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.
25 - Create a new GUI chapter explaining all the GUI programs.
26 - Tell the "restore" user when browsing is no longer possible.
27 - Add disk seeking on restore.
28 - Cancel command should include JobId in list of Jobs.
29 - Require restore via the restore command or make a restore Job
30 get the bootstrap file.
31 - Add dump of VolSessionId/Time and FileIndex with bls.
32 - Make bootstrap file handle multiple MediaTypes (SD)
33 - Add offline command to Bacula console.
34 - Implement WildFile and WildDir to solve problem of
35 saving only *.doc files.
36 - Add performance testing hooks
37 - Add Python writable variable for changing the Priority.
38 - Document that Bootstrap files can be written with cataloging
40 - Look at adding full Volume and Pool information to a Volume
41 label so that bscan can get *all* the info.
42 - Scratch Pool where the volumes can be re-assigned to any Pool.
43 - Fix orphanned buffers:
44 Orphaned buffer: 24 bytes allocated at line 808 of rufus-dir job.c
45 Orphaned buffer: 40 bytes allocated at line 45 of rufus-dir alist.c
46 - Upgrade to MySQL 4.1.1 See:
47 http://dev.mysql.com/doc/mysql/en/Server_SQL_mode.html
49 Enter MediaId or Volume name: 32
50 Enter new Volume name: DLT-20Dec04
51 Automatically selected Pool: Default
52 Connecting to Storage daemon DLTDrive at 192.168.68.104:9103 ...
53 Sending relabel command from "DLT-28Jun03" to "DLT-20Dec04" ...
54 block.c:552 Write error at 0:0 on device /dev/nst0. ERR=Bad file descriptor.
55 Error writing final EOF to tape. This tape may not be readable.
56 dev.c:1207 ioctl MTWEOF error on /dev/nst0. ERR=Permission denied.
57 askdir.c:219 NULL Volume name. This shouldn't happen!!!
58 3912 Failed to label Volume: ERR=dev.c:1207 ioctl MTWEOF error on /dev/nst0. ERR=Permission denied.
60 Label command failed for Volume DLT-20Dec04.
61 Do not forget to mount the drive!!!
62 - Bug: if a job is manually scheduled to run later, it does not appear
63 in any status report and cannot be cancelled.
65 Regression tests (Jo):
66 - Add Pool/Storage override regression test.
67 - Add delete JobId to regression.
68 - Add a regression test for dbcheck.
69 - New test to add bscan to four-concurrent-jobs regression,
70 i.e. after the four-concurrent jobs zap the
71 database as is done in the bscan-test, then use bscan to
72 restore the database, do a restore and compare with the
74 - Add restore of specific JobId to regression (item 3
75 on the restore prompt)
76 - Add IPv6 to regression
77 - Add database test to regression. Test each function like delete,
80 - AntiVir can slow down backups on Win32 systems.
81 - Win32 systems with FAT32 can be much slower than NTFS for
82 more than 1000 files per directory.
86 - Can one write tapes faster with 8192 byte block sizes?
87 - Specify a single directory to restore.
88 - FileSet implement "default = include" or "default = exclude"
89 as a solution to being able to include only certain files??????
90 - Implement Preben's suggestion to add
91 File System Types = ext2, ext3
92 to FileSets, thus simplifying backup of *all* local partitions.
93 - Implement Maximum Job Spool Size
94 - Document security problems with the same password for everyone in
95 rpm and Win32 releases.
96 - Browse generations of files.
97 - I've seen an error when my catalog's File table fills up. I
98 then have to recreate the File table with a larger maximum row
99 size. Relevant information is at
100 http://dev.mysql.com/doc/mysql/en/Full_table.html ; I think the
101 "Installing and Configuring MySQL" chapter should talk a bit
102 about this potential problem, and recommend a solution.
103 - For Solaris must use POSIX awk.
104 - Want speed of writing to tape while despooling.
105 - Supported autochanger:
113 Wangtek 6525ES (SCSI-1 QIC drive, 525MB), under Linux 2.4.something,
114 bacula 1.36.0/1 works with blocksize 16k INSIDE bacula-sd.conf.
115 - Add regex from http://www.pcre.org to Bacula for Win32.
116 - Use only shell tools no make in CDROM package.
117 - Document PostgreSQL performance problems bug 131.
119 - Document that ChangerDevice is used for Alert command.
120 - Include within include does it work?
121 - Implement a Pool of type Cleaning?
122 - Implement VolReadTime and VolWriteTime in SD
123 - Modify Backing up Your Database to include a bootstrap file.
124 - Think about making certain database errors fatal.
125 - Look at correcting the time jump in the scheduler for daylight
126 savings time changes.
127 - Add a "real" timer to network connections.
128 - Write non-optimized bsrs from the JobMedia and Media records,
129 even after Files are pruned.
130 - Promote to Full = Time period
131 - Despool attributes simultaneously with data in a separate
132 thread, rejoined at end of data spooling.
133 - Implement Files/Bytes,... stats for restore job.
134 - Implement Total Bytes Written, ... for restore job.
135 - Check dates entered by user for correctness (month/day/... ranges)
136 - Compress restore Volume listing by date and first file.
137 - Look at patches/bacula_db.b2z postgresql that loops during restore.
139 - Perhaps add read/write programs and/or plugins to FileSets.
140 - How to handle backing up portables ...
141 - Add some sort of guaranteed Interval for upgrading jobs.
142 - Can we write the state file after every job terminates? On Win32
143 the system crashes and the state file is not updated.
146 Documentation to do: (any release a little bit at a time)
147 - Doc to do unmount before removing magazine.
148 - Alternative to static linking "ldd prog" save all binaries listed,
149 restore them and point LD_LIBRARY_PATH to them.
150 - Document add "</dev/null >/dev/null 2>&1" to the bacula-fd command line
151 - Document query file format.
152 - Add more documentation for bsr files.
153 - Document problems with Verify and pruning.
154 - Document how to use multiple databases.
155 - VXA drives have a "cleaning required"
156 indicator, but Exabyte recommends preventive cleaning after every 75
159 In this context, it should be noted that Exabyte has a command-line
160 vxatool utility available for free download. (The current version is
161 vxatool-3.72.) It can get diagnostic info, read, write and erase tapes,
162 test the drive, unload tapes, change drive settings, flash new firmware,
164 Of particular interest in this context is that vxatool <device> -i will
165 report, among other details, the time since last cleaning in tape motion
166 minutes. This information can be retrieved (and settings changed, for
167 that matter) through the generic-SCSI device even when Bacula has the
168 regular tape device locked. (Needless to say, I don't recommend
169 changing tape settings while a job is running.)
170 - Lookup HP cleaning recommendations.
171 - Lookup HP tape replacement recommendations (see trouble shooting autochanger)
172 - Document doing table repair
175 Testing to do: (painful)
178 For 1.37 Testing/Documentation:
179 - Add bandwidth limiting.
180 - Add acks every once and a while from the SD to keep
181 the line from timing out.
182 - When an error in input occurs and conio beeps, you can back
183 up through the prompt.
184 - Detect fixed tape block mode during positioning by looking at
185 block numbers in btape "test". Possibly adjust in Bacula.
186 - Fix list volumes to output volume retention in some other
187 units, perhaps via a directive.
188 - If opening a tape in read/write mode fails attempt to open
189 it in read-only mode, and mark the tape for read only.
190 - Add a read-only mode to the mount option.
191 - Allow Simultaneous Priorities = yes => run up to Max concurrent jobs even
192 with multiple priorities.
193 - If you use restore replace=never, the directory attributes for
194 non-existent directories will not be restored properly.
196 - see lzma401.zip in others directory for new compression
198 - Minimal autochanger handling in Bacula and in btape.
199 - Look into how tar does not save sockets and the possiblity of
200 not saving them in Bacula (Martin Simmons reported this).
201 - Add All Local Partitions = yes to new style saves.
202 - localmounts=`awk '/ext/ { print $2 }' /proc/mounts` # or whatever
203 find $localmounts -xdev -type s -ls
204 - Fix restore jobs so that multiple jobs can run if they
205 are not using the same tape(s).
206 - Allow the user to select JobType for manual pruning/purging.
207 - Look at adding Client run command that will use the
208 port opened by the client.
209 - bscan does not put first of two volumes back with all info in
211 - Implement the FreeBSD nodump flag in chflags.
212 - Figure out how to make named console messages go only to that
213 console and to the non-restricted console (new console class?).
214 - Make restricted console prompt for password if *ask* is set or
215 perhaps if password is undefined.
216 - Implement "from ISO-date/time every x hours/days/weeks/months" in
218 ==== from Marc Schoechlin
219 - the help-command should be more verbose
220 (it should explain the paramters of the different
222 -> it´s time-comsuming to consult the manual anytime
223 you need a special parameter
224 -> maybe it´s more easy to maintain this, if the
225 descriptions of that commands are outsourced to
227 - the cd-command should allow complete paths
228 i.e. cd /foo/bar/foo/bar
229 -> if a customer mails me the path to a certain file,
230 it´s faster to enter the specified directory
231 - if the password is not configured in bconsole.conf
232 you should be asked for it.
233 -> sometimes you like to do restore on a customer-machine
234 which shouldn´t know the password for bacula.
235 -> adding the password to the file favours admins
236 to forget to remove the password after usage
238 the protection of that file is less important
239 - long-listed-output of commands should be scrollable
240 like the unix more/less-command does
241 -> if someone runs 200 and more machines, the lists could
242 be a little long and complex
243 - command-output should be shown column by column
244 to reduce scrolling and to increase clarity
246 - lsmark should list the selected files with full
248 - wildcards for selecting and file and directories would be nice
249 - any actions should be interuptable with STRG+C
250 - command-expansion would be pretty cool
252 - Compaction of Disk space by "migrating" Volumes that have pruned
253 Jobs (what criteria? size, #jobs, time).
254 - Add prune all command
255 - Document fact that purge can destroy a part of a restore by purging
256 one volume while others remain valid -- perhaps mark Jobs.
257 - Add multiple-media-types.txt
258 - look at mxt-changer.html
259 - Make ? do a help command (no return needed).
260 - Implement restore directory.
261 - Document streams and how to implement them.
262 - Try not to re-backup a file if a new hard link is added.
263 - Add feature to backup hard links only, but not the data.
264 - Fix stream handling to be simpler.
265 - Add Priority and Bootstrap to Run a Job.
266 - Eliminate Restore "Run Restore Job" prompt by allowing new "run command
268 - Remove View FileSet button from Run a Job dialog.
269 - Handle prompt for restore job at end of Restore command.
270 - Add display of total selected files to Restore window.
271 - Add tree pane to left of window.
272 - Add progress meter.
273 - Max wait time or max run time causes seg fault -- see runtime-bug.txt
274 - Document writing to a CD/DVD with Bacula.
275 - Add a "base" package to the window installer for pthreadsVCE.dll
276 which is needed by all packages.
277 - Add message to user to check for fixed block size when the forward
278 space test fails in btape.
279 - When unmarking a directory check if all files below are unmarked and
280 then remove the + flag -- in the restore tree.
281 - Possibly implement: Action = Unmount Device="TapeDrive1" in Admin jobs.
282 - Setup lrrd graphs: (http://www.linpro.no/projects/lrrd/) Mike Acar.
283 - Revisit the question of multiple Volumes (disk) on a single device.
284 - Add a block copy option to bcopy.
285 - Investigate adding Mac Resource Forks.
286 - Finish work on Gnome restore GUI.
287 - Fix "llist jobid=xx" where no fileset or client exists.
289 it seems to be complaining about 12:00pm which should be a valid 12
290 hour time. I changed the time to 11:59am and everything works fine.
291 Also 12:00am works fine. 0:00pm also works (which I don't think
292 should). None of the values 12:00pm - 12:59pm work for that matter.
293 - Add level to estimate command.
294 - For each job type (Admin, Restore, ...) require only the really necessary
296 - Fix option 2 of restore -- list where file is backed up -- require Client,
297 then list last 20 backups.
298 - Pass Director resource name as an option to the Console.
299 - Add a "batch" mode to the Console (no unsolicited queries, ...).
300 - Add a .list all files in the restore tree (probably also a list all files)
301 Do both a long and short form.
302 - Allow browsing the catalog to see all versions of a file (with
303 stat data on each file).
304 - Restore attributes of directory if replace=never set but directory
306 - Use SHA1 on authentication if possible.
307 - See comtest-xxx.zip for Windows code to talk to USB.
308 - Add Events and Perl scripting.
309 - Add John's appended files:
310 Appended = { /files/server/logs/http/*log }
311 and such files would be treated as follows.On a FULL backup, they would
312 be backed up like any other file.On an INCREMENTAL backup, where a
313 previous INCREMENTAL or FULL was already in thecatalogue and the length
314 of the file wasgreater than the length of the last backup, only thedata
315 added since the last backup will be dumped.On an INCREMENTAL backup, if
316 the length of the file is less than thelength of the file with the same
317 name last backed up, the completefile is dumped.On Windows systems, with
318 creation date of files, we can be evensmarter about this and not count
319 entirely upon the length.On a restore, the full and all incrementals
320 since it will beapplied in sequence to restore the file.
321 - Allow for optional cancelling of SD and FD in case DIR
322 gets a fatal error. Requested by Jesse Guardiani <jesse@wingnet.net>
323 - Add "limit=n" for "list jobs"
324 - Check new HAVE_WIN32 open bits.
325 - Check if the tape has moved before writing.
326 - Handling removable disks -- see below:
327 - Keep track of tape use time, and report when cleaning is necessary.
328 - Add FromClient and ToClient keywords on restore command (or
329 BackupClient RestoreClient).
330 - Implement a JobSet, which groups any number of jobs. If the
331 JobSet is started, all the jobs are started together.
332 Allow Pool, Level, and Schedule overrides.
333 - Enhance cancel to timeout BSOCK packets after a specific delay.
334 - Do scheduling by UTC using gmtime_r() in run_conf, scheduler, and
335 ua_status.!!! Thanks to Alan Brown for this tip.
336 - Look at updating Volume Jobs so that Max Volume Jobs = 1 will work
337 correctly for multiple simultaneous jobs.
338 - Correct code so that FileSet MD5 is calculated for < and | filename
340 - Implement the Media record flag that indicates that the Volume does disk
342 - Implement VolAddr, which is used when Volume is addressed like a disk,
343 and form it from VolFile and VolBlock.
344 - Make multiple restore jobs for multiple media types specifying
345 the proper storage type.
346 - Implement MediaType keyword in bsr?
347 - Fix fast block rejection (stored/read_record.c:118). It passes a null
348 pointer (rec) to try_repositioning().
349 - Look at extracting Win data from BackupRead.
350 - Implement RestoreJobRetention? Maybe better "JobRetention" in a Job,
351 which would take precidence over the Catalog "JobRetention".
352 - Implement Label Format in Add and Label console commands.
353 - Possibly up network buffers to 65K. Put on variable.
354 - Put email tape request delays on one or more variables. User wants
355 to cancel the job after a certain time interval. Maximum Mount Wait?
356 - Job, Client, Device, Pool, or Volume?
357 Is it possible to make this a directive which is *optional* in multiple
358 resources, like Level? If so, I think I'd make it an optional directive
359 in Job, Client, and Pool, with precedence such that Job overrides Client
360 which in turn overrides Pool.
362 - New Storage specifications:
363 - Want to write to multiple storage devices simultaneously
364 - Want to write to multiple storage devices sequentially (in one job)
365 - Want to read/write simultaneously
366 - Key is MediaType -- it must match
368 Passed to SD as a sort of BSR record called Storage Specification
372 MediaType -> Next MediaType
374 Device -> Next Device
376 Allow multiple Storage specifications
384 Allow Multiple Pool specifications (note, Pool currently
386 Allow Multiple MediaType specifications in Dir conf
387 Allow Multiple Device specifications in Dir conf
388 Perhaps keep this in a single SSR
389 Tie a Volume to a specific device by using a MediaType that
390 is contained in only one device.
391 In SD allow Device to have Multiple MediaTypes
393 - Ideas from Jerry Scharf:
394 First let's point out some big pluses that bacula has for this
396 more importantly it's active. Thank you so much for that
397 even more important, it's not flaky
398 it has an open access catalog, opening many possibilities
399 it's pushing toward heterogeneous systems capability
401 Macintosh file client
402 macs are an interesting niche, but I fear a server is a rathole
403 working bare iron recovery for windows
404 the option for inc/diff backups not reset on fileset revision
405 a) use both change and inode update time against base time
406 b) do the full catalog check (expensive but accurate)
407 sizing guide (how much system is needed to back up N systems/files)
408 consultants on using bacula in building a disaster recovery system
410 or how to get at fancy things that one could do with bacula
411 logwatch code for bacula logs (or similar)
412 linux distro inclusion of bacula (brings good and bad, but necessary)
413 win2k/XP server capability (icky but you asked)
414 support for Oracle database ??
416 - Look at adding SQL server and Exchange support for Windows.
417 - Each DVD-RAM disk would be a volume, just like each tape is
418 a volume. It's a 4.7GB media with random access, but there's nothing about
419 it that I can see that makes it so different than a tape from bacula's
420 perspective. Why couldn't I back up to a bare floppy as a volume (ignoring
422 - Make dev->file and dev->block_num signed integers so that -1 can
423 be an invalid value which happens with BSR.
424 - Create VolAddr for disk files in place of VolFile and VolBlock. This
425 is needed to properly specify ranges.
426 - Add progress of files/bytes to SD and FD.
427 - Print warning message if FileId > 4 billion
428 - do a "messages" before the first prompt in Console
429 - Add a date and time stamp at the beginning of every line in the
430 Job report (Volker Sauer).
431 - Client does not show busy during Estimate command.
432 - Implement Console mtx commands.
433 - Implement a Mount Command and an Unmount Command where
434 the users could specify a system command to be performed
435 to do the mount, after which Bacula could attempt to
436 read the device. This is for Removeable media such as a CDROM.
437 - Most likely, this mount command would be invoked explicitly
438 by the user using the current Console "mount" and "unmount"
439 commands -- the Storage Daemon would do the right thing
440 depending on the exact nature of the device.
441 - As with tape drives, when Bacula wanted a new removable
442 disk mounted, it would unmount the old one, and send a message
443 to the user, who would then use "mount" as described above
444 once he had actually inserted the disk.
445 - Implement dump/print label to UA
446 - Spool to disk only when the tape is full, then when a tape is hung move
448 - bextract is sending everything to the log file ****FIXME****
449 - Allow multiple Storage specifications (or multiple names on
450 a single Storage specification) in the Job record. Thus a job
451 can be backed up to a number of storage devices.
452 - Implement some way for the File daemon to contact the Director
453 to start a job or pass its DHCP obtained IP number.
454 - Implement a query tape prompt/replace feature for a console
455 - Copy console @ code to gnome2-console
456 - Make AES the only encryption algorithm see
457 http://csrc.nist.gov/CryptoToolkit/aes/). It's
458 an officially adopted standard, has survived peer
459 review, and provides keys up to 256 bits.
460 - Take a careful look at SetACL http://setacl.sourceforge.net
461 - Take a careful look at Level for the estimate command, maybe make
462 it a command line option.
463 - Add Volume name to "I cannot write on this volume because"
464 - Make tree walk routines like cd, ls, ... more user friendly
465 by handling spaces better.
466 - Make sure that Bacula rechecks the tape after the 20 min wait.
467 - Set IO_NOWAIT on Bacula TCP/IP packets.
468 - Try doing a raw partition backup and restore by mounting a
471 Yes, it would allow to highly automatic the request for new tapes. If a
472 tape is empty, bacula reads the barcodes (native or simulated), and if
473 an unused tape is found, it runs the label command with all the
474 necessary parameters.
476 By the way can bacula automatically "move" an empty/purged volume say
477 in the "short" pool to the "long" pool if this pool runs out of volume
479 - What to do about "list files job=xxx".
480 - Make Dmsg look at global before calling subroutine.
481 - Get and test MySQL 4.0
482 - Look at how fuser works and /proc/PID/fd that is how Nic found the
483 file descriptor leak in Bacula.
484 - Implement WrapCounters in Counters.
485 - Add heartbeat from FD to SD if hb interval expires.
486 - Can we dynamically change FileSets?
487 - If pool specified to label command and Label Format is specified,
488 automatically generate the Volume name.
489 - Why can't SQL do the filename sort for restore?
490 - Look into Microsoft Volume Shadowcopy Service VSS for backing
491 up system state components (Active Directory, System Volume, ...)
492 - Add ExhautiveRestoreSearch
493 - Look at the possibility of loading only the necessary
494 data into the restore tree (i.e. do it one directory at a
495 time as the user walks through the tree).
496 - Possibly use the hash code if the user selects all for a restore command.
497 - Fix "restore all" to bypass building the tree.
498 - Prohibit backing up archive device (findlib/find_one.c:128)
499 - Implement Release Device in the Job resource to unmount a drive.
500 - Implement Acquire Device in the Job resource to mount a drive,
501 be sure this works with admin jobs so that the user can get
502 prompted to insert the correct tape. Possibly some way to say to
503 run the job but don't save the files.
504 - Make things like list where a file is saved case independent for
507 - Use autochanger to handle multiple devices.
508 - Add SuSE install doc to list.
509 - Check and rechedk "Invalid block number"
510 - User wants to NOT backup up certain big files (email files).
511 - Maybe remove multiple simultaneous devices code in SD.
512 - On Windows with very long path names, it may be impossible to create
513 a file (and thus restore it) because the total length is too long.
514 We must cd into the directory then create the file without the
516 - Implement a Recycle command
517 - Add client name to cram-md5 challenge so Director can immediately
518 verify if it is the correct client.
519 - Audit all UA commands to ensure that we always prompt where possible.
520 - Make bootstrap filename unique.
521 - Test a second language e.g. french.
522 - Start working on Base jobs.
523 - Implement UnsavedFiles DB record.
524 - Implement argc/argv for daemon command line scanning using table driven
526 - From Phil Stracchino:
527 It would probably be a per-client option, and would be called
528 something like, say, "Automatically purge obsoleted jobs". What it
529 would do is, when you successfully complete a Differential backup of a
530 client, it would automatically purge all Incremental backups for that
531 client that are rendered redundant by that Differential. Likewise,
532 when a Full backup on a client completed, it would automatically purge
533 all Differential and Incremental jobs obsoleted by that Full backup.
534 This would let people minimize the number of tapes they're keeping on
535 hand without having to master the art of retention times.
536 - Implement a M_SECURITY message class.
537 - When doing a Backup send all attributes back to the Director, who
538 would then figure out what files have been deleted.
539 - Currently in mount.c:236 the SD simply creates a Volume. It should have
540 explicit permission to do so. It should also mark the tape in error
541 if there is an error.
542 - SD Bytes Read is wrong.
543 - Investigate doing RAW backup of Win32 partition.
544 - Cancel waiting for Client connect in SD if FD goes away.
546 - Implement timeout in response() when it should come quickly.
547 - Implement console @echo command.
548 - Implement a Slot priority (loaded/not loaded).
549 - Implement "vacation" Incremental only saves.
550 - Implement single pane restore (much like the Gftp panes).
551 - Implement Automatic Mount even in operator wait.
552 - Implement create "FileSet"?
553 - Implement restore "current system", but take all files without
554 doing selection tree -- so that jobs without File records can
556 - Add prefixlinks to where or not where absolute links to FD.
557 - Issue message to mount a new tape before the rewind.
558 - Simplified client job initiation for portables.
559 - If SD cannot open a drive, make it periodically retry.
560 - Add more of the config info to the tape label.
562 - If tape is marked read-only, then try opening it read-only rather than
563 failing, and remember that it cannot be written.
564 - Refine SD waiting output:
565 Device is being positioned
566 > Device is being positioned for append
567 > Device is being positioned to file x
569 - Figure out some way to estimate output size and to avoid splitting
570 a backup across two Volumes -- this could be useful for writing CDROMs
571 where you really prefer not to have it split -- not serious.
572 - Have SD compute MD5 or SHA1 and compare to what FD computes.
573 - Make VolumeToCatalog calculate an MD5 or SHA1 from the
574 actual data on the Volume and compare it.
575 - Implement Bacula plugins -- design API
576 - Make bcopy read through bad tape records.
577 - Program files (i.e. execute a program to read/write files).
578 Pass read date of last backup, size of file last time.
579 - Add Signature type to File DB record.
580 - CD into subdirectory when open()ing files for backup to
581 speed up things. Test with testfind().
582 - Priority job to go to top of list.
583 - Why are save/restore of device different sizes (sparse?) Yup! Fix it.
584 - Implement some way for the Console to dynamically create a job.
585 - Restore to a particular time -- e.g. before date, after date.
586 - Solaris -I on tar for include list
587 - Need a verbose mode in restore, perhaps to bsr.
588 - bscan without -v is too quiet -- perhaps show jobs.
589 - Add code to reject whole blocks if not wanted on restore.
590 - Check if we can increase Bacula FD priorty in Win2000
591 - Make sure the MaxVolFiles is fully implemented in SD
592 - Check if both CatalogFiles and UseCatalog are set to SD.
593 - Figure out how to do a bare metal Windows restore
594 - Possibly add email to Watchdog if drive is unmounted too
595 long and a job is waiting on the drive.
596 - After unmount, if restore job started, ask to mount.
597 - Convert all %x substitution variables, which are hard to remember
598 and read to %(variable-name). Idea from TMDA.
599 - Remove NextId for SQLite. Optimize.
600 - Move all SQL statements into a single location.
601 - Add UA rc and history files.
602 - put termcap (used by console) in ./configure and
603 allow -with-termcap-dir.
604 - Fix Autoprune for Volumes to respect need for full save.
605 - Fix Win32 config file definition name on /install
606 - Compare tape to Client files (attributes, or attributes and data)
607 - Make all database Ids 64 bit.
608 - Write an applet for Linux.
609 - Allow console commands to detach or run in background.
610 - Fix status delay on storage daemon during rewind.
611 - Add SD message variables to control operator wait time
612 - Maximum Operator Wait
613 - Minimum Message Interval
614 - Maximum Message Interval
615 - Send Operator message when cannot read tape label.
616 - Verify level=Volume (scan only), level=Data (compare of data to file).
617 Verify level=Catalog, level=InitCatalog
619 - Add keyword search to show command in Console.
620 - Events : tape has more than xxx bytes.
621 - Complete code in Bacula Resources -- this will permit
622 reading a new config file at any time.
623 - Handle ctl-c in Console
624 - Implement script driven addition of File daemon to config files.
625 - Think about how to make Bacula work better with File (non-tape) archives.
626 - Write Unix emulator for Windows.
627 - Put memory utilization in Status output of each daemon
628 if full status requested or if some level of debug on.
629 - Make database type selectable by .conf files i.e. at runtime
630 - Set flag for uname -a. Add to Volume label.
631 - Restore files modified after date
632 - SET LD_RUN_PATH=$HOME/mysql/lib/mysql
633 - Remove duplicate fields from jcr (e.g. jcr.level and jcr.jr.Level, ...).
634 - Timout a job or terminate if link goes down, or reopen link and query.
635 - Concept of precious tapes (cannot be reused).
636 - Make bcopy copy with a single tape drive.
637 - Permit changing ownership during restore.
640 > My suggestion: Add a feature on the systray menu-icon menu to request
641 > an immediate backup now. This would be useful for laptop users who may
642 > not be on the network when the regular scheduled backup is run.
644 > My wife's suggestion: Add a setting to the win32 client to allow it to
645 > shut down the machine after backup is complete (after, of course,
646 > displaying a "System will shut down in one minute, click here to cancel"
647 > warning dialog). This would be useful for sites that want user
648 > woorkstations to be shut down overnight to save power.
651 - Autolabel should be specified by DIR instead of SD.
654 - AutoScan (check checksum of tape)
655 - Format command = "format /dev/nst0"
659 - Seek resolution (usually corresponds to buffer size)
660 - EODErrorCode=ENOSPC or code
661 - Partial Read error code
662 - Partial write error code
663 - Nonformatted read error
664 - Nonformatted write error
665 - WriteProtected error
669 - IgnoreCloseErrors=yes
679 - FD sends unsaved file list to Director at end of job (see
681 - File daemon should build list of files skipped, and then
682 at end of save retry and report any errors.
683 - Write a Storage daemon that uses pipes and
684 standard Unix programs to write to the tape.
686 - Need something that monitors the JCR queue and
687 times out jobs by asking the deamons where they are.
688 - Enhance Jmsg code to permit buffering and saving to disk.
689 - device driver = "xxxx" for drives.
691 - Ensure that /dev/null works
692 - Need report class for messages. Perhaps
693 report resource where report=group of messages
694 - enhance scan_attrib and rename scan_jobtype, and
695 fill in code for "since" option
696 - Director needs a time after which the report status is sent
697 anyway -- or better yet, a retry time for the job.
698 - Don't reschedule a job if previous incarnation is still running.
699 - Some way to automatically backup everything is needed????
700 - Need a structure for pending actions:
702 - termination status (part of buffered msgs?)
704 Read, Write, Clean, Delete
705 - Login to Bacula; Bacula users with different permissions:
706 owner, group, user, quotas
707 - Store info on each file system type (probably in the job header on tape.
708 This could be the output of df; or perhaps some sort of /etc/mtab record.
710 ========= ideas ===============
711 From: "Jerry K. Schieffer" <jerry@skylinetechnology.com>
712 To: <kern@sibbald.com>
713 Subject: RE: [Bacula-users] future large programming jobs
714 Date: Thu, 26 Feb 2004 11:34:54 -0600
716 I noticed the subject thread and thought I would offer the following
717 merely as sources of ideas, i.e. something to think about, not even as
718 strong as a request. In my former life (before retiring) I often
719 dealt with backups and storage management issues/products as a
720 developer and as a consultant. I am currently migrating my personal
721 network from amanda to bacula specifically because of the ability to
722 cross media boundaries during storing backups.
723 Are you familiar with the commercial product called ADSM (I think IBM
724 now sells it under the Tivoli label)? It has a couple of interesting
725 ideas that may apply to the following topics.
727 1. Migration: Consider that when you need to restore a system, there
728 may be pressure to hurry. If all the information for a single client
729 can eventually end up on the same media (and in chronological order),
730 the restore is facillitated by not having to search past information
731 from other clients. ADSM has the concept of "client affinity" that
732 may be associated with it's storage pools. It seems to me that this
733 concept (as an optional feature) might fit in your architecture for
736 ADSM also has the concept of defining one or more storage pools as
737 "copy pools" (almost mirrors, but only in the sense of contents).
738 These pools provide the ability to have duplicte data stored both
739 onsite and offsite. The copy process can be scheduled to be handled
740 by their storage manager during periods when there is no backup
741 activity. Again, the migration process might be a place to consider
742 implementing something like this.
745 > It strikes me that it would be very nice to be able to do things
747 > have the Job(s) backing up the machines run, and once they have all
748 > completed, start a migration job to copy the data from disks Volumes
750 > a tape library and then to offsite storage. Maybe this can already
752 > done with some careful scheduling and Job prioritzation; the events
753 > mechanism described below would probably make it very easy.
755 This is the goal. In the first step (before events), you simply
757 the Migration to tape later.
759 2. Base jobs: In ADSM, each copy of each stored file is tracked in
760 the database. Once a file (unique by path and metadata such as dates,
761 size, ownership, etc.) is in a copy pool, no more copies are made. In
762 other words, when you start ADSM, it begins like your concept of a
763 base job. After that it is in the "incremental" mode. You can
764 configure the number of "generations" of files to be retained, plus a
765 retention date after which even old generations are purged. The
766 database tracks the contents of media and projects the percentage of
767 each volume that is valid. When the valid content of a volume drops
768 below a configured percentage, the valid data are migrated to another
769 volume and the old volume is marked as empty. Note, this requires
770 ADSM to have an idea of the contents of a client, i.e. marking the
771 database when an existing file was deleted, but this would solve your
772 issue of restoring a client without restoring deleted files.
774 This is pretty far from what bacula now does, but if you are going to
775 rip things up for Base jobs,.....
776 Also, the benefits of this are huge for very large shops, especially
777 with media robots, but are a pain for shops with manual media
781 > Base jobs sound pretty useful, but I'm not dying for them.
783 Nobody is dying for them, but when you see what it does, you will die
786 3. Restoring deleted files: Since I think my comments in (2) above
787 have low probability of implementation, I'll also suggest that you
788 could approach the issue of deleted files by a mechanism of having the
789 fd report to the dir, a list of all files on the client for every
790 backup job. The dir could note in the database entry for each file
791 the date that the file was seen. Then if a restore as of date X takes
792 place, only files that exist from before X until after X would be
793 restored. Probably the major cost here is the extra date container in
794 each row of the files table.
796 Thanks for "listening". I hope some of this helps. If you want to
797 contact me, please send me an email - I read some but not all of the
798 mailing list traffic and might miss a reply there.
800 Please accept my compliments for bacula. It is doing a great job for
801 me!! I sympathize with you in the need to wrestle with excelence in
802 execution vs. excelence in feature inclusion.
807 ==============================
810 - Design at hierarchial storage for Bacula. Migration and Clone.
811 - Implement FSM (File System Modules).
812 - Audit M_ error codes to ensure they are correct and consistent.
813 - Add variable break characters to lex analyzer.
814 Either a bit mask or a string of chars so that
815 the caller can change the break characters.
816 - Make a single T_BREAK to replace T_COMMA, etc.
817 - Ensure that File daemon and Storage daemon can
818 continue a save if the Director goes down (this
819 is NOT currently the case). Must detect socket error,
820 buffer messages for later.
821 - Enhance time/duration input to allow multiple qualifiers e.g. 3d2h
822 - Add ability to backup to two Storage devices (two SD sessions) at
823 the same time -- e.g. onsite, offsite.
824 - Add the ability to consolidate old backup sets (basically do a restore
825 to tape and appropriately update the catalog). Compress Volume sets.
826 Might need to spool via file is only one drive is available.
827 - Compress or consolidate Volumes of old possibly deleted files. Perhaps
828 someway to do so with every volume that has less than x% valid
832 Migration: Move a backup from one Volume to another
833 Clone: Copy a backup -- two Volumes
835 Bacula Migration is based on Jobs (apparently Networker is file by file).
837 Migration triggered by:
841 Highwater mark (keep total size)
846 ======================================================
848 It is somewhat like a Full save becomes an incremental since
849 the Base job (or jobs) plus other non-base files.
851 - A Base backup is same as Full backup, just different type.
852 - New BaseFiles table that contains:
854 BaseJobId - Base JobId referenced for this FileId (needed ???)
855 JobId - JobId currently running
856 FileId - File not backed up, exists in Base Job
857 FileIndex - FileIndex from Base Job.
858 i.e. for each base file that exists but is not saved because
859 it has not changed, the File daemon sends the JobId, BaseId,
860 FileId, FileIndex back to the Director who creates the DB entry.
861 - To initiate a Base save, the Director sends the FD
862 the FileId, and full filename for each file in the Base.
863 - When the FD finds a Base file, he requests the Director to
864 send him the full File entry (stat packet plus MD5), or
865 conversely, the FD sends it to the Director and the Director
866 says yes or no. This can be quite rapid if the FileId is kept
867 by the FD for each Base Filename.
868 - It is probably better to have the comparison done by the FD
869 despite the fact that the File entry must be sent across the
871 - An alternative would be to send the FD the whole File entry
872 from the start. The disadvantage is that it requires a lot of
873 space. The advantage is that it requires less communications
875 - The Job record must be updated to indicate that one or more
877 - At end of Job, FD returns:
878 1. Count of base files/bytes not written to tape (i.e. matches)
879 2. Count of base file that were saved i.e. had changed.
880 - No tape record would be written for a Base file that matches, in the
881 same way that no tape record is written for Incremental jobs where
882 the file is not saved because it is unchanged.
883 - On a restore, all the Base file records must explicitly be
884 found from the BaseFile tape. I.e. for each Full save that is marked
885 to have one or more Base Jobs, search the BaseFile for all occurrences
887 - An optimization might be to make the BaseFile have:
893 This would avoid the need to explicitly fetch each File record for
894 the Base job. The Base Job record will be fetched to get the
895 VolSessionId and VolSessionTime.
896 =========================================================
899 ==========================================================
901 For each Incremental job that is run, there may be files that
902 were found but not saved because they were locked (this applies
903 only to Windows). Such a system could send back to the Director
904 a list of Unsaved files.
906 - New UnSavedFiles table that contains:
910 - Then in the next Incremental job, the list of Unsaved Files will be
911 feed to the FD, who will ensure that they are explicitly chosen even
912 if standard date/time check would not have selected them.
913 =============================================================
916 Done: (see kernsdone for more)
919 - John's Full save failed with 1.32c FD and 1.31 Dir no FD status,
920 and no error message.
921 - Add fd and st as Console keywords.
922 - Recycling volume with a Slot requires an operator intervention:
923 rufus-dir: Start Backup JobId 18, Job=kernsave.2003-11-01_21.23.52
924 rufus-dir: Pruned 1 Job on Volume Vol01 from catalog.
925 rufus-dir: There are no Jobs associated with Volume Vol01. Marking it purged.
926 rufus-dir: Recycled volume "Vol01"
927 rufus-sd: Please mount Volume "Vol01" on Storage Device "DDS-4" for Job kernsave.2003-11-01_21.23.52
928 Use "mount" command to release Job.
929 - Implement Dan's bacula script (email of 26 Oct).
930 - Add JobName= to VerifyToCatalog so that all verifies can be done at the end.
931 - Edit the Client/Storage name into authentication failure messages.
932 - Fix packet too big problem. This is most likely a Windows TCP stack
934 - Implement ClientRunBeforeJob and ClientRunAfterJob.
935 - Implement forward spacing block/file: position_device(bsr) --
936 just before read_block_from_device();
939 Multiple drive autochanger data: see Alan Brown
940 mtx -f xxx unloadStorage Element 1 is Already Full(drive 0 was empty)
941 Unloading Data Transfer Element into Storage Element 1...source Element
944 (drive 0 was empty and so was slot 1)
945 > mtx -f xxx load 15 0
946 no response, just returns to the command prompt when complete.
947 > mtx -f xxx status Storage Changer /dev/changer:2 Drives, 60 Slots ( 2 Import/Export )
948 Data Transfer Element 0:Full (Storage Element 15 Loaded):VolumeTag = HX001
949 Data Transfer Element 1:Empty
950 Storage Element 1:Empty
951 Storage Element 2:Full :VolumeTag=HX002
952 Storage Element 3:Full :VolumeTag=HX003
953 Storage Element 4:Full :VolumeTag=HX004
954 Storage Element 5:Full :VolumeTag=HX005
955 Storage Element 6:Full :VolumeTag=HX006
956 Storage Element 7:Full :VolumeTag=HX007
957 Storage Element 8:Full :VolumeTag=HX008
958 Storage Element 9:Full :VolumeTag=HX009
959 Storage Element 10:Full :VolumeTag=HX010
960 Storage Element 11:Empty
961 Storage Element 12:Empty
962 Storage Element 13:Empty
963 Storage Element 14:Empty
964 Storage Element 15:Empty
965 Storage Element 16:Empty....
966 Storage Element 28:Empty
967 Storage Element 29:Full :VolumeTag=CLNU01L1
968 Storage Element 30:Empty....
969 Storage Element 57:Empty
970 Storage Element 58:Full :VolumeTag=NEX261L2
971 Storage Element 59 IMPORT/EXPORT:Empty
972 Storage Element 60 IMPORT/EXPORT:Empty
974 Unloading Data Transfer Element into Storage Element 15...done
976 (just to verify it remembers where it came from, however it can be
977 overrriden with mtx unload {slotnumber} to go to any storage slot.)
979 There needs to be a table of drive # to devices somewhere - If there are
980 multiple changers or drives there may not be a 1:1 correspondance between
981 changer drive number and system device name - and depending on the way the
982 drives are hooked up to scsi busses, they may not be linearly numbered
983 from an offset point either.something like
985 Autochanger drives = 2
986 Autochanger drive 0 = /dev/nst1
987 Autochanger drive 1 = /dev/nst2
988 IMHO, it would be _safest_ to use explicit mtx unload commands at all
989 times, not just for multidrive changers. For a 1 drive changer, that's
995 MTX's manpage (1.2.15):
996 unload [<slotnum>] [ <drivenum> ]
997 Unloads media from drive <drivenum> into slot
998 <slotnum>. If <drivenum> is omitted, defaults to
999 drive 0 (as do all commands). If <slotnum> is
1000 omitted, defaults to the slot that the drive was
1001 loaded from. Note that there's currently no way
1002 to say 'unload drive 1's media to the slot it
1003 came from', other than to explicitly use that
1004 slot number as the destination.AB
1010 undef# camcontrol devlist
1011 <WANGTEK 51000 SCSI M74H 12B3> at scbus0 target 2 lun 0 (pass0,sa0)
1012 <ARCHIVE 4586XX 28887-XXX 4BGD> at scbus0 target 4 lun 0 (pass1,sa1)
1013 <ARCHIVE 4586XX 28887-XXX 4BGD> at scbus0 target 4 lun 1 (pass2)
1015 tapeinfo -f /dev/sg0 with a bad tape in drive 1:
1016 [kern@rufus mtx-1.2.17kes]$ ./tapeinfo -f /dev/sg0
1017 Product Type: Tape Drive
1019 Product ID: 'C5713A '
1021 Attached Changer: No
1022 TapeAlert[3]: Hard Error: Uncorrectable read/write error.
1023 TapeAlert[20]: Clean Now: The tape drive neads cleaning NOW.
1030 Medium Type: Not Loaded
1033 DataCompEnabled: yes
1034 DataCompCapable: yes
1035 DataDeCompEnabled: yes
1042 Handling removable disks
1044 From: Karl Cunningham <karlc@keckec.com>
1046 My backups are only to hard disk these days, in removable bays. This is my
1047 idea of how a backup to hard disk would work more smoothly. Some of these
1048 things Bacula does already, but I mention them for completeness. If others
1049 have better ways to do this, I'd like to hear about it.
1051 1. Accommodate several disks, rotated similar to how tapes are. Identified
1052 by partition volume ID or perhaps by the name of a subdirectory.
1053 2. Abort & notify the admin if the wrong disk is in the bay.
1054 3. Write backups to different subdirectories for each machine to be backed
1056 4. Volumes (files) get created as needed in the proper subdirectory, one
1058 5. When a disk is recycled, remove or zero all old backup files. This is
1059 important as the disk being recycled may be close to full. This may be
1060 better done manually since the backup files for many machines may be
1061 scattered in many subdirectories.
1066 - Save mount point for directories not traversed with onefs=yes.
1067 - Add seconds to start and end times in the Job report output.
1068 - if 2 concurrent backups are attempted on the same tape
1069 drive (autoloader) into different tape pools, one of them will exit
1070 fatally instead of halting until the drive is idle
1071 - Update StartTime if job held in Job Queue.
1072 - Look at www.nu2.nu/pebuilder as a helper for full windows
1073 bare metal restore. (done by Scott)