4 Documentation to do: (any release a little bit at a time)
5 - Document running a test version.
6 - Document query file format.
7 - Document static linking
8 - Document problems with Verify and pruning.
9 - Document how to use multiple databases.
10 - For FreeBSD typical /dev/nrsa0 and for mtx /dev/pass1
11 - VXA drives have a "cleaning required"
12 indicator, but Exabyte recommends preventive cleaning after every 75
15 In this context, it should be noted that Exabyte has a command-line
16 vxatool utility available for free download. (The current version is
17 vxatool-3.72.) It can get diagnostic info, read, write and erase tapes,
18 test the drive, unload tapes, change drive settings, flash new firmware,
20 Of particular interest in this context is that vxatool <device> -i will
21 report, among other details, the time since last cleaning in tape motion
22 minutes. This information can be retrieved (and settings changed, for
23 that matter) through the generic-SCSI device even when Bacula has the
24 regular tape device locked. (Needless to say, I don't recommend
25 changing tape settings while a job is running.)
26 - Lookup HP cleaning recommendations.
27 - Lookup HP tape replacement recommendations (see trouble shooting autochanger)
28 - Create a man page for each binary (Debian package requirement).
30 Testing to do: (painful)
31 - that ALL console command line options work and are always implemented
32 - blocksize recognition code.
33 - Test if rewind at end of tape waits for tape to rewind.
36 For 1.33 Testing/Documentation:
37 - Document new alias records in Director. SDAddress SDDeviceName, SDPassword.
38 FDPassword, FDAddress, DBAddress, DBPort, DBPassword.
39 - Document new Include/Exclude ...
40 - Add test of exclusion, test multiple Include {} statements.
41 - Add counter variable test.
42 - Document ln -sf /usr/lib/libncurses.so /usr/lib/libtermcap.so
43 and install the esound-dev package for compiling Console on SuSE.
44 - Add an example of using a FIFO in dirdconf.wml
45 - Add an item to the FAQ about running jobs in different timezones.
46 - Add some examples of job editing codes.
48 - Document Dan's new --with-dir-user, ... options.
50 - Figure out how to use ssh or stunnel to protect Bacula communications.
51 Add Dan's work to manual
53 - Add db check test to regression. Test each function like delete,
55 - Add subsections to the Disaster Recovery index section.
58 - Look at Dan's field width problems in PostgreSQL.
59 - Look at effect of removing GROUP BYs.
60 - In restore take all filesets with same base name.
61 - From Alan Brown <ajb2@mssl.ucl.ac.uk>
62 BTW, there's a make install bug in 1.33 - with --enable-gnome,
63 gnome-console is built, but the binary and .conf are not being installed.
64 - Make Scheduler sort jobs by StartTime, Priority.
65 - Make sure smtp and any other useful program is executable by the world
66 in case Bacula is not running as root.
67 - Look at updating Volume Jobs so that Max Volume Jobs = 1 will work
68 correctly for multiple simultaneous jobs.
69 - Correct code so that FileSet MD5 is calculated for < and | filename
71 - Permit Bacula and apcupsd donations.
72 - Mark Volume in error on error from WEOF.
73 - Why does Bacula need the drive open to do "autochanger list" ?
74 - Add a .list all files in the restore tree (probably also a list all files)
75 Do both a long and short form.
76 - Add a Media record flag that indicates that the Volume does disk
78 - Implement VolAddr, which is used when Volume is addressed like a disk,
79 and form it from VolFile and VolBlock.
80 - Make multiple restore jobs for multiple media types specifying
81 the proper storage type.
82 - Implement MediaType keyword in bsr?
83 - Fix fast block rejection (stored/read_record.c:118). It passes a null
84 pointer (rec) to try_repositioning().
85 - Look at extracting Win data from BackupRead.
86 - Having dashes in filenames apparently creates problems for restore
88 - Add data compare on write/read in btape "test".
89 - Implement RestoreJobRetention? Maybe better "JobRetention" in a Job,
90 which would take precidence over the Catalog "JobRetention".
91 - Implement Label Format in Add and Label console commands.
92 - Possibly up network buffers to 65K. Put on variable.
93 - Put email tape request delays on one or more variables. User wants
94 to cancel the job after a certain time interval. Maximum Mount Wait?
95 Job, Client, Device, Pool, or Volume?
96 Is it possible to make this a directive which is *optional* in multiple
97 resources, like Level? If so, I think I'd make it an optional directive
98 in Job, Client, and Pool, with precedence such that Job overrides Client
99 which in turn overrides Pool.
100 - Fix Ctl-C crashing the Console (readline?).
101 - Finish work on conio.c
102 - To pass Include 1 or two letter commands
103 I Name Include name - first record
104 B Name Base name - repeat
107 E zzz Exclude expression (wild card)
109 D "reader" Reader program
110 T "writer" Writer program
111 O Options In current commpressed format
112 (compression, signature, onefs, recurse, sparse,
113 replace, verify options, ...)
115 B BaseName Start second option set any letter
118 F Number Number of filenames to follow
123 F Number Number of filenames to follow
126 - New Storage specifications:
127 Passed to SD as a sort of BSR record called Storage Specification
131 MediaType -> Next MediaType
133 Device -> Next Device
134 Write Copy Resource that makes a copy of a resource.
136 Allow multiple Storage specifications
144 Allow Multiple Pool specifications (note, Pool currently
146 Allow Multiple MediaType specifications
147 Allow Multiple Device specifications
148 Perhaps keep this in a single SSR
149 Tie a Volume to a specific device by using a MediaType that
150 is contained in only one device.
151 In SD allow Device to have Multiple MediaTypes
154 - Look at adding SQL server and Exchange support for Windows.
155 - Restore: Enter Filename: 'C:/Documents and Settings/Comercial/My
156 Documents/MOP/formulário de registro BELAS ARTES.doc' causes Bacula to
158 - Each DVD-RAM disk would be a volume, just like each tape is
159 a volume. It's a 4.7GB media with random access, but there's nothing about
160 it that I can see that makes it so different than a tape from bacula's
161 perspective. Why couldn't I back up to a bare floppy as a volume (ignoring
163 - Make dev->file and dev->block_num signed integers so that -1 can
164 be an invalid value which happens with BSR.
165 - Create VolAddr for disk files in place of VolFile and VolBlock. This
166 is needed to properly specify ranges.
167 - Print bsmtp output to job report so that problems will be seen.
168 - Pass the number of files to be restored to the FD for reporting
169 - Add progress of files/bytes to SD and FD.
170 - Don't continue Restore if no files selected.
171 - Print warning message if FileId > 4 billion
172 - do a "messages" before the first prompt in Console
173 - Add a date and time stamp at the beginning of every line in the
174 Job report (Volker Sauer).
175 - Client does not show busy during Estimate command.
176 - Implement Console mtx commands.
177 - Look at 2Gb limit for SQLite.
178 - Implement 3 Pools for a Job:
181 Full Backup Pool = xxx
182 Incremental Backup Pool = yyy
183 Differential Backup Pool = zzz
185 - Add a default DB password to MySQL.
186 GRANT all privileges ON bacula.* TO bacula@localhost IDENTIFIED BY
189 - Define week of year for scheduler. W01, W02, ...
190 Week 01 of a year is per definition the first week that has the
191 Thursday in this year, which is equivalent to the week that contains the
192 fourth day of January. In other words, the first week of a new year is
193 the week that has the majority of its days in the new year. Week 01
194 might also contain days from the previous year and the week before week
195 01 of a year is the last week (52 or 53) of the previous year even if it
196 contains days from the new year. A week starts with Monday (day 1) and
197 ends with Sunday (day 7). For example, the first week of the year 1997
198 lasts from 1996-12-30 to 1997-01-05 and can be written in standard
203 The week notation can also be extended by a number indicating the day
204 of the week. For example, the day 1996-12-31, which is the Tuesday (day
205 2) of the first week of 1997, can also be written as
207 1997-W01-2 or 1997W012
209 - Implement a Mount Command and an Unmount Command where
210 the users could specify a system command to be performed
211 to do the mount, after which Bacula could attempt to
212 read the device. This is for Removeable media such as a CDROM.
213 - Most likely, this mount command would be invoked explicitly
214 by the user using the current Console "mount" and "unmount"
215 commands -- the Storage Daemon would do the right thing
216 depending on the exact nature of the device.
217 - As with tape drives, when Bacula wanted a new removable
218 disk mounted, it would unmount the old one, and send a message
219 to the user, who would then use "mount" as described above
220 once he had actually inserted the disk.
221 - Implement dump/print label to UA
222 - Implement disk spooling. Two parts: 1. Spool to disk then
223 immediately to tape to speed up tape operations. 2. Spool to
224 disk only when the tape is full, then when a tape is hung move
226 - Scratch Pool where the volumes can be re-assigned to any Pool.
227 - bextract is sending everything to the log file ****FIXME****
228 - Add Progress command that periodically reports the progress of
230 - Restrict characters permitted in a Resource name, and don't permit
232 - Allow multiple Storage specifications (or multiple names on
233 a single Storage specification) in the Job record. Thus a job
234 can be backed up to a number of storage devices.
235 - Implement some way for the File daemon to contact the Director
236 to start a job or pass its DHCP obtained IP number.
237 - Implement multiple Consoles.
238 - Implement a query tape prompt/replace feature for a console
240 Two jobs ready to go, first one blocked waiting for media
241 Cancel 2nd job ("waiting execution" one)
244 - Copy console @ code to gnome2-console
245 - Make AES the only encryption algorithm see
246 http://csrc.nist.gov/CryptoToolkit/aes/). It's
247 an officially adopted standard, has survived peer
248 review, and provides keys up to 256 bits.
249 - Add ctl-c to console to stop current command and discard buffered
251 - Estimate to Tibs never returns.
252 - Think about how space could be freed up on a tape -- perhaps this
253 is a Merge or Compact feature that is needed.
254 - Modify FileSet, did not upgrade the current Increment job, but
255 waited for the next job to be upgraded.
256 - Take a careful look at SetACL http://setacl.sourceforge.net
257 - Implement a where command for the tree telling where a file
259 - Take a careful look at Level for the estimate command, maybe make
260 it a command line option.
261 - Add Volume name to "I cannot write on this volume because"
262 - Make restore job check if all the files are actually restored.
263 - Make tree walk routines like cd, ls, ... more user friendly
264 by handling spaces better.
265 - Write your PID file and chown root:wheel before drop.
266 - Make sure there is no symlink in a file before creating a
268 - Look at mktemp or mkstemp(3).
269 mktemp and mkstemp create files with predictable names too. That's
270 not the vulnerability. The vulnerability is in creating files without
271 using the O_EXCL flag, which means "only create this file if it doesn't
272 exist, including if the file is a dangling symlink."
274 It is *NOT* enough to do the equivalent of
276 if doesn't exist $filename
277 then create $filename
279 because between the test and the create another process could have
280 gotten the CPU and created the file. You must use atomic functions
281 (those that don't get interrupted by other processes) and O_EXCL is
282 the only way for this particular example.
283 - Automatically create pools, but instead of looking for what
284 in in Job records, walk through the pool resources.
285 - Check and double check tree code, why does it take so long?
286 - Add device name to "Current Volume not acceptable because ..."
287 - Make sure that Bacula rechecks the tape after the 20 min wait.
288 - Set IO_NOWAIT on Bacula TCP/IP packets.
289 - Try doing a raw partition backup and restore by mounting a
291 - Report CVS problems to SourceForge.
292 - Implement .consolerc for Console
294 Yes, it would allow to highly automatic the request for new tapes. If a
295 tape is empty, bacula reads the barcodes (native or simulated), and if
296 an unused tape is found, it runs the label command with all the
297 necessary parameters.
299 By the way can bacula automatically "move" an empty/purged volume say
300 in the "short" pool to the "long" pool if this pool runs out of volume
302 - Either restrict the characters in a name, or fix the problem
303 emailing with names containing / (smtp command line breaks).
304 - Eliminate orphaned jobs: dbcheck, normal pruning, delete job command.
305 Hm. Well, there are the remaining orphaned job records:
307 | 105 | Llioness Save | 0000-00-00 00:00:00 | B | D | 0 | 0 | f |
308 | 110 | Llioness Save | 0000-00-00 00:00:00 | B | I | 0 | 0 | f |
309 | 115 | Llioness Save | 2003-09-10 02:22:03 | B | I | 0 | 0 | A |
310 | 128 | Catalog Save | 2003-09-11 03:53:32 | B | I | 0 | 0 | C |
311 | 131 | Catalog Save | 0000-00-00 00:00:00 | B | I | 0 | 0 | f |
313 As you can see, three of the five are failures. I already deleted the
314 one restore and one other failure using the by-client option. Deciding
315 what is an orphaned job is a tricky problem though, I agree. All these
316 records have or had 0 files/ 0 bytes, except for the restore. With no
317 files, of course, I don't know of the job ever actually becomes
318 associated with a Volume.
320 (I'm not sure if this is documented anywhere -- what are the meanings of
321 all the possible JobStatus codes?)
323 Looking at my database, it appears to me as though all the "orphaned"
324 jobs fit into one of two categories:
326 1) The Job record has a StartTime but no EndTime, and the job is not
329 2) The Job record has an EndTime, indicating that it completed, but
330 it has no associated JobMedia record.
333 This does suggest an approach. If failed jobs (or jobs that, for some
334 other reason, write no files) are associated with a volume via a
335 JobMedia record, then they should be purged when the associated volume
336 is purged. I see two ways to handle jobs that are NOT associated with a
339 1) purge them automatically whenever any volume is manually purged;
341 2) add an option to the purge command to manually purge all jobs with
342 no associated volume.
344 I think Restore jobs also fall into category 2 above .... so one might
345 want to make that "The Job record has an EndTime,, but no associated
346 JobMedia record, and is not a Restore job."
347 - make "btape /tmp" work.
348 - Make sure a rescheduled job is properly reported by status.
349 - Walk through the Pool records rather than the Job records
350 in dird.c to create/update pools.
351 - What to do about "list files job=xxx".
352 - Implement scan: for every slot it finds, zero the slot of
353 Volume other volume having that slot.
354 - When job rescheduled, status gives is waiting for Client Rufus
355 to connect to Storage File. Dir needs to inform SD that job
357 - Fix get_storage_from_media_type (ua_restore) to use command line
359 - Don't print "Warning: Wrong Volume mounted ..." if mounting second volume.
360 - Make Dmsg look at global before calling subroutine.
361 - Enable trace output at runtime for Win32
362 - Available volumes for autochangers (see patrick@baanboard.com 3 Sep 03
363 and 4 Sep) scan slots.
364 - Get and test MySQL 4.0
365 - Do a complete audit of all pthreads_mutex, cond, ... to ensure that
366 any that are dynamically initialized are destroyed when no longer used.
367 - Write a mini-readline with history and editing.
368 - Look at how fuser works and /proc/PID/fd that is how Nic found the
369 file descriptor leak in Bacula.
370 - Implement WrapCounters in Counters.
371 - Turn on SIGHUP in dird.c and test.
372 - Use system dependent calls to get more precise info on tape errors.
373 - Add heartbeat from FD to SD if hb interval expires.
374 - Suppress read error on blank tape when doing a label.
375 - Can we dynamically change FileSets?
376 - If pool specified to label command and Label Format is specified,
377 automatically generate the Volume name.
378 - Take a careful look a the Basic recycling algorithm. When Bacula
379 chooses, the order should be:
381 - Look for Recycle or Purged
384 Instead of using lowest media Id, find the least recently used
387 When the tape is mounted and Bacula requests the status
388 - Do everything possible to use it.
390 Define a "available" status, which is the currently mounted
391 Volume and all volumes that are currently in the autochanger.
393 - Why can't SQL do the filename sort for restore?
394 - Is a pool specification really needed for a restore? Yes, and
395 you may want to exclude archive Pools.
396 - Look at libkse (man kse) for FreeBSD threading.
397 - Look into Microsoft Volume Shadowcopy Service VSS for backing
398 up system state components (Active Directory, System Volume, ...)
399 - Add ExhautiveRestoreSearch
400 - Look at the possibility of loading only the necessary
401 data into the restore tree (i.e. do it one directory at a
402 time as the user walks through the tree).
403 - Possibly use the hash code if the user selects all for a restore command.
404 - Orphaned Dir buffer at parse_conf.c:373 => store_dir
405 - Add Console usr permissions.
406 - Fix "restore all" to bypass building the tree.
407 - Fix restore to list errors if Invalid block found, and if # files
408 restored does not match # expected.
409 - Prohibit backing up archive device (findlib/find_one.c:128)
410 - Implement Release Device in the Job resource to unmount a drive.
411 - Implement Acquire Device in the Job resource to mount a drive,
412 be sure this works with admin jobs so that the user can get
413 prompted to insert the correct tape. Possibly some way to say to
414 run the job but don't save the files.
415 - Implement FileOptions (see end of this document)
416 - Make things like list where a file is saved case independent for
419 - Implement a PostgreSQL driver.
420 - Bacula needs to propagate SD errors.
421 > > cluster-dir: Start Backup JobId 252, Job=REUTERS.2003-08-11_15.04.12
422 > > prod4-sd: REUTERS.2003-08-11_15.04.12 Error: Write error on device
423 /dev/nst0. ERR=Input/output error.
424 > > prod4-sd: REUTERS.2003-08-11_15.04.12 Error: Re-read of last block failed.
425 Last block=5162 Current block=5164.
426 > > prod4-sd: End of medium on Volume "REU007" Bytes=16,303,521,933
428 - Use autochanger to handle multiple devices.
429 - Add SuSE install doc to list.
430 - Check and rechedk "Invalid block number"
431 - Make bextract release the drive properly between tapes
432 so that an autochanger can be made to work.
433 - User wants to NOT backup up certain big files (email files).
434 - Maybe remove multiple simultaneous devices code in SD.
435 - On Windows with very long path names, it may be impossible to create
436 a file (and thus restore it) because the total length is too long.
437 We must cd into the directory then create the file without the
439 - lstat() is not going to work on Win32 for testing date.
440 - Something is not right in last block of fill command.
441 - Implement a Recycle command
442 - Add FileSet to command line arguments for restore.
443 - Add client name to cram-md5 challenge so Director can immediately
444 verify if it is the correct client.
445 - Add JobLevel in FD status (but make sure it is defined).
446 - Audit all UA commands to ensure that we always prompt where possible.
447 - Check Jmsg in bnet, may not work, must dup bsock.
448 - Suppress Job Name in Jmsg for console
449 - Create Pools that are referenced in a Run statement at startup if possible.
450 - Use runbeforejob to unload, then reload a volume previously used,
451 then the next job run gets an error reading the drive.
452 - Make bootstrap filename unique.
453 - Test a second language e.g. french.
454 - Start working on Base jobs.
455 - Make "make binary-release" work from any directory.
456 - Implement UnsavedFiles DB record.
457 - Implement argc/argv for daemon command line scanning using table driven
459 - Implement table driven single argc/argv scanner to pickup all arguments.
460 Much like xxx_conf.c scan table.
461 keyword, handler(store_routine), store_address, code, flags, default.
462 - From Phil Stracchino:
463 It would probably be a per-client option, and would be called
464 something like, say, "Automatically purge obsoleted jobs". What it
465 would do is, when you successfully complete a Differential backup of a
466 client, it would automatically purge all Incremental backups for that
467 client that are rendered redundant by that Differential. Likewise,
468 when a Full backup on a client completed, it would automatically purge
469 all Differential and Incremental jobs obsoleted by that Full backup.
470 This would let people minimize the number of tapes they're keeping on
471 hand without having to master the art of retention times.
472 - Implement a M_SECURITY message class.
473 - When doing a Backup send all attributes back to the Director, who
474 would then figure out what files have been deleted.
475 - Currently in mount.c:236 the SD simply creates a Volume. It should have
476 explicit permission to do so. It should also mark the tape in error
477 if there is an error.
478 - Make sure all restore counters are working correctly in the FD.
479 - SD Bytes Read is wrong.
480 - Look at ALL higher level routines that call block.c to be sure
481 they don't expect something in errmsg.
482 - Investigate doing RAW backup of Win32 partition.
483 - Add thread specific data to hold the jcr -- send error messages from
484 low level routines by accessing it and using Jmsg().
485 - Cancel waiting for Client connect in SD if FD goes away.
487 - Examine Bare Metal restore problem (a FD crash exists somewhere ...).
488 - Implement timeout in response() when it should come quickly.
489 - Implement console @echo command.
490 - Implement a Slot priority (loaded/not loaded).
491 - Implement "vacation" Incremental only saves.
492 - Implement single pane restore (much like the Gftp panes).
493 - Implement Automatic Mount even in operator wait.
494 - Implement create "FileSet"?
495 - Fix watchdog pthread crash on Win32 (this is pthread_kill() Cygwin bug)
496 - Implement "scratch pool" where tapes are defined and can be
497 taken by any pool that needs them.
498 - Implement restore "current system", but take all files without
499 doing selection tree -- so that jobs without File records can
501 - Implement a relocatable bacula.spec
502 - Add prefixlinks to where or not where absolute links to FD.
503 - Issue message to mount a new tape before the rewind.
504 - Simplified client job initiation for portables.
505 - If SD cannot open a drive, make it periodically retry.
506 - Add more of the config info to the tape label.
508 - If tape is marked read-only, then try opening it read-only rather than
509 failing, and remember that it cannot be written.
510 - Refine SD waiting output:
511 Device is being positioned
512 > Device is being positioned for append
513 > Device is being positioned to file x
515 - Figure out some way to estimate output size and to avoid splitting
516 a backup across two Volumes -- this could be useful for writing CDROMs
517 where you really prefer not to have it split -- not serious.
518 - Have SD compute MD5 or SHA1 and compare to what FD computes.
519 - Make VolumeToCatalog calculate an MD5 or SHA1 from the
520 actual data on the Volume and compare it.
521 - Implement Bacula plugins -- design API
522 - Make bcopy read through bad tape records.
523 - Program files (i.e. execute a program to read/write files).
524 Pass read date of last backup, size of file last time.
525 - Add Signature type to File DB record.
526 - CD into subdirectory when open()ing files for backup to
527 speed up things. Test with testfind().
528 - Priority job to go to top of list.
529 - Why are save/restore of device different sizes (sparse?) Yup! Fix it.
530 - Implement some way for the Console to dynamically create a job.
531 - Restore to a particular time -- e.g. before date, after date.
532 - Solaris -I on tar for include list
533 - Need a verbose mode in restore, perhaps to bsr.
534 - bscan without -v is too quiet -- perhaps show jobs.
535 - Add code to reject whole blocks if not wanted on restore.
536 - Check if we can increase Bacula FD priorty in Win2000
537 - Make sure the MaxVolFiles is fully implemented in SD
538 - Check if both CatalogFiles and UseCatalog are set to SD.
539 - Figure out how to do a bare metal Windows restore
540 - Possibly add email to Watchdog if drive is unmounted too
541 long and a job is waiting on the drive.
542 - Restore program that errs in SD due to no tape, reports
543 OK incorrectly in output.
544 - After unmount, if restore job started, ask to mount.
545 - Convert all %x substitution variables, which are hard to remember
546 and read to %(variable-name). Idea from TMDA.
547 - Remove NextId for SQLite. Optimize.
548 - Move all SQL statements into a single location.
549 - Add UA rc and history files.
550 - put termcap (used by console) in ./configure and
551 allow -with-termcap-dir.
552 - Enhance time and size scanning routines.
553 - Fix Autoprune for Volumes to respect need for full save.
554 - Fix Win32 config file definition name on /install
555 - Compare tape to Client files (attributes, or attributes and data)
556 - Make all database Ids 64 bit.
557 - Write an applet for Linux.
558 - Allow console commands to detach or run in background.
559 - Fix status delay on storage daemon during rewind.
560 - Add SD message variables to control operator wait time
561 - Maximum Operator Wait
562 - Minimum Message Interval
563 - Maximum Message Interval
564 - Send Operator message when cannot read tape label.
565 - Verify level=Volume (scan only), level=Data (compare of data to file).
566 Verify level=Catalog, level=InitCatalog
568 - Add keyword search to show command in Console.
569 - Events : tape has more than xxx bytes.
570 - Complete code in Bacula Resources -- this will permit
571 reading a new config file at any time.
572 - Handle ctl-c in Console
573 - Implement script driven addition of File daemon to config files.
574 - Think about how to make Bacula work better with File (non-tape) archives.
575 - Write Unix emulator for Windows.
576 - Put memory utilization in Status output of each daemon
577 if full status requested or if some level of debug on.
578 - Make database type selectable by .conf files i.e. at runtime
579 - Set flag for uname -a. Add to Volume label.
580 - Implement throttled work queue.
581 - Restore files modified after date
582 - SET LD_RUN_PATH=$HOME/mysql/lib/mysql
583 - Implement Restore FileSet=
584 - Create a protocol.h and protocol.c where all protocol messages
586 - Remove duplicate fields from jcr (e.g. jcr.level and jcr.jr.Level, ...).
587 - Timout a job or terminate if link goes down, or reopen link and query.
588 - Concept of precious tapes (cannot be reused).
589 - Make bcopy copy with a single tape drive.
590 - Permit changing ownership during restore.
593 > My suggestion: Add a feature on the systray menu-icon menu to request
594 > an immediate backup now. This would be useful for laptop users who may
595 > not be on the network when the regular scheduled backup is run.
597 > My wife's suggestion: Add a setting to the win32 client to allow it to
598 > shut down the machine after backup is complete (after, of course,
599 > displaying a "System will shut down in one minute, click here to cancel"
600 > warning dialog). This would be useful for sites that want user
601 > woorkstations to be shut down overnight to save power.
604 - From Terry Manderson <terry@apnic.net>
605 jobdefs { # new structure
606 name = "monthlyUnixBoxen"
618 jobdefs = "genericUnixSet"
624 jobdefs = "genericUnixSet"
630 jobdefs = "UnixDevelBoxSet"
635 - Autolabel should be specified by DIR instead of SD.
638 - AutoScan (check checksum of tape)
639 - Format command = "format /dev/nst0"
643 - Seek resolution (usually corresponds to buffer size)
644 - EODErrorCode=ENOSPC or code
645 - Partial Read error code
646 - Partial write error code
647 - Nonformatted read error
648 - Nonformatted write error
649 - WriteProtected error
653 - IgnoreCloseErrors=yes
663 - FD sends unsaved file list to Director at end of job (see
665 - File daemon should build list of files skipped, and then
666 at end of save retry and report any errors.
667 - Write a Storage daemon that uses pipes and
668 standard Unix programs to write to the tape.
670 - Need something that monitors the JCR queue and
671 times out jobs by asking the deamons where they are.
672 - Enhance Jmsg code to permit buffering and saving to disk.
673 - device driver = "xxxx" for drives.
675 - Ensure that /dev/null works
676 - Need report class for messages. Perhaps
677 report resource where report=group of messages
678 - enhance scan_attrib and rename scan_jobtype, and
679 fill in code for "since" option
680 - Director needs a time after which the report status is sent
681 anyway -- or better yet, a retry time for the job.
682 - Don't reschedule a job if previous incarnation is still running.
683 - Some way to automatically backup everything is needed????
684 - Need a structure for pending actions:
686 - termination status (part of buffered msgs?)
688 Read, Write, Clean, Delete
689 - Login to Bacula; Bacula users with different permissions:
690 owner, group, user, quotas
691 - Store info on each file system type (probably in the job header on tape.
692 This could be the output of df; or perhaps some sort of /etc/mtab record.
695 - Design at hierarchial storage for Bacula. Migration and Clone.
696 - Implement FSM (File System Modules).
697 - Audit M_ error codes to ensure they are correct and consistent.
698 - Add variable break characters to lex analyzer.
699 Either a bit mask or a string of chars so that
700 the caller can change the break characters.
701 - Make a single T_BREAK to replace T_COMMA, etc.
702 - Ensure that File daemon and Storage daemon can
703 continue a save if the Director goes down (this
704 is NOT currently the case). Must detect socket error,
705 buffer messages for later.
706 - Enhance time/duration input to allow multiple qualifiers e.g. 3d2h
707 - Add ability to backup to two Storage devices (two SD sessions) at
708 the same time -- e.g. onsite, offsite.
709 - Add the ability to consolidate old backup sets (basically do a restore
710 to tape and appropriately update the catalog). Compress Volume sets.
711 Might need to spool via file is only one drive is available.
712 - Compress or consolidate Volumes of old possibly deleted files. Perhaps
713 someway to do so with every volume that has less than x% valid
717 Migration: Move a backup from one Volume to another
718 Clone: Copy a backup -- two Volumes
720 Bacula Migration is based on Jobs (apparently Networker is file by file).
722 Migration triggered by:
726 Highwater mark (keep total size)
731 ======================================================
733 It is somewhat like a Full save becomes an incremental since
734 the Base job (or jobs) plus other non-base files.
736 - A Base backup is same as Full backup, just different type.
737 - New BaseFiles table that contains:
739 BaseJobId - Base JobId referenced for this FileId (needed ???)
740 JobId - JobId currently running
741 FileId - File not backed up, exists in Base Job
742 FileIndex - FileIndex from Base Job.
743 i.e. for each base file that exists but is not saved because
744 it has not changed, the File daemon sends the JobId, BaseId,
745 FileId, FileIndex back to the Director who creates the DB entry.
746 - To initiate a Base save, the Director sends the FD
747 the FileId, and full filename for each file in the Base.
748 - When the FD finds a Base file, he requests the Director to
749 send him the full File entry (stat packet plus MD5), or
750 conversely, the FD sends it to the Director and the Director
751 says yes or no. This can be quite rapid if the FileId is kept
752 by the FD for each Base Filename.
753 - It is probably better to have the comparison done by the FD
754 despite the fact that the File entry must be sent across the
756 - An alternative would be to send the FD the whole File entry
757 from the start. The disadvantage is that it requires a lot of
758 space. The advantage is that it requires less communications
760 - The Job record must be updated to indicate that one or more
762 - At end of Job, FD returns:
763 1. Count of base files/bytes not written to tape (i.e. matches)
764 2. Count of base file that were saved i.e. had changed.
765 - No tape record would be written for a Base file that matches, in the
766 same way that no tape record is written for Incremental jobs where
767 the file is not saved because it is unchanged.
768 - On a restore, all the Base file records must explicitly be
769 found from the BaseFile tape. I.e. for each Full save that is marked
770 to have one or more Base Jobs, search the BaseFile for all occurrences
772 - An optimization might be to make the BaseFile have:
778 This would avoid the need to explicitly fetch each File record for
779 the Base job. The Base Job record will be fetched to get the
780 VolSessionId and VolSessionTime.
781 =========================================================
785 =============================================================
787 Request For Comments For File Backup Options
790 Subject: File Backup Options
793 A few days ago, a Bacula user who is backing up to file volumes and
794 using compression asked if it was possible to suppress compressing
795 all .gz files since it was a waste of CPU time. Although Bacula
796 currently permits using different options (compression, ...) on
797 a directory by directory basis, it cannot do it on a file by
798 file basis, which is clearly what was desired.
800 Proposed Implementation:
801 To solve this problem, I propose the following:
803 - Add a new Director resource type called Options.
805 - The Options resource will have records for all
806 options that can currently be specified on the Include record
807 (in a FileSet). Examples below.
809 - The Options resource will permit an exclude option as well
810 as a number of additional options.
812 - The heart of the Options resource is the ability to
813 supply any number of Match records which specify POSIX
814 regular expressions. These Match regular expressions are
815 applied to the fully qualified filename (path and all). If
816 one matches, then the Options will be used.
818 - When an Match specification matches an included file, the
819 options specified in the Options resource will override
820 the default options specified on the Include record.
822 - Include records will be modified to permit referencing one or
823 more Options resources. The Options will be used
824 in the order listed on the Include record and the first
825 one that matches will be applied.
827 - Options (or specifications) currently supplied on the Include
828 record will be deprecated (i.e. removed in a later version a
829 year or so from now).
831 - The Exclude record will be deprecated as the same functionality
832 can be obtained by using an Exclude = yes in the Options.
835 The following records can appear in the Options resource. An
836 asterisk preceding the name indicates a feature not currently
839 - Regexp "xxx" - Match regular expression
840 - Wild "xxx" - Do a wild card match
843 - Compression= (GZIP, ...)
844 - Signature= (MD5, SHA1, ...)
846 - OneFs= (yes/no) - remain on one filesystem
847 - Recurse= (yes/no) - recurse into subdirectories
848 - Sparse= (yes/no) - do sparse file backup
849 - *Exclude= (yes/no) - exclude file from being saved
850 - *Reader= (filename) - external read (backup) program
851 - *Plugin= (filename) - read/write plugin module
853 - Include= (yes/no) - Include the file matched no additional
854 patterns are applied.
857 - verify= (ipnougsamc5) - verify options
860 - replace= (always/ifnewer/ifolder/never) - replace options currently
862 - *Writer= (filename) - external write (restore) program
866 Currently options specifying compression, MD5 signatures, recursion,
867 ... of a FileSet are supplied on the Include record. These will now
868 all be collected into a Options resource, which will be
869 specified in the Include in place of the options. Multiple Options
870 may be specified. Since the Options may contain regular expressions
871 that are applied to the full filename, this will give the ability
872 to specify backup options on a file by file basis to whatever level
881 Include = compression=GZIP signature=MD5 {
893 Wild = /*.?*/ # matches all files.
898 That's a lot more to do the same thing, but it gives the ability to
899 apply options on a file by file basis. For example, suppose you
900 want to compress all files but not any file with extensions .gz or .Z.
901 In that case, you will need to group two sets of options using
902 the Options resource. Files may be anywhere except in an
903 option set??? All OptionSets apply to all files in the order
904 the OptionSets were specified. To have files included with
905 different option sets without using wild-cards, use two or more
906 Includes -- each one is handled in turn using only the files and
907 optionsets specified in the include.
914 # Note multiple Matches are ORed
915 Wild = "*.gz" # matches .gz files
916 Wild = "*.Z" # matches .Z files
921 Wild = "*.?*" # matches all files
927 Now, since the no Compression option is specified in the
928 first group of Options, *.gz or *.Z file will have an MD5 signature computed,
929 but will not be compressed. For all other files, the *.gz *.Z will not
930 match, so the second group of options will be used which will include GZIP
934 - Is it necessary to provide some means of ANDing regular expressions
935 and negation? (not currently planned)
937 e.g. Wild = /*.gz/ && !/big.gz/
939 - I see that Networker has a "null" module which, if specified, does not
940 backup the file, but does make an record of the file in the catalog
941 so that the catalog will reflect an exact picture of the filesystem.
942 The result is that the file can be "seen" when "browsing" the save
943 sets, but it cannot be restored.
945 Is this really useful? Should it be implemented in Bacula?
948 After implementing the above, the user will be able to specify
949 on a file by file basis (using regular expressions) what options are
950 applied for the backup.
953 =============================================
955 ==========================================================
957 For each Incremental job that is run, there may be files that
958 were found but not saved because they were locked (this applies
959 only to Windows). Such a system could send back to the Director
960 a list of Unsaved files.
962 - New UnSavedFiles table that contains:
966 - Then in the next Incremental job, the list of Unsaved Files will be
967 feed to the FD, who will ensure that they are explicitly chosen even
968 if standard date/time check would not have selected them.
969 =============================================================
972 Done: (see kernsdone for more)
975 - John's Full save failed with 1.32c FD and 1.31 Dir no FD status,
976 and no error message.
977 - Add fd and st as Console keywords.
978 - Recycling volume with a Slot requires an operator intervention:
979 rufus-dir: Start Backup JobId 18, Job=kernsave.2003-11-01_21.23.52
980 rufus-dir: Pruned 1 Job on Volume Vol01 from catalog.
981 rufus-dir: There are no Jobs associated with Volume Vol01. Marking it purged.
982 rufus-dir: Recycled volume "Vol01"
983 rufus-sd: Please mount Volume "Vol01" on Storage Device "DDS-4" for Job kernsave.2003-11-01_21.23.52
984 Use "mount" command to release Job.
985 - Implement Dan's bacula script (email of 26 Oct).
986 - Add JobName= to VerifyToCatalog so that all verifies can be done at the end.
987 - Edit the Client/Storage name into authentication failure messages.
988 - Fix packet too big problem. This is most likely a Windows TCP stack
990 - Implement ClientRunBeforeJob and ClientRunAfterJob.
991 - Implement forward spacing block/file: position_device(bsr) --
992 just before read_block_from_device();
994 - Change console to bconsole.
995 - Change smtp to bsmtp.
996 - Fix time difference problem between Bacula and Client
997 so that everything is in GMT.
998 - Fix TimeZone problem!
999 - Mount a tape that is not right for the job (wrong # files on tape)
1000 Bacula asks for another tape, fix problems with first tape and
1001 say "mount". All works OK, but status shows:
1002 Device /dev/nst0 open but no Bacula volume is mounted.
1003 Total Bytes=1,153,820,213 Blocks=17,888 Bytes/block=64,502
1004 Positioned at File=9 Block=3,951
1005 Full Backup job Rufus.2003-10-26_16.45.31 using Volume "DLT-24Oct03" on device /dev/nst0
1006 Files=21,003 Bytes=253,954,408 Bytes/sec=2,919,016
1007 FDReadSeqNo=192,134 in_msg=129830 out_msg=5 fd=7
1008 - Upgrade to cygwin 1.5
1009 - Optimize fsf not to read.
1010 - Use ioctl() fsf if it exists. Figure out where we are from
1011 the mt_status command. Use slow fsf only if other does not work.
1012 - Enhance "update slots" to include a "scan" feature
1013 scan 1; scan 1-5; scan 1,2,4 ... to update the catalog
1014 - Allow a slot or range of slots on the label barcodes command.
1015 - Finish implementation of Verify=DiskToCatalog
1016 - Make sure that Volumes are recycled based on "Least recently used"
1017 rather than lowest MediaId.
1018 - Add flag to write only one EOF mark on the tape.
1019 - Implement autochanger testing in btape "test" command.
1020 - Implement lmark to list everyfile marked.
1021 - Make mark/unmark report how many files marked/unmarked.
1022 - Keep last 5 or 10 completed jobs and show them in a similar list.
1023 - Make a Running Jobs: output similar to current Scheduled Jobs:
1024 - Change "create_media_record in bscan to use Archive instead of Full.
1025 - Have some way to estimate the restore size or have it printed.
1026 - Volume problems occurs if you have valid volume, written, then it is
1027 truncated. You get 12-Nov-2003 11:48 rufus-sd: kernsave.2003-11-12_11.48.09 Warning: mount.c:228 Volume on /tmp is not a Bacula labeled Volume, because:
1028 block.c:640 Read zero bytes on device /tmp.
1029 - Make sure that 64 bit I/O packets are used on Cygwin.
1030 - Add to supported autochangers
1032 Auto-Changer : QUALSTAR TLS-4210
1033 Manufufactur : Qualstar
1034 Tapes : 12 (AIT1: 36GB, AIT2: 50GB all uncompressed)
1035 Drives : 2xAIT2 (installed in the Qualstar: SONY SDX-500C AIT2)
1036 - Document estimate command in tree.
1037 - Document lsmark command in tree.
1038 - Setup a standard job that builds a bootstrap file and saves
1039 it with the catalog database.
1040 - See if a restore job can add a file to the tape (prohibit this).