5 - Implement SIGHUP in Dir
6 - Implement fast tree insert (doubly linked list?)
7 - Do tape alerts -- see tapealert.txt
8 - When restore started from console, report jobid.
9 - Feedback while the tree is being built.
10 - Add reporting in attr despooling.
11 - Win32 inc problem when new directory added.
13 The error I when I installed 1.34.2 clients:
15 linked to missing export KERNEL32.DLL:GetFileAttributesExA.
17 Documentation from Christopher Hull on new client job requests:
19 The command can be anything that cmd.exe or command.com will recognize
20 as a executable file. Specifiying the executable's extention is
21 optional, unless there is an ambiguity. (i.e. ls.bat, ls.exe)
23 The System %Path% will be searched for the command. (under the
24 envrionment variable dialog you have have both System Environment and
25 User Environment, we believe that only the System environment will be
26 available to bacual-fd, if it is running as a service.)
28 System environment varaible can be called out using the %var% syntax and
29 used as either part of the command name or arguments.
31 When specifiying a full path to an executable if the path or executable
32 name contains whitespace or special characters they will need to be
33 quoted. Arguments containing whitespace or special characters will also
36 ClientRunBeforeJob = "\"C:/Program Files/Software
37 Vendor/Executable\" /arg1 /arg2 \"foo bar\""
39 The special characters &()[]{}^=;!'+,`~ will need to be quoted if part
40 of a filename or argument.
42 If someone is logged in a blank "command" window running the commands
43 will be present during the execution of the command.
46 Documentation to do: (any release a little bit at a time)
47 - Document query file format.
48 - Document problems with Verify and pruning.
49 - Document how to use multiple databases.
50 - VXA drives have a "cleaning required"
51 indicator, but Exabyte recommends preventive cleaning after every 75
54 In this context, it should be noted that Exabyte has a command-line
55 vxatool utility available for free download. (The current version is
56 vxatool-3.72.) It can get diagnostic info, read, write and erase tapes,
57 test the drive, unload tapes, change drive settings, flash new firmware,
59 Of particular interest in this context is that vxatool <device> -i will
60 report, among other details, the time since last cleaning in tape motion
61 minutes. This information can be retrieved (and settings changed, for
62 that matter) through the generic-SCSI device even when Bacula has the
63 regular tape device locked. (Needless to say, I don't recommend
64 changing tape settings while a job is running.)
65 - Lookup HP cleaning recommendations.
66 - Lookup HP tape replacement recommendations (see trouble shooting autochanger)
67 - Add more obvious documentation of @ for conf files.
68 - Document doing table repair
70 Testing to do: (painful)
72 - blocksize recognition code.
73 - Test if rewind at end of tape waits for tape to rewind.
76 For 1.33 Testing/Documentation:
77 - Figure out how to use ssh or stunnel to protect Bacula communications.
78 Add Dan's work to manual See ssl.txt
79 - Add db check test to regression. Test each function like delete,
81 - If you use restore replace=never, the directory attributes for
82 non-existent directories will not be restored properly.
85 - Add multiple-media-types.txt
86 - look at mxt-changer.html
87 - Document a get out of jail procedure if everything breaks if you lost/broke
89 - Make ? do a help command (no return needed).
90 - Implement restore directory.
91 - Add All Local Partitions = yes to new style saves.
92 - Document streams and how to implement them.
93 - Possibly implement "Ensure Full Backup = yes" looks for a failed full backup
94 and upgrades the current backup if one exists.
95 - Check that barcode reading and update slots scan works.
96 - Try not to re-backup a file if a new hard link is added.
97 - Add feature to backup hard links only, but not the data.
98 - Add "All Local = yes" option to save to include all local partitions.
99 - Fix stream handling to be simpler.
100 - Add Priority and Bootstrap to Run a Job.
101 - Eliminate Restore "Run Restore Job" prompt by allowing new "run command
103 - Remove View FileSet button from Run a Job dialog.
104 - Don't count higher level non-saved directories in Files expected
106 - Handle prompt for restore job at end of Restore command.
107 - Add display of total selected files to Restore window.
108 - Add tree pane to left of window.
109 - Add progress meter.
110 - Polling does not work for restore. It tries a number of times,
111 gives up, and crashes the FD.
112 - Lock jcr_chain when doing attach/detach in acquire.c
113 - Add assert in free_jcr if attach/detach chain active.
114 - Max wait time or max run time causes seg fault -- see runtime-bug.txt
115 - Document writing to a CD/DVD with Bacula.
116 - Add check for tape alerts.
117 - Add a "base" package to the window installer for pthreadsVCE.dll
118 which is needed by all packages.
119 - Add message to user to check for fixed block size when the forward
120 space test fails in btape.
121 - When unmarking a directory check if all files below are unmarked and
122 then remove the + flag -- in the restore tree.
123 - Possibly implement: Action = Unmount Device="TapeDrive1" in Admin jobs.
124 - Setup lrrd graphs: (http://www.linpro.no/projects/lrrd/) Mike Acar.
125 - Revisit the question of multiple Volumes (disk) on a single device.
126 - Finish SIGHUP work.
127 - Check that all change in wait status in the SD are
128 signaled to the Director.
129 - Add a block copy option to bcopy.
130 - Investigate adding Mac Resource Forks.
131 - Finish work on Gnome restore GUI.
132 - Fix "llist jobid=xx" where no fileset or client exists.
133 - Check pruning of restore jobs.
135 it seems to be complaining about 12:00pm which should be a valid 12
136 hour time. I changed the time to 11:59am and everything works fine.
137 Also 12:00am works fine. 0:00pm also works (which I don't think
138 should). None of the values 12:00pm - 12:59pm work for that matter.
139 - Add level to estimate command.
140 - For each job type (Admin, Restore, ...) require only the really necessary
142 - Fix option 2 of restore -- list where file is backed up -- require Client,
143 then list last 20 backups.
144 - Add all pools in Dir conf to DB also update them to catch changed
145 LabelFormats and such.
146 - Update volumes FromPool=xxx does all volumes.
147 - Pass Director resource name as an option to the Console.
148 - Add a "batch" mode to the Console (no unsolicited queries, ...).
149 - Add code to check for tape alerts -- tapeinfo.
150 - Make sure list of Volumes needed is in correct order for restore.
152 - Remove paths (and files that reference them) that have no trailing slash
153 in dbcheck -- or add a trailing slash.
154 - Remove Filenames (and files that reference them) that have a trailing
155 slash in dbcheck -- or remove the trailing slash.
156 - Remove orphaned paths/filenames by copying them to a new table with a
157 reference count, then mark all referenced files/paths and remove unreferenced
159 - Add a .list all files in the restore tree (probably also a list all files)
160 Do both a long and short form.
161 - Allow browsing the catalog to see all versions of a file (with
162 stat data on each file).
163 - Restore attributes of directory if replace=never set but directory
165 - Allow "delete job jobid=xxx,yyy,aaa-bbb" i.e. list + ranges.
166 - Use SHA1 on authentication if possible.
167 - See comtest-xxx.zip for Windows code to talk to USB.
168 - Make btape accept Device Names in addition to Archive names.
169 - Add Events and Perl scripting.
170 - Add John's appended files:
171 Appended = { /files/server/logs/http/*log }
172 and such files would be treated as follows.On a FULL backup, they would
173 be backed up like any other file.On an INCREMENTAL backup, where a
174 previous INCREMENTAL or FULL was already in thecatalogue and the length
175 of the file wasgreater than the length of the last backup, only thedata
176 added since the last backup will be dumped.On an INCREMENTAL backup, if
177 the length of the file is less than thelength of the file with the same
178 name last backed up, the completefile is dumped.On Windows systems, with
179 creation date of files, we can be evensmarter about this and not count
180 entirely upon the length.On a restore, the full and all incrementals
181 since it will beapplied in sequence to restore the file.
182 - Add a regression test for dbcheck.
183 - Add disk seeking on restore. - Allow
184 for optional cancelling of SD and FD in case DIR
185 gets a fatal error. Requested by Jesse Guardiani <jesse@wingnet.net>
186 - Bizarre message: Error: Could not open WriteBootstrap file:
187 - Build console in client only build.
188 - Add "limit=n" for "list jobs"
189 - Check new HAVE_WIN32 open bits.
190 - Check if the tape has moved before writing.
191 - Handling removable disks -- see below:
192 - Multiple drive autochanger support -- see below.
193 - Keep track of tape use time, and report when cleaning is necessary.
194 - Fix FreeBSD mt_count problem.
195 - Add FromClient and ToClient keywords on restore command (or
196 BackupClient RestoreClient).
197 - Automatic "update slots" on user configuration directive when a
199 - Implement a JobSet, which groups any number of jobs. If the
200 JobSet is started, all the jobs are started together.
201 Allow Pool, Level, and Schedule overrides.
202 - Enhance cancel to timeout BSOCK packets after a specific delay.
203 - When I restore to Windows the Created, Accessed and Modifiedtimes are
204 those of the time of the restore, not those of the originalfile.
205 The dates you will find in your restore log seem to be the original
207 - Volume "add"ed to Pool gets recycled in first use. VolBytes=0
208 - If a tape is recycled while it is mounted, Stanislav Tvrudy must do an
209 additional mount to deblock the job.
211 bscan: sql_update.c:65 UPDATE File SET MD5='Ij+5kwN6TFIxK+8l8+/I+A' WHERE FileId=0
212 bscan: bscan.c:1074 Could not add MD5/SHA1 to File record. ERR=sql_update.c:65 Update problem: affected_rows=0
213 - Do scheduling by UTC using gmtime_r() in run_conf, scheduler, and
214 ua_status.!!! Thanks to Alan Brown for this tip.
215 - Look at updating Volume Jobs so that Max Volume Jobs = 1 will work
216 correctly for multiple simultaneous jobs.
217 - Correct code so that FileSet MD5 is calculated for < and | filename
219 - Mark Volume in error on error from WEOF.
220 - Implement the Media record flag that indicates that the Volume does disk
222 - Implement VolAddr, which is used when Volume is addressed like a disk,
223 and form it from VolFile and VolBlock.
224 - Make multiple restore jobs for multiple media types specifying
225 the proper storage type.
226 - Implement MediaType keyword in bsr?
227 - Fix fast block rejection (stored/read_record.c:118). It passes a null
228 pointer (rec) to try_repositioning().
229 - Look at extracting Win data from BackupRead.
230 - Having dashes in filenames apparently creates problems for restore
231 by filename??? hard to believe.
232 - Implement RestoreJobRetention? Maybe better "JobRetention" in a Job,
233 which would take precidence over the Catalog "JobRetention".
234 - Implement Label Format in Add and Label console commands.
235 - Possibly up network buffers to 65K. Put on variable.
236 - Put email tape request delays on one or more variables. User wants
237 to cancel the job after a certain time interval. Maximum Mount Wait?
238 - Job, Client, Device, Pool, or Volume?
239 Is it possible to make this a directive which is *optional* in multiple
240 resources, like Level? If so, I think I'd make it an optional directive
241 in Job, Client, and Pool, with precedence such that Job overrides Client
242 which in turn overrides Pool.
243 - Print a message when a job starts if the conf file is not current.
245 - Spooling ideas taken from Volker Sauer's and other's emails:
246 > IMHO job spooling should be turned on
252 > where and 2) overrides 1) and 3) is independent.
254 Yes, this is the minimum that I think is necessary.
258 > It should be switched by job, because the job that backs up the machine
259 > with the bacula-sd on doesn't need spooling.
260 > It should be switched by schedule, because for full-backups I don't need
261 > spooling, so I can switch it off (because the network faster then the
264 True, with the exception that if you have enough disk spool space,
265 and you want to run concurrent jobs, spooling can eliminate the block
266 interleaving restore inefficiencies.
268 > And you should be able to turn it of by sd for sd-machines with low disk
269 > capacity or if you just don't need or want this feature.
272 > - definitly the possibility for multipe spool direcories
274 Having multiple directories is no problem -- having different maximum
275 sizes creates specification problems. At some point, I will probably
276 have a common SD pool of spool directories as well as a set of
277 private spool directories for each device. The first implementation
278 will be a set of private spool directories for each device since
279 managing a global pool with a bunch of threads writing into the same
280 directory is *much* more complicated and prone to error.
282 > - the ability to spool parts of a backup (not the whole client)
284 This may change in the future, but for the moment, it will spool
285 either to a job high water mark, or until the directory is full
286 (reaches max spool size or I/O error). It will then write to tape,
287 truncate the spool file, and begin spooling again.
289 > - spooling while writing to tape
291 Not within a job, but yes, if you run concurrent jobs -- each is a
292 different thread. Within a job could be a feature, but *much* later.
294 > - parallel spooling (like parallel jobs/ concurrent jobs) of clients
296 Yes, this is one of my main motivations for doing it (aside from
297 eliminating tape "shoe shine" during incremental backups.
299 > - flushing a backup that only went to disk (like amflush in amanda)
301 This will be a future feature, since spooling is different from backing
302 up to disk. The future feature will be "migration" which will move a job
303 from one backup Volume to another.
305 - New Storage specifications:
306 Passed to SD as a sort of BSR record called Storage Specification
310 MediaType -> Next MediaType
312 Device -> Next Device
313 Write Copy Resource that makes a copy of a resource.
315 Allow multiple Storage specifications
323 Allow Multiple Pool specifications (note, Pool currently
325 Allow Multiple MediaType specifications
326 Allow Multiple Device specifications
327 Perhaps keep this in a single SSR
328 Tie a Volume to a specific device by using a MediaType that
329 is contained in only one device.
330 In SD allow Device to have Multiple MediaTypes
333 - Look at www.nu2.nu/pebuilder as a helper for full windows
335 Ideas from Jerry Scharf:
336 First let's point out some big pluses that bacula has for this
338 more importantly it's active. Thank you so much for that
339 even more important, it's not flaky
340 it has an open access catalog, opening many possibilities
341 it's pushing toward heterogeneous systems capability
343 I don't remember an include file directive for config files
344 (not filesets, actual config directives)
345 can you check the configs without starting the daemon?
346 some warnings about possible common mistakes
348 doing the testing and blessing of concurrent backup writes
349 this is absolutely necessary in the enterprise
350 easy user recovery GUI with full access checking
351 Macintosh file client
352 macs are an interesting niche, but I fear a server is a rathole
353 working bare iron recovery for windows
354 much better handling on running config changes
355 thinking through the logic of what happens to jobs in progress
356 the option for inc/diff backups not reset on fileset revision
357 a) use both change and inode update time against base time
358 b) do the full catalog check (expensive but accurate)
359 sizing guide (how much system is needed to back up N systems/files)
360 consultants on using bacula in building a disaster recovery system
362 or how to get at fancy things that one could do with bacula
363 logwatch code for bacula logs (or similar)
364 linux distro inclusion of bacula (brings good and bad, but necessary)
365 win2k/XP server capability (icky but you asked)
366 support for Oracle database ??
368 - Look at adding SQL server and Exchange support for Windows.
369 - Restore: Enter Filename: 'C:/Documents and Settings/Comercial/My
370 Documents/MOP/formulário de registro BELAS ARTES.doc' causes Bacula to
372 - Each DVD-RAM disk would be a volume, just like each tape is
373 a volume. It's a 4.7GB media with random access, but there's nothing about
374 it that I can see that makes it so different than a tape from bacula's
375 perspective. Why couldn't I back up to a bare floppy as a volume (ignoring
377 - Make dev->file and dev->block_num signed integers so that -1 can
378 be an invalid value which happens with BSR.
379 - Create VolAddr for disk files in place of VolFile and VolBlock. This
380 is needed to properly specify ranges.
381 - Print bsmtp output to job report so that problems will be seen.
382 - Pass the number of files to be restored to the FD for reporting
383 - Add progress of files/bytes to SD and FD.
384 - Don't continue Restore if no files selected.
385 - Print warning message if FileId > 4 billion
386 - do a "messages" before the first prompt in Console
387 - Add a date and time stamp at the beginning of every line in the
388 Job report (Volker Sauer).
389 - Client does not show busy during Estimate command.
390 - Implement Console mtx commands.
391 - Add a default DB password to MySQL.
392 GRANT all privileges ON bacula.* TO bacula@localhost IDENTIFIED BY
395 - Implement a Mount Command and an Unmount Command where
396 the users could specify a system command to be performed
397 to do the mount, after which Bacula could attempt to
398 read the device. This is for Removeable media such as a CDROM.
399 - Most likely, this mount command would be invoked explicitly
400 by the user using the current Console "mount" and "unmount"
401 commands -- the Storage Daemon would do the right thing
402 depending on the exact nature of the device.
403 - As with tape drives, when Bacula wanted a new removable
404 disk mounted, it would unmount the old one, and send a message
405 to the user, who would then use "mount" as described above
406 once he had actually inserted the disk.
407 - Implement dump/print label to UA
408 - Implement disk spooling. Two parts: 1. Spool to disk then
409 immediately to tape to speed up tape operations. 2. Spool to
410 disk only when the tape is full, then when a tape is hung move
412 - Scratch Pool where the volumes can be re-assigned to any Pool.
413 - bextract is sending everything to the log file ****FIXME****
414 - Add Progress command that periodically reports the progress of
416 - Restrict characters permitted in a Resource name, and don't permit
418 - Allow multiple Storage specifications (or multiple names on
419 a single Storage specification) in the Job record. Thus a job
420 can be backed up to a number of storage devices.
421 - Implement some way for the File daemon to contact the Director
422 to start a job or pass its DHCP obtained IP number.
423 - Implement multiple Consoles.
424 - Implement a query tape prompt/replace feature for a console
426 Two jobs ready to go, first one blocked waiting for media
427 Cancel 2nd job ("waiting execution" one)
430 - Copy console @ code to gnome2-console
431 - Make AES the only encryption algorithm see
432 http://csrc.nist.gov/CryptoToolkit/aes/). It's
433 an officially adopted standard, has survived peer
434 review, and provides keys up to 256 bits.
435 - Think about how space could be freed up on a tape -- perhaps this
436 is a Merge or Compact feature that is needed.
437 - Modify FileSet, did not upgrade the current Increment job, but
438 waited for the next job to be upgraded.
439 - Take a careful look at SetACL http://setacl.sourceforge.net
440 - Implement a where command for the tree telling where a file
442 - Take a careful look at Level for the estimate command, maybe make
443 it a command line option.
444 - Add Volume name to "I cannot write on this volume because"
445 - Make tree walk routines like cd, ls, ... more user friendly
446 by handling spaces better.
447 - Write your PID file and chown root:wheel before drop.
448 - Make sure there is no symlink in a file before creating a
450 - Look at mktemp or mkstemp(3).
451 mktemp and mkstemp create files with predictable names too. That's
452 not the vulnerability. The vulnerability is in creating files without
453 using the O_EXCL flag, which means "only create this file if it doesn't
454 exist, including if the file is a dangling symlink."
456 It is *NOT* enough to do the equivalent of
458 if doesn't exist $filename
459 then create $filename
461 because between the test and the create another process could have
462 gotten the CPU and created the file. You must use atomic functions
463 (those that don't get interrupted by other processes) and O_EXCL is
464 the only way for this particular example.
465 - Automatically create pools, but instead of looking for what
466 in in Job records, walk through the pool resources.
467 - Check and double check tree code, why does it take so long?
468 - Add device name to "Current Volume not acceptable because ..."
469 - Make sure that Bacula rechecks the tape after the 20 min wait.
470 - Set IO_NOWAIT on Bacula TCP/IP packets.
471 - Try doing a raw partition backup and restore by mounting a
474 Yes, it would allow to highly automatic the request for new tapes. If a
475 tape is empty, bacula reads the barcodes (native or simulated), and if
476 an unused tape is found, it runs the label command with all the
477 necessary parameters.
479 By the way can bacula automatically "move" an empty/purged volume say
480 in the "short" pool to the "long" pool if this pool runs out of volume
482 - Eliminate orphaned jobs: dbcheck, normal pruning, delete job command.
483 Hm. Well, there are the remaining orphaned job records:
485 | 105 | Llioness Save | 0000-00-00 00:00:00 | B | D | 0 | 0 | f |
486 | 110 | Llioness Save | 0000-00-00 00:00:00 | B | I | 0 | 0 | f |
487 | 115 | Llioness Save | 2003-09-10 02:22:03 | B | I | 0 | 0 | A |
488 | 128 | Catalog Save | 2003-09-11 03:53:32 | B | I | 0 | 0 | C |
489 | 131 | Catalog Save | 0000-00-00 00:00:00 | B | I | 0 | 0 | f |
491 As you can see, three of the five are failures. I already deleted the
492 one restore and one other failure using the by-client option. Deciding
493 what is an orphaned job is a tricky problem though, I agree. All these
494 records have or had 0 files/ 0 bytes, except for the restore. With no
495 files, of course, I don't know of the job ever actually becomes
496 associated with a Volume.
498 (I'm not sure if this is documented anywhere -- what are the meanings of
499 all the possible JobStatus codes?)
501 Looking at my database, it appears to me as though all the "orphaned"
502 jobs fit into one of two categories:
504 1) The Job record has a StartTime but no EndTime, and the job is not
507 2) The Job record has an EndTime, indicating that it completed, but
508 it has no associated JobMedia record.
511 This does suggest an approach. If failed jobs (or jobs that, for some
512 other reason, write no files) are associated with a volume via a
513 JobMedia record, then they should be purged when the associated volume
514 is purged. I see two ways to handle jobs that are NOT associated with a
517 1) purge them automatically whenever any volume is manually purged;
519 2) add an option to the purge command to manually purge all jobs with
520 no associated volume.
522 I think Restore jobs also fall into category 2 above .... so one might
523 want to make that "The Job record has an EndTime,, but no associated
524 JobMedia record, and is not a Restore job."
525 - make "btape /tmp" work.
526 - Make sure a rescheduled job is properly reported by status.
527 - Walk through the Pool records rather than the Job records
528 in dird.c to create/update pools.
529 - What to do about "list files job=xxx".
530 - When job rescheduled, status gives is waiting for Client Rufus
531 to connect to Storage File. Dir needs to inform SD that job
533 - Make Dmsg look at global before calling subroutine.
534 - Enable trace output at runtime for Win32
535 - Available volumes for autochangers (see patrick@baanboard.com 3 Sep 03
536 and 4 Sep) scan slots.
537 - Get and test MySQL 4.0
538 - Do a complete audit of all pthreads_mutex, cond, ... to ensure that
539 any that are dynamically initialized are destroyed when no longer used.
540 - Look at how fuser works and /proc/PID/fd that is how Nic found the
541 file descriptor leak in Bacula.
542 - Implement WrapCounters in Counters.
543 - Turn on SIGHUP in dird.c and test.
544 - Use system dependent calls to get more precise info on tape errors.
545 - Add heartbeat from FD to SD if hb interval expires.
546 - Suppress read error on blank tape when doing a label.
547 - Can we dynamically change FileSets?
548 - If pool specified to label command and Label Format is specified,
549 automatically generate the Volume name.
550 - Why can't SQL do the filename sort for restore?
551 - Look at libkse (man kse) for FreeBSD threading.
552 - Look into Microsoft Volume Shadowcopy Service VSS for backing
553 up system state components (Active Directory, System Volume, ...)
554 - Add ExhautiveRestoreSearch
555 - Look at the possibility of loading only the necessary
556 data into the restore tree (i.e. do it one directory at a
557 time as the user walks through the tree).
558 - Possibly use the hash code if the user selects all for a restore command.
559 - Orphaned Dir buffer at parse_conf.c:373 => store_dir
560 - Fix "restore all" to bypass building the tree.
561 - Prohibit backing up archive device (findlib/find_one.c:128)
562 - Implement Release Device in the Job resource to unmount a drive.
563 - Implement Acquire Device in the Job resource to mount a drive,
564 be sure this works with admin jobs so that the user can get
565 prompted to insert the correct tape. Possibly some way to say to
566 run the job but don't save the files.
567 - Implement FileOptions (see end of this document)
568 - Make things like list where a file is saved case independent for
571 - Bacula needs to propagate SD errors.
572 > > cluster-dir: Start Backup JobId 252, Job=REUTERS.2003-08-11_15.04.12
573 > > prod4-sd: REUTERS.2003-08-11_15.04.12 Error: Write error on device
574 /dev/nst0. ERR=Input/output error.
575 > > prod4-sd: REUTERS.2003-08-11_15.04.12 Error: Re-read of last block failed.
576 Last block=5162 Current block=5164.
577 > > prod4-sd: End of medium on Volume "REU007" Bytes=16,303,521,933
579 - Use autochanger to handle multiple devices.
580 - Add SuSE install doc to list.
581 - Check and rechedk "Invalid block number"
582 - Make bextract release the drive properly between tapes
583 so that an autochanger can be made to work.
584 - User wants to NOT backup up certain big files (email files).
585 - Maybe remove multiple simultaneous devices code in SD.
586 - On Windows with very long path names, it may be impossible to create
587 a file (and thus restore it) because the total length is too long.
588 We must cd into the directory then create the file without the
590 - lstat() is not going to work on Win32 for testing date.
591 - Implement a Recycle command
592 - Add client name to cram-md5 challenge so Director can immediately
593 verify if it is the correct client.
594 - Add JobLevel in FD status (but make sure it is defined).
595 - Audit all UA commands to ensure that we always prompt where possible.
596 - Check Jmsg in bnet, may not work, must dup bsock.
597 - Suppress Job Name in Jmsg for console
598 - Create Pools that are referenced in a Run statement at startup if possible.
599 - Use runbeforejob to unload, then reload a volume previously used,
600 then the next job run gets an error reading the drive.
601 - Make bootstrap filename unique.
602 - Test a second language e.g. french.
603 - Start working on Base jobs.
604 - Make "make binary-release" work from any directory.
605 - Implement UnsavedFiles DB record.
606 - Implement argc/argv for daemon command line scanning using table driven
608 - Implement table driven single argc/argv scanner to pickup all arguments.
609 Much like xxx_conf.c scan table.
610 keyword, handler(store_routine), store_address, code, flags, default.
611 - From Phil Stracchino:
612 It would probably be a per-client option, and would be called
613 something like, say, "Automatically purge obsoleted jobs". What it
614 would do is, when you successfully complete a Differential backup of a
615 client, it would automatically purge all Incremental backups for that
616 client that are rendered redundant by that Differential. Likewise,
617 when a Full backup on a client completed, it would automatically purge
618 all Differential and Incremental jobs obsoleted by that Full backup.
619 This would let people minimize the number of tapes they're keeping on
620 hand without having to master the art of retention times.
621 - Implement a M_SECURITY message class.
622 - When doing a Backup send all attributes back to the Director, who
623 would then figure out what files have been deleted.
624 - Currently in mount.c:236 the SD simply creates a Volume. It should have
625 explicit permission to do so. It should also mark the tape in error
626 if there is an error.
627 - Make sure all restore counters are working correctly in the FD.
628 - SD Bytes Read is wrong.
629 - Look at ALL higher level routines that call block.c to be sure
630 they don't expect something in errmsg.
631 - Investigate doing RAW backup of Win32 partition.
632 - Add thread specific data to hold the jcr -- send error messages from
633 low level routines by accessing it and using Jmsg().
634 - Cancel waiting for Client connect in SD if FD goes away.
636 - Examine Bare Metal restore problem (a FD crash exists somewhere ...).
637 - Implement timeout in response() when it should come quickly.
638 - Implement console @echo command.
639 - Implement a Slot priority (loaded/not loaded).
640 - Implement "vacation" Incremental only saves.
641 - Implement single pane restore (much like the Gftp panes).
642 - Implement Automatic Mount even in operator wait.
643 - Implement create "FileSet"?
644 - Fix watchdog pthread crash on Win32 (this is pthread_kill() Cygwin bug)
645 - Implement "scratch pool" where tapes are defined and can be
646 taken by any pool that needs them.
647 - Implement restore "current system", but take all files without
648 doing selection tree -- so that jobs without File records can
650 - Add prefixlinks to where or not where absolute links to FD.
651 - Issue message to mount a new tape before the rewind.
652 - Simplified client job initiation for portables.
653 - If SD cannot open a drive, make it periodically retry.
654 - Add more of the config info to the tape label.
656 - If tape is marked read-only, then try opening it read-only rather than
657 failing, and remember that it cannot be written.
658 - Refine SD waiting output:
659 Device is being positioned
660 > Device is being positioned for append
661 > Device is being positioned to file x
663 - Figure out some way to estimate output size and to avoid splitting
664 a backup across two Volumes -- this could be useful for writing CDROMs
665 where you really prefer not to have it split -- not serious.
666 - Have SD compute MD5 or SHA1 and compare to what FD computes.
667 - Make VolumeToCatalog calculate an MD5 or SHA1 from the
668 actual data on the Volume and compare it.
669 - Implement Bacula plugins -- design API
670 - Make bcopy read through bad tape records.
671 - Program files (i.e. execute a program to read/write files).
672 Pass read date of last backup, size of file last time.
673 - Add Signature type to File DB record.
674 - CD into subdirectory when open()ing files for backup to
675 speed up things. Test with testfind().
676 - Priority job to go to top of list.
677 - Why are save/restore of device different sizes (sparse?) Yup! Fix it.
678 - Implement some way for the Console to dynamically create a job.
679 - Restore to a particular time -- e.g. before date, after date.
680 - Solaris -I on tar for include list
681 - Need a verbose mode in restore, perhaps to bsr.
682 - bscan without -v is too quiet -- perhaps show jobs.
683 - Add code to reject whole blocks if not wanted on restore.
684 - Check if we can increase Bacula FD priorty in Win2000
685 - Make sure the MaxVolFiles is fully implemented in SD
686 - Check if both CatalogFiles and UseCatalog are set to SD.
687 - Figure out how to do a bare metal Windows restore
688 - Possibly add email to Watchdog if drive is unmounted too
689 long and a job is waiting on the drive.
690 - Restore program that errs in SD due to no tape, reports
691 OK incorrectly in output.
692 - After unmount, if restore job started, ask to mount.
693 - Convert all %x substitution variables, which are hard to remember
694 and read to %(variable-name). Idea from TMDA.
695 - Remove NextId for SQLite. Optimize.
696 - Move all SQL statements into a single location.
697 - Add UA rc and history files.
698 - put termcap (used by console) in ./configure and
699 allow -with-termcap-dir.
700 - Fix Autoprune for Volumes to respect need for full save.
701 - Fix Win32 config file definition name on /install
702 - Compare tape to Client files (attributes, or attributes and data)
703 - Make all database Ids 64 bit.
704 - Write an applet for Linux.
705 - Allow console commands to detach or run in background.
706 - Fix status delay on storage daemon during rewind.
707 - Add SD message variables to control operator wait time
708 - Maximum Operator Wait
709 - Minimum Message Interval
710 - Maximum Message Interval
711 - Send Operator message when cannot read tape label.
712 - Verify level=Volume (scan only), level=Data (compare of data to file).
713 Verify level=Catalog, level=InitCatalog
715 - Add keyword search to show command in Console.
716 - Events : tape has more than xxx bytes.
717 - Complete code in Bacula Resources -- this will permit
718 reading a new config file at any time.
719 - Handle ctl-c in Console
720 - Implement script driven addition of File daemon to config files.
721 - Think about how to make Bacula work better with File (non-tape) archives.
722 - Write Unix emulator for Windows.
723 - Put memory utilization in Status output of each daemon
724 if full status requested or if some level of debug on.
725 - Make database type selectable by .conf files i.e. at runtime
726 - Set flag for uname -a. Add to Volume label.
727 - Implement throttled work queue.
728 - Restore files modified after date
729 - SET LD_RUN_PATH=$HOME/mysql/lib/mysql
730 - Implement Restore FileSet=
731 - Create a protocol.h and protocol.c where all protocol messages
733 - Remove duplicate fields from jcr (e.g. jcr.level and jcr.jr.Level, ...).
734 - Timout a job or terminate if link goes down, or reopen link and query.
735 - Concept of precious tapes (cannot be reused).
736 - Make bcopy copy with a single tape drive.
737 - Permit changing ownership during restore.
740 > My suggestion: Add a feature on the systray menu-icon menu to request
741 > an immediate backup now. This would be useful for laptop users who may
742 > not be on the network when the regular scheduled backup is run.
744 > My wife's suggestion: Add a setting to the win32 client to allow it to
745 > shut down the machine after backup is complete (after, of course,
746 > displaying a "System will shut down in one minute, click here to cancel"
747 > warning dialog). This would be useful for sites that want user
748 > woorkstations to be shut down overnight to save power.
751 - Autolabel should be specified by DIR instead of SD.
754 - AutoScan (check checksum of tape)
755 - Format command = "format /dev/nst0"
759 - Seek resolution (usually corresponds to buffer size)
760 - EODErrorCode=ENOSPC or code
761 - Partial Read error code
762 - Partial write error code
763 - Nonformatted read error
764 - Nonformatted write error
765 - WriteProtected error
769 - IgnoreCloseErrors=yes
779 - FD sends unsaved file list to Director at end of job (see
781 - File daemon should build list of files skipped, and then
782 at end of save retry and report any errors.
783 - Write a Storage daemon that uses pipes and
784 standard Unix programs to write to the tape.
786 - Need something that monitors the JCR queue and
787 times out jobs by asking the deamons where they are.
788 - Enhance Jmsg code to permit buffering and saving to disk.
789 - device driver = "xxxx" for drives.
791 - Ensure that /dev/null works
792 - Need report class for messages. Perhaps
793 report resource where report=group of messages
794 - enhance scan_attrib and rename scan_jobtype, and
795 fill in code for "since" option
796 - Director needs a time after which the report status is sent
797 anyway -- or better yet, a retry time for the job.
798 - Don't reschedule a job if previous incarnation is still running.
799 - Some way to automatically backup everything is needed????
800 - Need a structure for pending actions:
802 - termination status (part of buffered msgs?)
804 Read, Write, Clean, Delete
805 - Login to Bacula; Bacula users with different permissions:
806 owner, group, user, quotas
807 - Store info on each file system type (probably in the job header on tape.
808 This could be the output of df; or perhaps some sort of /etc/mtab record.
811 - Design at hierarchial storage for Bacula. Migration and Clone.
812 - Implement FSM (File System Modules).
813 - Audit M_ error codes to ensure they are correct and consistent.
814 - Add variable break characters to lex analyzer.
815 Either a bit mask or a string of chars so that
816 the caller can change the break characters.
817 - Make a single T_BREAK to replace T_COMMA, etc.
818 - Ensure that File daemon and Storage daemon can
819 continue a save if the Director goes down (this
820 is NOT currently the case). Must detect socket error,
821 buffer messages for later.
822 - Enhance time/duration input to allow multiple qualifiers e.g. 3d2h
823 - Add ability to backup to two Storage devices (two SD sessions) at
824 the same time -- e.g. onsite, offsite.
825 - Add the ability to consolidate old backup sets (basically do a restore
826 to tape and appropriately update the catalog). Compress Volume sets.
827 Might need to spool via file is only one drive is available.
828 - Compress or consolidate Volumes of old possibly deleted files. Perhaps
829 someway to do so with every volume that has less than x% valid
833 Migration: Move a backup from one Volume to another
834 Clone: Copy a backup -- two Volumes
836 Bacula Migration is based on Jobs (apparently Networker is file by file).
838 Migration triggered by:
842 Highwater mark (keep total size)
847 ======================================================
849 It is somewhat like a Full save becomes an incremental since
850 the Base job (or jobs) plus other non-base files.
852 - A Base backup is same as Full backup, just different type.
853 - New BaseFiles table that contains:
855 BaseJobId - Base JobId referenced for this FileId (needed ???)
856 JobId - JobId currently running
857 FileId - File not backed up, exists in Base Job
858 FileIndex - FileIndex from Base Job.
859 i.e. for each base file that exists but is not saved because
860 it has not changed, the File daemon sends the JobId, BaseId,
861 FileId, FileIndex back to the Director who creates the DB entry.
862 - To initiate a Base save, the Director sends the FD
863 the FileId, and full filename for each file in the Base.
864 - When the FD finds a Base file, he requests the Director to
865 send him the full File entry (stat packet plus MD5), or
866 conversely, the FD sends it to the Director and the Director
867 says yes or no. This can be quite rapid if the FileId is kept
868 by the FD for each Base Filename.
869 - It is probably better to have the comparison done by the FD
870 despite the fact that the File entry must be sent across the
872 - An alternative would be to send the FD the whole File entry
873 from the start. The disadvantage is that it requires a lot of
874 space. The advantage is that it requires less communications
876 - The Job record must be updated to indicate that one or more
878 - At end of Job, FD returns:
879 1. Count of base files/bytes not written to tape (i.e. matches)
880 2. Count of base file that were saved i.e. had changed.
881 - No tape record would be written for a Base file that matches, in the
882 same way that no tape record is written for Incremental jobs where
883 the file is not saved because it is unchanged.
884 - On a restore, all the Base file records must explicitly be
885 found from the BaseFile tape. I.e. for each Full save that is marked
886 to have one or more Base Jobs, search the BaseFile for all occurrences
888 - An optimization might be to make the BaseFile have:
894 This would avoid the need to explicitly fetch each File record for
895 the Base job. The Base Job record will be fetched to get the
896 VolSessionId and VolSessionTime.
897 =========================================================
900 ==========================================================
902 For each Incremental job that is run, there may be files that
903 were found but not saved because they were locked (this applies
904 only to Windows). Such a system could send back to the Director
905 a list of Unsaved files.
907 - New UnSavedFiles table that contains:
911 - Then in the next Incremental job, the list of Unsaved Files will be
912 feed to the FD, who will ensure that they are explicitly chosen even
913 if standard date/time check would not have selected them.
914 =============================================================
917 Done: (see kernsdone for more)
920 - John's Full save failed with 1.32c FD and 1.31 Dir no FD status,
921 and no error message.
922 - Add fd and st as Console keywords.
923 - Recycling volume with a Slot requires an operator intervention:
924 rufus-dir: Start Backup JobId 18, Job=kernsave.2003-11-01_21.23.52
925 rufus-dir: Pruned 1 Job on Volume Vol01 from catalog.
926 rufus-dir: There are no Jobs associated with Volume Vol01. Marking it purged.
927 rufus-dir: Recycled volume "Vol01"
928 rufus-sd: Please mount Volume "Vol01" on Storage Device "DDS-4" for Job kernsave.2003-11-01_21.23.52
929 Use "mount" command to release Job.
930 - Implement Dan's bacula script (email of 26 Oct).
931 - Add JobName= to VerifyToCatalog so that all verifies can be done at the end.
932 - Edit the Client/Storage name into authentication failure messages.
933 - Fix packet too big problem. This is most likely a Windows TCP stack
935 - Implement ClientRunBeforeJob and ClientRunAfterJob.
936 - Implement forward spacing block/file: position_device(bsr) --
937 just before read_block_from_device();
940 Multiple drive autochanger data: see Alan Brown
941 mtx -f xxx unloadStorage Element 1 is Already Full(drive 0 was empty)
942 Unloading Data Transfer Element into Storage Element 1...source Element
945 (drive 0 was empty and so was slot 1)
946 > mtx -f xxx load 15 0
947 no response, just returns to the command prompt when complete.
948 > mtx -f xxx status Storage Changer /dev/changer:2 Drives, 60 Slots ( 2 Import/Export )
949 Data Transfer Element 0:Full (Storage Element 15 Loaded):VolumeTag = HX001
950 Data Transfer Element 1:Empty
951 Storage Element 1:Empty
952 Storage Element 2:Full :VolumeTag=HX002
953 Storage Element 3:Full :VolumeTag=HX003
954 Storage Element 4:Full :VolumeTag=HX004
955 Storage Element 5:Full :VolumeTag=HX005
956 Storage Element 6:Full :VolumeTag=HX006
957 Storage Element 7:Full :VolumeTag=HX007
958 Storage Element 8:Full :VolumeTag=HX008
959 Storage Element 9:Full :VolumeTag=HX009
960 Storage Element 10:Full :VolumeTag=HX010
961 Storage Element 11:Empty
962 Storage Element 12:Empty
963 Storage Element 13:Empty
964 Storage Element 14:Empty
965 Storage Element 15:Empty
966 Storage Element 16:Empty....
967 Storage Element 28:Empty
968 Storage Element 29:Full :VolumeTag=CLNU01L1
969 Storage Element 30:Empty....
970 Storage Element 57:Empty
971 Storage Element 58:Full :VolumeTag=NEX261L2
972 Storage Element 59 IMPORT/EXPORT:Empty
973 Storage Element 60 IMPORT/EXPORT:Empty
975 Unloading Data Transfer Element into Storage Element 15...done
977 (just to verify it remembers where it came from, however it can be
978 overrriden with mtx unload {slotnumber} to go to any storage slot.)
980 There needs to be a table of drive # to devices somewhere - If there are
981 multiple changers or drives there may not be a 1:1 correspondance between
982 changer drive number and system device name - and depending on the way the
983 drives are hooked up to scsi busses, they may not be linearly numbered
984 from an offset point either.something like
986 Autochanger drives = 2
987 Autochanger drive 0 = /dev/nst1
988 Autochanger drive 1 = /dev/nst2
989 IMHO, it would be _safest_ to use explicit mtx unload commands at all
990 times, not just for multidrive changers. For a 1 drive changer, that's
996 MTX's manpage (1.2.15):
997 unload [<slotnum>] [ <drivenum> ]
998 Unloads media from drive <drivenum> into slot
999 <slotnum>. If <drivenum> is omitted, defaults to
1000 drive 0 (as do all commands). If <slotnum> is
1001 omitted, defaults to the slot that the drive was
1002 loaded from. Note that there's currently no way
1003 to say 'unload drive 1's media to the slot it
1004 came from', other than to explicitly use that
1005 slot number as the destination.AB
1011 undef# camcontrol devlist
1012 <WANGTEK 51000 SCSI M74H 12B3> at scbus0 target 2 lun 0 (pass0,sa0)
1013 <ARCHIVE 4586XX 28887-XXX 4BGD> at scbus0 target 4 lun 0 (pass1,sa1)
1014 <ARCHIVE 4586XX 28887-XXX 4BGD> at scbus0 target 4 lun 1 (pass2)
1016 tapeinfo -f /dev/sg0 with a bad tape in drive 1:
1017 [kern@rufus mtx-1.2.17kes]$ ./tapeinfo -f /dev/sg0
1018 Product Type: Tape Drive
1020 Product ID: 'C5713A '
1022 Attached Changer: No
1023 TapeAlert[3]: Hard Error: Uncorrectable read/write error.
1024 TapeAlert[20]: Clean Now: The tape drive neads cleaning NOW.
1031 Medium Type: Not Loaded
1034 DataCompEnabled: yes
1035 DataCompCapable: yes
1036 DataDeCompEnabled: yes
1043 Handling removable disks
1045 From: Karl Cunningham <karlc@keckec.com>
1047 My backups are only to hard disk these days, in removable bays. This is my
1048 idea of how a backup to hard disk would work more smoothly. Some of these
1049 things Bacula does already, but I mention them for completeness. If others
1050 have better ways to do this, I'd like to hear about it.
1052 1. Accommodate several disks, rotated similar to how tapes are. Identified
1053 by partition volume ID or perhaps by the name of a subdirectory.
1054 2. Abort & notify the admin if the wrong disk is in the bay.
1055 3. Write backups to different subdirectories for each machine to be backed
1057 4. Volumes (files) get created as needed in the proper subdirectory, one
1059 5. When a disk is recycled, remove or zero all old backup files. This is
1060 important as the disk being recycled may be close to full. This may be
1061 better done manually since the backup files for many machines may be
1062 scattered in many subdirectories.
1067 - To pass Include 1 or two letter commands
1068 I Name Include name - first record
1069 B Name Base name - repeat
1072 E zzz Exclude expression (wild card)
1074 D "reader" Reader program
1075 T "writer" Writer program
1076 O Options In current commpressed format
1077 (compression, signature, onefs, recurse, sparse,
1078 replace, verify options, ...)
1080 B BaseName Start second option set any letter
1083 F Number Number of filenames to follow
1088 F Number Number of filenames to follow
1091 - Add regression of btape "fill"
1093 - Implement "label slots=1,2... pool=yyy barcodes"