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