8 Data encryption Meno Abels (see projects)
9 Communication encryption Meno Abels
10 Version 1.37 Kern (see below)
11 ========================================================
14 - Fix orphanned buffers:
15 Orphaned buffer: 24 bytes allocated at line 808 of rufus-dir job.c
16 Orphaned buffer: 40 bytes allocated at line 45 of rufus-dir alist.c
17 - Add dump of VolSessionId/Time and FileIndex with bls.
20 #3 Migration (Move, Copy, Archive Jobs)
21 #4 Embedded Python Scripting
22 #5 Events that call a Python program
23 #6 Select one from among Multiple Storage Devices for Job
24 #7 Single Job Writing to Multiple Storage Devices
29 - Add performance testing hooks
30 - if 2 concurrent backups are attempted on the same tape
31 drive (autoloader) into different tape pools, one of them will exit
32 fatally instead of halting until the drive is idle
33 - Add seconds to start and end times in the Job report output.
34 - Add disk seeking on restore.
35 - Document that ChangerDevice is used for Alert command.
36 - Include within include does it work?
37 - Implement a Pool of type Cleaning?
38 - Implement VolReadTime and VolWriteTime in SD
39 - Modify Backing up Your Database to include a bootstrap file.
40 - Add restore of specific JobId to regression.
41 - Think about making certain database errors fatal.
42 - Look at correcting the time jump in the sceduler for daylight
44 - Add a "real" timer to network connections.
45 - Tell the "restore" user when browsing is no longer possible.
46 - Write non-optimized bsrs from the JobMedia and Media records,
47 even after Files are pruned.
48 - Document that Bootstrap files can be written with cataloging
50 - Look at adding full Volume and Pool information to a Volume
51 label so that bscan can get *all* the info.
52 - Promote to Full = Time period
53 - Scratch Pool where the volumes can be re-assigned to any Pool.
54 - Update StartTime if job held in Job Queue.
55 - Despool attributes simultaneously with data in a separate
56 thread, rejoined at end of data spooling.
57 - Implement Files/Bytes,... stats for restore job.
58 - Implement Total Bytes Written, ... for restore job.
59 - Check dates entered by user for correctness (month/day/... ranges)
60 - Compress restore Volume listing by date and first file.
61 - Add Pool/Storage override regression test.
62 - Add delete JobId to regression.
63 - Add bscan to four-concurrent-jobs regression.
64 - Look at patches/bacula_db.b2z postgresql that loops during restore.
66 - Add IPv6 to regression
67 - Perhaps add read/write programs and/or plugins to FileSets.
68 - How to handle backing up portables ...
69 - Add some sort of guaranteed Interval for upgrading jobs.
70 - Can we write the state file after every job terminates? On Win32
71 the system crashes and the state file is not updated.
74 Documentation to do: (any release a little bit at a time)
75 - Doc to do unmount before removing magazine.
76 - Alternative to static linking "ldd prog" save all binaries listed,
77 restore them and point LD_LIBRARY_PATH to them.
78 - Document add "</dev/null >/dev/null 2>&1" to the bacula-fd command line
79 - Document query file format.
80 - Add more documentation for bsr files.
81 - Document problems with Verify and pruning.
82 - Document how to use multiple databases.
83 - VXA drives have a "cleaning required"
84 indicator, but Exabyte recommends preventive cleaning after every 75
87 In this context, it should be noted that Exabyte has a command-line
88 vxatool utility available for free download. (The current version is
89 vxatool-3.72.) It can get diagnostic info, read, write and erase tapes,
90 test the drive, unload tapes, change drive settings, flash new firmware,
92 Of particular interest in this context is that vxatool <device> -i will
93 report, among other details, the time since last cleaning in tape motion
94 minutes. This information can be retrieved (and settings changed, for
95 that matter) through the generic-SCSI device even when Bacula has the
96 regular tape device locked. (Needless to say, I don't recommend
97 changing tape settings while a job is running.)
98 - Lookup HP cleaning recommendations.
99 - Lookup HP tape replacement recommendations (see trouble shooting autochanger)
100 - Document doing table repair
103 Testing to do: (painful)
106 For 1.37 Testing/Documentation:
107 - Add bandwidth limiting.
108 - Add acks every once and a while from the SD to keep
109 the line from timing out.
110 - When an error in input occurs and conio beeps, you can back
111 up through the prompt.
112 - Detect fixed tape block mode during positioning by looking at
113 block numbers in btape "test". Possibly adjust in Bacula.
114 - Fix list volumes to output volume retention in some other
115 units, perhaps via a directive.
116 - If opening a tape in read/write mode fails attempt to open
117 it in read-only mode, and mark the tape for read only.
118 - Add a read-only mode to the mount option.
119 - Allow Simultaneous Priorities = yes => run up to Max concurrent jobs even
120 with multiple priorities.
121 - Add db check test to regression. Test each function like delete,
123 - If you use restore replace=never, the directory attributes for
124 non-existent directories will not be restored properly.
126 - see lzma401.zip in others directory for new compression
128 - Minimal autochanger handling in Bacula and in btape.
129 - Look into how tar does not save sockets and the possiblity of
130 not saving them in Bacula (Martin Simmons reported this).
131 - Add All Local Partitions = yes to new style saves.
132 - localmounts=`awk '/ext/ { print $2 }' /proc/mounts` # or whatever
133 find $localmounts -xdev -type s -ls
134 - Fix restore jobs so that multiple jobs can run if they
135 are not using the same tape(s).
136 - Allow the user to select JobType for manual pruning/purging.
137 - Look at adding Client run command that will use the
138 port opened by the client.
139 - bscan does not put first of two volumes back with all info in
141 - Implement the FreeBSD nodump flag in chflags.
142 - Figure out how to make named console messages go only to that
143 console and to the non-restricted console (new console class?).
144 - Make restricted console prompt for password if *ask* is set or
145 perhaps if password is undefined.
146 - Implement "from ISO-date/time every x hours/days/weeks/months" in
148 ==== from Marc Schoechlin
149 - the help-command should be more verbose
150 (it should explain the paramters of the different
152 -> it´s time-comsuming to consult the manual anytime
153 you need a special parameter
154 -> maybe it´s more easy to maintain this, if the
155 descriptions of that commands are outsourced to
157 - the cd-command should allow complete paths
158 i.e. cd /foo/bar/foo/bar
159 -> if a customer mails me the path to a certain file,
160 it´s faster to enter the specified directory
161 - if the password is not configured in bconsole.conf
162 you should be asked for it.
163 -> sometimes you like to do restore on a customer-machine
164 which shouldn´t know the password for bacula.
165 -> adding the password to the file favours admins
166 to forget to remove the password after usage
168 the protection of that file is less important
169 - long-listed-output of commands should be scrollable
170 like the unix more/less-command does
171 -> if someone runs 200 and more machines, the lists could
172 be a little long and complex
173 - command-output should be shown column by column
174 to reduce scrolling and to increase clarity
176 - lsmark should list the selected files with full
178 - wildcards for selecting and file and directories would be nice
179 - any actions should be interuptable with STRG+C
180 - command-expansion would be pretty cool
182 - Compaction of Disk space by "migrating" Volumes that have pruned
183 Jobs (what criteria? size, #jobs, time).
184 - Add prune all command
185 - Document fact that purge can destroy a part of a restore by purging
186 one volume while others remain valid -- perhaps mark Jobs.
187 - Add multiple-media-types.txt
188 - look at mxt-changer.html
189 - Make ? do a help command (no return needed).
190 - Implement restore directory.
191 - Document streams and how to implement them.
192 - Try not to re-backup a file if a new hard link is added.
193 - Add feature to backup hard links only, but not the data.
194 - Fix stream handling to be simpler.
195 - Add Priority and Bootstrap to Run a Job.
196 - Eliminate Restore "Run Restore Job" prompt by allowing new "run command
198 - Remove View FileSet button from Run a Job dialog.
199 - Handle prompt for restore job at end of Restore command.
200 - Add display of total selected files to Restore window.
201 - Add tree pane to left of window.
202 - Add progress meter.
203 - Max wait time or max run time causes seg fault -- see runtime-bug.txt
204 - Document writing to a CD/DVD with Bacula.
205 - Add a "base" package to the window installer for pthreadsVCE.dll
206 which is needed by all packages.
207 - Add message to user to check for fixed block size when the forward
208 space test fails in btape.
209 - When unmarking a directory check if all files below are unmarked and
210 then remove the + flag -- in the restore tree.
211 - Possibly implement: Action = Unmount Device="TapeDrive1" in Admin jobs.
212 - Setup lrrd graphs: (http://www.linpro.no/projects/lrrd/) Mike Acar.
213 - Revisit the question of multiple Volumes (disk) on a single device.
214 - Add a block copy option to bcopy.
215 - Investigate adding Mac Resource Forks.
216 - Finish work on Gnome restore GUI.
217 - Fix "llist jobid=xx" where no fileset or client exists.
219 it seems to be complaining about 12:00pm which should be a valid 12
220 hour time. I changed the time to 11:59am and everything works fine.
221 Also 12:00am works fine. 0:00pm also works (which I don't think
222 should). None of the values 12:00pm - 12:59pm work for that matter.
223 - Add level to estimate command.
224 - For each job type (Admin, Restore, ...) require only the really necessary
226 - Fix option 2 of restore -- list where file is backed up -- require Client,
227 then list last 20 backups.
228 - Pass Director resource name as an option to the Console.
229 - Add a "batch" mode to the Console (no unsolicited queries, ...).
230 - Add a .list all files in the restore tree (probably also a list all files)
231 Do both a long and short form.
232 - Allow browsing the catalog to see all versions of a file (with
233 stat data on each file).
234 - Restore attributes of directory if replace=never set but directory
236 - Use SHA1 on authentication if possible.
237 - See comtest-xxx.zip for Windows code to talk to USB.
238 - Add Events and Perl scripting.
239 - Add John's appended files:
240 Appended = { /files/server/logs/http/*log }
241 and such files would be treated as follows.On a FULL backup, they would
242 be backed up like any other file.On an INCREMENTAL backup, where a
243 previous INCREMENTAL or FULL was already in thecatalogue and the length
244 of the file wasgreater than the length of the last backup, only thedata
245 added since the last backup will be dumped.On an INCREMENTAL backup, if
246 the length of the file is less than thelength of the file with the same
247 name last backed up, the completefile is dumped.On Windows systems, with
248 creation date of files, we can be evensmarter about this and not count
249 entirely upon the length.On a restore, the full and all incrementals
250 since it will beapplied in sequence to restore the file.
251 - Add a regression test for dbcheck.
252 - Allow for optional cancelling of SD and FD in case DIR
253 gets a fatal error. Requested by Jesse Guardiani <jesse@wingnet.net>
254 - Add "limit=n" for "list jobs"
255 - Check new HAVE_WIN32 open bits.
256 - Check if the tape has moved before writing.
257 - Handling removable disks -- see below:
258 - Keep track of tape use time, and report when cleaning is necessary.
259 - Add FromClient and ToClient keywords on restore command (or
260 BackupClient RestoreClient).
261 - Implement a JobSet, which groups any number of jobs. If the
262 JobSet is started, all the jobs are started together.
263 Allow Pool, Level, and Schedule overrides.
264 - Enhance cancel to timeout BSOCK packets after a specific delay.
265 - Do scheduling by UTC using gmtime_r() in run_conf, scheduler, and
266 ua_status.!!! Thanks to Alan Brown for this tip.
267 - Look at updating Volume Jobs so that Max Volume Jobs = 1 will work
268 correctly for multiple simultaneous jobs.
269 - Correct code so that FileSet MD5 is calculated for < and | filename
271 - Implement the Media record flag that indicates that the Volume does disk
273 - Implement VolAddr, which is used when Volume is addressed like a disk,
274 and form it from VolFile and VolBlock.
275 - Make multiple restore jobs for multiple media types specifying
276 the proper storage type.
277 - Implement MediaType keyword in bsr?
278 - Fix fast block rejection (stored/read_record.c:118). It passes a null
279 pointer (rec) to try_repositioning().
280 - Look at extracting Win data from BackupRead.
281 - Implement RestoreJobRetention? Maybe better "JobRetention" in a Job,
282 which would take precidence over the Catalog "JobRetention".
283 - Implement Label Format in Add and Label console commands.
284 - Possibly up network buffers to 65K. Put on variable.
285 - Put email tape request delays on one or more variables. User wants
286 to cancel the job after a certain time interval. Maximum Mount Wait?
287 - Job, Client, Device, Pool, or Volume?
288 Is it possible to make this a directive which is *optional* in multiple
289 resources, like Level? If so, I think I'd make it an optional directive
290 in Job, Client, and Pool, with precedence such that Job overrides Client
291 which in turn overrides Pool.
293 - New Storage specifications:
294 - Want to write to multiple storage devices simultaneously
295 - Want to write to multiple storage devices sequentially (in one job)
296 - Want to read/write simultaneously
297 - Key is MediaType -- it must match
299 Passed to SD as a sort of BSR record called Storage Specification
303 MediaType -> Next MediaType
305 Device -> Next Device
307 Allow multiple Storage specifications
315 Allow Multiple Pool specifications (note, Pool currently
317 Allow Multiple MediaType specifications in Dir conf
318 Allow Multiple Device specifications in Dir conf
319 Perhaps keep this in a single SSR
320 Tie a Volume to a specific device by using a MediaType that
321 is contained in only one device.
322 In SD allow Device to have Multiple MediaTypes
324 - Look at www.nu2.nu/pebuilder as a helper for full windows
326 - Ideas from Jerry Scharf:
327 First let's point out some big pluses that bacula has for this
329 more importantly it's active. Thank you so much for that
330 even more important, it's not flaky
331 it has an open access catalog, opening many possibilities
332 it's pushing toward heterogeneous systems capability
334 Macintosh file client
335 macs are an interesting niche, but I fear a server is a rathole
336 working bare iron recovery for windows
337 the option for inc/diff backups not reset on fileset revision
338 a) use both change and inode update time against base time
339 b) do the full catalog check (expensive but accurate)
340 sizing guide (how much system is needed to back up N systems/files)
341 consultants on using bacula in building a disaster recovery system
343 or how to get at fancy things that one could do with bacula
344 logwatch code for bacula logs (or similar)
345 linux distro inclusion of bacula (brings good and bad, but necessary)
346 win2k/XP server capability (icky but you asked)
347 support for Oracle database ??
349 - Look at adding SQL server and Exchange support for Windows.
350 - Each DVD-RAM disk would be a volume, just like each tape is
351 a volume. It's a 4.7GB media with random access, but there's nothing about
352 it that I can see that makes it so different than a tape from bacula's
353 perspective. Why couldn't I back up to a bare floppy as a volume (ignoring
355 - Make dev->file and dev->block_num signed integers so that -1 can
356 be an invalid value which happens with BSR.
357 - Create VolAddr for disk files in place of VolFile and VolBlock. This
358 is needed to properly specify ranges.
359 - Add progress of files/bytes to SD and FD.
360 - Print warning message if FileId > 4 billion
361 - do a "messages" before the first prompt in Console
362 - Add a date and time stamp at the beginning of every line in the
363 Job report (Volker Sauer).
364 - Client does not show busy during Estimate command.
365 - Implement Console mtx commands.
366 - Implement a Mount Command and an Unmount Command where
367 the users could specify a system command to be performed
368 to do the mount, after which Bacula could attempt to
369 read the device. This is for Removeable media such as a CDROM.
370 - Most likely, this mount command would be invoked explicitly
371 by the user using the current Console "mount" and "unmount"
372 commands -- the Storage Daemon would do the right thing
373 depending on the exact nature of the device.
374 - As with tape drives, when Bacula wanted a new removable
375 disk mounted, it would unmount the old one, and send a message
376 to the user, who would then use "mount" as described above
377 once he had actually inserted the disk.
378 - Implement dump/print label to UA
379 - Spool to disk only when the tape is full, then when a tape is hung move
381 - bextract is sending everything to the log file ****FIXME****
382 - Allow multiple Storage specifications (or multiple names on
383 a single Storage specification) in the Job record. Thus a job
384 can be backed up to a number of storage devices.
385 - Implement some way for the File daemon to contact the Director
386 to start a job or pass its DHCP obtained IP number.
387 - Implement a query tape prompt/replace feature for a console
388 - Copy console @ code to gnome2-console
389 - Make AES the only encryption algorithm see
390 http://csrc.nist.gov/CryptoToolkit/aes/). It's
391 an officially adopted standard, has survived peer
392 review, and provides keys up to 256 bits.
393 - Take a careful look at SetACL http://setacl.sourceforge.net
394 - Take a careful look at Level for the estimate command, maybe make
395 it a command line option.
396 - Add Volume name to "I cannot write on this volume because"
397 - Make tree walk routines like cd, ls, ... more user friendly
398 by handling spaces better.
399 - Make sure that Bacula rechecks the tape after the 20 min wait.
400 - Set IO_NOWAIT on Bacula TCP/IP packets.
401 - Try doing a raw partition backup and restore by mounting a
404 Yes, it would allow to highly automatic the request for new tapes. If a
405 tape is empty, bacula reads the barcodes (native or simulated), and if
406 an unused tape is found, it runs the label command with all the
407 necessary parameters.
409 By the way can bacula automatically "move" an empty/purged volume say
410 in the "short" pool to the "long" pool if this pool runs out of volume
412 - What to do about "list files job=xxx".
413 - Make Dmsg look at global before calling subroutine.
414 - Get and test MySQL 4.0
415 - Look at how fuser works and /proc/PID/fd that is how Nic found the
416 file descriptor leak in Bacula.
417 - Implement WrapCounters in Counters.
418 - Add heartbeat from FD to SD if hb interval expires.
419 - Can we dynamically change FileSets?
420 - If pool specified to label command and Label Format is specified,
421 automatically generate the Volume name.
422 - Why can't SQL do the filename sort for restore?
423 - Look into Microsoft Volume Shadowcopy Service VSS for backing
424 up system state components (Active Directory, System Volume, ...)
425 - Add ExhautiveRestoreSearch
426 - Look at the possibility of loading only the necessary
427 data into the restore tree (i.e. do it one directory at a
428 time as the user walks through the tree).
429 - Possibly use the hash code if the user selects all for a restore command.
430 - Fix "restore all" to bypass building the tree.
431 - Prohibit backing up archive device (findlib/find_one.c:128)
432 - Implement Release Device in the Job resource to unmount a drive.
433 - Implement Acquire Device in the Job resource to mount a drive,
434 be sure this works with admin jobs so that the user can get
435 prompted to insert the correct tape. Possibly some way to say to
436 run the job but don't save the files.
437 - Make things like list where a file is saved case independent for
440 - Use autochanger to handle multiple devices.
441 - Add SuSE install doc to list.
442 - Check and rechedk "Invalid block number"
443 - User wants to NOT backup up certain big files (email files).
444 - Maybe remove multiple simultaneous devices code in SD.
445 - On Windows with very long path names, it may be impossible to create
446 a file (and thus restore it) because the total length is too long.
447 We must cd into the directory then create the file without the
449 - Implement a Recycle command
450 - Add client name to cram-md5 challenge so Director can immediately
451 verify if it is the correct client.
452 - Audit all UA commands to ensure that we always prompt where possible.
453 - Make bootstrap filename unique.
454 - Test a second language e.g. french.
455 - Start working on Base jobs.
456 - Implement UnsavedFiles DB record.
457 - Implement argc/argv for daemon command line scanning using table driven
459 - From Phil Stracchino:
460 It would probably be a per-client option, and would be called
461 something like, say, "Automatically purge obsoleted jobs". What it
462 would do is, when you successfully complete a Differential backup of a
463 client, it would automatically purge all Incremental backups for that
464 client that are rendered redundant by that Differential. Likewise,
465 when a Full backup on a client completed, it would automatically purge
466 all Differential and Incremental jobs obsoleted by that Full backup.
467 This would let people minimize the number of tapes they're keeping on
468 hand without having to master the art of retention times.
469 - Implement a M_SECURITY message class.
470 - When doing a Backup send all attributes back to the Director, who
471 would then figure out what files have been deleted.
472 - Currently in mount.c:236 the SD simply creates a Volume. It should have
473 explicit permission to do so. It should also mark the tape in error
474 if there is an error.
475 - SD Bytes Read is wrong.
476 - Investigate doing RAW backup of Win32 partition.
477 - Cancel waiting for Client connect in SD if FD goes away.
479 - Implement timeout in response() when it should come quickly.
480 - Implement console @echo command.
481 - Implement a Slot priority (loaded/not loaded).
482 - Implement "vacation" Incremental only saves.
483 - Implement single pane restore (much like the Gftp panes).
484 - Implement Automatic Mount even in operator wait.
485 - Implement create "FileSet"?
486 - Implement restore "current system", but take all files without
487 doing selection tree -- so that jobs without File records can
489 - Add prefixlinks to where or not where absolute links to FD.
490 - Issue message to mount a new tape before the rewind.
491 - Simplified client job initiation for portables.
492 - If SD cannot open a drive, make it periodically retry.
493 - Add more of the config info to the tape label.
495 - If tape is marked read-only, then try opening it read-only rather than
496 failing, and remember that it cannot be written.
497 - Refine SD waiting output:
498 Device is being positioned
499 > Device is being positioned for append
500 > Device is being positioned to file x
502 - Figure out some way to estimate output size and to avoid splitting
503 a backup across two Volumes -- this could be useful for writing CDROMs
504 where you really prefer not to have it split -- not serious.
505 - Have SD compute MD5 or SHA1 and compare to what FD computes.
506 - Make VolumeToCatalog calculate an MD5 or SHA1 from the
507 actual data on the Volume and compare it.
508 - Implement Bacula plugins -- design API
509 - Make bcopy read through bad tape records.
510 - Program files (i.e. execute a program to read/write files).
511 Pass read date of last backup, size of file last time.
512 - Add Signature type to File DB record.
513 - CD into subdirectory when open()ing files for backup to
514 speed up things. Test with testfind().
515 - Priority job to go to top of list.
516 - Why are save/restore of device different sizes (sparse?) Yup! Fix it.
517 - Implement some way for the Console to dynamically create a job.
518 - Restore to a particular time -- e.g. before date, after date.
519 - Solaris -I on tar for include list
520 - Need a verbose mode in restore, perhaps to bsr.
521 - bscan without -v is too quiet -- perhaps show jobs.
522 - Add code to reject whole blocks if not wanted on restore.
523 - Check if we can increase Bacula FD priorty in Win2000
524 - Make sure the MaxVolFiles is fully implemented in SD
525 - Check if both CatalogFiles and UseCatalog are set to SD.
526 - Figure out how to do a bare metal Windows restore
527 - Possibly add email to Watchdog if drive is unmounted too
528 long and a job is waiting on the drive.
529 - After unmount, if restore job started, ask to mount.
530 - Convert all %x substitution variables, which are hard to remember
531 and read to %(variable-name). Idea from TMDA.
532 - Remove NextId for SQLite. Optimize.
533 - Move all SQL statements into a single location.
534 - Add UA rc and history files.
535 - put termcap (used by console) in ./configure and
536 allow -with-termcap-dir.
537 - Fix Autoprune for Volumes to respect need for full save.
538 - Fix Win32 config file definition name on /install
539 - Compare tape to Client files (attributes, or attributes and data)
540 - Make all database Ids 64 bit.
541 - Write an applet for Linux.
542 - Allow console commands to detach or run in background.
543 - Fix status delay on storage daemon during rewind.
544 - Add SD message variables to control operator wait time
545 - Maximum Operator Wait
546 - Minimum Message Interval
547 - Maximum Message Interval
548 - Send Operator message when cannot read tape label.
549 - Verify level=Volume (scan only), level=Data (compare of data to file).
550 Verify level=Catalog, level=InitCatalog
552 - Add keyword search to show command in Console.
553 - Events : tape has more than xxx bytes.
554 - Complete code in Bacula Resources -- this will permit
555 reading a new config file at any time.
556 - Handle ctl-c in Console
557 - Implement script driven addition of File daemon to config files.
558 - Think about how to make Bacula work better with File (non-tape) archives.
559 - Write Unix emulator for Windows.
560 - Put memory utilization in Status output of each daemon
561 if full status requested or if some level of debug on.
562 - Make database type selectable by .conf files i.e. at runtime
563 - Set flag for uname -a. Add to Volume label.
564 - Restore files modified after date
565 - SET LD_RUN_PATH=$HOME/mysql/lib/mysql
566 - Remove duplicate fields from jcr (e.g. jcr.level and jcr.jr.Level, ...).
567 - Timout a job or terminate if link goes down, or reopen link and query.
568 - Concept of precious tapes (cannot be reused).
569 - Make bcopy copy with a single tape drive.
570 - Permit changing ownership during restore.
573 > My suggestion: Add a feature on the systray menu-icon menu to request
574 > an immediate backup now. This would be useful for laptop users who may
575 > not be on the network when the regular scheduled backup is run.
577 > My wife's suggestion: Add a setting to the win32 client to allow it to
578 > shut down the machine after backup is complete (after, of course,
579 > displaying a "System will shut down in one minute, click here to cancel"
580 > warning dialog). This would be useful for sites that want user
581 > woorkstations to be shut down overnight to save power.
584 - Autolabel should be specified by DIR instead of SD.
587 - AutoScan (check checksum of tape)
588 - Format command = "format /dev/nst0"
592 - Seek resolution (usually corresponds to buffer size)
593 - EODErrorCode=ENOSPC or code
594 - Partial Read error code
595 - Partial write error code
596 - Nonformatted read error
597 - Nonformatted write error
598 - WriteProtected error
602 - IgnoreCloseErrors=yes
612 - FD sends unsaved file list to Director at end of job (see
614 - File daemon should build list of files skipped, and then
615 at end of save retry and report any errors.
616 - Write a Storage daemon that uses pipes and
617 standard Unix programs to write to the tape.
619 - Need something that monitors the JCR queue and
620 times out jobs by asking the deamons where they are.
621 - Enhance Jmsg code to permit buffering and saving to disk.
622 - device driver = "xxxx" for drives.
624 - Ensure that /dev/null works
625 - Need report class for messages. Perhaps
626 report resource where report=group of messages
627 - enhance scan_attrib and rename scan_jobtype, and
628 fill in code for "since" option
629 - Director needs a time after which the report status is sent
630 anyway -- or better yet, a retry time for the job.
631 - Don't reschedule a job if previous incarnation is still running.
632 - Some way to automatically backup everything is needed????
633 - Need a structure for pending actions:
635 - termination status (part of buffered msgs?)
637 Read, Write, Clean, Delete
638 - Login to Bacula; Bacula users with different permissions:
639 owner, group, user, quotas
640 - Store info on each file system type (probably in the job header on tape.
641 This could be the output of df; or perhaps some sort of /etc/mtab record.
643 ========= ideas ===============
644 From: "Jerry K. Schieffer" <jerry@skylinetechnology.com>
645 To: <kern@sibbald.com>
646 Subject: RE: [Bacula-users] future large programming jobs
647 Date: Thu, 26 Feb 2004 11:34:54 -0600
649 I noticed the subject thread and thought I would offer the following
650 merely as sources of ideas, i.e. something to think about, not even as
651 strong as a request. In my former life (before retiring) I often
652 dealt with backups and storage management issues/products as a
653 developer and as a consultant. I am currently migrating my personal
654 network from amanda to bacula specifically because of the ability to
655 cross media boundaries during storing backups.
656 Are you familiar with the commercial product called ADSM (I think IBM
657 now sells it under the Tivoli label)? It has a couple of interesting
658 ideas that may apply to the following topics.
660 1. Migration: Consider that when you need to restore a system, there
661 may be pressure to hurry. If all the information for a single client
662 can eventually end up on the same media (and in chronological order),
663 the restore is facillitated by not having to search past information
664 from other clients. ADSM has the concept of "client affinity" that
665 may be associated with it's storage pools. It seems to me that this
666 concept (as an optional feature) might fit in your architecture for
669 ADSM also has the concept of defining one or more storage pools as
670 "copy pools" (almost mirrors, but only in the sense of contents).
671 These pools provide the ability to have duplicte data stored both
672 onsite and offsite. The copy process can be scheduled to be handled
673 by their storage manager during periods when there is no backup
674 activity. Again, the migration process might be a place to consider
675 implementing something like this.
678 > It strikes me that it would be very nice to be able to do things
680 > have the Job(s) backing up the machines run, and once they have all
681 > completed, start a migration job to copy the data from disks Volumes
683 > a tape library and then to offsite storage. Maybe this can already
685 > done with some careful scheduling and Job prioritzation; the events
686 > mechanism described below would probably make it very easy.
688 This is the goal. In the first step (before events), you simply
690 the Migration to tape later.
692 2. Base jobs: In ADSM, each copy of each stored file is tracked in
693 the database. Once a file (unique by path and metadata such as dates,
694 size, ownership, etc.) is in a copy pool, no more copies are made. In
695 other words, when you start ADSM, it begins like your concept of a
696 base job. After that it is in the "incremental" mode. You can
697 configure the number of "generations" of files to be retained, plus a
698 retention date after which even old generations are purged. The
699 database tracks the contents of media and projects the percentage of
700 each volume that is valid. When the valid content of a volume drops
701 below a configured percentage, the valid data are migrated to another
702 volume and the old volume is marked as empty. Note, this requires
703 ADSM to have an idea of the contents of a client, i.e. marking the
704 database when an existing file was deleted, but this would solve your
705 issue of restoring a client without restoring deleted files.
707 This is pretty far from what bacula now does, but if you are going to
708 rip things up for Base jobs,.....
709 Also, the benefits of this are huge for very large shops, especially
710 with media robots, but are a pain for shops with manual media
714 > Base jobs sound pretty useful, but I'm not dying for them.
716 Nobody is dying for them, but when you see what it does, you will die
719 3. Restoring deleted files: Since I think my comments in (2) above
720 have low probability of implementation, I'll also suggest that you
721 could approach the issue of deleted files by a mechanism of having the
722 fd report to the dir, a list of all files on the client for every
723 backup job. The dir could note in the database entry for each file
724 the date that the file was seen. Then if a restore as of date X takes
725 place, only files that exist from before X until after X would be
726 restored. Probably the major cost here is the extra date container in
727 each row of the files table.
729 Thanks for "listening". I hope some of this helps. If you want to
730 contact me, please send me an email - I read some but not all of the
731 mailing list traffic and might miss a reply there.
733 Please accept my compliments for bacula. It is doing a great job for
734 me!! I sympathize with you in the need to wrestle with excelence in
735 execution vs. excelence in feature inclusion.
740 ==============================
743 - Design at hierarchial storage for Bacula. Migration and Clone.
744 - Implement FSM (File System Modules).
745 - Audit M_ error codes to ensure they are correct and consistent.
746 - Add variable break characters to lex analyzer.
747 Either a bit mask or a string of chars so that
748 the caller can change the break characters.
749 - Make a single T_BREAK to replace T_COMMA, etc.
750 - Ensure that File daemon and Storage daemon can
751 continue a save if the Director goes down (this
752 is NOT currently the case). Must detect socket error,
753 buffer messages for later.
754 - Enhance time/duration input to allow multiple qualifiers e.g. 3d2h
755 - Add ability to backup to two Storage devices (two SD sessions) at
756 the same time -- e.g. onsite, offsite.
757 - Add the ability to consolidate old backup sets (basically do a restore
758 to tape and appropriately update the catalog). Compress Volume sets.
759 Might need to spool via file is only one drive is available.
760 - Compress or consolidate Volumes of old possibly deleted files. Perhaps
761 someway to do so with every volume that has less than x% valid
765 Migration: Move a backup from one Volume to another
766 Clone: Copy a backup -- two Volumes
768 Bacula Migration is based on Jobs (apparently Networker is file by file).
770 Migration triggered by:
774 Highwater mark (keep total size)
779 ======================================================
781 It is somewhat like a Full save becomes an incremental since
782 the Base job (or jobs) plus other non-base files.
784 - A Base backup is same as Full backup, just different type.
785 - New BaseFiles table that contains:
787 BaseJobId - Base JobId referenced for this FileId (needed ???)
788 JobId - JobId currently running
789 FileId - File not backed up, exists in Base Job
790 FileIndex - FileIndex from Base Job.
791 i.e. for each base file that exists but is not saved because
792 it has not changed, the File daemon sends the JobId, BaseId,
793 FileId, FileIndex back to the Director who creates the DB entry.
794 - To initiate a Base save, the Director sends the FD
795 the FileId, and full filename for each file in the Base.
796 - When the FD finds a Base file, he requests the Director to
797 send him the full File entry (stat packet plus MD5), or
798 conversely, the FD sends it to the Director and the Director
799 says yes or no. This can be quite rapid if the FileId is kept
800 by the FD for each Base Filename.
801 - It is probably better to have the comparison done by the FD
802 despite the fact that the File entry must be sent across the
804 - An alternative would be to send the FD the whole File entry
805 from the start. The disadvantage is that it requires a lot of
806 space. The advantage is that it requires less communications
808 - The Job record must be updated to indicate that one or more
810 - At end of Job, FD returns:
811 1. Count of base files/bytes not written to tape (i.e. matches)
812 2. Count of base file that were saved i.e. had changed.
813 - No tape record would be written for a Base file that matches, in the
814 same way that no tape record is written for Incremental jobs where
815 the file is not saved because it is unchanged.
816 - On a restore, all the Base file records must explicitly be
817 found from the BaseFile tape. I.e. for each Full save that is marked
818 to have one or more Base Jobs, search the BaseFile for all occurrences
820 - An optimization might be to make the BaseFile have:
826 This would avoid the need to explicitly fetch each File record for
827 the Base job. The Base Job record will be fetched to get the
828 VolSessionId and VolSessionTime.
829 =========================================================
832 ==========================================================
834 For each Incremental job that is run, there may be files that
835 were found but not saved because they were locked (this applies
836 only to Windows). Such a system could send back to the Director
837 a list of Unsaved files.
839 - New UnSavedFiles table that contains:
843 - Then in the next Incremental job, the list of Unsaved Files will be
844 feed to the FD, who will ensure that they are explicitly chosen even
845 if standard date/time check would not have selected them.
846 =============================================================
849 Done: (see kernsdone for more)
852 - John's Full save failed with 1.32c FD and 1.31 Dir no FD status,
853 and no error message.
854 - Add fd and st as Console keywords.
855 - Recycling volume with a Slot requires an operator intervention:
856 rufus-dir: Start Backup JobId 18, Job=kernsave.2003-11-01_21.23.52
857 rufus-dir: Pruned 1 Job on Volume Vol01 from catalog.
858 rufus-dir: There are no Jobs associated with Volume Vol01. Marking it purged.
859 rufus-dir: Recycled volume "Vol01"
860 rufus-sd: Please mount Volume "Vol01" on Storage Device "DDS-4" for Job kernsave.2003-11-01_21.23.52
861 Use "mount" command to release Job.
862 - Implement Dan's bacula script (email of 26 Oct).
863 - Add JobName= to VerifyToCatalog so that all verifies can be done at the end.
864 - Edit the Client/Storage name into authentication failure messages.
865 - Fix packet too big problem. This is most likely a Windows TCP stack
867 - Implement ClientRunBeforeJob and ClientRunAfterJob.
868 - Implement forward spacing block/file: position_device(bsr) --
869 just before read_block_from_device();
872 Multiple drive autochanger data: see Alan Brown
873 mtx -f xxx unloadStorage Element 1 is Already Full(drive 0 was empty)
874 Unloading Data Transfer Element into Storage Element 1...source Element
877 (drive 0 was empty and so was slot 1)
878 > mtx -f xxx load 15 0
879 no response, just returns to the command prompt when complete.
880 > mtx -f xxx status Storage Changer /dev/changer:2 Drives, 60 Slots ( 2 Import/Export )
881 Data Transfer Element 0:Full (Storage Element 15 Loaded):VolumeTag = HX001
882 Data Transfer Element 1:Empty
883 Storage Element 1:Empty
884 Storage Element 2:Full :VolumeTag=HX002
885 Storage Element 3:Full :VolumeTag=HX003
886 Storage Element 4:Full :VolumeTag=HX004
887 Storage Element 5:Full :VolumeTag=HX005
888 Storage Element 6:Full :VolumeTag=HX006
889 Storage Element 7:Full :VolumeTag=HX007
890 Storage Element 8:Full :VolumeTag=HX008
891 Storage Element 9:Full :VolumeTag=HX009
892 Storage Element 10:Full :VolumeTag=HX010
893 Storage Element 11:Empty
894 Storage Element 12:Empty
895 Storage Element 13:Empty
896 Storage Element 14:Empty
897 Storage Element 15:Empty
898 Storage Element 16:Empty....
899 Storage Element 28:Empty
900 Storage Element 29:Full :VolumeTag=CLNU01L1
901 Storage Element 30:Empty....
902 Storage Element 57:Empty
903 Storage Element 58:Full :VolumeTag=NEX261L2
904 Storage Element 59 IMPORT/EXPORT:Empty
905 Storage Element 60 IMPORT/EXPORT:Empty
907 Unloading Data Transfer Element into Storage Element 15...done
909 (just to verify it remembers where it came from, however it can be
910 overrriden with mtx unload {slotnumber} to go to any storage slot.)
912 There needs to be a table of drive # to devices somewhere - If there are
913 multiple changers or drives there may not be a 1:1 correspondance between
914 changer drive number and system device name - and depending on the way the
915 drives are hooked up to scsi busses, they may not be linearly numbered
916 from an offset point either.something like
918 Autochanger drives = 2
919 Autochanger drive 0 = /dev/nst1
920 Autochanger drive 1 = /dev/nst2
921 IMHO, it would be _safest_ to use explicit mtx unload commands at all
922 times, not just for multidrive changers. For a 1 drive changer, that's
928 MTX's manpage (1.2.15):
929 unload [<slotnum>] [ <drivenum> ]
930 Unloads media from drive <drivenum> into slot
931 <slotnum>. If <drivenum> is omitted, defaults to
932 drive 0 (as do all commands). If <slotnum> is
933 omitted, defaults to the slot that the drive was
934 loaded from. Note that there's currently no way
935 to say 'unload drive 1's media to the slot it
936 came from', other than to explicitly use that
937 slot number as the destination.AB
943 undef# camcontrol devlist
944 <WANGTEK 51000 SCSI M74H 12B3> at scbus0 target 2 lun 0 (pass0,sa0)
945 <ARCHIVE 4586XX 28887-XXX 4BGD> at scbus0 target 4 lun 0 (pass1,sa1)
946 <ARCHIVE 4586XX 28887-XXX 4BGD> at scbus0 target 4 lun 1 (pass2)
948 tapeinfo -f /dev/sg0 with a bad tape in drive 1:
949 [kern@rufus mtx-1.2.17kes]$ ./tapeinfo -f /dev/sg0
950 Product Type: Tape Drive
952 Product ID: 'C5713A '
955 TapeAlert[3]: Hard Error: Uncorrectable read/write error.
956 TapeAlert[20]: Clean Now: The tape drive neads cleaning NOW.
963 Medium Type: Not Loaded
968 DataDeCompEnabled: yes
975 Handling removable disks
977 From: Karl Cunningham <karlc@keckec.com>
979 My backups are only to hard disk these days, in removable bays. This is my
980 idea of how a backup to hard disk would work more smoothly. Some of these
981 things Bacula does already, but I mention them for completeness. If others
982 have better ways to do this, I'd like to hear about it.
984 1. Accommodate several disks, rotated similar to how tapes are. Identified
985 by partition volume ID or perhaps by the name of a subdirectory.
986 2. Abort & notify the admin if the wrong disk is in the bay.
987 3. Write backups to different subdirectories for each machine to be backed
989 4. Volumes (files) get created as needed in the proper subdirectory, one
991 5. When a disk is recycled, remove or zero all old backup files. This is
992 important as the disk being recycled may be close to full. This may be
993 better done manually since the backup files for many machines may be
994 scattered in many subdirectories.
999 - Add "Rerun failed levels = yes/no" to Job resource.
1000 - Fix CDROM make script to permit picking up already installed
1002 - Add date/time to each Jmsg.