacula console.
133 Enter MediaId or Volume name: 32
134 Enter new Volume name: DLT-20Dec04
135 Automatically selected Pool: Default
136 Connecting to Storage daemon DLTDrive at 192.168.68.104:9103 ...
137 Sending relabel command from "DLT-28Jun03" to "DLT-20Dec04" ...
138 block.c:552 Write error at 0:0 on device /dev/nst0. ERR=Bad file descriptor.
139 Error writing final EOF to tape. This tape may not be readable.
140 dev.c:1207 ioctl MTWEOF error on /dev/nst0. ERR=Permission denied.
141 askdir.c:219 NULL Volume name. This shouldn't happen!!!
142 3912 Failed to label Volume: ERR=dev.c:1207 ioctl MTWEOF error on /dev/nst0. ERR=Permission denied.
143 Label command failed for Volume DLT-20Dec04.
144 Do not forget to mount the drive!!!
145 - Bug: if a job is manually scheduled to run later, it does not appear
146 in any status report and cannot be cancelled.
148 ==== Keeping track of deleted files ====
149 My "trick" for keeping track of deletions is the following.
150 Assuming the user turns on this option, after all the files
151 have been backed up, but before the job has terminated, the
152 FD will make a pass through all the files and send their
153 names to the DIR (*exactly* the same as what a Verify job
154 currently does). This will probably be done at the same
155 time the files are being sent to the SD avoiding a second
156 pass. The DIR will then compare that to what is stored in
157 the catalog. Any files in the catalog but not in what the
158 FD sent will receive a catalog File entry that indicates
159 that at that point in time the file was deleted.
161 During a restore, any file initially picked up by some
162 backup (Full, ...) then subsequently having a File entry
163 marked "delete" will be removed from the tree, so will not
164 be restored. If a file with the same name is later OK it
165 will be inserted in the tree -- this already happens. All
166 will be consistent except for possible changes during the
169 Since I'm on the subject, some of you may be wondering what
170 the utility of the in memory tree is if you are going to
171 restore everything (at least it comes up from time to time
172 on the list). Well, it is still *very* useful because it
173 allows only the last item found for a particular filename
174 (full path) to be entered into the tree, and thus if a file
175 is backed up 10 times, only the last copy will be restored.
176 I recently (last Friday) restored a complete directory, and
177 the Full and all the Differential and Incremental backups
178 spanned 3 Volumes. The first Volume was not even mounted
179 because all the files had been updated and hence backed up
180 since the Full backup was made. In this case, the tree
181 saved me a *lot* of time.
183 Make sure this information is stored on the tape too so
184 that it can be restored directly from the tape.
186 Comments from Martin Simmons (I think they are all covered):
187 Ok, that should cover the basics. There are few issues though:
189 - Restore will depend on the catalog. I think it is better to include the
190 extra data in the backup as well, so it can be seen by bscan and bextract.
192 - I'm not sure if it will preserve multiple hard links to the same inode. Or
193 maybe adding or removing links will cause the data to be dumped again?
195 - I'm not sure if it will handle renamed directories. Possibly it will work
196 by dumping the whole tree under a renamed directory?
198 - It remains to be seen how the backup performance of the DIR's will be
199 affected when comparing the catalog for a large filesystem.
204 - Add Pool/Storage override regression test.
205 - Add delete JobId to regression.
206 - Add a regression test for dbcheck.
207 - New test to add bscan to four-concurrent-jobs regression,
208 i.e. after the four-concurrent jobs zap the
209 database as is done in the bscan-test, then use bscan to
210 restore the database, do a restore and compare with the
212 - Add restore of specific JobId to regression (item 3
213 on the restore prompt)
214 - Add IPv6 to regression
215 - Add database test to regression. Test each function like delete,
218 - AntiVir can slow down backups on Win32 systems.
219 - Win32 systems with FAT32 can be much slower than NTFS for
220 more than 1000 files per directory.
224 - A HOLD command to stop all jobs from starting.
225 - A PAUSE command to pause all running jobs ==> release the
227 - Media Type = LTO,LTO-2,LTO-3
228 Media Type Read = LTO,LTO2,LTO3
229 Media Type Write = LTO2, LTO3
231 === From Carsten Menke <bootsy52@gmx.net>
233 Following is a list of what I think in the situations where I'm faced with,
234 could be a usefull enhancement to bacula, which I'm certain other users will
235 benefit from as well.
237 1. NextJob/NextJobs Directive within a Job Resource in the form of
238 NextJobs = job1,job2.
241 I currently solved the problem with running multiple jobs each after each
242 by setting the Max Wait Time for a job to 8 hours, and give
243 the jobs different Priorities. However, there scenarios where
244 1 Job is directly depending on another job, so if the former job fails,
245 the job after it needn't to be run
246 while maybe other jobs should run despite of that
249 A Backup Job and a Verify job, if the backup job fails there is no need to run
250 the verify job, as the backup job already failed. However, one may like
251 to backup the Catalog to disk despite of that the main backup job failed.
254 I see that this is related to the Event Handlers which are on the ToDo
255 list, also it is maybe a good idea to check for the return value and
256 execute different actions based on the return value
259 3. offline capability to bconsole
262 Currently I use a script which I execute within the last Job via the
263 RunAfterJob Directive, to release and eject the tape.
264 So I have to call bconsole "release=Storage-Name" and afterwards
265 mt -f /dev/nst0 eject to get the tape out.
267 If I have multiple Storage Devices, than these may not be /dev/nst0 and
268 I have to modify the script or call it with parameters etc.
269 This would actually not be needed, as everything is already defined
270 in bacula-sd.conf and if I can invoke bconsole with the
271 storage name via $1 in the script than I'm done and information is
274 4. %s for Storage Name added to the chars being substituted in "RunAfterJob"
278 For the reason mentioned in 3. to have the ability to call a
279 script with /scripts/foobar %s and in the script use $1
280 to pass the Storage Name to bconsole
282 5. Setting Volume State within a Job Resource
285 Instead of using "Maximum Volume Jobs" in the Pool Resource,
286 I would have the possibilty to define
287 in a Job Resource that after this certain job is run, the Volume State
288 should be set to "Volume State = Used", this give more flexibility (IMHO).
290 6. Localization of Bacula Messages
293 Unfortunatley many,many people I work with don't speak english very well.
294 So if at least the Reporting messages would be localized then they
295 would understand that they have to change the tape,etc. etc.
297 I volunteer to do the german translations, and if I can convince my wife also
298 french and Morre (western african language).
300 7. OK, this is evil, probably bound to security risks and maybe not possible
301 due to the design of bacula.
303 Implementation of Backtics ( `command` ) for shell comand execution to
304 the "Label Format" Directive.
308 Currently I have defined BACULA_DAY_OF_WEEK="day1|day2..." resulting in
309 Label Format = "HolyBackup-${BACULA_DAY_OF_WEEK[${WeekDay}]}". If I could
310 use backticks than I could use "Label Format = HolyBackup-`date +%A` to have
311 the localized name for the day of the week appended to the
312 format string. Then I have the tape labeled automatically with weekday
313 name in the correct language.
315 - Yes, that is surely the case. I probably should turn those into Warning
316 errors. In addition, you just made me think that it might not be bad to
317 add an option to check the file size after backing up the file and
318 report if it changes. This would be done as an option because it would
321 Kern, good idea. If you do do that, mention in the output: file
322 shrunk, or file expanded, just to make it obvious to the user
323 (without having to the refer to file size), just how the file size
326 Would this option be for all file, or just one file? Or a fileset?
327 - Make output from status use html table tags for nicely
328 presenting in a browser.
329 - Can one write tapes faster with 8192 byte block sizes?
330 - Document security problems with the same password for everyone in
331 rpm and Win32 releases.
332 - Browse generations of files.
333 - I've seen an error when my catalog's File table fills up. I
334 then have to recreate the File table with a larger maximum row
335 size. Relevant information is at
336 http://dev.mysql.com/doc/mysql/en/Full_table.html ; I think the
337 "Installing and Configuring MySQL" chapter should talk a bit
338 about this potential problem, and recommend a solution.
339 - For Solaris must use POSIX awk.
340 - Want speed of writing to tape while despooling.
341 - Supported autochanger:
349 Wangtek 6525ES (SCSI-1 QIC drive, 525MB), under Linux 2.4.something,
350 bacula 1.36.0/1 works with blocksize 16k INSIDE bacula-sd.conf.
351 - Add regex from http://www.pcre.org to Bacula for Win32.
352 - Use only shell tools no make in CDROM package.
353 - Include within include does it work?
354 - Implement a Pool of type Cleaning?
355 - Implement VolReadTime and VolWriteTime in SD
356 - Modify Backing up Your Database to include a bootstrap file.
357 - Think about making certain database errors fatal.
358 - Look at correcting the time jump in the scheduler for daylight
359 savings time changes.
360 - Add a "real" timer to network connections.
361 - Promote to Full = Time period
362 - Check dates entered by user for correctness (month/day/... ranges)
363 - Compress restore Volume listing by date and first file.
364 - Look at patches/bacula_db.b2z postgresql that loops during restore.
366 - Perhaps add read/write programs and/or plugins to FileSets.
367 - How to handle backing up portables ...
368 - Add some sort of guaranteed Interval for upgrading jobs.
369 - Can we write the state file after every job terminates? On Win32
370 the system crashes and the state file is not updated.
373 Documentation to do: (any release a little bit at a time)
374 - Doc to do unmount before removing magazine.
375 - Alternative to static linking "ldd prog" save all binaries listed,
376 restore them and point LD_LIBRARY_PATH to them.
377 - Document add "</dev/null >/dev/null 2>&1" to the bacula-fd command line
378 - Document query file format.
379 - Add more documentation for bsr files.
380 - Document problems with Verify and pruning.
381 - Document how to use multiple databases.
382 - VXA drives have a "cleaning required"
383 indicator, but Exabyte recommends preventive cleaning after every 75
386 In this context, it should be noted that Exabyte has a command-line
387 vxatool utility available for free download. (The current version is
388 vxatool-3.72.) It can get diagnostic info, read, write and erase tapes,
389 test the drive, unload tapes, change drive settings, flash new firmware,
391 Of particular interest in this context is that vxatool <device> -i will
392 report, among other details, the time since last cleaning in tape motion
393 minutes. This information can be retrieved (and settings changed, for
394 that matter) through the generic-SCSI device even when Bacula has the
395 regular tape device locked. (Needless to say, I don't recommend
396 changing tape settings while a job is running.)
397 - Lookup HP cleaning recommendations.
398 - Lookup HP tape replacement recommendations (see trouble shooting autochanger)
399 - Document doing table repair
402 ===================================
403 - Add macro expansions in JobDefs.
404 Run Before Job = "SomeFile %{Level} %{Client}"
405 Write Bootstrap="/some/dir/%{JobName}_%{Client}.bsr"
406 - Use non-blocking network I/O but if no data is available, use
408 - Use gather write() for network I/O.
409 - Autorestart on crash.
410 - Add bandwidth limiting.
411 - Add acks every once and a while from the SD to keep
412 the line from timing out.
413 - When an error in input occurs and conio beeps, you can back
414 up through the prompt.
415 - Detect fixed tape block mode during positioning by looking at
416 block numbers in btape "test". Possibly adjust in Bacula.
417 - Fix list volumes to output volume retention in some other
418 units, perhaps via a directive.
419 - If opening a tape in read/write mode fails attempt to open
420 it in read-only mode, and mark the tape for read only.
421 - Allow Simultaneous Priorities = yes => run up to Max concurrent jobs even
422 with multiple priorities.
423 - If you use restore replace=never, the directory attributes for
424 non-existent directories will not be restored properly.
426 - see lzma401.zip in others directory for new compression
428 - Minimal autochanger handling in Bacula and in btape.
429 - Look into how tar does not save sockets and the possiblity of
430 not saving them in Bacula (Martin Simmons reported this).
431 - Fix restore jobs so that multiple jobs can run if they
432 are not using the same tape(s).
433 - Allow the user to select JobType for manual pruning/purging.
434 - bscan does not put first of two volumes back with all info in
436 - Implement the FreeBSD nodump flag in chflags.
437 - Figure out how to make named console messages go only to that
438 console and to the non-restricted console (new console class?).
439 - Make restricted console prompt for password if *ask* is set or
440 perhaps if password is undefined.
441 - Implement "from ISO-date/time every x hours/days/weeks/months" in
444 ==== from Marc Schoechlin
445 - the help-command should be more verbose
446 (it should explain the paramters of the different
448 -> its time-comsuming to consult the manual anytime
449 you need a special parameter
450 -> maybe its more easy to maintain this, if the
451 descriptions of that commands are outsourced to
453 - the cd-command should allow complete paths
454 i.e. cd /foo/bar/foo/bar
455 -> if a customer mails me the path to a certain file,
456 its faster to enter the specified directory
457 - if the password is not configured in bconsole.conf
458 you should be asked for it.
459 -> sometimes you like to do restore on a customer-machine
460 which shouldnt know the password for bacula.
461 -> adding the password to the file favours admins
462 to forget to remove the password after usage
464 the protection of that file is less important
465 - long-listed-output of commands should be scrollable
466 like the unix more/less-command does
467 -> if someone runs 200 and more machines, the lists could
468 be a little long and complex
469 - command-output should be shown column by column
470 to reduce scrolling and to increase clarity
472 - lsmark should list the selected files with full
474 - wildcards for selecting and file and directories would be nice
475 - any actions should be interuptable with STRG+C
476 - command-expansion would be pretty cool
478 - When the replace Never option is set, new directory permissions
479 are not restored. See bug 213. To fix this requires creating a
480 list of newly restored directories so that those directory
481 permissions *can* be restored.
482 - Compaction of Disk space by "migrating" Volumes that have pruned
483 Jobs (what criteria? size, #jobs, time).
484 - Add prune all command
485 - Document fact that purge can destroy a part of a restore by purging
486 one volume while others remain valid -- perhaps mark Jobs.
487 - Add multiple-media-types.txt
488 - look at mxt-changer.html
489 - Make ? do a help command (no return needed).
490 - Implement restore directory.
491 - Document streams and how to implement them.
492 - Try not to re-backup a file if a new hard link is added.
493 - Add feature to backup hard links only, but not the data.
494 - Fix stream handling to be simpler.
495 - Add Priority and Bootstrap to Run a Job.
496 - Eliminate Restore "Run Restore Job" prompt by allowing new "run command
498 - Remove View FileSet button from Run a Job dialog.
499 - Handle prompt for restore job at end of Restore command.
500 - Add display of total selected files to Restore window.
501 - Add tree pane to left of window.
502 - Add progress meter.
503 - Max wait time or max run time causes seg fault -- see runtime-bug.txt
504 - Document writing to a CD/DVD with Bacula.
505 - Add a "base" package to the window installer for pthreadsVCE.dll
506 which is needed by all packages.
507 - Add message to user to check for fixed block size when the forward
508 space test fails in btape.
509 - When unmarking a directory check if all files below are unmarked and
510 then remove the + flag -- in the restore tree.
511 - Possibly implement: Action = Unmount Device="TapeDrive1" in Admin jobs.
512 - Setup lrrd graphs: (http://www.linpro.no/projects/lrrd/) Mike Acar.
513 - Revisit the question of multiple Volumes (disk) on a single device.
514 - Add a block copy option to bcopy.
515 - Investigate adding Mac Resource Forks.
516 - Finish work on Gnome restore GUI.
517 - Fix "llist jobid=xx" where no fileset or client exists.
518 - For each job type (Admin, Restore, ...) require only the really necessary
519 fields.- Pass Director resource name as an option to the Console.
520 - Add a "batch" mode to the Console (no unsolicited queries, ...).
521 - Add a .list all files in the restore tree (probably also a list all files)
522 Do both a long and short form.
523 - Allow browsing the catalog to see all versions of a file (with
524 stat data on each file).
525 - Restore attributes of directory if replace=never set but directory
527 - Use SHA1 on authentication if possible.
528 - See comtest-xxx.zip for Windows code to talk to USB.
529 - Add John's appended files:
530 Appended = { /files/server/logs/http/*log }
531 and such files would be treated as follows.On a FULL backup, they would
532 be backed up like any other file.On an INCREMENTAL backup, where a
533 previous INCREMENTAL or FULL was already in thecatalogue and the length
534 of the file wasgreater than the length of the last backup, only thedata
535 added since the last backup will be dumped.On an INCREMENTAL backup, if
536 the length of the file is less than thelength of the file with the same
537 name last backed up, the completefile is dumped.On Windows systems, with
538 creation date of files, we can be evensmarter about this and not count
539 entirely upon the length.On a restore, the full and all incrementals
540 since it will beapplied in sequence to restore the file.
541 - Check new HAVE_WIN32 open bits.
542 - Check if the tape has moved before writing.
543 - Handling removable disks -- see below:
544 - Keep track of tape use time, and report when cleaning is necessary.
545 - Add FromClient and ToClient keywords on restore command (or
546 BackupClient RestoreClient).
547 - Implement a JobSet, which groups any number of jobs. If the
548 JobSet is started, all the jobs are started together.
549 Allow Pool, Level, and Schedule overrides.
550 - Enhance cancel to timeout BSOCK packets after a specific delay.
551 - Do scheduling by UTC using gmtime_r() in run_conf, scheduler, and
552 ua_status.!!! Thanks to Alan Brown for this tip.
553 - Look at updating Volume Jobs so that Max Volume Jobs = 1 will work
554 correctly for multiple simultaneous jobs.
555 - Correct code so that FileSet MD5 is calculated for < and | filename
557 - Implement the Media record flag that indicates that the Volume does disk
559 - Implement VolAddr, which is used when Volume is addressed like a disk,
560 and form it from VolFile and VolBlock.
561 - Make multiple restore jobs for multiple media types specifying
562 the proper storage type.
563 - Fix fast block rejection (stored/read_record.c:118). It passes a null
564 pointer (rec) to try_repositioning().
565 - Look at extracting Win data from BackupRead.
566 - Implement RestoreJobRetention? Maybe better "JobRetention" in a Job,
567 which would take precidence over the Catalog "JobRetention".
568 - Implement Label Format in Add and Label console commands.
569 - Possibly up network buffers to 65K. Put on variable.
570 - Put email tape request delays on one or more variables. User wants
571 to cancel the job after a certain time interval. Maximum Mount Wait?
572 - Job, Client, Device, Pool, or Volume?
573 Is it possible to make this a directive which is *optional* in multiple
574 resources, like Level? If so, I think I'd make it an optional directive
575 in Job, Client, and Pool, with precedence such that Job overrides Client
576 which in turn overrides Pool.
578 - New Storage specifications:
579 - Want to write to multiple storage devices simultaneously
580 - Want to write to multiple storage devices sequentially (in one job)
581 - Want to read/write simultaneously
582 - Key is MediaType -- it must match
584 Passed to SD as a sort of BSR record called Storage Specification
588 MediaType -> Next MediaType
590 Device -> Next Device
592 Allow multiple Storage specifications
600 Allow Multiple Pool specifications (note, Pool currently
602 Allow Multiple MediaType specifications in Dir conf
603 Allow Multiple Device specifications in Dir conf
604 Perhaps keep this in a single SSR
605 Tie a Volume to a specific device by using a MediaType that
606 is contained in only one device.
607 In SD allow Device to have Multiple MediaTypes
609 - Ideas from Jerry Scharf:
610 First let's point out some big pluses that bacula has for this
612 more importantly it's active. Thank you so much for that
613 even more important, it's not flaky
614 it has an open access catalog, opening many possibilities
615 it's pushing toward heterogeneous systems capability
617 Macintosh file client
618 macs are an interesting niche, but I fear a server is a rathole
619 working bare iron recovery for windows
620 the option for inc/diff backups not reset on fileset revision
621 a) use both change and inode update time against base time
622 b) do the full catalog check (expensive but accurate)
623 sizing guide (how much system is needed to back up N systems/files)
624 consultants on using bacula in building a disaster recovery system
626 or how to get at fancy things that one could do with bacula
627 logwatch code for bacula logs (or similar)
628 linux distro inclusion of bacula (brings good and bad, but necessary)
629 win2k/XP server capability (icky but you asked)
630 support for Oracle database ??
632 - Look at adding SQL server and Exchange support for Windows.
633 - Make dev->file and dev->block_num signed integers so that -1 can
634 be an invalid value which happens with BSR.
635 - Create VolAddr for disk files in place of VolFile and VolBlock. This
636 is needed to properly specify ranges.
637 - Add progress of files/bytes to SD and FD.
638 - Print warning message if FileId > 4 billion
639 - do a "messages" before the first prompt in Console
640 - Client does not show busy during Estimate command.
641 - Implement Console mtx commands.
642 - Implement a Mount Command and an Unmount Command where
643 the users could specify a system command to be performed
644 to do the mount, after which Bacula could attempt to
645 read the device. This is for Removeable media such as a CDROM.
646 - Most likely, this mount command would be invoked explicitly
647 by the user using the current Console "mount" and "unmount"
648 commands -- the Storage Daemon would do the right thing
649 depending on the exact nature of the device.
650 - As with tape drives, when Bacula wanted a new removable
651 disk mounted, it would unmount the old one, and send a message
652 to the user, who would then use "mount" as described above
653 once he had actually inserted the disk.
654 - Implement dump/print label to UA
655 - Spool to disk only when the tape is full, then when a tape is hung move
657 - bextract is sending everything to the log file ****FIXME****
658 - Allow multiple Storage specifications (or multiple names on
659 a single Storage specification) in the Job record. Thus a job
660 can be backed up to a number of storage devices.
661 - Implement some way for the File daemon to contact the Director
662 to start a job or pass its DHCP obtained IP number.
663 - Implement a query tape prompt/replace feature for a console
664 - Copy console @ code to gnome2-console
665 - Make AES the only encryption algorithm see
666 http://csrc.nist.gov/CryptoToolkit/aes/). It's
667 an officially adopted standard, has survived peer
668 review, and provides keys up to 256 bits.
669 - Take a careful look at SetACL http://setacl.sourceforge.net
670 - Make tree walk routines like cd, ls, ... more user friendly
671 by handling spaces better.
672 - Make sure that Bacula rechecks the tape after the 20 min wait.
673 - Set IO_NOWAIT on Bacula TCP/IP packets.
674 - Try doing a raw partition backup and restore by mounting a
677 Yes, it would allow to highly automatic the request for new tapes. If a
678 tape is empty, bacula reads the barcodes (native or simulated), and if
679 an unused tape is found, it runs the label command with all the
680 necessary parameters.
682 By the way can bacula automatically "move" an empty/purged volume say
683 in the "short" pool to the "long" pool if this pool runs out of volume
685 - What to do about "list files job=xxx".
686 - Get and test MySQL 4.0
687 - Look at how fuser works and /proc/PID/fd that is how Nic found the
688 file descriptor leak in Bacula.
689 - Implement WrapCounters in Counters.
690 - Add heartbeat from FD to SD if hb interval expires.
691 - Can we dynamically change FileSets?
692 - If pool specified to label command and Label Format is specified,
693 automatically generate the Volume name.
694 - Why can't SQL do the filename sort for restore?
695 - Add ExhautiveRestoreSearch
696 - Look at the possibility of loading only the necessary
697 data into the restore tree (i.e. do it one directory at a
698 time as the user walks through the tree).
699 - Possibly use the hash code if the user selects all for a restore command.
700 - Fix "restore all" to bypass building the tree.
701 - Prohibit backing up archive device (findlib/find_one.c:128)
702 - Implement Release Device in the Job resource to unmount a drive.
703 - Implement Acquire Device in the Job resource to mount a drive,
704 be sure this works with admin jobs so that the user can get
705 prompted to insert the correct tape. Possibly some way to say to
706 run the job but don't save the files.
707 - Make things like list where a file is saved case independent for
710 - Use autochanger to handle multiple devices.
711 - On Windows with very long path names, it may be impossible to create
712 a file (and thus restore it) because the total length is too long.
713 We must cd into the directory then create the file without the
715 - Implement a Recycle command
716 - Test a second language e.g. french.
717 - Start working on Base jobs.
718 - Implement UnsavedFiles DB record.
719 - From Phil Stracchino:
720 It would probably be a per-client option, and would be called
721 something like, say, "Automatically purge obsoleted jobs". What it
722 would do is, when you successfully complete a Differential backup of a
723 client, it would automatically purge all Incremental backups for that
724 client that are rendered redundant by that Differential. Likewise,
725 when a Full backup on a client completed, it would automatically purge
726 all Differential and Incremental jobs obsoleted by that Full backup.
727 This would let people minimize the number of tapes they're keeping on
728 hand without having to master the art of retention times.
729 - When doing a Backup send all attributes back to the Director, who
730 would then figure out what files have been deleted.
731 - Currently in mount.c:236 the SD simply creates a Volume. It should have
732 explicit permission to do so. It should also mark the tape in error
733 if there is an error.
734 - Cancel waiting for Client connect in SD if FD goes away.
736 - Implement timeout in response() when it should come quickly.
737 - Implement a Slot priority (loaded/not loaded).
738 - Implement "vacation" Incremental only saves.
739 - Implement create "FileSet"?
740 - Add prefixlinks to where or not where absolute links to FD.
741 - Issue message to mount a new tape before the rewind.
742 - Simplified client job initiation for portables.
743 - If SD cannot open a drive, make it periodically retry.
744 - Add more of the config info to the tape label.
746 - If tape is marked read-only, then try opening it read-only rather than
747 failing, and remember that it cannot be written.
748 - Refine SD waiting output:
749 Device is being positioned
750 > Device is being positioned for append
751 > Device is being positioned to file x
753 - Figure out some way to estimate output size and to avoid splitting
754 a backup across two Volumes -- this could be useful for writing CDROMs
755 where you really prefer not to have it split -- not serious.
756 - Have SD compute MD5 or SHA1 and compare to what FD computes.
757 - Make VolumeToCatalog calculate an MD5 or SHA1 from the
758 actual data on the Volume and compare it.
759 - Implement Bacula plugins -- design API
760 - Make bcopy read through bad tape records.
761 - Program files (i.e. execute a program to read/write files).
762 Pass read date of last backup, size of file last time.
763 - Add Signature type to File DB record.
764 - CD into subdirectory when open()ing files for backup to
765 speed up things. Test with testfind().
766 - Priority job to go to top of list.
767 - Why are save/restore of device different sizes (sparse?) Yup! Fix it.
768 - Implement some way for the Console to dynamically create a job.
769 - Solaris -I on tar for include list
770 - Need a verbose mode in restore, perhaps to bsr.
771 - bscan without -v is too quiet -- perhaps show jobs.
772 - Add code to reject whole blocks if not wanted on restore.
773 - Check if we can increase Bacula FD priorty in Win2000
774 - Make sure the MaxVolFiles is fully implemented in SD
775 - Check if both CatalogFiles and UseCatalog are set to SD.
776 - Possibly add email to Watchdog if drive is unmounted too
777 long and a job is waiting on the drive.
778 - After unmount, if restore job started, ask to mount.
779 - Add UA rc and history files.
780 - put termcap (used by console) in ./configure and
781 allow -with-termcap-dir.
782 - Fix Autoprune for Volumes to respect need for full save.
783 - Compare tape to Client files (attributes, or attributes and data)
784 - Make all database Ids 64 bit.
785 - Allow console commands to detach or run in background.
786 - Fix status delay on storage daemon during rewind.
787 - Add SD message variables to control operator wait time
788 - Maximum Operator Wait
789 - Minimum Message Interval
790 - Maximum Message Interval
791 - Send Operator message when cannot read tape label.
792 - Verify level=Volume (scan only), level=Data (compare of data to file).
793 Verify level=Catalog, level=InitCatalog
795 - Add keyword search to show command in Console.
796 - Events : tape has more than xxx bytes.
797 - Complete code in Bacula Resources -- this will permit
798 reading a new config file at any time.
799 - Handle ctl-c in Console
800 - Implement script driven addition of File daemon to config files.
801 - Think about how to make Bacula work better with File (non-tape) archives.
802 - Write Unix emulator for Windows.
803 - Put memory utilization in Status output of each daemon
804 if full status requested or if some level of debug on.
805 - Make database type selectable by .conf files i.e. at runtime
806 - Set flag for uname -a. Add to Volume label.
807 - Restore files modified after date
808 - SET LD_RUN_PATH=$HOME/mysql/lib/mysql
809 - Remove duplicate fields from jcr (e.g. jcr.level and jcr.jr.Level, ...).
810 - Timout a job or terminate if link goes down, or reopen link and query.
811 - Concept of precious tapes (cannot be reused).
812 - Make bcopy copy with a single tape drive.
813 - Permit changing ownership during restore.
816 > My suggestion: Add a feature on the systray menu-icon menu to request
817 > an immediate backup now. This would be useful for laptop users who may
818 > not be on the network when the regular scheduled backup is run.
820 > My wife's suggestion: Add a setting to the win32 client to allow it to
821 > shut down the machine after backup is complete (after, of course,
822 > displaying a "System will shut down in one minute, click here to cancel"
823 > warning dialog). This would be useful for sites that want user
824 > woorkstations to be shut down overnight to save power.
827 - Autolabel should be specified by DIR instead of SD.
830 - AutoScan (check checksum of tape)
831 - Format command = "format /dev/nst0"
835 - Seek resolution (usually corresponds to buffer size)
836 - EODErrorCode=ENOSPC or code
837 - Partial Read error code
838 - Partial write error code
839 - Nonformatted read error
840 - Nonformatted write error
841 - WriteProtected error
845 - IgnoreCloseErrors=yes
855 - FD sends unsaved file list to Director at end of job (see
857 - File daemon should build list of files skipped, and then
858 at end of save retry and report any errors.
859 - Write a Storage daemon that uses pipes and
860 standard Unix programs to write to the tape.
862 - Need something that monitors the JCR queue and
863 times out jobs by asking the deamons where they are.
864 - Enhance Jmsg code to permit buffering and saving to disk.
865 - device driver = "xxxx" for drives.
867 - Ensure that /dev/null works
868 - Need report class for messages. Perhaps
869 report resource where report=group of messages
870 - enhance scan_attrib and rename scan_jobtype, and
871 fill in code for "since" option
872 - Director needs a time after which the report status is sent
873 anyway -- or better yet, a retry time for the job.
874 - Don't reschedule a job if previous incarnation is still running.
875 - Some way to automatically backup everything is needed????
876 - Need a structure for pending actions:
878 - termination status (part of buffered msgs?)
880 Read, Write, Clean, Delete
881 - Login to Bacula; Bacula users with different permissions:
882 owner, group, user, quotas
883 - Store info on each file system type (probably in the job header on tape.
884 This could be the output of df; or perhaps some sort of /etc/mtab record.
886 ========= ideas ===============
887 From: "Jerry K. Schieffer" <jerry@skylinetechnology.com>
888 To: <kern@sibbald.com>
889 Subject: RE: [Bacula-users] future large programming jobs
890 Date: Thu, 26 Feb 2004 11:34:54 -0600
892 I noticed the subject thread and thought I would offer the following
893 merely as sources of ideas, i.e. something to think about, not even as
894 strong as a request. In my former life (before retiring) I often
895 dealt with backups and storage management issues/products as a
896 developer and as a consultant. I am currently migrating my personal
897 network from amanda to bacula specifically because of the ability to
898 cross media boundaries during storing backups.
899 Are you familiar with the commercial product called ADSM (I think IBM
900 now sells it under the Tivoli label)? It has a couple of interesting
901 ideas that may apply to the following topics.
903 1. Migration: Consider that when you need to restore a system, there
904 may be pressure to hurry. If all the information for a single client
905 can eventually end up on the same media (and in chronological order),
906 the restore is facillitated by not having to search past information
907 from other clients. ADSM has the concept of "client affinity" that
908 may be associated with it's storage pools. It seems to me that this
909 concept (as an optional feature) might fit in your architecture for
912 ADSM also has the concept of defining one or more storage pools as
913 "copy pools" (almost mirrors, but only in the sense of contents).
914 These pools provide the ability to have duplicte data stored both
915 onsite and offsite. The copy process can be scheduled to be handled
916 by their storage manager during periods when there is no backup
917 activity. Again, the migration process might be a place to consider
918 implementing something like this.
921 > It strikes me that it would be very nice to be able to do things
923 > have the Job(s) backing up the machines run, and once they have all
924 > completed, start a migration job to copy the data from disks Volumes
926 > a tape library and then to offsite storage. Maybe this can already
928 > done with some careful scheduling and Job prioritzation; the events
929 > mechanism described below would probably make it very easy.
931 This is the goal. In the first step (before events), you simply
933 the Migration to tape later.
935 2. Base jobs: In ADSM, each copy of each stored file is tracked in
936 the database. Once a file (unique by path and metadata such as dates,
937 size, ownership, etc.) is in a copy pool, no more copies are made. In
938 other words, when you start ADSM, it begins like your concept of a
939 base job. After that it is in the "incremental" mode. You can
940 configure the number of "generations" of files to be retained, plus a
941 retention date after which even old generations are purged. The
942 database tracks the contents of media and projects the percentage of
943 each volume that is valid. When the valid content of a volume drops
944 below a configured percentage, the valid data are migrated to another
945 volume and the old volume is marked as empty. Note, this requires
946 ADSM to have an idea of the contents of a client, i.e. marking the
947 database when an existing file was deleted, but this would solve your
948 issue of restoring a client without restoring deleted files.
950 This is pretty far from what bacula now does, but if you are going to
951 rip things up for Base jobs,.....
952 Also, the benefits of this are huge for very large shops, especially
953 with media robots, but are a pain for shops with manual media
957 > Base jobs sound pretty useful, but I'm not dying for them.
959 Nobody is dying for them, but when you see what it does, you will die
962 3. Restoring deleted files: Since I think my comments in (2) above
963 have low probability of implementation, I'll also suggest that you
964 could approach the issue of deleted files by a mechanism of having the
965 fd report to the dir, a list of all files on the client for every
966 backup job. The dir could note in the database entry for each file
967 the date that the file was seen. Then if a restore as of date X takes
968 place, only files that exist from before X until after X would be
969 restored. Probably the major cost here is the extra date container in
970 each row of the files table.
972 Thanks for "listening". I hope some of this helps. If you want to
973 contact me, please send me an email - I read some but not all of the
974 mailing list traffic and might miss a reply there.
976 Please accept my compliments for bacula. It is doing a great job for
977 me!! I sympathize with you in the need to wrestle with excelence in
978 execution vs. excelence in feature inclusion.
983 ==============================
986 - Design at hierarchial storage for Bacula. Migration and Clone.
987 - Implement FSM (File System Modules).
988 - Audit M_ error codes to ensure they are correct and consistent.
989 - Add variable break characters to lex analyzer.
990 Either a bit mask or a string of chars so that
991 the caller can change the break characters.
992 - Make a single T_BREAK to replace T_COMMA, etc.
993 - Ensure that File daemon and Storage daemon can
994 continue a save if the Director goes down (this
995 is NOT currently the case). Must detect socket error,
996 buffer messages for later.
997 - Enhance time/duration input to allow multiple qualifiers e.g. 3d2h
998 - Add ability to backup to two Storage devices (two SD sessions) at
999 the same time -- e.g. onsite, offsite.
1000 - Add the ability to consolidate old backup sets (basically do a restore
1001 to tape and appropriately update the catalog). Compress Volume sets.
1002 Might need to spool via file is only one drive is available.
1003 - Compress or consolidate Volumes of old possibly deleted files. Perhaps
1004 someway to do so with every volume that has less than x% valid
1008 Migration: Move a backup from one Volume to another
1009 Clone: Copy a backup -- two Volumes
1011 Bacula Migration is based on Jobs (apparently Networker is file by file).
1013 Migration triggered by:
1017 Highwater mark (keep total size)
1022 ======================================================
1024 It is somewhat like a Full save becomes an incremental since
1025 the Base job (or jobs) plus other non-base files.
1027 - A Base backup is same as Full backup, just different type.
1028 - New BaseFiles table that contains:
1030 BaseJobId - Base JobId referenced for this FileId (needed ???)
1031 JobId - JobId currently running
1032 FileId - File not backed up, exists in Base Job
1033 FileIndex - FileIndex from Base Job.
1034 i.e. for each base file that exists but is not saved because
1035 it has not changed, the File daemon sends the JobId, BaseId,
1036 FileId, FileIndex back to the Director who creates the DB entry.
1037 - To initiate a Base save, the Director sends the FD
1038 the FileId, and full filename for each file in the Base.
1039 - When the FD finds a Base file, he requests the Director to
1040 send him the full File entry (stat packet plus MD5), or
1041 conversely, the FD sends it to the Director and the Director
1042 says yes or no. This can be quite rapid if the FileId is kept
1043 by the FD for each Base Filename.
1044 - It is probably better to have the comparison done by the FD
1045 despite the fact that the File entry must be sent across the
1047 - An alternative would be to send the FD the whole File entry
1048 from the start. The disadvantage is that it requires a lot of
1049 space. The advantage is that it requires less communications
1051 - The Job record must be updated to indicate that one or more
1053 - At end of Job, FD returns:
1054 1. Count of base files/bytes not written to tape (i.e. matches)
1055 2. Count of base file that were saved i.e. had changed.
1056 - No tape record would be written for a Base file that matches, in the
1057 same way that no tape record is written for Incremental jobs where
1058 the file is not saved because it is unchanged.
1059 - On a restore, all the Base file records must explicitly be
1060 found from the BaseFile tape. I.e. for each Full save that is marked
1061 to have one or more Base Jobs, search the BaseFile for all occurrences
1063 - An optimization might be to make the BaseFile have:
1069 This would avoid the need to explicitly fetch each File record for
1070 the Base job. The Base Job record will be fetched to get the
1071 VolSessionId and VolSessionTime.
1072 =========================================================
1075 ==========================================================
1077 For each Incremental job that is run, there may be files that
1078 were found but not saved because they were locked (this applies
1079 only to Windows). Such a system could send back to the Director
1080 a list of Unsaved files.
1082 - New UnSavedFiles table that contains:
1086 - Then in the next Incremental job, the list of Unsaved Files will be
1087 feed to the FD, who will ensure that they are explicitly chosen even
1088 if standard date/time check would not have selected them.
1089 =============================================================
1093 Multiple drive autochanger data: see Alan Brown
1094 mtx -f xxx unloadStorage Element 1 is Already Full(drive 0 was empty)
1095 Unloading Data Transfer Element into Storage Element 1...source Element
1096 Address 480 is Empty
1098 (drive 0 was empty and so was slot 1)
1099 > mtx -f xxx load 15 0
1100 no response, just returns to the command prompt when complete.
1101 > mtx -f xxx status Storage Changer /dev/changer:2 Drives, 60 Slots ( 2 Import/Export )
1102 Data Transfer Element 0:Full (Storage Element 15 Loaded):VolumeTag = HX001
1103 Data Transfer Element 1:Empty
1104 Storage Element 1:Empty
1105 Storage Element 2:Full :VolumeTag=HX002
1106 Storage Element 3:Full :VolumeTag=HX003
1107 Storage Element 4:Full :VolumeTag=HX004
1108 Storage Element 5:Full :VolumeTag=HX005
1109 Storage Element 6:Full :VolumeTag=HX006
1110 Storage Element 7:Full :VolumeTag=HX007
1111 Storage Element 8:Full :VolumeTag=HX008
1112 Storage Element 9:Full :VolumeTag=HX009
1113 Storage Element 10:Full :VolumeTag=HX010
1114 Storage Element 11:Empty
1115 Storage Element 12:Empty
1116 Storage Element 13:Empty
1117 Storage Element 14:Empty
1118 Storage Element 15:Empty
1119 Storage Element 16:Empty....
1120 Storage Element 28:Empty
1121 Storage Element 29:Full :VolumeTag=CLNU01L1
1122 Storage Element 30:Empty....
1123 Storage Element 57:Empty
1124 Storage Element 58:Full :VolumeTag=NEX261L2
1125 Storage Element 59 IMPORT/EXPORT:Empty
1126 Storage Element 60 IMPORT/EXPORT:Empty
1128 Unloading Data Transfer Element into Storage Element 15...done
1130 (just to verify it remembers where it came from, however it can be
1131 overrriden with mtx unload {slotnumber} to go to any storage slot.)
1133 There needs to be a table of drive # to devices somewhere - If there are
1134 multiple changers or drives there may not be a 1:1 correspondance between
1135 changer drive number and system device name - and depending on the way the
1136 drives are hooked up to scsi busses, they may not be linearly numbered
1137 from an offset point either.something like
1139 Autochanger drives = 2
1140 Autochanger drive 0 = /dev/nst1
1141 Autochanger drive 1 = /dev/nst2
1142 IMHO, it would be _safest_ to use explicit mtx unload commands at all
1143 times, not just for multidrive changers. For a 1 drive changer, that's
1149 MTX's manpage (1.2.15):
1150 unload [<slotnum>] [ <drivenum> ]
1151 Unloads media from drive <drivenum> into slot
1152 <slotnum>. If <drivenum> is omitted, defaults to
1153 drive 0 (as do all commands). If <slotnum> is
1154 omitted, defaults to the slot that the drive was
1155 loaded from. Note that there's currently no way
1156 to say 'unload drive 1's media to the slot it
1157 came from', other than to explicitly use that
1158 slot number as the destination.AB
1164 undef# camcontrol devlist
1165 <WANGTEK 51000 SCSI M74H 12B3> at scbus0 target 2 lun 0 (pass0,sa0)
1166 <ARCHIVE 4586XX 28887-XXX 4BGD> at scbus0 target 4 lun 0 (pass1,sa1)
1167 <ARCHIVE 4586XX 28887-XXX 4BGD> at scbus0 target 4 lun 1 (pass2)
1169 tapeinfo -f /dev/sg0 with a bad tape in drive 1:
1170 [kern@rufus mtx-1.2.17kes]$ ./tapeinfo -f /dev/sg0
1171 Product Type: Tape Drive
1173 Product ID: 'C5713A '
1175 Attached Changer: No
1176 TapeAlert[3]: Hard Error: Uncorrectable read/write error.
1177 TapeAlert[20]: Clean Now: The tape drive neads cleaning NOW.
1184 Medium Type: Not Loaded
1187 DataCompEnabled: yes
1188 DataCompCapable: yes
1189 DataDeCompEnabled: yes
1196 Handling removable disks
1198 From: Karl Cunningham <karlc@keckec.com>
1200 My backups are only to hard disk these days, in removable bays. This is my
1201 idea of how a backup to hard disk would work more smoothly. Some of these
1202 things Bacula does already, but I mention them for completeness. If others
1203 have better ways to do this, I'd like to hear about it.
1205 1. Accommodate several disks, rotated similar to how tapes are. Identified
1206 by partition volume ID or perhaps by the name of a subdirectory.
1207 2. Abort & notify the admin if the wrong disk is in the bay.
1208 3. Write backups to different subdirectories for each machine to be backed
1210 4. Volumes (files) get created as needed in the proper subdirectory, one
1212 5. When a disk is recycled, remove or zero all old backup files. This is
1213 important as the disk being recycled may be close to full. This may be
1214 better done manually since the backup files for many machines may be
1215 scattered in many subdirectories.
1220 - Save mount point for directories not traversed with onefs=yes.
1221 - Add seconds to start and end times in the Job report output.
1222 - if 2 concurrent backups are attempted on the same tape
1223 drive (autoloader) into different tape pools, one of them will exit
1224 fatally instead of halting until the drive is idle
1225 - Update StartTime if job held in Job Queue.
1226 - Look at www.nu2.nu/pebuilder as a helper for full windows
1227 bare metal restore. (done by Scott)
1228 - Fix orphanned buffers:
1229 Orphaned buffer: 24 bytes allocated at line 808 of rufus-dir job.c
1230 Orphaned buffer: 40 bytes allocated at line 45 of rufus-dir alist.c
1231 - Implement Preben's suggestion to add
1232 File System Types = ext2, ext3
1233 to FileSets, thus simplifying backup of *all* local partitions.
1234 - Try to open a device on each Job if it was not opened
1235 when the SD started.
1236 - Add dump of VolSessionId/Time and FileIndex with bls.
1237 - If Bacula does not find the right tape in the Autochanger,
1238 then mark the tape in error and move on rather than asking
1239 for operator intervention.
1240 - Cancel command should include JobId in list of Jobs.
1241 - Add performance testing hooks
1242 - Bootstrap from JobMedia records.
1243 - Implement WildFile and WildDir to solve problem of
1244 saving only *.doc files.
1246 Please use the "label" command to create a new Volume for:
1247 Storage: DDS-4-changer
1251 The defined Storage resources are:
1252 - Copy Changer Device and Changer Command from Autochanger
1253 to Device resource in SD if none given in Device resource.
1254 - 1. Automatic use of more than one drive in an autochanger (done)
1255 - 2. Automatic selection of the correct drive for each Job (i.e.
1256 selects a drive with an appropriate Volume for the Job) (done)
1257 - 6. Allow multiple simultaneous Jobs referencing the same pool write
1258 to several tapes (some new directive(s) are are probably needed for
1261 - Key on Storage rather than Pool (done)
1262 - Allow multiple drives to use same Pool (change jobq.c DIR) (done).
1263 - Synchronize multiple drives so that not more
1264 than one loads a tape and any time (done)
1265 - 4. Use Changer Device and Changer Command specified in the
1266 Autochanger resource, if none is found in the Device resource.
1267 You can continue to specify them in the Device resource if you want
1268 or need them to be different for each device.
1269 - 5. Implement a new Device directive (perhaps "Autoselect = yes/no")
1270 that can allow a Device be part of an Autochanger, and hence the changer
1271 script protected, but if set to no, will prevent the Device from being
1272 automatically selected from the changer. This allows the device to
1273 be directly accessed through its Device name, but not through the
1275 #6 Select one from among Multiple Storage Devices for Job
1276 #5 Events that call a Python program
1277 (Implemented in Dir/SD)
1278 - Make sure the Device name is in the Query packet returned.
1279 - Don't start a second file job if one is already running.
1280 - Implement EOF/EOV labels for ANSI labels
1281 - Implement IBM labels.
1282 - When Python creates a new label, the tape is immediately
1283 recycled and no label created. This happens when using
1284 autolabeling -- even when Python doesn't generate the name.
1285 - Scratch Pool where the volumes can be re-assigned to any Pool.
1286 - 28-Mar 23:19 rufus-sd: acquire.c:379 Device "DDS-4" (/dev/nst0)
1287 is busy reading. Job 6 canceled.
1288 - Remove separate thread for opening devices in SD. On the other
1289 hand, don't block waiting for open() for devices.
1290 - Fix code to either handle updating NumVol or to calculate it in
1292 - Ensure that you cannot exclude a directory or a file explicitly
1294 #4 Embedded Python Scripting
1295 (Implemented in Dir/SD/FD)
1296 - Add Python writable variable for changing the Priority,
1297 Client, Storage, JobStatus (error), ...
1300 - Add disk seeking on restore; turn off seek on tapes.
1302 - Look at dird_conf.c:1000: warning: `int size'
1303 might be used uninitialized in this function
1304 - Indicate when a Job is purged/pruned during restore.
1305 - Implement some way to turn off automatic pruning in Jobs.
1306 - Implement a way an Admin Job can prune, possibly multiple
1307 clients -- Python script?
1308 - Look at Preben's acl.c error handling code.
1309 - SD crashes after a tape restore then doing a backup.
1310 - If drive is opened read/write, close it and re-open
1311 read-only if doing a restore, and vice-versa.
1313 data-fd: RestoreFiles.2004-12-07_15.56.42 Error:
1314 > ..\findlib\../../findlib/create_file.c:275 Could not open e:/: ERR=Der
1315 > Prozess kann nicht auf die Datei zugreifen, da sie von einem anderen
1316 > Prozess verwendet wird.
1317 Restore restores all files, but then fails at the end trying
1318 to set the attributes of e:
1319 from failed jobs.- Resolve the problem between Device name and Archive name,
1320 and fix SD messages.
1321 - Tell the "restore" user when browsing is no longer possible.
1322 - Add a restore directory-x
1323 - Write non-optimized bsrs from the JobMedia and Media records,
1324 even after Files are pruned.
1325 - Delete Stripe and Copy from VolParams to save space.
1326 - Fix option 2 of restore -- list where file is backed up -- require Client,
1327 then list last 20 backups.
1328 - Finish implementation of passing all Storage and Device needs to
1330 - Move test for max wait time exceeded in job.c up -- Peter's idea.
1331 ## Consider moving docs to their own project.
1332 ## Move rescue to its own project.
1333 - Add client version to the Client name line that prints in
1335 - Fix the Rescue CDROM.
1336 - By the way: on page http://www.bacula.org/?page=tapedrives , at the
1337 bottom, the link to "Tape Testing Chapter" is broken. It goes to
1338 /html-manual/... while the others point to /rel-manual/...
1339 - Device resource needs the "name" of the SD.
1340 - Specify a single directory to restore.
1341 - Implement MediaType keyword in bsr?
1342 - Add a date and time stamp at the beginning of every line in the
1343 Job report (Volker Sauer).
1344 - Add level to estimate command.
1345 - Add "limit=n" for "list jobs"
1346 - Make bootstrap filename unique.
1347 - Make Dmsg look at global before calling subroutine.
1349 it seems to be complaining about 12:00pm which should be a valid 12
1350 hour time. I changed the time to 11:59am and everything works fine.
1351 Also 12:00am works fine. 0:00pm also works (which I don't think
1352 should). None of the values 12:00pm - 12:59pm work for that matter.
1353 - Require restore via the restore command or make a restore Job
1354 get the bootstrap file.
1355 - Implement Maximum Job Spool Size
1356 - Fix 3993 error in SD. It forgets to look at autochanger
1357 resource for device command, ...
1358 - 3. Prevent two drives requesting the same Volume in any given
1359 autochanger, by checking if a Volume is mounted on another drive
1361 - Upgrade to MySQL 4.1.12 See:
1362 http://dev.mysql.com/doc/mysql/en/Server_SQL_mode.html
1363 - Add # Job Level date to bsr file
1364 - Implement "PreferMountedVolumes = yes|no" in Job resource.
1365 ## Integrate web-bacula into a new Bacula project with
1367 - Cleaning tapes should have Status "Cleaning" rather than append.
1368 - Make sure that Python has access to Client address/port so that
1369 it can check if Clients are alive.
1370 - Review all items in "restore".
1371 - Fix PostgreSQL GROUP BY problems in restore.
1372 - Fix PostgreSQL sql problems in bugs.
1374 04-Jul 13:01 MainSD: Rufus.2005-07-04_01.05.02 Warning: Director wanted Volume
1376 Current Volume "DLT-04Jul05" not acceptable because:
1377 1997 Volume "DLT-13Feb04" not in catalog.
1378 04-Jul 13:01 MainSD: Please mount Volume "DLT-04Jul05" on Storage Device
1379 "HP DLT 80" (/dev/nst0) for Job Rufus.2005-07-04_01.05.02
1380 ## Create a new GUI chapter explaining all the GUI programs.
1381 - Make "update slots" when pointing to Autochanger, remove
1382 all Volumes from other drives. "update slots all-drives"?
1383 No, this is done by modifying mtx-changer to list what is
1385 - Finish TLS implementation.
1386 - Port limiting -m in iptables to prevent DoS attacks
1387 could cause broken pipes on Bacula.
1388 6. Build and test the Volume Shadow Copy (VSS) for Win32.
1389 - Allow cancel of unknown Job
1390 - State not saved when closing Win32 FD by icon
1391 - bsr-opt-test fails. bsr deleted. Fix.
1392 - Move Python daemon variables from Job to Bacula object.
1393 WorkingDir, ConfigFile
1394 - Document that Bootstrap files can be written with cataloging
1396 - Document details of ANSI/IBM labels
1398 1) ADIC, DLT, FastStor 4000, 7*20GB
1399 - Linux Sony LIB-D81, AIT-3 library works.
1401 to activate, check or disable the hardware compression feature on my
1402 exb-8900 i use the exabyte "MammothTool" you can get it here:
1403 http://www.exabyte.com/support/online/downloads/index.cfm
1404 There is a solaris version of this tool. With option -C 0 or 1 you can
1405 disable or activate compression. Start this tool without any options for
1407 - Document Heartbeat Interval in the dealing with firewalls section.
1408 - Document new CDROM directory.
1409 - On Win32 working directory must have drive letter ????
1410 - On Win32 working directory must be writable by SYSTEM to
1412 - Document that ChangerDevice is used for Alert command.
1413 - Add better documentation on how restores can be done
1414 8. Take one more try at making DVD writing work (no go)
1415 7. Write a bacula-web document
1416 - Why isn't the DEVICE structure defined when doing
1418 - Multi-drive changer seems to only use drive 0
1419 Multiple drives don't seem to be opened.