4 Documentation to do: (any release a little bit at a time)
5 - DB upgrade to version 5 in bacula-1.27b, DB upgrade to
6 version 6 in 1.31; DB upgrade to version 7 in 1.33/4.
7 - Document running a test version.
8 - Document query file format.
9 - Document static linking
10 - Document problems with Verify and pruning.
11 - Document how to use multiple databases.
12 - For FreeBSD typical /dev/nrsa0 and for mtx /dev/pass1
13 - VXA drives have a "cleaning required"
14 indicator, but Exabyte recommends preventive cleaning after every 75
17 In this context, it should be noted that Exabyte has a command-line
18 vxatool utility available for free download. (The current version is
19 vxatool-3.72.) It can get diagnostic info, read, write and erase tapes,
20 test the drive, unload tapes, change drive settings, flash new firmware,
22 Of particular interest in this context is that vxatool <device> -i will
23 report, among other details, the time since last cleaning in tape motion
24 minutes. This information can be retrieved (and settings changed, for
25 that matter) through the generic-SCSI device even when Bacula has the
26 regular tape device locked. (Needless to say, I don't recommend
27 changing tape settings while a job is running.)
28 - Lookup HP cleaning recommendations.
29 - Lookup HP tape replacement recommendations (see trouble shooting autochanger)
30 - Create a man page for each binary (Debian package requirement).
32 Testing to do: (painful)
34 - that ALL console command line options work and are always implemented
35 - blocksize recognition code.
36 - Test if rewind at end of tape waits for tape to rewind.
39 For 1.33 Testing/Documentation:
40 - Document new Include/Exclude ...
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 This should read 'Document LDFLAGS="-L/usr/lib/termcap" ... '
45 - Add an example of using a FIFO in dirdconf.wml
46 - Add an item to the FAQ about running jobs in different timezones.
47 - 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 See ssl.txt
52 - Add db check test to regression. Test each function like delete,
54 - Add subsections to the Disaster Recovery index section.
55 - Document Pool keyword for restore.
56 - If you use restore replace=never, the directory attributes for
57 non-existent directories will not be restored properly.
58 - In the Bacula User Guide you write:"Note, one major disadvantage of
59 writing to a NFS mounted volume as I do isthat if the other machine goes
60 down, the OS will wait forever on the fopen()call that Bacula makes. As
61 a consequence, Bacula will completely stall untilthe machine exporting
62 the NSF mounts comes back up. If someone knows a wayaround this, please
63 let me know."I haven't tried using NFS in years, but I think that the
64 "soft" and "intr"remount options may well help you. The only way of
65 being sure would be totry it.See, for example,
66 http://howtos.linux.com/guides/nag2/x-087-2-nfs.mountd.shtml
67 - Add the following devices as working:
71 Overland LoaderXpress LTO (LVD)
72 Overland Neo2000 (LVD)
75 - Complete Win32 installer
76 - Finish work on Gnome restore GUI.
77 - On unknown client in restore "client=xxx"
78 Could not find Client "Matou": ERR=Query failed: DROP TABLE temp1:
79 ERR=no such table: temp1
80 - Fix "llist jobid=xx" where no fileset or client exists.
81 - Build console in client-only build.
82 - Test Qmsg() code to be used in bnet.c to prevent recursion. Queue the
83 message. If dequeueing toss the messages. Lock while dequeuing so that
84 it cannot be called recursively and set dequeuing flag.
85 - Test work on conio.c -- particularly linking.
86 - Phil says that Windows file sizes mismatch in Verify when they should,
87 and that either the file size or the catalog size was zero.
88 - Check time/dates printed during restore when using Win32 API.
92 it seems to be complaining about 12:00pm which should be a valid 12
93 hour time. I changed the time to 11:59am and everything works fine.
94 Also 12:00am works fine. 0:00pm also works (which I don't think
95 should). None of the values 12:00pm - 12:59pm work for that matter.
96 - Add level to estimate command.
97 - Fix option 2 of restore -- list where file is backed up -- require Client,
98 then list last 20 backups.
99 - Add all pools in Dir conf to DB also update them to catch changed
100 LabelFormats and such.
101 - Update volumes FromPool=xxx does all volumes.
102 - Once a job is canceled, make sure the status reflects that even if the
103 job is "stuck" in the run queue.
107 - Remove paths (and files that reference them) that have no trailing slash
108 in dbcheck -- or add a trailing slash.
109 - Remove Filenames (and files that reference them) that have a trailing
110 slash in dbcheck -- or remove the trailing slash.
111 - Remove orphaned paths/filenames by copying them to a new table with a
112 reference count, then mark all referenced files/paths and remove unreferenced
114 - Add a .list all files in the restore tree (probably also a list all files)
115 Do both a long and short form.
116 - Allow browsing the catalog to see all versions of a file (with
117 stat data on each file).
118 - Restore attributes of directory if replace=never set but directory
120 - Allow "delete job jobid=xxx,yyy,aaa-bbb" i.e. list + ranges.
121 - Use SHA1 on authentication if possible.
122 - See comtest-xxx.zip for Windows code to talk to USB.
123 - Make btape accept Device Names in addition to Archive names.
124 - Add Events and Perl scripting.
125 - Add John's appended files:
126 Appended = { /files/server/logs/http/*log }
127 and such files would be treated as follows.On a FULL backup, they would
128 be backed up like any other file.On an INCREMENTAL backup, where a
129 previous INCREMENTAL or FULL was already in thecatalogue and the length
130 of the file wasgreater than the length of the last backup, only thedata
131 added since the last backup will be dumped.On an INCREMENTAL backup, if
132 the length of the file is less than thelength of the file with the same
133 name last backed up, the completefile is dumped.On Windows systems, with
134 creation date of files, we can be evensmarter about this and not count
135 entirely upon the length.On a restore, the full and all incrementals
136 since it will beapplied in sequence to restore the file. - Add a
137 regression test for dbcheck. - Add disk seeking on restore. - Allow
138 for optional cancelling of SD and FD in case DIR
139 gets a fatal error. Requested by Jesse Guardiani <jesse@wingnet.net>
140 - Bizarre message: Error: Could not open WriteBootstrap file:
141 - Build console in client only build.
142 - Add "limit=n" for "list jobs"
143 - Check new HAVE_WIN32 open bits.
144 - Check if the tape has moved before writing.
145 - Handling removable disks -- see below:
146 - Multiple drive autochanger support -- see below.
147 - Keep track of tape use time, and report when cleaning is necessary.
148 - Fix FreeBSD mt_count problem.
149 - Add FromClient and ToClient keywords on restore command (or
150 BackupClient RestoreClient).
151 - Automatic "update slots" on user configuration directive when a
153 - Implement a JobSet, which groups any number of jobs. If the
154 JobSet is started, all the jobs are started together.
155 Allow Pool, Level, and Schedule overrides.
156 - Enhance cancel to timeout BSOCK packets after a specific delay.
157 - When I restore to Windows the Created, Accessed and Modifiedtimes are
158 those of the time of the restore, not those of the originalfile.
159 The dates you will find in your restore log seem to be the original
161 - Volume "add"ed to Pool gets recycled in first use. VolBytes=0
162 - Get rid of 0 dates in LastWritten, ...
163 - If a tape is recycled while it is mounted, Stanislav Tvrudy must do an
164 additional mount to deblock the job.
166 bscan: sql_update.c:65 UPDATE File SET MD5='Ij+5kwN6TFIxK+8l8+/I+A' WHERE FileId=0
167 bscan: bscan.c:1074 Could not add MD5/SHA1 to File record. ERR=sql_update.c:65 Update problem: affected_rows=0
168 - Do scheduling by UTC using gmtime_r() in run_conf, scheduler, and
169 ua_status.!!! Thanks to Alan Brown for this tip.
170 - Look at updating Volume Jobs so that Max Volume Jobs = 1 will work
171 correctly for multiple simultaneous jobs.
172 - Correct code so that FileSet MD5 is calculated for < and | filename
174 - Mark Volume in error on error from WEOF.
175 - Implement the Media record flag that indicates that the Volume does disk
177 - Implement VolAddr, which is used when Volume is addressed like a disk,
178 and form it from VolFile and VolBlock.
179 - Make multiple restore jobs for multiple media types specifying
180 the proper storage type.
181 - Implement MediaType keyword in bsr?
182 - Fix fast block rejection (stored/read_record.c:118). It passes a null
183 pointer (rec) to try_repositioning().
184 - Look at extracting Win data from BackupRead.
185 - Having dashes in filenames apparently creates problems for restore
186 by filename??? hard to believe.
187 - Implement RestoreJobRetention? Maybe better "JobRetention" in a Job,
188 which would take precidence over the Catalog "JobRetention".
189 - Implement Label Format in Add and Label console commands.
190 - Possibly up network buffers to 65K. Put on variable.
191 - Put email tape request delays on one or more variables. User wants
192 to cancel the job after a certain time interval. Maximum Mount Wait?
193 - Job, Client, Device, Pool, or Volume?
194 Is it possible to make this a directive which is *optional* in multiple
195 resources, like Level? If so, I think I'd make it an optional directive
196 in Job, Client, and Pool, with precedence such that Job overrides Client
197 which in turn overrides Pool.
198 - Print a message when a job starts if the conf file is not current.
199 - To pass Include 1 or two letter commands
200 I Name Include name - first record
201 B Name Base name - repeat
204 E zzz Exclude expression (wild card)
206 D "reader" Reader program
207 T "writer" Writer program
208 O Options In current commpressed format
209 (compression, signature, onefs, recurse, sparse,
210 replace, verify options, ...)
212 B BaseName Start second option set any letter
215 F Number Number of filenames to follow
220 F Number Number of filenames to follow
224 - Spooling ideas taken from Volker Sauer's and other's emails:
225 > IMHO job spooling should be turned on
231 > where and 2) overrides 1) and 3) is independent.
233 Yes, this is the minimum that I think is necessary.
237 > It should be switched by job, because the job that backs up the machine
238 > with the bacula-sd on doesn't need spooling.
239 > It should be switched by schedule, because for full-backups I don't need
240 > spooling, so I can switch it off (because the network faster then the
243 True, with the exception that if you have enough disk spool space,
244 and you want to run concurrent jobs, spooling can eliminate the block
245 interleaving restore inefficiencies.
247 > And you should be able to turn it of by sd for sd-machines with low disk
248 > capacity or if you just don't need or want this feature.
251 > - definitly the possibility for multipe spool direcories
253 Having multiple directories is no problem -- having different maximum
254 sizes creates specification problems. At some point, I will probably
255 have a common SD pool of spool directories as well as a set of
256 private spool directories for each device. The first implementation
257 will be a set of private spool directories for each device since
258 managing a global pool with a bunch of threads writing into the same
259 directory is *much* more complicated and prone to error.
261 > - the ability to spool parts of a backup (not the whole client)
263 This may change in the future, but for the moment, it will spool
264 either to a job high water mark, or until the directory is full
265 (reaches max spool size or I/O error). It will then write to tape,
266 truncate the spool file, and begin spooling again.
268 > - spooling while writing to tape
270 Not within a job, but yes, if you run concurrent jobs -- each is a
271 different thread. Within a job could be a feature, but *much* later.
273 > - parallel spooling (like parallel jobs/ concurrent jobs) of clients
275 Yes, this is one of my main motivations for doing it (aside from
276 eliminating tape "shoe shine" during incremental backups.
278 > - flushing a backup that only went to disk (like amflush in amanda)
280 This will be a future feature, since spooling is different from backing
281 up to disk. The future feature will be "migration" which will move a job
282 from one backup Volume to another.
284 - New Storage specifications:
285 Passed to SD as a sort of BSR record called Storage Specification
289 MediaType -> Next MediaType
291 Device -> Next Device
292 Write Copy Resource that makes a copy of a resource.
294 Allow multiple Storage specifications
302 Allow Multiple Pool specifications (note, Pool currently
304 Allow Multiple MediaType specifications
305 Allow Multiple Device specifications
306 Perhaps keep this in a single SSR
307 Tie a Volume to a specific device by using a MediaType that
308 is contained in only one device.
309 In SD allow Device to have Multiple MediaTypes
312 - Look at www.nu2.nu/pebuilder as a helper for full windows
314 Ideas from Jerry Scharf:
315 First let's point out some big pluses that bacula has for this
317 more importantly it's active. Thank you so much for that
318 even more important, it's not flaky
319 it has an open access catalog, opening many possibilities
320 it's pushing toward heterogeneous systems capability
322 I don't remember an include file directive for config files
323 (not filesets, actual config directives)
324 can you check the configs without starting the daemon?
325 some warnings about possible common mistakes
327 doing the testing and blessing of concurrent backup writes
328 this is absolutely necessary in the enterprise
329 easy user recovery GUI with full access checking
330 Macintosh file client
331 macs are an interesting niche, but I fear a server is a rathole
332 working bare iron recovery for windows
333 much better handling on running config changes
334 thinking through the logic of what happens to jobs in progress
335 the option for inc/diff backups not reset on fileset revision
336 a) use both change and inode update time against base time
337 b) do the full catalog check (expensive but accurate)
338 sizing guide (how much system is needed to back up N systems/files)
339 consultants on using bacula in building a disaster recovery system
341 or how to get at fancy things that one could do with bacula
342 logwatch code for bacula logs (or similar)
343 linux distro inclusion of bacula (brings good and bad, but necessary)
344 win2k/XP server capability (icky but you asked)
345 support for Oracle database ??
347 - Look at adding SQL server and Exchange support for Windows.
348 - Restore: Enter Filename: 'C:/Documents and Settings/Comercial/My
349 Documents/MOP/formulário de registro BELAS ARTES.doc' causes Bacula to
351 - Each DVD-RAM disk would be a volume, just like each tape is
352 a volume. It's a 4.7GB media with random access, but there's nothing about
353 it that I can see that makes it so different than a tape from bacula's
354 perspective. Why couldn't I back up to a bare floppy as a volume (ignoring
356 - Make dev->file and dev->block_num signed integers so that -1 can
357 be an invalid value which happens with BSR.
358 - Create VolAddr for disk files in place of VolFile and VolBlock. This
359 is needed to properly specify ranges.
360 - Print bsmtp output to job report so that problems will be seen.
361 - Pass the number of files to be restored to the FD for reporting
362 - Add progress of files/bytes to SD and FD.
363 - Don't continue Restore if no files selected.
364 - Print warning message if FileId > 4 billion
365 - do a "messages" before the first prompt in Console
366 - Add a date and time stamp at the beginning of every line in the
367 Job report (Volker Sauer).
368 - Client does not show busy during Estimate command.
369 - Implement Console mtx commands.
370 - Add a default DB password to MySQL.
371 GRANT all privileges ON bacula.* TO bacula@localhost IDENTIFIED BY
374 - Implement a Mount Command and an Unmount Command where
375 the users could specify a system command to be performed
376 to do the mount, after which Bacula could attempt to
377 read the device. This is for Removeable media such as a CDROM.
378 - Most likely, this mount command would be invoked explicitly
379 by the user using the current Console "mount" and "unmount"
380 commands -- the Storage Daemon would do the right thing
381 depending on the exact nature of the device.
382 - As with tape drives, when Bacula wanted a new removable
383 disk mounted, it would unmount the old one, and send a message
384 to the user, who would then use "mount" as described above
385 once he had actually inserted the disk.
386 - Implement dump/print label to UA
387 - Implement disk spooling. Two parts: 1. Spool to disk then
388 immediately to tape to speed up tape operations. 2. Spool to
389 disk only when the tape is full, then when a tape is hung move
391 - Scratch Pool where the volumes can be re-assigned to any Pool.
392 - bextract is sending everything to the log file ****FIXME****
393 - Add Progress command that periodically reports the progress of
395 - Restrict characters permitted in a Resource name, and don't permit
397 - Allow multiple Storage specifications (or multiple names on
398 a single Storage specification) in the Job record. Thus a job
399 can be backed up to a number of storage devices.
400 - Implement some way for the File daemon to contact the Director
401 to start a job or pass its DHCP obtained IP number.
402 - Implement multiple Consoles.
403 - Implement a query tape prompt/replace feature for a console
405 Two jobs ready to go, first one blocked waiting for media
406 Cancel 2nd job ("waiting execution" one)
409 - Copy console @ code to gnome2-console
410 - Make AES the only encryption algorithm see
411 http://csrc.nist.gov/CryptoToolkit/aes/). It's
412 an officially adopted standard, has survived peer
413 review, and provides keys up to 256 bits.
414 - Think about how space could be freed up on a tape -- perhaps this
415 is a Merge or Compact feature that is needed.
416 - Modify FileSet, did not upgrade the current Increment job, but
417 waited for the next job to be upgraded.
418 - Take a careful look at SetACL http://setacl.sourceforge.net
419 - Implement a where command for the tree telling where a file
421 - Take a careful look at Level for the estimate command, maybe make
422 it a command line option.
423 - Add Volume name to "I cannot write on this volume because"
424 - Make tree walk routines like cd, ls, ... more user friendly
425 by handling spaces better.
426 - Write your PID file and chown root:wheel before drop.
427 - Make sure there is no symlink in a file before creating a
429 - Look at mktemp or mkstemp(3).
430 mktemp and mkstemp create files with predictable names too. That's
431 not the vulnerability. The vulnerability is in creating files without
432 using the O_EXCL flag, which means "only create this file if it doesn't
433 exist, including if the file is a dangling symlink."
435 It is *NOT* enough to do the equivalent of
437 if doesn't exist $filename
438 then create $filename
440 because between the test and the create another process could have
441 gotten the CPU and created the file. You must use atomic functions
442 (those that don't get interrupted by other processes) and O_EXCL is
443 the only way for this particular example.
444 - Automatically create pools, but instead of looking for what
445 in in Job records, walk through the pool resources.
446 - Check and double check tree code, why does it take so long?
447 - Add device name to "Current Volume not acceptable because ..."
448 - Make sure that Bacula rechecks the tape after the 20 min wait.
449 - Set IO_NOWAIT on Bacula TCP/IP packets.
450 - Try doing a raw partition backup and restore by mounting a
453 Yes, it would allow to highly automatic the request for new tapes. If a
454 tape is empty, bacula reads the barcodes (native or simulated), and if
455 an unused tape is found, it runs the label command with all the
456 necessary parameters.
458 By the way can bacula automatically "move" an empty/purged volume say
459 in the "short" pool to the "long" pool if this pool runs out of volume
461 - Eliminate orphaned jobs: dbcheck, normal pruning, delete job command.
462 Hm. Well, there are the remaining orphaned job records:
464 | 105 | Llioness Save | 0000-00-00 00:00:00 | B | D | 0 | 0 | f |
465 | 110 | Llioness Save | 0000-00-00 00:00:00 | B | I | 0 | 0 | f |
466 | 115 | Llioness Save | 2003-09-10 02:22:03 | B | I | 0 | 0 | A |
467 | 128 | Catalog Save | 2003-09-11 03:53:32 | B | I | 0 | 0 | C |
468 | 131 | Catalog Save | 0000-00-00 00:00:00 | B | I | 0 | 0 | f |
470 As you can see, three of the five are failures. I already deleted the
471 one restore and one other failure using the by-client option. Deciding
472 what is an orphaned job is a tricky problem though, I agree. All these
473 records have or had 0 files/ 0 bytes, except for the restore. With no
474 files, of course, I don't know of the job ever actually becomes
475 associated with a Volume.
477 (I'm not sure if this is documented anywhere -- what are the meanings of
478 all the possible JobStatus codes?)
480 Looking at my database, it appears to me as though all the "orphaned"
481 jobs fit into one of two categories:
483 1) The Job record has a StartTime but no EndTime, and the job is not
486 2) The Job record has an EndTime, indicating that it completed, but
487 it has no associated JobMedia record.
490 This does suggest an approach. If failed jobs (or jobs that, for some
491 other reason, write no files) are associated with a volume via a
492 JobMedia record, then they should be purged when the associated volume
493 is purged. I see two ways to handle jobs that are NOT associated with a
496 1) purge them automatically whenever any volume is manually purged;
498 2) add an option to the purge command to manually purge all jobs with
499 no associated volume.
501 I think Restore jobs also fall into category 2 above .... so one might
502 want to make that "The Job record has an EndTime,, but no associated
503 JobMedia record, and is not a Restore job."
504 - make "btape /tmp" work.
505 - Make sure a rescheduled job is properly reported by status.
506 - Walk through the Pool records rather than the Job records
507 in dird.c to create/update pools.
508 - What to do about "list files job=xxx".
509 - When job rescheduled, status gives is waiting for Client Rufus
510 to connect to Storage File. Dir needs to inform SD that job
512 - Make Dmsg look at global before calling subroutine.
513 - Enable trace output at runtime for Win32
514 - Available volumes for autochangers (see patrick@baanboard.com 3 Sep 03
515 and 4 Sep) scan slots.
516 - Get and test MySQL 4.0
517 - Do a complete audit of all pthreads_mutex, cond, ... to ensure that
518 any that are dynamically initialized are destroyed when no longer used.
519 - Look at how fuser works and /proc/PID/fd that is how Nic found the
520 file descriptor leak in Bacula.
521 - Implement WrapCounters in Counters.
522 - Turn on SIGHUP in dird.c and test.
523 - Use system dependent calls to get more precise info on tape errors.
524 - Add heartbeat from FD to SD if hb interval expires.
525 - Suppress read error on blank tape when doing a label.
526 - Can we dynamically change FileSets?
527 - If pool specified to label command and Label Format is specified,
528 automatically generate the Volume name.
529 - Why can't SQL do the filename sort for restore?
530 - Look at libkse (man kse) for FreeBSD threading.
531 - Look into Microsoft Volume Shadowcopy Service VSS for backing
532 up system state components (Active Directory, System Volume, ...)
533 - Add ExhautiveRestoreSearch
534 - Look at the possibility of loading only the necessary
535 data into the restore tree (i.e. do it one directory at a
536 time as the user walks through the tree).
537 - Possibly use the hash code if the user selects all for a restore command.
538 - Orphaned Dir buffer at parse_conf.c:373 => store_dir
539 - Fix "restore all" to bypass building the tree.
540 - Prohibit backing up archive device (findlib/find_one.c:128)
541 - Implement Release Device in the Job resource to unmount a drive.
542 - Implement Acquire Device in the Job resource to mount a drive,
543 be sure this works with admin jobs so that the user can get
544 prompted to insert the correct tape. Possibly some way to say to
545 run the job but don't save the files.
546 - Implement FileOptions (see end of this document)
547 - Make things like list where a file is saved case independent for
550 - Bacula needs to propagate SD errors.
551 > > cluster-dir: Start Backup JobId 252, Job=REUTERS.2003-08-11_15.04.12
552 > > prod4-sd: REUTERS.2003-08-11_15.04.12 Error: Write error on device
553 /dev/nst0. ERR=Input/output error.
554 > > prod4-sd: REUTERS.2003-08-11_15.04.12 Error: Re-read of last block failed.
555 Last block=5162 Current block=5164.
556 > > prod4-sd: End of medium on Volume "REU007" Bytes=16,303,521,933
558 - Use autochanger to handle multiple devices.
559 - Add SuSE install doc to list.
560 - Check and rechedk "Invalid block number"
561 - Make bextract release the drive properly between tapes
562 so that an autochanger can be made to work.
563 - User wants to NOT backup up certain big files (email files).
564 - Maybe remove multiple simultaneous devices code in SD.
565 - On Windows with very long path names, it may be impossible to create
566 a file (and thus restore it) because the total length is too long.
567 We must cd into the directory then create the file without the
569 - lstat() is not going to work on Win32 for testing date.
570 - Implement a Recycle command
571 - Add client name to cram-md5 challenge so Director can immediately
572 verify if it is the correct client.
573 - Add JobLevel in FD status (but make sure it is defined).
574 - Audit all UA commands to ensure that we always prompt where possible.
575 - Check Jmsg in bnet, may not work, must dup bsock.
576 - Suppress Job Name in Jmsg for console
577 - Create Pools that are referenced in a Run statement at startup if possible.
578 - Use runbeforejob to unload, then reload a volume previously used,
579 then the next job run gets an error reading the drive.
580 - Make bootstrap filename unique.
581 - Test a second language e.g. french.
582 - Start working on Base jobs.
583 - Make "make binary-release" work from any directory.
584 - Implement UnsavedFiles DB record.
585 - Implement argc/argv for daemon command line scanning using table driven
587 - Implement table driven single argc/argv scanner to pickup all arguments.
588 Much like xxx_conf.c scan table.
589 keyword, handler(store_routine), store_address, code, flags, default.
590 - From Phil Stracchino:
591 It would probably be a per-client option, and would be called
592 something like, say, "Automatically purge obsoleted jobs". What it
593 would do is, when you successfully complete a Differential backup of a
594 client, it would automatically purge all Incremental backups for that
595 client that are rendered redundant by that Differential. Likewise,
596 when a Full backup on a client completed, it would automatically purge
597 all Differential and Incremental jobs obsoleted by that Full backup.
598 This would let people minimize the number of tapes they're keeping on
599 hand without having to master the art of retention times.
600 - Implement a M_SECURITY message class.
601 - When doing a Backup send all attributes back to the Director, who
602 would then figure out what files have been deleted.
603 - Currently in mount.c:236 the SD simply creates a Volume. It should have
604 explicit permission to do so. It should also mark the tape in error
605 if there is an error.
606 - Make sure all restore counters are working correctly in the FD.
607 - SD Bytes Read is wrong.
608 - Look at ALL higher level routines that call block.c to be sure
609 they don't expect something in errmsg.
610 - Investigate doing RAW backup of Win32 partition.
611 - Add thread specific data to hold the jcr -- send error messages from
612 low level routines by accessing it and using Jmsg().
613 - Cancel waiting for Client connect in SD if FD goes away.
615 - Examine Bare Metal restore problem (a FD crash exists somewhere ...).
616 - Implement timeout in response() when it should come quickly.
617 - Implement console @echo command.
618 - Implement a Slot priority (loaded/not loaded).
619 - Implement "vacation" Incremental only saves.
620 - Implement single pane restore (much like the Gftp panes).
621 - Implement Automatic Mount even in operator wait.
622 - Implement create "FileSet"?
623 - Fix watchdog pthread crash on Win32 (this is pthread_kill() Cygwin bug)
624 - Implement "scratch pool" where tapes are defined and can be
625 taken by any pool that needs them.
626 - Implement restore "current system", but take all files without
627 doing selection tree -- so that jobs without File records can
629 - Add prefixlinks to where or not where absolute links to FD.
630 - Issue message to mount a new tape before the rewind.
631 - Simplified client job initiation for portables.
632 - If SD cannot open a drive, make it periodically retry.
633 - Add more of the config info to the tape label.
635 - If tape is marked read-only, then try opening it read-only rather than
636 failing, and remember that it cannot be written.
637 - Refine SD waiting output:
638 Device is being positioned
639 > Device is being positioned for append
640 > Device is being positioned to file x
642 - Figure out some way to estimate output size and to avoid splitting
643 a backup across two Volumes -- this could be useful for writing CDROMs
644 where you really prefer not to have it split -- not serious.
645 - Have SD compute MD5 or SHA1 and compare to what FD computes.
646 - Make VolumeToCatalog calculate an MD5 or SHA1 from the
647 actual data on the Volume and compare it.
648 - Implement Bacula plugins -- design API
649 - Make bcopy read through bad tape records.
650 - Program files (i.e. execute a program to read/write files).
651 Pass read date of last backup, size of file last time.
652 - Add Signature type to File DB record.
653 - CD into subdirectory when open()ing files for backup to
654 speed up things. Test with testfind().
655 - Priority job to go to top of list.
656 - Why are save/restore of device different sizes (sparse?) Yup! Fix it.
657 - Implement some way for the Console to dynamically create a job.
658 - Restore to a particular time -- e.g. before date, after date.
659 - Solaris -I on tar for include list
660 - Need a verbose mode in restore, perhaps to bsr.
661 - bscan without -v is too quiet -- perhaps show jobs.
662 - Add code to reject whole blocks if not wanted on restore.
663 - Check if we can increase Bacula FD priorty in Win2000
664 - Make sure the MaxVolFiles is fully implemented in SD
665 - Check if both CatalogFiles and UseCatalog are set to SD.
666 - Figure out how to do a bare metal Windows restore
667 - Possibly add email to Watchdog if drive is unmounted too
668 long and a job is waiting on the drive.
669 - Restore program that errs in SD due to no tape, reports
670 OK incorrectly in output.
671 - After unmount, if restore job started, ask to mount.
672 - Convert all %x substitution variables, which are hard to remember
673 and read to %(variable-name). Idea from TMDA.
674 - Remove NextId for SQLite. Optimize.
675 - Move all SQL statements into a single location.
676 - Add UA rc and history files.
677 - put termcap (used by console) in ./configure and
678 allow -with-termcap-dir.
679 - Fix Autoprune for Volumes to respect need for full save.
680 - Fix Win32 config file definition name on /install
681 - Compare tape to Client files (attributes, or attributes and data)
682 - Make all database Ids 64 bit.
683 - Write an applet for Linux.
684 - Allow console commands to detach or run in background.
685 - Fix status delay on storage daemon during rewind.
686 - Add SD message variables to control operator wait time
687 - Maximum Operator Wait
688 - Minimum Message Interval
689 - Maximum Message Interval
690 - Send Operator message when cannot read tape label.
691 - Verify level=Volume (scan only), level=Data (compare of data to file).
692 Verify level=Catalog, level=InitCatalog
694 - Add keyword search to show command in Console.
695 - Events : tape has more than xxx bytes.
696 - Complete code in Bacula Resources -- this will permit
697 reading a new config file at any time.
698 - Handle ctl-c in Console
699 - Implement script driven addition of File daemon to config files.
700 - Think about how to make Bacula work better with File (non-tape) archives.
701 - Write Unix emulator for Windows.
702 - Put memory utilization in Status output of each daemon
703 if full status requested or if some level of debug on.
704 - Make database type selectable by .conf files i.e. at runtime
705 - Set flag for uname -a. Add to Volume label.
706 - Implement throttled work queue.
707 - Restore files modified after date
708 - SET LD_RUN_PATH=$HOME/mysql/lib/mysql
709 - Implement Restore FileSet=
710 - Create a protocol.h and protocol.c where all protocol messages
712 - Remove duplicate fields from jcr (e.g. jcr.level and jcr.jr.Level, ...).
713 - Timout a job or terminate if link goes down, or reopen link and query.
714 - Concept of precious tapes (cannot be reused).
715 - Make bcopy copy with a single tape drive.
716 - Permit changing ownership during restore.
719 > My suggestion: Add a feature on the systray menu-icon menu to request
720 > an immediate backup now. This would be useful for laptop users who may
721 > not be on the network when the regular scheduled backup is run.
723 > My wife's suggestion: Add a setting to the win32 client to allow it to
724 > shut down the machine after backup is complete (after, of course,
725 > displaying a "System will shut down in one minute, click here to cancel"
726 > warning dialog). This would be useful for sites that want user
727 > woorkstations to be shut down overnight to save power.
730 - From Terry Manderson <terry@apnic.net>
731 jobdefs { # new structure
732 name = "monthlyUnixBoxen"
744 jobdefs = "genericUnixSet"
750 jobdefs = "genericUnixSet"
756 jobdefs = "UnixDevelBoxSet"
761 - Autolabel should be specified by DIR instead of SD.
764 - AutoScan (check checksum of tape)
765 - Format command = "format /dev/nst0"
769 - Seek resolution (usually corresponds to buffer size)
770 - EODErrorCode=ENOSPC or code
771 - Partial Read error code
772 - Partial write error code
773 - Nonformatted read error
774 - Nonformatted write error
775 - WriteProtected error
779 - IgnoreCloseErrors=yes
789 - FD sends unsaved file list to Director at end of job (see
791 - File daemon should build list of files skipped, and then
792 at end of save retry and report any errors.
793 - Write a Storage daemon that uses pipes and
794 standard Unix programs to write to the tape.
796 - Need something that monitors the JCR queue and
797 times out jobs by asking the deamons where they are.
798 - Enhance Jmsg code to permit buffering and saving to disk.
799 - device driver = "xxxx" for drives.
801 - Ensure that /dev/null works
802 - Need report class for messages. Perhaps
803 report resource where report=group of messages
804 - enhance scan_attrib and rename scan_jobtype, and
805 fill in code for "since" option
806 - Director needs a time after which the report status is sent
807 anyway -- or better yet, a retry time for the job.
808 - Don't reschedule a job if previous incarnation is still running.
809 - Some way to automatically backup everything is needed????
810 - Need a structure for pending actions:
812 - termination status (part of buffered msgs?)
814 Read, Write, Clean, Delete
815 - Login to Bacula; Bacula users with different permissions:
816 owner, group, user, quotas
817 - Store info on each file system type (probably in the job header on tape.
818 This could be the output of df; or perhaps some sort of /etc/mtab record.
821 - Design at hierarchial storage for Bacula. Migration and Clone.
822 - Implement FSM (File System Modules).
823 - Audit M_ error codes to ensure they are correct and consistent.
824 - Add variable break characters to lex analyzer.
825 Either a bit mask or a string of chars so that
826 the caller can change the break characters.
827 - Make a single T_BREAK to replace T_COMMA, etc.
828 - Ensure that File daemon and Storage daemon can
829 continue a save if the Director goes down (this
830 is NOT currently the case). Must detect socket error,
831 buffer messages for later.
832 - Enhance time/duration input to allow multiple qualifiers e.g. 3d2h
833 - Add ability to backup to two Storage devices (two SD sessions) at
834 the same time -- e.g. onsite, offsite.
835 - Add the ability to consolidate old backup sets (basically do a restore
836 to tape and appropriately update the catalog). Compress Volume sets.
837 Might need to spool via file is only one drive is available.
838 - Compress or consolidate Volumes of old possibly deleted files. Perhaps
839 someway to do so with every volume that has less than x% valid
843 Migration: Move a backup from one Volume to another
844 Clone: Copy a backup -- two Volumes
846 Bacula Migration is based on Jobs (apparently Networker is file by file).
848 Migration triggered by:
852 Highwater mark (keep total size)
857 ======================================================
859 It is somewhat like a Full save becomes an incremental since
860 the Base job (or jobs) plus other non-base files.
862 - A Base backup is same as Full backup, just different type.
863 - New BaseFiles table that contains:
865 BaseJobId - Base JobId referenced for this FileId (needed ???)
866 JobId - JobId currently running
867 FileId - File not backed up, exists in Base Job
868 FileIndex - FileIndex from Base Job.
869 i.e. for each base file that exists but is not saved because
870 it has not changed, the File daemon sends the JobId, BaseId,
871 FileId, FileIndex back to the Director who creates the DB entry.
872 - To initiate a Base save, the Director sends the FD
873 the FileId, and full filename for each file in the Base.
874 - When the FD finds a Base file, he requests the Director to
875 send him the full File entry (stat packet plus MD5), or
876 conversely, the FD sends it to the Director and the Director
877 says yes or no. This can be quite rapid if the FileId is kept
878 by the FD for each Base Filename.
879 - It is probably better to have the comparison done by the FD
880 despite the fact that the File entry must be sent across the
882 - An alternative would be to send the FD the whole File entry
883 from the start. The disadvantage is that it requires a lot of
884 space. The advantage is that it requires less communications
886 - The Job record must be updated to indicate that one or more
888 - At end of Job, FD returns:
889 1. Count of base files/bytes not written to tape (i.e. matches)
890 2. Count of base file that were saved i.e. had changed.
891 - No tape record would be written for a Base file that matches, in the
892 same way that no tape record is written for Incremental jobs where
893 the file is not saved because it is unchanged.
894 - On a restore, all the Base file records must explicitly be
895 found from the BaseFile tape. I.e. for each Full save that is marked
896 to have one or more Base Jobs, search the BaseFile for all occurrences
898 - An optimization might be to make the BaseFile have:
904 This would avoid the need to explicitly fetch each File record for
905 the Base job. The Base Job record will be fetched to get the
906 VolSessionId and VolSessionTime.
907 =========================================================
911 =============================================================
913 Request For Comments For File Backup Options
916 Subject: File Backup Options
919 A few days ago, a Bacula user who is backing up to file volumes and
920 using compression asked if it was possible to suppress compressing
921 all .gz files since it was a waste of CPU time. Although Bacula
922 currently permits using different options (compression, ...) on
923 a directory by directory basis, it cannot do it on a file by
924 file basis, which is clearly what was desired.
926 Proposed Implementation:
927 To solve this problem, I propose the following:
929 - Add a new Director resource type called Options.
931 - The Options resource will have records for all
932 options that can currently be specified on the Include record
933 (in a FileSet). Examples below.
935 - The Options resource will permit an exclude option as well
936 as a number of additional options.
938 - The heart of the Options resource is the ability to
939 supply any number of Match records which specify POSIX
940 regular expressions. These Match regular expressions are
941 applied to the fully qualified filename (path and all). If
942 one matches, then the Options will be used.
944 - When an Match specification matches an included file, the
945 options specified in the Options resource will override
946 the default options specified on the Include record.
948 - Include records will be modified to permit referencing one or
949 more Options resources. The Options will be used
950 in the order listed on the Include record and the first
951 one that matches will be applied.
953 - Options (or specifications) currently supplied on the Include
954 record will be deprecated (i.e. removed in a later version a
955 year or so from now).
957 - The Exclude record will be deprecated as the same functionality
958 can be obtained by using an Exclude = yes in the Options.
961 The following records can appear in the Options resource. An
962 asterisk preceding the name indicates a feature not currently
965 - Regexp "xxx" - Match regular expression
966 - Wild "xxx" - Do a wild card match
969 - Compression= (GZIP, ...)
970 - Signature= (MD5, SHA1, ...)
972 - OneFs= (yes/no) - remain on one filesystem
973 - Recurse= (yes/no) - recurse into subdirectories
974 - Sparse= (yes/no) - do sparse file backup
975 - *Exclude= (yes/no) - exclude file from being saved
976 - *Reader= (filename) - external read (backup) program
977 - *Plugin= (filename) - read/write plugin module
979 - Include= (yes/no) - Include the file matched no additional
980 patterns are applied.
983 - verify= (ipnougsamc5) - verify options
986 - replace= (always/ifnewer/ifolder/never) - replace options currently
988 - *Writer= (filename) - external write (restore) program
992 Currently options specifying compression, MD5 signatures, recursion,
993 ... of a FileSet are supplied on the Include record. These will now
994 all be collected into a Options resource, which will be
995 specified in the Include in place of the options. Multiple Options
996 may be specified. Since the Options may contain regular expressions
997 that are applied to the full filename, this will give the ability
998 to specify backup options on a file by file basis to whatever level
1007 Include = compression=GZIP signature=MD5 {
1019 Wild = /*.?*/ # matches all files.
1024 That's a lot more to do the same thing, but it gives the ability to
1025 apply options on a file by file basis. For example, suppose you
1026 want to compress all files but not any file with extensions .gz or .Z.
1027 In that case, you will need to group two sets of options using
1028 the Options resource. Files may be anywhere except in an
1029 option set??? All OptionSets apply to all files in the order
1030 the OptionSets were specified. To have files included with
1031 different option sets without using wild-cards, use two or more
1032 Includes -- each one is handled in turn using only the files and
1033 optionsets specified in the include.
1040 # Note multiple Matches are ORed
1041 Wild = "*.gz" # matches .gz files
1042 Wild = "*.Z" # matches .Z files
1047 Wild = "*.?*" # matches all files
1053 Now, since the no Compression option is specified in the
1054 first group of Options, *.gz or *.Z file will have an MD5 signature computed,
1055 but will not be compressed. For all other files, the *.gz *.Z will not
1056 match, so the second group of options will be used which will include GZIP
1060 - Is it necessary to provide some means of ANDing regular expressions
1061 and negation? (not currently planned)
1063 e.g. Wild = /*.gz/ && !/big.gz/
1065 - I see that Networker has a "null" module which, if specified, does not
1066 backup the file, but does make an record of the file in the catalog
1067 so that the catalog will reflect an exact picture of the filesystem.
1068 The result is that the file can be "seen" when "browsing" the save
1069 sets, but it cannot be restored.
1071 Is this really useful? Should it be implemented in Bacula?
1074 After implementing the above, the user will be able to specify
1075 on a file by file basis (using regular expressions) what options are
1076 applied for the backup.
1079 =============================================
1081 ==========================================================
1083 For each Incremental job that is run, there may be files that
1084 were found but not saved because they were locked (this applies
1085 only to Windows). Such a system could send back to the Director
1086 a list of Unsaved files.
1088 - New UnSavedFiles table that contains:
1092 - Then in the next Incremental job, the list of Unsaved Files will be
1093 feed to the FD, who will ensure that they are explicitly chosen even
1094 if standard date/time check would not have selected them.
1095 =============================================================
1098 Done: (see kernsdone for more)
1101 - John's Full save failed with 1.32c FD and 1.31 Dir no FD status,
1102 and no error message.
1103 - Add fd and st as Console keywords.
1104 - Recycling volume with a Slot requires an operator intervention:
1105 rufus-dir: Start Backup JobId 18, Job=kernsave.2003-11-01_21.23.52
1106 rufus-dir: Pruned 1 Job on Volume Vol01 from catalog.
1107 rufus-dir: There are no Jobs associated with Volume Vol01. Marking it purged.
1108 rufus-dir: Recycled volume "Vol01"
1109 rufus-sd: Please mount Volume "Vol01" on Storage Device "DDS-4" for Job kernsave.2003-11-01_21.23.52
1110 Use "mount" command to release Job.
1111 - Implement Dan's bacula script (email of 26 Oct).
1112 - Add JobName= to VerifyToCatalog so that all verifies can be done at the end.
1113 - Edit the Client/Storage name into authentication failure messages.
1114 - Fix packet too big problem. This is most likely a Windows TCP stack
1116 - Implement ClientRunBeforeJob and ClientRunAfterJob.
1117 - Implement forward spacing block/file: position_device(bsr) --
1118 just before read_block_from_device();
1121 Multiple drive autochanger data: see Alan Brown
1122 mtx -f xxx unloadStorage Element 1 is Already Full(drive 0 was empty)
1123 Unloading Data Transfer Element into Storage Element 1...source Element
1124 Address 480 is Empty
1126 (drive 0 was empty and so was slot 1)
1127 > mtx -f xxx load 15 0
1128 no response, just returns to the command prompt when complete.
1129 > mtx -f xxx status Storage Changer /dev/changer:2 Drives, 60 Slots ( 2 Import/Export )
1130 Data Transfer Element 0:Full (Storage Element 15 Loaded):VolumeTag = HX001
1131 Data Transfer Element 1:Empty
1132 Storage Element 1:Empty
1133 Storage Element 2:Full :VolumeTag=HX002
1134 Storage Element 3:Full :VolumeTag=HX003
1135 Storage Element 4:Full :VolumeTag=HX004
1136 Storage Element 5:Full :VolumeTag=HX005
1137 Storage Element 6:Full :VolumeTag=HX006
1138 Storage Element 7:Full :VolumeTag=HX007
1139 Storage Element 8:Full :VolumeTag=HX008
1140 Storage Element 9:Full :VolumeTag=HX009
1141 Storage Element 10:Full :VolumeTag=HX010
1142 Storage Element 11:Empty
1143 Storage Element 12:Empty
1144 Storage Element 13:Empty
1145 Storage Element 14:Empty
1146 Storage Element 15:Empty
1147 Storage Element 16:Empty....
1148 Storage Element 28:Empty
1149 Storage Element 29:Full :VolumeTag=CLNU01L1
1150 Storage Element 30:Empty....
1151 Storage Element 57:Empty
1152 Storage Element 58:Full :VolumeTag=NEX261L2
1153 Storage Element 59 IMPORT/EXPORT:Empty
1154 Storage Element 60 IMPORT/EXPORT:Empty
1156 Unloading Data Transfer Element into Storage Element 15...done
1158 (just to verify it remembers where it came from, however it can be
1159 overrriden with mtx unload {slotnumber} to go to any storage slot.)
1161 There needs to be a table of drive # to devices somewhere - If there are
1162 multiple changers or drives there may not be a 1:1 correspondance between
1163 changer drive number and system device name - and depending on the way the
1164 drives are hooked up to scsi busses, they may not be linearly numbered
1165 from an offset point either.something like
1167 Autochanger drives = 2
1168 Autochanger drive 0 = /dev/nst1
1169 Autochanger drive 1 = /dev/nst2
1170 IMHO, it would be _safest_ to use explicit mtx unload commands at all
1171 times, not just for multidrive changers. For a 1 drive changer, that's
1177 MTX's manpage (1.2.15):
1178 unload [<slotnum>] [ <drivenum> ]
1179 Unloads media from drive <drivenum> into slot
1180 <slotnum>. If <drivenum> is omitted, defaults to
1181 drive 0 (as do all commands). If <slotnum> is
1182 omitted, defaults to the slot that the drive was
1183 loaded from. Note that there's currently no way
1184 to say 'unload drive 1's media to the slot it
1185 came from', other than to explicitly use that
1186 slot number as the destination.AB
1192 undef# camcontrol devlist
1193 <WANGTEK 51000 SCSI M74H 12B3> at scbus0 target 2 lun 0 (pass0,sa0)
1194 <ARCHIVE 4586XX 28887-XXX 4BGD> at scbus0 target 4 lun 0 (pass1,sa1)
1195 <ARCHIVE 4586XX 28887-XXX 4BGD> at scbus0 target 4 lun 1 (pass2)
1197 tapeinfo -f /dev/sg0 with a bad tape in drive 1:
1198 [kern@rufus mtx-1.2.17kes]$ ./tapeinfo -f /dev/sg0
1199 Product Type: Tape Drive
1201 Product ID: 'C5713A '
1203 Attached Changer: No
1204 TapeAlert[3]: Hard Error: Uncorrectable read/write error.
1205 TapeAlert[20]: Clean Now: The tape drive neads cleaning NOW.
1212 Medium Type: Not Loaded
1215 DataCompEnabled: yes
1216 DataCompCapable: yes
1217 DataDeCompEnabled: yes
1224 Handling removable disks
1226 From: Karl Cunningham <karlc@keckec.com>
1228 My backups are only to hard disk these days, in removable bays. This is my
1229 idea of how a backup to hard disk would work more smoothly. Some of these
1230 things Bacula does already, but I mention them for completeness. If others
1231 have better ways to do this, I'd like to hear about it.
1233 1. Accommodate several disks, rotated similar to how tapes are. Identified
1234 by partition volume ID or perhaps by the name of a subdirectory.
1235 2. Abort & notify the admin if the wrong disk is in the bay.
1236 3. Write backups to different subdirectories for each machine to be backed
1238 4. Volumes (files) get created as needed in the proper subdirectory, one
1240 5. When a disk is recycled, remove or zero all old backup files. This is
1241 important as the disk being recycled may be close to full. This may be
1242 better done manually since the backup files for many machines may be
1243 scattered in many subdirectories.
1248 - Change console to bconsole.
1249 - Change smtp to bsmtp.
1250 - Fix time difference problem between Bacula and Client
1251 so that everything is in GMT.
1252 - Fix TimeZone problem!
1253 - Mount a tape that is not right for the job (wrong # files on tape)
1254 Bacula asks for another tape, fix problems with first tape and
1255 say "mount". All works OK, but status shows:
1256 Device /dev/nst0 open but no Bacula volume is mounted.
1257 Total Bytes=1,153,820,213 Blocks=17,888 Bytes/block=64,502
1258 Positioned at File=9 Block=3,951
1259 Full Backup job Rufus.2003-10-26_16.45.31 using Volume "DLT-24Oct03" on device /dev/nst0
1260 Files=21,003 Bytes=253,954,408 Bytes/sec=2,919,016
1261 FDReadSeqNo=192,134 in_msg=129830 out_msg=5 fd=7
1262 - Upgrade to cygwin 1.5
1263 - Optimize fsf not to read.
1264 - Use ioctl() fsf if it exists. Figure out where we are from
1265 the mt_status command. Use slow fsf only if other does not work.
1266 - Enhance "update slots" to include a "scan" feature
1267 scan 1; scan 1-5; scan 1,2,4 ... to update the catalog
1268 - Allow a slot or range of slots on the label barcodes command.
1269 - Finish implementation of Verify=DiskToCatalog
1270 - Make sure that Volumes are recycled based on "Least recently used"
1271 rather than lowest MediaId.
1272 - Add flag to write only one EOF mark on the tape.
1273 - Implement autochanger testing in btape "test" command.
1274 - Implement lmark to list everyfile marked.
1275 - Make mark/unmark report how many files marked/unmarked.
1276 - Keep last 5 or 10 completed jobs and show them in a similar list.
1277 - Make a Running Jobs: output similar to current Scheduled Jobs:
1278 - Change "create_media_record in bscan to use Archive instead of Full.
1279 - Have some way to estimate the restore size or have it printed.
1280 - Volume problems occurs if you have valid volume, written, then it is
1281 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:
1282 block.c:640 Read zero bytes on device /tmp.
1283 - Make sure that 64 bit I/O packets are used on Cygwin.
1284 - Add to supported autochangers
1286 Auto-Changer : QUALSTAR TLS-4210
1287 Manufufactur : Qualstar
1288 Tapes : 12 (AIT1: 36GB, AIT2: 50GB all uncompressed)
1289 Drives : 2xAIT2 (installed in the Qualstar: SONY SDX-500C AIT2)
1290 - Document estimate command in tree.
1291 - Document lsmark command in tree.
1292 - Setup a standard job that builds a bootstrap file and saves
1293 it with the catalog database.
1294 - See if a restore job can add a file to the tape (prohibit this).
1295 - Restrict characters permitted in a name.
1296 - In restore, provide option for limiting to a particular Pool.
1297 - In restore, list FileSets that only have different base names --
1298 i.e. any FileSet with the same name should be treated as the same.
1299 - Make Scheduler sort jobs by StartTime, Priority.
1300 - Make sure smtp and any other useful program is executable by the world
1301 in case Bacula is not running as root.
1302 - Look at Dan's field width problems in PostgreSQL.
1303 - Look at effect of removing GROUP BYs.
1304 - In restore take all filesets with same base name.
1305 - From Alan Brown <ajb2@mssl.ucl.ac.uk>
1306 BTW, there's a make install bug in 1.33 - with --enable-gnome,
1307 gnome-console is built, but the binary and .conf are not being installed.
1308 - Permit Bacula and apcupsd donations (not done for apcupsd).
1309 - Fix Ctl-C crashing the Console (readline?).
1310 - Look at code in recycle_oldes_purged_volume() recycle.c. Why not
1311 let SQL do ORDER BY LastWritten ASC?
1312 - Look at find_next_volume() algorithm. Currently, it selects:
1313 +---------+------------+---------------------+-----------+
1314 | MediaId | VolumeName | LastWritten | VolBytes |
1315 +---------+------------+---------------------+-----------+
1316 | 3 | Test13 | 0000-00-00 00:00:00 | 1 |
1317 | 4 | Test14 | 0000-00-00 00:00:00 | 1 |
1318 | 1 | test11 | 2003-12-03 18:39:55 | 4,004,926 |
1319 | 2 | test12 | 2004-01-04 15:25:56 | 2,078,691 |
1320 +---------+------------+---------------------+-----------+
1321 but perhaps it should fill already used Volumes first, and use
1322 Append volumes before Purged, or Recycled, ...
1323 - Possibly remove the "|| ap == NULL" on lines 123 and 207 of lib/var.c,
1324 which creates compile problems on alpha systems.
1325 var.c:123: no match for `va_list & == long int'
1326 - Check "restore" 3 (JobId), then it asks for Storage resource. Does
1327 it verify that the correct volume is chosen?
1328 - Make Bacula "poll a drive".
1329 - Notes for final checking of Nic's code:
1330 Could I get you to double check the switch () statements in the
1331 job_check_maxwaittime and job_check_maxruntime functions in
1333 - Define week of year for scheduler. W01, W02, ...
1334 Week 01 of a year is per definition the first week that has the
1335 Thursday in this year, which is equivalent to the week that contains the
1336 fourth day of January. In other words, the first week of a new year is
1337 the week that has the majority of its days in the new year. Week 01
1338 might also contain days from the previous year and the week before week
1339 01 of a year is the last week (52 or 53) of the previous year even if it
1340 contains days from the new year. A week starts with Monday (day 1) and
1341 ends with Sunday (day 7). For example, the first week of the year 1997
1342 lasts from 1996-12-30 to 1997-01-05 and can be written in standard
1345 The week notation can also be extended by a number indicating the day
1346 of the week. For example, the day 1996-12-31, which is the Tuesday (day
1347 2) of the first week of 1997, can also be written as
1348 1997-W01-2 or 1997W012
1349 - Either restrict the characters in a name, or fix the problem
1350 emailing with names containing / (smtp command line breaks).
1351 - Implement .consolerc for Console
1352 - Implement scan: for every slot it finds, zero the slot of
1353 Volume other volume having that slot.
1354 - Make restore job check if all the files are actually restored.
1355 - Look at 2Gb limit for SQLite.
1356 - Fix get_storage_from_media_type (ua_restore) to use command line
1358 - Don't print "Warning: Wrong Volume mounted ..." if mounting second volume.
1359 - Write a mini-readline with history and editing.
1360 - Take a careful look a the Basic recycling algorithm. When Bacula
1361 chooses, the order should be:
1363 - Look for Recycle or Purged
1366 Instead of using lowest media Id, find the least recently used
1369 When the tape is mounted and Bacula requests the status
1370 - Do everything possible to use it.
1372 Define a "available" status, which is the currently mounted
1373 Volume and all volumes that are currently in the autochanger.
1374 - Is a pool specification really needed for a restore? Yes, and
1375 you may want to exclude archive Pools.
1376 - Implement a PostgreSQL driver.
1377 - Fix restore to list errors if Invalid block found, and if # files
1378 restored does not match # expected.
1379 - Something is not right in last block of fill command.
1380 - Add FileSet to command line arguments for restore.
1381 - Enhance time and size scanning routines.
1382 - Add Console usr permissions -- do by adding filters for
1383 jobs, clients, storage, ...
1384 - Put max network buffer size on a directive.
1385 - Why does "mark cygwin" take so long!!!!!!!!
1386 - Implement alist processing for ACLs from Console.
1387 - When a file is set for restore, walk back up the chain of
1388 directories, setting them to be restored.
1389 - Figure out a way to set restore on a directory without recursively
1390 decending. (recurse off?).
1391 - Fix restore to only pull in last Differential and later Incrementals.
1392 - Implement 3 Pools for a Job:
1395 Full Backup Pool = xxx
1396 Incremental Backup Pool = yyy
1397 Differential Backup Pool = zzz
1399 - Look at ASSERT() at 384 src/lib/bnet.c
1400 - Dates are wrong in restore list from Win32 FD.
1401 - Dates are wrong in catalog from Win32 FD.
1402 - Remove h_errno from bnet.c by including proper header.
1403 - For "list jobs" order by EndTime.
1404 - Make two tape fill test work.
1405 - Add atime preservation.
1406 - Do not err job if could not write bootstrap file.
1407 - Save and restore last_job across executions.
1408 - Have each daemon save the last_jobs structure when exiting and
1409 read it back in when starting up.
1410 - "restore jobid=1 select" calls get_storage_xxx, which prints "JobId 1 is
1412 - Make column listing for running jobs JobId Level Type Started Name Status
1413 - Why does Bacula need the drive open to do "autochanger list" ?
1414 - Add data compare on write/read in btape "test".
1415 - Rescue builds incorrect script files on Rufus.
1416 - Release SQLite 2.8.9
1417 - During install, copy any console.conf to bconsole.conf.
1418 - Check: Run = Level=Differential feb-dec 1 at 1:05 to see if wday is empty.
1419 - Look at installation file permissions with Scott so that make install
1421 - Finish code passing files=nnn to restore start.
1422 - Add ctl-c to console to stop current command and discard buffered
1424 - Estimate to Tibs never returns.
1425 - Symbolic link a directory to another one, then backup the symbolic link.
1426 - Check and possibly fix problems with hard links.
1427 - Fix query buffer clobber ua_query.c
1428 - Allow "delete job jobid=xx jobid=xxx".
1429 - Update volume FromPool (or FromPool=xxx) refreshes the Volume defaults from Pool.
1430 - Implement multiple Volume in "purge jobs volume=".