8 Data encryption Meno Abels (see projects)
9 Communication encryption Meno Abels
10 Version 1.37 Kern (see below)
11 ========================================================
14 #3 Migration (Move, Copy, Archive Jobs)
15 #4 Embedded Python Scripting (implemented in Dir)
16 #5 Events that call a Python program (Implemented in Dir)
17 #6 Select one from among Multiple Storage Devices for Job
18 #7 Single Job Writing to Multiple Storage Devices
20 ## Integrate web-bacula into a new Bacula project with
22 ## Consider moving docs to their own project.
24 Suggestions for Preben:
25 - Look at adding Client run command that will use the
26 port opened by the client.
27 - Implement WildFile and WildDir to solve problem of
28 saving only *.doc files.
29 - Bootstrap from JobMedia records.
30 - Optimized bootstrap.
33 - See multiple-store.txt for Multiple Storage implementation
35 - Add Set Error from Python.
36 - Create a new GUI chapter explaining all the GUI programs.
37 - Tell the "restore" user when browsing is no longer possible.
38 - Add disk seeking on restore.
39 - Cancel command should include JobId in list of Jobs.
40 - Require restore via the restore command or make a restore Job
41 get the bootstrap file.
42 - Add dump of VolSessionId/Time and FileIndex with bls.
43 - Make bootstrap file handle multiple MediaTypes (SD)
44 - Add offline command to Bacula console.
45 - Add performance testing hooks
46 - Add Python writable variable for changing the Priority.
47 - Document that Bootstrap files can be written with cataloging
49 - Look at adding full Volume and Pool information to a Volume
50 label so that bscan can get *all* the info.
51 - Scratch Pool where the volumes can be re-assigned to any Pool.
52 - Upgrade to MySQL 4.1.1 See:
53 http://dev.mysql.com/doc/mysql/en/Server_SQL_mode.html
55 Enter MediaId or Volume name: 32
56 Enter new Volume name: DLT-20Dec04
57 Automatically selected Pool: Default
58 Connecting to Storage daemon DLTDrive at 192.168.68.104:9103 ...
59 Sending relabel command from "DLT-28Jun03" to "DLT-20Dec04" ...
60 block.c:552 Write error at 0:0 on device /dev/nst0. ERR=Bad file descriptor.
61 Error writing final EOF to tape. This tape may not be readable.
62 dev.c:1207 ioctl MTWEOF error on /dev/nst0. ERR=Permission denied.
63 askdir.c:219 NULL Volume name. This shouldn't happen!!!
64 3912 Failed to label Volume: ERR=dev.c:1207 ioctl MTWEOF error on /dev/nst0. ERR=Permission denied.
66 Label command failed for Volume DLT-20Dec04.
67 Do not forget to mount the drive!!!
68 - Bug: if a job is manually scheduled to run later, it does not appear
69 in any status report and cannot be cancelled.
71 Regression tests (Jo):
72 - Add Pool/Storage override regression test.
73 - Add delete JobId to regression.
74 - Add a regression test for dbcheck.
75 - New test to add bscan to four-concurrent-jobs regression,
76 i.e. after the four-concurrent jobs zap the
77 database as is done in the bscan-test, then use bscan to
78 restore the database, do a restore and compare with the
80 - Add restore of specific JobId to regression (item 3
81 on the restore prompt)
82 - Add IPv6 to regression
83 - Add database test to regression. Test each function like delete,
86 - AntiVir can slow down backups on Win32 systems.
87 - Win32 systems with FAT32 can be much slower than NTFS for
88 more than 1000 files per directory.
92 - Can one write tapes faster with 8192 byte block sizes?
93 - Specify a single directory to restore.
94 - Implement Maximum Job Spool Size
95 - Document security problems with the same password for everyone in
96 rpm and Win32 releases.
97 - Browse generations of files.
98 - I've seen an error when my catalog's File table fills up. I
99 then have to recreate the File table with a larger maximum row
100 size. Relevant information is at
101 http://dev.mysql.com/doc/mysql/en/Full_table.html ; I think the
102 "Installing and Configuring MySQL" chapter should talk a bit
103 about this potential problem, and recommend a solution.
104 - For Solaris must use POSIX awk.
105 - Want speed of writing to tape while despooling.
106 - Supported autochanger:
114 Wangtek 6525ES (SCSI-1 QIC drive, 525MB), under Linux 2.4.something,
115 bacula 1.36.0/1 works with blocksize 16k INSIDE bacula-sd.conf.
116 - Add regex from http://www.pcre.org to Bacula for Win32.
117 - Use only shell tools no make in CDROM package.
118 - Document PostgreSQL performance problems bug 131.
120 - Document that ChangerDevice is used for Alert command.
121 - Include within include does it work?
122 - Implement a Pool of type Cleaning?
123 - Implement VolReadTime and VolWriteTime in SD
124 - Modify Backing up Your Database to include a bootstrap file.
125 - Think about making certain database errors fatal.
126 - Look at correcting the time jump in the scheduler for daylight
127 savings time changes.
128 - Add a "real" timer to network connections.
129 - Write non-optimized bsrs from the JobMedia and Media records,
130 even after Files are pruned.
131 - Promote to Full = Time period
132 - Despool attributes simultaneously with data in a separate
133 thread, rejoined at end of data spooling.
134 - Implement Files/Bytes,... stats for restore job.
135 - Implement Total Bytes Written, ... for restore job.
136 - Check dates entered by user for correctness (month/day/... ranges)
137 - Compress restore Volume listing by date and first file.
138 - Look at patches/bacula_db.b2z postgresql that loops during restore.
140 - Perhaps add read/write programs and/or plugins to FileSets.
141 - How to handle backing up portables ...
142 - Add some sort of guaranteed Interval for upgrading jobs.
143 - Can we write the state file after every job terminates? On Win32
144 the system crashes and the state file is not updated.
147 Documentation to do: (any release a little bit at a time)
148 - Doc to do unmount before removing magazine.
149 - Alternative to static linking "ldd prog" save all binaries listed,
150 restore them and point LD_LIBRARY_PATH to them.
151 - Document add "</dev/null >/dev/null 2>&1" to the bacula-fd command line
152 - Document query file format.
153 - Add more documentation for bsr files.
154 - Document problems with Verify and pruning.
155 - Document how to use multiple databases.
156 - VXA drives have a "cleaning required"
157 indicator, but Exabyte recommends preventive cleaning after every 75
160 In this context, it should be noted that Exabyte has a command-line
161 vxatool utility available for free download. (The current version is
162 vxatool-3.72.) It can get diagnostic info, read, write and erase tapes,
163 test the drive, unload tapes, change drive settings, flash new firmware,
165 Of particular interest in this context is that vxatool <device> -i will
166 report, among other details, the time since last cleaning in tape motion
167 minutes. This information can be retrieved (and settings changed, for
168 that matter) through the generic-SCSI device even when Bacula has the
169 regular tape device locked. (Needless to say, I don't recommend
170 changing tape settings while a job is running.)
171 - Lookup HP cleaning recommendations.
172 - Lookup HP tape replacement recommendations (see trouble shooting autochanger)
173 - Document doing table repair
176 Testing to do: (painful)
179 For 1.37 Testing/Documentation:
180 - Add bandwidth limiting.
181 - Add acks every once and a while from the SD to keep
182 the line from timing out.
183 - When an error in input occurs and conio beeps, you can back
184 up through the prompt.
185 - Detect fixed tape block mode during positioning by looking at
186 block numbers in btape "test". Possibly adjust in Bacula.
187 - Fix list volumes to output volume retention in some other
188 units, perhaps via a directive.
189 - If opening a tape in read/write mode fails attempt to open
190 it in read-only mode, and mark the tape for read only.
191 - Add a read-only mode to the mount option.
192 - Allow Simultaneous Priorities = yes => run up to Max concurrent jobs even
193 with multiple priorities.
194 - If you use restore replace=never, the directory attributes for
195 non-existent directories will not be restored properly.
197 - see lzma401.zip in others directory for new compression
199 - Minimal autochanger handling in Bacula and in btape.
200 - Look into how tar does not save sockets and the possiblity of
201 not saving them in Bacula (Martin Simmons reported this).
202 - Fix restore jobs so that multiple jobs can run if they
203 are not using the same tape(s).
204 - Allow the user to select JobType for manual pruning/purging.
205 - bscan does not put first of two volumes back with all info in
207 - Implement the FreeBSD nodump flag in chflags.
208 - Figure out how to make named console messages go only to that
209 console and to the non-restricted console (new console class?).
210 - Make restricted console prompt for password if *ask* is set or
211 perhaps if password is undefined.
212 - Implement "from ISO-date/time every x hours/days/weeks/months" in
217 MA = (last_MA * 3 + rate) / 4
218 rate = (bytes - last_bytes) / (runtime - last_runtime)
220 ==== from Marc Schoechlin
221 - the help-command should be more verbose
222 (it should explain the paramters of the different
224 -> it´s time-comsuming to consult the manual anytime
225 you need a special parameter
226 -> maybe it´s more easy to maintain this, if the
227 descriptions of that commands are outsourced to
229 - the cd-command should allow complete paths
230 i.e. cd /foo/bar/foo/bar
231 -> if a customer mails me the path to a certain file,
232 it´s faster to enter the specified directory
233 - if the password is not configured in bconsole.conf
234 you should be asked for it.
235 -> sometimes you like to do restore on a customer-machine
236 which shouldn´t know the password for bacula.
237 -> adding the password to the file favours admins
238 to forget to remove the password after usage
240 the protection of that file is less important
241 - long-listed-output of commands should be scrollable
242 like the unix more/less-command does
243 -> if someone runs 200 and more machines, the lists could
244 be a little long and complex
245 - command-output should be shown column by column
246 to reduce scrolling and to increase clarity
248 - lsmark should list the selected files with full
250 - wildcards for selecting and file and directories would be nice
251 - any actions should be interuptable with STRG+C
252 - command-expansion would be pretty cool
254 - Compaction of Disk space by "migrating" Volumes that have pruned
255 Jobs (what criteria? size, #jobs, time).
256 - Add prune all command
257 - Document fact that purge can destroy a part of a restore by purging
258 one volume while others remain valid -- perhaps mark Jobs.
259 - Add multiple-media-types.txt
260 - look at mxt-changer.html
261 - Make ? do a help command (no return needed).
262 - Implement restore directory.
263 - Document streams and how to implement them.
264 - Try not to re-backup a file if a new hard link is added.
265 - Add feature to backup hard links only, but not the data.
266 - Fix stream handling to be simpler.
267 - Add Priority and Bootstrap to Run a Job.
268 - Eliminate Restore "Run Restore Job" prompt by allowing new "run command
270 - Remove View FileSet button from Run a Job dialog.
271 - Handle prompt for restore job at end of Restore command.
272 - Add display of total selected files to Restore window.
273 - Add tree pane to left of window.
274 - Add progress meter.
275 - Max wait time or max run time causes seg fault -- see runtime-bug.txt
276 - Document writing to a CD/DVD with Bacula.
277 - Add a "base" package to the window installer for pthreadsVCE.dll
278 which is needed by all packages.
279 - Add message to user to check for fixed block size when the forward
280 space test fails in btape.
281 - When unmarking a directory check if all files below are unmarked and
282 then remove the + flag -- in the restore tree.
283 - Possibly implement: Action = Unmount Device="TapeDrive1" in Admin jobs.
284 - Setup lrrd graphs: (http://www.linpro.no/projects/lrrd/) Mike Acar.
285 - Revisit the question of multiple Volumes (disk) on a single device.
286 - Add a block copy option to bcopy.
287 - Investigate adding Mac Resource Forks.
288 - Finish work on Gnome restore GUI.
289 - Fix "llist jobid=xx" where no fileset or client exists.
291 it seems to be complaining about 12:00pm which should be a valid 12
292 hour time. I changed the time to 11:59am and everything works fine.
293 Also 12:00am works fine. 0:00pm also works (which I don't think
294 should). None of the values 12:00pm - 12:59pm work for that matter.
295 - Add level to estimate command.
296 - For each job type (Admin, Restore, ...) require only the really necessary
298 - Fix option 2 of restore -- list where file is backed up -- require Client,
299 then list last 20 backups.
300 - Pass Director resource name as an option to the Console.
301 - Add a "batch" mode to the Console (no unsolicited queries, ...).
302 - Add a .list all files in the restore tree (probably also a list all files)
303 Do both a long and short form.
304 - Allow browsing the catalog to see all versions of a file (with
305 stat data on each file).
306 - Restore attributes of directory if replace=never set but directory
308 - Use SHA1 on authentication if possible.
309 - See comtest-xxx.zip for Windows code to talk to USB.
310 - Add Events and Perl scripting.
311 - Add John's appended files:
312 Appended = { /files/server/logs/http/*log }
313 and such files would be treated as follows.On a FULL backup, they would
314 be backed up like any other file.On an INCREMENTAL backup, where a
315 previous INCREMENTAL or FULL was already in thecatalogue and the length
316 of the file wasgreater than the length of the last backup, only thedata
317 added since the last backup will be dumped.On an INCREMENTAL backup, if
318 the length of the file is less than thelength of the file with the same
319 name last backed up, the completefile is dumped.On Windows systems, with
320 creation date of files, we can be evensmarter about this and not count
321 entirely upon the length.On a restore, the full and all incrementals
322 since it will beapplied in sequence to restore the file.
323 - Allow for optional cancelling of SD and FD in case DIR
324 gets a fatal error. Requested by Jesse Guardiani <jesse@wingnet.net>
325 - Add "limit=n" for "list jobs"
326 - Check new HAVE_WIN32 open bits.
327 - Check if the tape has moved before writing.
328 - Handling removable disks -- see below:
329 - Keep track of tape use time, and report when cleaning is necessary.
330 - Add FromClient and ToClient keywords on restore command (or
331 BackupClient RestoreClient).
332 - Implement a JobSet, which groups any number of jobs. If the
333 JobSet is started, all the jobs are started together.
334 Allow Pool, Level, and Schedule overrides.
335 - Enhance cancel to timeout BSOCK packets after a specific delay.
336 - Do scheduling by UTC using gmtime_r() in run_conf, scheduler, and
337 ua_status.!!! Thanks to Alan Brown for this tip.
338 - Look at updating Volume Jobs so that Max Volume Jobs = 1 will work
339 correctly for multiple simultaneous jobs.
340 - Correct code so that FileSet MD5 is calculated for < and | filename
342 - Implement the Media record flag that indicates that the Volume does disk
344 - Implement VolAddr, which is used when Volume is addressed like a disk,
345 and form it from VolFile and VolBlock.
346 - Make multiple restore jobs for multiple media types specifying
347 the proper storage type.
348 - Implement MediaType keyword in bsr?
349 - Fix fast block rejection (stored/read_record.c:118). It passes a null
350 pointer (rec) to try_repositioning().
351 - Look at extracting Win data from BackupRead.
352 - Implement RestoreJobRetention? Maybe better "JobRetention" in a Job,
353 which would take precidence over the Catalog "JobRetention".
354 - Implement Label Format in Add and Label console commands.
355 - Possibly up network buffers to 65K. Put on variable.
356 - Put email tape request delays on one or more variables. User wants
357 to cancel the job after a certain time interval. Maximum Mount Wait?
358 - Job, Client, Device, Pool, or Volume?
359 Is it possible to make this a directive which is *optional* in multiple
360 resources, like Level? If so, I think I'd make it an optional directive
361 in Job, Client, and Pool, with precedence such that Job overrides Client
362 which in turn overrides Pool.
364 - New Storage specifications:
365 - Want to write to multiple storage devices simultaneously
366 - Want to write to multiple storage devices sequentially (in one job)
367 - Want to read/write simultaneously
368 - Key is MediaType -- it must match
370 Passed to SD as a sort of BSR record called Storage Specification
374 MediaType -> Next MediaType
376 Device -> Next Device
378 Allow multiple Storage specifications
386 Allow Multiple Pool specifications (note, Pool currently
388 Allow Multiple MediaType specifications in Dir conf
389 Allow Multiple Device specifications in Dir conf
390 Perhaps keep this in a single SSR
391 Tie a Volume to a specific device by using a MediaType that
392 is contained in only one device.
393 In SD allow Device to have Multiple MediaTypes
395 - Ideas from Jerry Scharf:
396 First let's point out some big pluses that bacula has for this
398 more importantly it's active. Thank you so much for that
399 even more important, it's not flaky
400 it has an open access catalog, opening many possibilities
401 it's pushing toward heterogeneous systems capability
403 Macintosh file client
404 macs are an interesting niche, but I fear a server is a rathole
405 working bare iron recovery for windows
406 the option for inc/diff backups not reset on fileset revision
407 a) use both change and inode update time against base time
408 b) do the full catalog check (expensive but accurate)
409 sizing guide (how much system is needed to back up N systems/files)
410 consultants on using bacula in building a disaster recovery system
412 or how to get at fancy things that one could do with bacula
413 logwatch code for bacula logs (or similar)
414 linux distro inclusion of bacula (brings good and bad, but necessary)
415 win2k/XP server capability (icky but you asked)
416 support for Oracle database ??
418 - Look at adding SQL server and Exchange support for Windows.
419 - Each DVD-RAM disk would be a volume, just like each tape is
420 a volume. It's a 4.7GB media with random access, but there's nothing about
421 it that I can see that makes it so different than a tape from bacula's
422 perspective. Why couldn't I back up to a bare floppy as a volume (ignoring
424 - Make dev->file and dev->block_num signed integers so that -1 can
425 be an invalid value which happens with BSR.
426 - Create VolAddr for disk files in place of VolFile and VolBlock. This
427 is needed to properly specify ranges.
428 - Add progress of files/bytes to SD and FD.
429 - Print warning message if FileId > 4 billion
430 - do a "messages" before the first prompt in Console
431 - Add a date and time stamp at the beginning of every line in the
432 Job report (Volker Sauer).
433 - Client does not show busy during Estimate command.
434 - Implement Console mtx commands.
435 - Implement a Mount Command and an Unmount Command where
436 the users could specify a system command to be performed
437 to do the mount, after which Bacula could attempt to
438 read the device. This is for Removeable media such as a CDROM.
439 - Most likely, this mount command would be invoked explicitly
440 by the user using the current Console "mount" and "unmount"
441 commands -- the Storage Daemon would do the right thing
442 depending on the exact nature of the device.
443 - As with tape drives, when Bacula wanted a new removable
444 disk mounted, it would unmount the old one, and send a message
445 to the user, who would then use "mount" as described above
446 once he had actually inserted the disk.
447 - Implement dump/print label to UA
448 - Spool to disk only when the tape is full, then when a tape is hung move
450 - bextract is sending everything to the log file ****FIXME****
451 - Allow multiple Storage specifications (or multiple names on
452 a single Storage specification) in the Job record. Thus a job
453 can be backed up to a number of storage devices.
454 - Implement some way for the File daemon to contact the Director
455 to start a job or pass its DHCP obtained IP number.
456 - Implement a query tape prompt/replace feature for a console
457 - Copy console @ code to gnome2-console
458 - Make AES the only encryption algorithm see
459 http://csrc.nist.gov/CryptoToolkit/aes/). It's
460 an officially adopted standard, has survived peer
461 review, and provides keys up to 256 bits.
462 - Take a careful look at SetACL http://setacl.sourceforge.net
463 - Take a careful look at Level for the estimate command, maybe make
464 it a command line option.
465 - Add Volume name to "I cannot write on this volume because"
466 - Make tree walk routines like cd, ls, ... more user friendly
467 by handling spaces better.
468 - Make sure that Bacula rechecks the tape after the 20 min wait.
469 - Set IO_NOWAIT on Bacula TCP/IP packets.
470 - Try doing a raw partition backup and restore by mounting a
473 Yes, it would allow to highly automatic the request for new tapes. If a
474 tape is empty, bacula reads the barcodes (native or simulated), and if
475 an unused tape is found, it runs the label command with all the
476 necessary parameters.
478 By the way can bacula automatically "move" an empty/purged volume say
479 in the "short" pool to the "long" pool if this pool runs out of volume
481 - What to do about "list files job=xxx".
482 - Make Dmsg look at global before calling subroutine.
483 - Get and test MySQL 4.0
484 - Look at how fuser works and /proc/PID/fd that is how Nic found the
485 file descriptor leak in Bacula.
486 - Implement WrapCounters in Counters.
487 - Add heartbeat from FD to SD if hb interval expires.
488 - Can we dynamically change FileSets?
489 - If pool specified to label command and Label Format is specified,
490 automatically generate the Volume name.
491 - Why can't SQL do the filename sort for restore?
492 - Look into Microsoft Volume Shadowcopy Service VSS for backing
493 up system state components (Active Directory, System Volume, ...)
494 - Add ExhautiveRestoreSearch
495 - Look at the possibility of loading only the necessary
496 data into the restore tree (i.e. do it one directory at a
497 time as the user walks through the tree).
498 - Possibly use the hash code if the user selects all for a restore command.
499 - Fix "restore all" to bypass building the tree.
500 - Prohibit backing up archive device (findlib/find_one.c:128)
501 - Implement Release Device in the Job resource to unmount a drive.
502 - Implement Acquire Device in the Job resource to mount a drive,
503 be sure this works with admin jobs so that the user can get
504 prompted to insert the correct tape. Possibly some way to say to
505 run the job but don't save the files.
506 - Make things like list where a file is saved case independent for
509 - Use autochanger to handle multiple devices.
510 - Add SuSE install doc to list.
511 - Check and rechedk "Invalid block number"
512 - User wants to NOT backup up certain big files (email files).
513 - Maybe remove multiple simultaneous devices code in SD.
514 - On Windows with very long path names, it may be impossible to create
515 a file (and thus restore it) because the total length is too long.
516 We must cd into the directory then create the file without the
518 - Implement a Recycle command
519 - Add client name to cram-md5 challenge so Director can immediately
520 verify if it is the correct client.
521 - Audit all UA commands to ensure that we always prompt where possible.
522 - Make bootstrap filename unique.
523 - Test a second language e.g. french.
524 - Start working on Base jobs.
525 - Implement UnsavedFiles DB record.
526 - Implement argc/argv for daemon command line scanning using table driven
528 - From Phil Stracchino:
529 It would probably be a per-client option, and would be called
530 something like, say, "Automatically purge obsoleted jobs". What it
531 would do is, when you successfully complete a Differential backup of a
532 client, it would automatically purge all Incremental backups for that
533 client that are rendered redundant by that Differential. Likewise,
534 when a Full backup on a client completed, it would automatically purge
535 all Differential and Incremental jobs obsoleted by that Full backup.
536 This would let people minimize the number of tapes they're keeping on
537 hand without having to master the art of retention times.
538 - Implement a M_SECURITY message class.
539 - When doing a Backup send all attributes back to the Director, who
540 would then figure out what files have been deleted.
541 - Currently in mount.c:236 the SD simply creates a Volume. It should have
542 explicit permission to do so. It should also mark the tape in error
543 if there is an error.
544 - SD Bytes Read is wrong.
545 - Investigate doing RAW backup of Win32 partition.
546 - Cancel waiting for Client connect in SD if FD goes away.
548 - Implement timeout in response() when it should come quickly.
549 - Implement console @echo command.
550 - Implement a Slot priority (loaded/not loaded).
551 - Implement "vacation" Incremental only saves.
552 - Implement single pane restore (much like the Gftp panes).
553 - Implement Automatic Mount even in operator wait.
554 - Implement create "FileSet"?
555 - Implement restore "current system", but take all files without
556 doing selection tree -- so that jobs without File records can
558 - Add prefixlinks to where or not where absolute links to FD.
559 - Issue message to mount a new tape before the rewind.
560 - Simplified client job initiation for portables.
561 - If SD cannot open a drive, make it periodically retry.
562 - Add more of the config info to the tape label.
564 - If tape is marked read-only, then try opening it read-only rather than
565 failing, and remember that it cannot be written.
566 - Refine SD waiting output:
567 Device is being positioned
568 > Device is being positioned for append
569 > Device is being positioned to file x
571 - Figure out some way to estimate output size and to avoid splitting
572 a backup across two Volumes -- this could be useful for writing CDROMs
573 where you really prefer not to have it split -- not serious.
574 - Have SD compute MD5 or SHA1 and compare to what FD computes.
575 - Make VolumeToCatalog calculate an MD5 or SHA1 from the
576 actual data on the Volume and compare it.
577 - Implement Bacula plugins -- design API
578 - Make bcopy read through bad tape records.
579 - Program files (i.e. execute a program to read/write files).
580 Pass read date of last backup, size of file last time.
581 - Add Signature type to File DB record.
582 - CD into subdirectory when open()ing files for backup to
583 speed up things. Test with testfind().
584 - Priority job to go to top of list.
585 - Why are save/restore of device different sizes (sparse?) Yup! Fix it.
586 - Implement some way for the Console to dynamically create a job.
587 - Restore to a particular time -- e.g. before date, after date.
588 - Solaris -I on tar for include list
589 - Need a verbose mode in restore, perhaps to bsr.
590 - bscan without -v is too quiet -- perhaps show jobs.
591 - Add code to reject whole blocks if not wanted on restore.
592 - Check if we can increase Bacula FD priorty in Win2000
593 - Make sure the MaxVolFiles is fully implemented in SD
594 - Check if both CatalogFiles and UseCatalog are set to SD.
595 - Figure out how to do a bare metal Windows restore
596 - Possibly add email to Watchdog if drive is unmounted too
597 long and a job is waiting on the drive.
598 - After unmount, if restore job started, ask to mount.
599 - Convert all %x substitution variables, which are hard to remember
600 and read to %(variable-name). Idea from TMDA.
601 - Remove NextId for SQLite. Optimize.
602 - Move all SQL statements into a single location.
603 - Add UA rc and history files.
604 - put termcap (used by console) in ./configure and
605 allow -with-termcap-dir.
606 - Fix Autoprune for Volumes to respect need for full save.
607 - Fix Win32 config file definition name on /install
608 - Compare tape to Client files (attributes, or attributes and data)
609 - Make all database Ids 64 bit.
610 - Write an applet for Linux.
611 - Allow console commands to detach or run in background.
612 - Fix status delay on storage daemon during rewind.
613 - Add SD message variables to control operator wait time
614 - Maximum Operator Wait
615 - Minimum Message Interval
616 - Maximum Message Interval
617 - Send Operator message when cannot read tape label.
618 - Verify level=Volume (scan only), level=Data (compare of data to file).
619 Verify level=Catalog, level=InitCatalog
621 - Add keyword search to show command in Console.
622 - Events : tape has more than xxx bytes.
623 - Complete code in Bacula Resources -- this will permit
624 reading a new config file at any time.
625 - Handle ctl-c in Console
626 - Implement script driven addition of File daemon to config files.
627 - Think about how to make Bacula work better with File (non-tape) archives.
628 - Write Unix emulator for Windows.
629 - Put memory utilization in Status output of each daemon
630 if full status requested or if some level of debug on.
631 - Make database type selectable by .conf files i.e. at runtime
632 - Set flag for uname -a. Add to Volume label.
633 - Restore files modified after date
634 - SET LD_RUN_PATH=$HOME/mysql/lib/mysql
635 - Remove duplicate fields from jcr (e.g. jcr.level and jcr.jr.Level, ...).
636 - Timout a job or terminate if link goes down, or reopen link and query.
637 - Concept of precious tapes (cannot be reused).
638 - Make bcopy copy with a single tape drive.
639 - Permit changing ownership during restore.
642 > My suggestion: Add a feature on the systray menu-icon menu to request
643 > an immediate backup now. This would be useful for laptop users who may
644 > not be on the network when the regular scheduled backup is run.
646 > My wife's suggestion: Add a setting to the win32 client to allow it to
647 > shut down the machine after backup is complete (after, of course,
648 > displaying a "System will shut down in one minute, click here to cancel"
649 > warning dialog). This would be useful for sites that want user
650 > woorkstations to be shut down overnight to save power.
653 - Autolabel should be specified by DIR instead of SD.
656 - AutoScan (check checksum of tape)
657 - Format command = "format /dev/nst0"
661 - Seek resolution (usually corresponds to buffer size)
662 - EODErrorCode=ENOSPC or code
663 - Partial Read error code
664 - Partial write error code
665 - Nonformatted read error
666 - Nonformatted write error
667 - WriteProtected error
671 - IgnoreCloseErrors=yes
681 - FD sends unsaved file list to Director at end of job (see
683 - File daemon should build list of files skipped, and then
684 at end of save retry and report any errors.
685 - Write a Storage daemon that uses pipes and
686 standard Unix programs to write to the tape.
688 - Need something that monitors the JCR queue and
689 times out jobs by asking the deamons where they are.
690 - Enhance Jmsg code to permit buffering and saving to disk.
691 - device driver = "xxxx" for drives.
693 - Ensure that /dev/null works
694 - Need report class for messages. Perhaps
695 report resource where report=group of messages
696 - enhance scan_attrib and rename scan_jobtype, and
697 fill in code for "since" option
698 - Director needs a time after which the report status is sent
699 anyway -- or better yet, a retry time for the job.
700 - Don't reschedule a job if previous incarnation is still running.
701 - Some way to automatically backup everything is needed????
702 - Need a structure for pending actions:
704 - termination status (part of buffered msgs?)
706 Read, Write, Clean, Delete
707 - Login to Bacula; Bacula users with different permissions:
708 owner, group, user, quotas
709 - Store info on each file system type (probably in the job header on tape.
710 This could be the output of df; or perhaps some sort of /etc/mtab record.
712 ========= ideas ===============
713 From: "Jerry K. Schieffer" <jerry@skylinetechnology.com>
714 To: <kern@sibbald.com>
715 Subject: RE: [Bacula-users] future large programming jobs
716 Date: Thu, 26 Feb 2004 11:34:54 -0600
718 I noticed the subject thread and thought I would offer the following
719 merely as sources of ideas, i.e. something to think about, not even as
720 strong as a request. In my former life (before retiring) I often
721 dealt with backups and storage management issues/products as a
722 developer and as a consultant. I am currently migrating my personal
723 network from amanda to bacula specifically because of the ability to
724 cross media boundaries during storing backups.
725 Are you familiar with the commercial product called ADSM (I think IBM
726 now sells it under the Tivoli label)? It has a couple of interesting
727 ideas that may apply to the following topics.
729 1. Migration: Consider that when you need to restore a system, there
730 may be pressure to hurry. If all the information for a single client
731 can eventually end up on the same media (and in chronological order),
732 the restore is facillitated by not having to search past information
733 from other clients. ADSM has the concept of "client affinity" that
734 may be associated with it's storage pools. It seems to me that this
735 concept (as an optional feature) might fit in your architecture for
738 ADSM also has the concept of defining one or more storage pools as
739 "copy pools" (almost mirrors, but only in the sense of contents).
740 These pools provide the ability to have duplicte data stored both
741 onsite and offsite. The copy process can be scheduled to be handled
742 by their storage manager during periods when there is no backup
743 activity. Again, the migration process might be a place to consider
744 implementing something like this.
747 > It strikes me that it would be very nice to be able to do things
749 > have the Job(s) backing up the machines run, and once they have all
750 > completed, start a migration job to copy the data from disks Volumes
752 > a tape library and then to offsite storage. Maybe this can already
754 > done with some careful scheduling and Job prioritzation; the events
755 > mechanism described below would probably make it very easy.
757 This is the goal. In the first step (before events), you simply
759 the Migration to tape later.
761 2. Base jobs: In ADSM, each copy of each stored file is tracked in
762 the database. Once a file (unique by path and metadata such as dates,
763 size, ownership, etc.) is in a copy pool, no more copies are made. In
764 other words, when you start ADSM, it begins like your concept of a
765 base job. After that it is in the "incremental" mode. You can
766 configure the number of "generations" of files to be retained, plus a
767 retention date after which even old generations are purged. The
768 database tracks the contents of media and projects the percentage of
769 each volume that is valid. When the valid content of a volume drops
770 below a configured percentage, the valid data are migrated to another
771 volume and the old volume is marked as empty. Note, this requires
772 ADSM to have an idea of the contents of a client, i.e. marking the
773 database when an existing file was deleted, but this would solve your
774 issue of restoring a client without restoring deleted files.
776 This is pretty far from what bacula now does, but if you are going to
777 rip things up for Base jobs,.....
778 Also, the benefits of this are huge for very large shops, especially
779 with media robots, but are a pain for shops with manual media
783 > Base jobs sound pretty useful, but I'm not dying for them.
785 Nobody is dying for them, but when you see what it does, you will die
788 3. Restoring deleted files: Since I think my comments in (2) above
789 have low probability of implementation, I'll also suggest that you
790 could approach the issue of deleted files by a mechanism of having the
791 fd report to the dir, a list of all files on the client for every
792 backup job. The dir could note in the database entry for each file
793 the date that the file was seen. Then if a restore as of date X takes
794 place, only files that exist from before X until after X would be
795 restored. Probably the major cost here is the extra date container in
796 each row of the files table.
798 Thanks for "listening". I hope some of this helps. If you want to
799 contact me, please send me an email - I read some but not all of the
800 mailing list traffic and might miss a reply there.
802 Please accept my compliments for bacula. It is doing a great job for
803 me!! I sympathize with you in the need to wrestle with excelence in
804 execution vs. excelence in feature inclusion.
809 ==============================
812 - Design at hierarchial storage for Bacula. Migration and Clone.
813 - Implement FSM (File System Modules).
814 - Audit M_ error codes to ensure they are correct and consistent.
815 - Add variable break characters to lex analyzer.
816 Either a bit mask or a string of chars so that
817 the caller can change the break characters.
818 - Make a single T_BREAK to replace T_COMMA, etc.
819 - Ensure that File daemon and Storage daemon can
820 continue a save if the Director goes down (this
821 is NOT currently the case). Must detect socket error,
822 buffer messages for later.
823 - Enhance time/duration input to allow multiple qualifiers e.g. 3d2h
824 - Add ability to backup to two Storage devices (two SD sessions) at
825 the same time -- e.g. onsite, offsite.
826 - Add the ability to consolidate old backup sets (basically do a restore
827 to tape and appropriately update the catalog). Compress Volume sets.
828 Might need to spool via file is only one drive is available.
829 - Compress or consolidate Volumes of old possibly deleted files. Perhaps
830 someway to do so with every volume that has less than x% valid
834 Migration: Move a backup from one Volume to another
835 Clone: Copy a backup -- two Volumes
837 Bacula Migration is based on Jobs (apparently Networker is file by file).
839 Migration triggered by:
843 Highwater mark (keep total size)
848 ======================================================
850 It is somewhat like a Full save becomes an incremental since
851 the Base job (or jobs) plus other non-base files.
853 - A Base backup is same as Full backup, just different type.
854 - New BaseFiles table that contains:
856 BaseJobId - Base JobId referenced for this FileId (needed ???)
857 JobId - JobId currently running
858 FileId - File not backed up, exists in Base Job
859 FileIndex - FileIndex from Base Job.
860 i.e. for each base file that exists but is not saved because
861 it has not changed, the File daemon sends the JobId, BaseId,
862 FileId, FileIndex back to the Director who creates the DB entry.
863 - To initiate a Base save, the Director sends the FD
864 the FileId, and full filename for each file in the Base.
865 - When the FD finds a Base file, he requests the Director to
866 send him the full File entry (stat packet plus MD5), or
867 conversely, the FD sends it to the Director and the Director
868 says yes or no. This can be quite rapid if the FileId is kept
869 by the FD for each Base Filename.
870 - It is probably better to have the comparison done by the FD
871 despite the fact that the File entry must be sent across the
873 - An alternative would be to send the FD the whole File entry
874 from the start. The disadvantage is that it requires a lot of
875 space. The advantage is that it requires less communications
877 - The Job record must be updated to indicate that one or more
879 - At end of Job, FD returns:
880 1. Count of base files/bytes not written to tape (i.e. matches)
881 2. Count of base file that were saved i.e. had changed.
882 - No tape record would be written for a Base file that matches, in the
883 same way that no tape record is written for Incremental jobs where
884 the file is not saved because it is unchanged.
885 - On a restore, all the Base file records must explicitly be
886 found from the BaseFile tape. I.e. for each Full save that is marked
887 to have one or more Base Jobs, search the BaseFile for all occurrences
889 - An optimization might be to make the BaseFile have:
895 This would avoid the need to explicitly fetch each File record for
896 the Base job. The Base Job record will be fetched to get the
897 VolSessionId and VolSessionTime.
898 =========================================================
901 ==========================================================
903 For each Incremental job that is run, there may be files that
904 were found but not saved because they were locked (this applies
905 only to Windows). Such a system could send back to the Director
906 a list of Unsaved files.
908 - New UnSavedFiles table that contains:
912 - Then in the next Incremental job, the list of Unsaved Files will be
913 feed to the FD, who will ensure that they are explicitly chosen even
914 if standard date/time check would not have selected them.
915 =============================================================
918 Done: (see kernsdone for more)
921 - John's Full save failed with 1.32c FD and 1.31 Dir no FD status,
922 and no error message.
923 - Add fd and st as Console keywords.
924 - Recycling volume with a Slot requires an operator intervention:
925 rufus-dir: Start Backup JobId 18, Job=kernsave.2003-11-01_21.23.52
926 rufus-dir: Pruned 1 Job on Volume Vol01 from catalog.
927 rufus-dir: There are no Jobs associated with Volume Vol01. Marking it purged.
928 rufus-dir: Recycled volume "Vol01"
929 rufus-sd: Please mount Volume "Vol01" on Storage Device "DDS-4" for Job kernsave.2003-11-01_21.23.52
930 Use "mount" command to release Job.
931 - Implement Dan's bacula script (email of 26 Oct).
932 - Add JobName= to VerifyToCatalog so that all verifies can be done at the end.
933 - Edit the Client/Storage name into authentication failure messages.
934 - Fix packet too big problem. This is most likely a Windows TCP stack
936 - Implement ClientRunBeforeJob and ClientRunAfterJob.
937 - Implement forward spacing block/file: position_device(bsr) --
938 just before read_block_from_device();
941 Multiple drive autochanger data: see Alan Brown
942 mtx -f xxx unloadStorage Element 1 is Already Full(drive 0 was empty)
943 Unloading Data Transfer Element into Storage Element 1...source Element
946 (drive 0 was empty and so was slot 1)
947 > mtx -f xxx load 15 0
948 no response, just returns to the command prompt when complete.
949 > mtx -f xxx status Storage Changer /dev/changer:2 Drives, 60 Slots ( 2 Import/Export )
950 Data Transfer Element 0:Full (Storage Element 15 Loaded):VolumeTag = HX001
951 Data Transfer Element 1:Empty
952 Storage Element 1:Empty
953 Storage Element 2:Full :VolumeTag=HX002
954 Storage Element 3:Full :VolumeTag=HX003
955 Storage Element 4:Full :VolumeTag=HX004
956 Storage Element 5:Full :VolumeTag=HX005
957 Storage Element 6:Full :VolumeTag=HX006
958 Storage Element 7:Full :VolumeTag=HX007
959 Storage Element 8:Full :VolumeTag=HX008
960 Storage Element 9:Full :VolumeTag=HX009
961 Storage Element 10:Full :VolumeTag=HX010
962 Storage Element 11:Empty
963 Storage Element 12:Empty
964 Storage Element 13:Empty
965 Storage Element 14:Empty
966 Storage Element 15:Empty
967 Storage Element 16:Empty....
968 Storage Element 28:Empty
969 Storage Element 29:Full :VolumeTag=CLNU01L1
970 Storage Element 30:Empty....
971 Storage Element 57:Empty
972 Storage Element 58:Full :VolumeTag=NEX261L2
973 Storage Element 59 IMPORT/EXPORT:Empty
974 Storage Element 60 IMPORT/EXPORT:Empty
976 Unloading Data Transfer Element into Storage Element 15...done
978 (just to verify it remembers where it came from, however it can be
979 overrriden with mtx unload {slotnumber} to go to any storage slot.)
981 There needs to be a table of drive # to devices somewhere - If there are
982 multiple changers or drives there may not be a 1:1 correspondance between
983 changer drive number and system device name - and depending on the way the
984 drives are hooked up to scsi busses, they may not be linearly numbered
985 from an offset point either.something like
987 Autochanger drives = 2
988 Autochanger drive 0 = /dev/nst1
989 Autochanger drive 1 = /dev/nst2
990 IMHO, it would be _safest_ to use explicit mtx unload commands at all
991 times, not just for multidrive changers. For a 1 drive changer, that's
997 MTX's manpage (1.2.15):
998 unload [<slotnum>] [ <drivenum> ]
999 Unloads media from drive <drivenum> into slot
1000 <slotnum>. If <drivenum> is omitted, defaults to
1001 drive 0 (as do all commands). If <slotnum> is
1002 omitted, defaults to the slot that the drive was
1003 loaded from. Note that there's currently no way
1004 to say 'unload drive 1's media to the slot it
1005 came from', other than to explicitly use that
1006 slot number as the destination.AB
1012 undef# camcontrol devlist
1013 <WANGTEK 51000 SCSI M74H 12B3> at scbus0 target 2 lun 0 (pass0,sa0)
1014 <ARCHIVE 4586XX 28887-XXX 4BGD> at scbus0 target 4 lun 0 (pass1,sa1)
1015 <ARCHIVE 4586XX 28887-XXX 4BGD> at scbus0 target 4 lun 1 (pass2)
1017 tapeinfo -f /dev/sg0 with a bad tape in drive 1:
1018 [kern@rufus mtx-1.2.17kes]$ ./tapeinfo -f /dev/sg0
1019 Product Type: Tape Drive
1021 Product ID: 'C5713A '
1023 Attached Changer: No
1024 TapeAlert[3]: Hard Error: Uncorrectable read/write error.
1025 TapeAlert[20]: Clean Now: The tape drive neads cleaning NOW.
1032 Medium Type: Not Loaded
1035 DataCompEnabled: yes
1036 DataCompCapable: yes
1037 DataDeCompEnabled: yes
1044 Handling removable disks
1046 From: Karl Cunningham <karlc@keckec.com>
1048 My backups are only to hard disk these days, in removable bays. This is my
1049 idea of how a backup to hard disk would work more smoothly. Some of these
1050 things Bacula does already, but I mention them for completeness. If others
1051 have better ways to do this, I'd like to hear about it.
1053 1. Accommodate several disks, rotated similar to how tapes are. Identified
1054 by partition volume ID or perhaps by the name of a subdirectory.
1055 2. Abort & notify the admin if the wrong disk is in the bay.
1056 3. Write backups to different subdirectories for each machine to be backed
1058 4. Volumes (files) get created as needed in the proper subdirectory, one
1060 5. When a disk is recycled, remove or zero all old backup files. This is
1061 important as the disk being recycled may be close to full. This may be
1062 better done manually since the backup files for many machines may be
1063 scattered in many subdirectories.
1068 - Save mount point for directories not traversed with onefs=yes.
1069 - Add seconds to start and end times in the Job report output.
1070 - if 2 concurrent backups are attempted on the same tape
1071 drive (autoloader) into different tape pools, one of them will exit
1072 fatally instead of halting until the drive is idle
1073 - Update StartTime if job held in Job Queue.
1074 - Look at www.nu2.nu/pebuilder as a helper for full windows
1075 bare metal restore. (done by Scott)
1076 - Fix orphanned buffers:
1077 Orphaned buffer: 24 bytes allocated at line 808 of rufus-dir job.c
1078 Orphaned buffer: 40 bytes allocated at line 45 of rufus-dir alist.c
1079 - Implement Preben's suggestion to add
1080 File System Types = ext2, ext3
1081 to FileSets, thus simplifying backup of *all* local partitions.